Research ArticleComparison of Different Strategies for SelectionAdaptation ofMixed Microbial Cultures Able to Ferment Crude GlycerolDerived from Second-Generation Biodiesel
C Varrone12 T M B Heggeset3 S B Le3 T Haugen3 S Markussen3
I V Skiadas12 and H N Gavala12
1Department of Chemistry and Biosciences Aalborg University 2350 Copenhagen Denmark2Department of Chemical and Biochemical Engineering Technical University of Denmark 2800 Lyngby Denmark3Biotechnology and Nanomedicine SINTEF Materials and Chemistry 7465 Trondheim Norway
Correspondence should be addressed to C Varrone cristianovarronegmailcom
Received 29 May 2015 Accepted 12 July 2015
Academic Editor Abd El-Latif Hesham
Copyright copy 2015 C Varrone et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
Objective of this study was the selection and adaptation of mixed microbial cultures (MMCs) able to ferment crude glycerolgenerated from animal fat-based biodiesel and produce building-blocks and green chemicals Various adaptation strategies havebeen investigated for the enrichment of suitable and stable MMC trying to overcome inhibition problems and enhance substratedegradation efficiency as well as generation of soluble fermentation products Repeated transfers in small batches and fed-batchconditions have been applied comparing the use of different inoculum growth media and Kinetic Control The adaptation ofactivated sludge inoculum was performed successfully and continued unhindered for several months The best results showed asubstrate degradation efficiency of almost 100 (about 10 gL glycerol in 21 h) and different dominant metabolic products wereobtained depending on the selection strategy (mainly 13-propanediol ethanol or butyrate) On the other hand anaerobic sludgeexhibited inactivation after a few transfers To circumvent this problem fed-batch mode was used as an alternative adaptationstrategy which led to effective substrate degradation and high 13-propanediol and butyrate production Changes in microbialcomposition were monitored by means of Next Generation Sequencing revealing a dominance of glycerol consuming speciessuch as Clostridium Klebsiella and Escherichia
1 Introduction
The exponential growth of biodiesel production in the lastdecade has led to a concomitant increase in crude glycerol[1 2] Hence new uses of crude glycerol are required inorder to overcome the problem of glycerol glut Meth-ods for glycerol utilization or disposal include combustioncomposting anaerobic digestion animal feed and thermo-chemical or biological conversion to value-added products[3] New methods for the valorization of glycerol involvethe bioconversion into biofuels and green chemicals whichmight provide several advantages compared to some of theabove-mentioned methods Environmental biotechnologiesare thus going to provide a significant contribution to tacklethe challenge of a more efficient use of by-products and
waste streams In this frame a so-called ldquoecobiotechnologicalapproachrdquo has been recently proposed as an interesting toolfor a more effective exploitation of wastes and wastewaters[4]
As stated by Johnson and colleagues [5] ecobiotechnol-ogy aims at applying ldquoprocesses based on openmixed culturesand ecological selection principles (rather than genetic ormetabolic engineering) thus combining the methodologyof environmental biotechnology with the goals of industrialbiotechnologyrdquo Some recent studies have started to applysuch principles also to the valorization of crude glycerolshowing interesting results in terms of conversion efficien-cies and decreased substrate and operating costs (no substratepretreatment no sterilization etc) mainly due to lowerenergy consumption [2 4 6 7] Glycerol fermentation can
Hindawi Publishing CorporationBioMed Research InternationalVolume 2015 Article ID 932934 14 pageshttpdxdoiorg1011552015932934
2 BioMed Research International
lead to the production of several useful metabolites such asalcohols (ie ethanol and butanol) 13-propanediol (13 PD)23-butanediol (23 BD) hydrogen polyhydroxyalkanoates(PHA) and volatile fatty acids (VFAs) [8ndash13] The latter rep-resent important bulk chemicals and preferred substrates formany bioprocesses [14] Interestingly they are also known tobe preferred substrates for enhanced polyhydroxyalkanoates(PHA) production [15] and in principle they might be usedfor a 2-stage process for the bioconversion of glycerol intoVFAs followed by PHA production
Thus in recent years the glycerol glut problem hasled to several studies on the conversion of crude glycerolHowever valorization of crude glycerol derived from second-generation (2G) biodiesel has been scarcely investigated andto our knowledge bioconversion of crude glycerol from theprocessing of animal fat derived biodiesel has been reportedonly by Sarma and colleagues [16] so far On the other handproduction of 2G biodiesel is expected to increase in the nearfuture due to incentives Europe for instance has proposedsubsidies for the production of biofuels produced fromwaste feedstocks (ie ldquomultiple accounting mechanismrdquoRenewable Energy Directive 200928EC) thus leading toan enhanced production of crude glycerol derived from 2Gbiodiesel
Nevertheless the use of such a substrate containinghigh amounts of contaminants such as soaps and long chainfatty acids (LCFA) salts ashes and methanol can stronglyinterfere with or even inhibit the microbial growth andconversion efficiency especially in the case of pure strains[17 18] In fact crude glycerol derived from complex wastematerials such as meat processing and restaurant wasteis considered to have even more impurities (very highamount of sulfur and LCFA very low pH etc) than thecrude glycerol derived from pure substrates [16] For thisreason most studies working with pure strains focus onthe use of purified glycerol This allows for higher substrateconversion efficiency but significantly increases processingcosts [19] A very important step to reduce costs relatedto the conversion of glycerol would therefore be to usecrude glycerol directly without previous pretreatment Thismight be achieved by using selectedmixedmicrobial cultures(MMCs) Since sterile cultivation enables an easy way ofcontrolling microbial growth and product formation mostindustrial biotechnological processes today are based on asingle microbial strain Nonetheless there are many caseswhere the utilization of mixed cultures andor coculturesappears to be advantageous over a singlemicroorganism [20]
The ability of the selected MMC to create synergisticeffects can help degrading complex substrates with differentgrades of impurities also in nonsterile conditions MMCcan thus utilize a wide variety of complex substrates richin nutrients but also potentially inhibiting effluents This isparticularly advantageous if industrial waste feedstock con-taining compounds of undefined composition are used [21]In fact unlike monocultures MMCs show a complementarymetabolism and are able to utilize different carbon sourcesFor this reason they are considered by several authors to beof special interest in the fermentative processes [5 22 23]representing a promising alternative approach [5] in some
Table 1 Crude glycerol characteristics
Content Typical valuesRaw glycerine 75Fat 10Methanol lt1Sulphur 1-2Moisture 10Ash 5Density 12ndash13 KgLpH 15
cases even showing better performances than pure strains[24] Therefore a new promising direction in environmentalbiotechnology is to apply the principles of ecobiotechnologyand adaptive laboratory evolution to develop a mixed micro-bial population selected to achieve a higher production yieldand which would have unique metabolic capacities [25] atlower operational costs [6 26]
The objective of this study is the selection and adaptationof MMC able to ferment crude glycerol generated fromanimal fat derived biodiesel and produce building-blocksand green chemicals Various adaptation strategies havebeen investigated for the enrichment of suitable and stableMMCs trying to overcome inhibition problems and enhancesubstrate degradation efficiency as well as production ofsoluble fermentation products
2 Material and Methods
21 Choice of Crude Glycerol Unless differently stated non-pretreated crude glycerol provided by Daka Biodiesel (Den-mark) obtained from the transesterification of butcherywaste (based on animal fat categories 1 and 2 according tothe EU regulation number 10692009 and 1422011) was usedThe main characteristics of this type of crude glycerol arereported in Table 1
22 Experimental Plan The enrichment and selection wereperformed in small batches through repeated transfers of dif-ferent inocula in order to compare their performances Eachexperiment was performed in triplicate Activated sludgeand anaerobic sludge were used as inoculum source Thelatter underwent heat-shock treatment and the fermentationperformance was compared to the nonpretreated sludgeHeat-shock allows selecting for spore forming bacteria (typ-ically Gram-positive bacteria such as Clostridia which areabundant in anaerobic sludge and are well-known in darkfermentation processes) while getting rid of methanogensActivated sludge instead is mainly made of enterobacteriatypically nonspore forming bacteria which would be inhib-ited by the heat-shock Enterobacteria are considered to bean important component in dark fermentation processesand the heat shock would lead to a reduction of additionalfermentation pathways [27] Moreover the activated sludge isnot anaerobic and does not favor the growth ofmethanogens
BioMed Research International 3
Fed-batch
Starting inoculum Anaerobicsludge sludge
Activated
HeatshockPretreatment No No
Medium BA MM BA MM BA MM
Transfer KC EF KC EF KC EF KC EF KC EF KC EF
Figure 1 Transfer scheme for the selection and enrichment in batchconditions KC = Kinetic Control EF = End of Fermentation MM=Minimal Medium BA = BA medium
and thus the heat-shock treatment would not be necessary orbeneficial
Two different growth media were used for the enrich-ment containing 10 gL glycerol a medium rich in tracemetals vitamins and growth factors (BA) and a MinimalMedium (MM) which does not include yeast extract tryp-tone vitamin or mineral solutions
Transfers 10 inoculum was used in 125mL vials of 40mLworking volume The experiments were performed to com-pare the efficiency of two enrichment strategies (a) KineticControl (KC) and (b) non-Kinetic Control in which theinoculum was transferred only at the End of Fermentation(EF)
Kinetic Control Transfers occurred during the (late) expo-nential growth phase in rapid successions (after 21 h fermen-tation)
End of Fermentation The transfers occurred after 72 h whenno more fermentation gases were produced A scheme ofthe experimental inoculum transfers is presented in Figure 1In addition fed-batch experiments (400mL working volumein 1 L serum bottle) and enrichment on hexane-pretreatedcrude glycerol were also performed using anaerobic sludgeas starting inoculum
Liquid and gas samples were collected on a regular basis
221 Microorganisms Storage and Activation MMCs ob-tained during the exponential growth phase were stored inthe freezer at minus18∘C and periodically refreshed Prior to usethe frozen mixed culture was transferred to the refrigeratorat 4∘C for 2 hours and then for an additional hour at roomtemperature before being inoculated Activation was per-formed in the same conditions as the respective enrichmentand 10 vv inoculum was transferred into fresh medium
after 21 hours (in case of Kinetic Control experiments) or 72hours (in case of End of Fermentation experiments)
222 Batch Experiments 125mL serum vials were used forbatch experimentation to enrich the (activated or anaero-bic) sludge through repeated transfers into fresh mediumaccording to the transfer scheme shown in Figure 1 36mLgrowth medium (either MM or BA medium) containingaround 10 gL glycerol was flushed for 5 minutes with amixture of 80N
2and 20CO
2 in order to obtain anaerobic
conditions prior to inoculation and incubated at 37∘C usingan orbital shaker at 150 rpm Gas and liquid samples werecollected before transferring 10 vv of fermentation broth(representing the new inoculum) into fresh medium Alltransfer steps were performed in triplicate
223 Hexane Pretreatment of Crude Glycerol Enrichmentof (heat-shock treated) anaerobic sludge was also performed(in the same batch conditions described in Section 222)using hexane-pretreated crude glycerol The extraction stepwas applied in order to reduce the concentration of lipidsand (long chain) fatty acids present in the crude glycerol(coming from fat derived biodiesel) and evaluate its potentialinhibitory effect on the microbial growth Hexane pretreat-ment was performed as described by Anand and Saxena [28]and the batch transfers were performed with Kinetic Control(every 21 h)
224 Fed-Batch Experiments Repeated fed-batch culturewas used for the enrichment of heat-shock treated anaerobicsludge in a 1 L serum vial with 300mL work solutioncontaining 90 anaerobic sludge and 10 BA medium witharound 10 gL (nonpretreated) glycerol The serum vial wasflushed for 15 minutes with a mixture of 80 N
2and 20
CO2 in order to obtain anaerobic conditions and incubated
at 37∘C and 150 rpm Every day an aliquot of around 30mLwas collected and substituted with an equivalent amount offresh BA medium containing 10 gL glycerol Gas and liquidsamples were collected prior to this operation
23 Media Composition
231 Minimal Medium Minimal Medium (MM) is a verysimple growth medium containing per litre of distilledwater 10 g glycerol 34 g K
2HPO4sdot3H2O 13 g KH
2PO4 2 g
(NH4)2SO4 02 g MgSO
4sdot7H2O 20mg CaCl
2sdot2H2O and
5mg FeSO4sdot7H2O [29]
For cultivation 36mL of medium was dispensed into125mL serum bottles and sealed with butyl rubber stoppersSubsequently the medium was flushed with a mixture ofnitrogen and CO
2(80 20 vv) for 5 minutes and inoculated
with 4mL inoculum (10 vv inoculum) before being incu-bated at 37∘C with continuous stirring (150 rpm) Initial pHwas 7
232 Rich Medium A complete synthetic medium for an-aerobes (referred to as BA medium [30]) which con-tains salts vitamins and trace elements beside pH buffers
4 BioMed Research International
and reducing agents was also used The medium wasprepared from the following stock solutions (contain-ing per litre of distilled water) (A) 100 g NH
4Cl 10 g
NaCl 10 g MgCl2sdot6H2O and 5 g CaCl
2sdot2H2O (B) 200 g
K2HPO4sdot3H2O (C) trace metal and selenite solution 2 g
FeCl2sdot4H2O 005 g H
3BO3 005 g ZnCl
2 0038 g CuCl
2sdot2
H2O 005 g MnCl
2sdot4H2O 005 g (NH
4)6Mo7O24sdot4H2O
005 g AlCl3 005 g CoCl
2sdot6H2O 0092 g NiCl
2sdot6H2O 05 g
ethylenediaminetetraacetate 1mL concentrated HCl and01 g Na
2SeO3sdot5H2O (D) 52 g NaHCO
3 and (E) vitamin
mixture according to Wolin et al [31]974mL of redistilled water was added to the following
stock solutions A 10mL B 2mL C 1mL D 50mL and E1mL [30]
24 Inocula Activated sludgewas collected from thewastew-ater treatment plant of Daka Biodiesel Denmark as it wasanticipated that it should be already enriched inmicrobes ableto use glycerol and lipid substances as carbon source
Anaerobic sludge was obtained from the LundtofteWastewater Treatment plant (Denmark) and supplementedwith the effluent of a lab-scale anaerobic digester (5050 vv)treating swine manure
The heat-shock pretreatment was obtained by heatingthe anaerobic sludge mixture for 15 minutes at 90∘C whileflushing with the N
2-CO2mixture
25 Analytical Methods Detection and quantification ofglycerol ethanol 13-propanediol and lactic acid wereobtained with a HPLC equipped with a refractive index andAminex HPX-87H column (BioRad) at 60∘C A solutionof 4mM H
2SO4was used as an eluent at a flow rate of
06mLminSamples for HPLC analysis were centrifuged at
10000 rpm for 10min filtered through a 045 120583mmembranefilter and finally acidified with a 10 ww solution of H
2SO4
For the quantification of volatile fatty acids (VFAs)filtered samples were acidified with H
3PO4(30 120583L of 17
H3PO4was added in 1mL of sample) and analyzed on a gas
chromatograph (PerkinElmer Clarus 400) equipped with aflame ionization detector and a capillary column (AgilentHP-FFAP 30m long 053mm inner diameter) The oven wasprogrammed to start with 105∘C (for 3minutes) followed by aramp that reaches 130∘Cat a rate of 8∘Cmin and subsequently230∘C (held for 3min) at a rate of 45∘Cmin Nitrogen wasused as the carrier gas at 13mLmin the injector temperaturewas set at 240∘C and the detector at 230∘C
The total volume of gas production was measured using awater displacement system [32]
Hydrogen content in the produced gas was measuredwith a gas chromatograph (SRI GC model 310) equippedwith a thermal conductivity detector and a packed column(Porapak-Q length 6 ft and inner diameter 21mm) Thevolume of H
2produced in sealed vials during glycerol fer-
mentation tests was calculated by the mass balance equation[33]
Multivariate data analysis was performed usingUnscram-bler X 101 software (by Camo) A Principal Component
Analysis (PCA) [34] was chosen as a tool to explore the bigdatamatrix obtained from themain fermentation parametersmonitored during the enrichments
26 Next Generation Sequencing DNA was extracted fromthe pellets of 5mL crude samples using the PowerSoil DNAIsolation Kit (MoBio) according to the standard procedureSequencing amplicon librarieswere generated byPCR follow-ing the ldquo16S Metagenomic Sequencing Library PreparationPreparing 16S Ribosomal RNA Gene Amplicons for theIllumina MiSeq Systemrdquo protocol (Illumina part number15044223 rev B) Internal parts of the 16S ribosomal RNA(rRNA) gene covering variable regions V3 and V4 werePCR-amplified with the KAPA HiFi HotStart ReadyMix(KAPA Biosystems) and the primers 51015840-TCGTCGGCAGC-GTCAGATGTGTATAAGAGACAGCCTACGGGNGG-CWGCAG-31015840 and 51015840-GTCTCGTGGGCTCGGAGATG-TGTATAAGAGACAGGACTACHVGGGTATCTAATCC-31015840 and purifiedwith theAgencourt AMPure XP kit (BeckmanCoulter Genomics) The Nextera XT Index Kit was used toadd sequencing adapters and multiplexing indices PooledDNA libraries were sequenced on a MiSeq sequencer (Illu-mina) using theMiSeq Reagent Kit v3 in the 2sdot300 bp paired-end mode
Sequencing reads were demultiplexed trimmed andOTU-classified using the Metagenomics Workflow of theMiSeq Reporter Software v23 (Illumina) This workflowuses an Illumina proprietary classification algorithm andan Illumina-curated version of the Greengenes 135 (May2013) taxonomy database which covers 3 kingdoms 33 phyla74 classes 148 orders 321 families 1086 genera and 6466species
Due to the relatively high number of unclassified readsfound at the species level comparisons between samples arepresented at the genus level while comparisons at the speciesfamily order class and phylum level are available as supp-lementary information (in Supplementary Material availableonline at httpdxdoiorg1011552015932934) Sequencingreads have been deposited to the sequence read archive ofNCBI under the Bioprojects PRJNA285034 (httpwwwncbinlmnihgovbioproject285034) and PRJNA284929 (httpwwwncbinlmnihgovbioproject284929)
3 Results and Discussion
31 Enrichment in Batch Conditions
311 Activated Sludge Based on the experimental scheme(Figure 1) 12 different selection conditionswere tested in trip-licate The enrichment using activated sludge showed goodresults in terms of substrate degradation and it continuedunhindered for several transfers with no evident inhibition(due to the use of crude glycerol) This actually indicatedthe possibility to increase the substrate concentration infuture studiesThe best results obtained in terms of substratedegradation efficiency (practically reaching 100) and biogasproduction were observed with MM-KC This experimentalcondition led to the highest ethanol production converting
BioMed Research International 5
about 10 gL glycerol in 21 h (maximum yield = 46 gg) witha concomitant 13 PD yield of approximately 3 gg After 16transfers however the distribution of the main metaboliteschanged with 13 PD becoming the dominant one andshowing an increase in butyrate during the last transfers
MM-EF also showed a high substrate degradation effi-ciency and (with exception of transfers 5ndash7) the mainmetabolites were represented by 13 PD and butyrate Thiscondition performed the best butyrate production with amaximum yield of 33 gg (from 85 gL glycerol in 72 hfermentation) together with 13 PD yield of 47 gg
The use of BAmedium (experiments 3 and 4) seemed notto favor solventogenesis pathway (almost no ethanol produc-tion was observed) while 13 PD was still by far the mainmetabolite (with an average production of 367plusmn056 gL and399plusmn074 gL for KC and EF resp) followed by butyrate andacetate Also in this case the End of Fermentation seemed tofavor butyrate production with a yield reaching up to 299 gg(from 77 gL glycerol in 72 h fermentation) in BA-EF
Hydrogen in the biogas was rather modest in allexperiments reaching in most cases around 20
The distribution ofmainmetabolites and substrate degra-dation () observed during the enrichment process withactivated sludge are shown in Figures 2(a) and 2(b)
Principal Component Analysis based on the completedatamatrix of 240 samples with 11 variables showed clear dif-ferences between the tested enrichment strategies (Figure 3)with EF closer related to butyrate (especially MM-EF) andBA-KC closer related to acetate In general the first PrincipalComponent (PC) showed an increase of ethanol and hydro-gen moving towards the right while the second PC showedan increase of butyrate productionmoving upwardsThe firstPC roughly separated EF and KC (with the exception ofMM-EF) while the second PC separated MM from BA
Furthermore a comparison of the correlation loadingsobtained with the data of the four enrichment conditions(MM-KC MM-EF BA-KC and BA-EF) separately showedthat only in the case of BA butyric acid was related to H
2
production (Figure 4) as would be expected from a directglycerol conversion into butyrate In fact glycerol conversionto butyric acid has a theoretically yield of 2molmol [35]
Interestingly in the case of MM butyrate production wasnegatively correlated with lactic and acetic acid and alsowith hydrogen in MM-EF while it was positively correlatedwith hydrogen production when using BA medium thusimplying a secondary fermentation (sensu Agler et al [21])(a butyrate production which does not come directly fromglycerol conversion)
There might be several possible pathways leading tobutyrate production through the conversion of lactate andacetate [36] besides the above-mentioned conversion ofglycerol Some examples are provided in
Lactate+ 04Acetate+ 07H+
997888rarr 07Butyrate+ 06H2 +CO2 + 04H2O
ΔG = minus1839
(1)
Lactate+Acetate+H+
997888rarr Butyrate+ 08H2 + 14CO2 + 06H2O
ΔG = minus594
(2)
2Lactate+H+ 997888rarr Butyrate+ 2H2+ 2CO2
ΔG = minus641(3)
It is also worth noting that Zhu and Yang [37] observed ametabolic shift from butyrate formation to lactate and acetateat pH lt 63 associated with decreased activities of phos-photransbutyrylase and NAD-independent lactate dehydro-genase and increased activities of phosphotransacetylase andlactate dehydrogenase Our batch experiments were operatedwithout pH control starting at pH 7 and typically endingat around 48 due to glycerol acidification Therefore it islikely that such a metabolic shift was also involved in ourfermentation tests
312 Anaerobic Sludge Differently from activated sludge theenrichment of anaerobic sludge in batch conditions showed aclear inhibition regardless of the selection strategy (BA andMMgrowthmedium EF or KC transfers)The inhibitionwaspresumably related to the high concentration of LCFA andthe negative interaction with the cell membranes of Gram-positive anaerobic bacteria of the anaerobic sludge ratherthan product inhibition In fact even after centrifuging theinoculum washing away the supernatant and resuspendingthe pellet into freshmedium (thus washing away extracellularsoluble metabolites) no recovery of the fermentation wasachieved Addition of specific elements such as yeast extractor vitamin andmineral solution did not have any effect either
The distribution of main metabolites and fraction of H2
(in the headspace) detected during the enrichment processwith anaerobic sludge are shown in Figure 5 The use ofMM (without nutrient supplements) led to inactivation afteronly 1 transfer while BA reached 6-7 transfers before beinginhibited (Figure 5(a)) Nonpretreated sludge (Figure 5(b))showed a high production of propionic acid while withheat-shock treated sludge (Figure 5(c)) butyric acid was thedominant metabolite The latter condition was chosen for analternative selection strategy using fed-batch conditions
313 Hexane-Pretreated Glycerol Tests As mentioned aboveheat-shock treated (HS) inoculum was chosen for furtherexperimentation The possible inhibiting effect of LCFA andldquolipidic compoundsrdquo was evaluated in the following test Thehypothesis was that the animal fat derived crude glycerolwould contain inhibiting amounts of LCFA which mightnegatively interfere with the membrane of Gram-positivebacteria of the anaerobic sludge Activated sludge was notincluded in this test since it did not show any inhibition
Nonextracted crude glycerol showed an organic carboncontent expressed as chemical oxygen demand (COD) of1309 plusmn 32 g CODL while the extracted crude glycerol was1172 plusmn 12 g CODL thus suggesting that approximately 137 gCODL of ldquolipidic compoundsrdquo was removed (which would
6 BioMed Research International
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(2) MM-EF gas products
0005101520253035404550
(gL
)
(2) MM-EF liquid products
0005101520253035404550
(gL
)
(1) MM-KC liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(1) MM-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(a)
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(4) BA-EF gas products
0005101520253035404550
(gL
)
(3) BA-KC liquid products
0005101520253035404550
(gL
)
(4) BA-EF liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(3) BA-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(b)
Figure 2 Fermentation products monitored during the enrichment of activated sludge in batch conditions through repeated transfers usingMM (a) and BA (b) medium (1) MM-KC = Minimal Medium with Kinetic Control (21 h) (2) MM-EF = Minimal Medium with End ofFermentation (72 h) (3) BA-KC = Basal Medium with Kinetic Control (21 h) (4) BA-EF = Basal Medium with End of Fermentation (72 h)
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
2 BioMed Research International
lead to the production of several useful metabolites such asalcohols (ie ethanol and butanol) 13-propanediol (13 PD)23-butanediol (23 BD) hydrogen polyhydroxyalkanoates(PHA) and volatile fatty acids (VFAs) [8ndash13] The latter rep-resent important bulk chemicals and preferred substrates formany bioprocesses [14] Interestingly they are also known tobe preferred substrates for enhanced polyhydroxyalkanoates(PHA) production [15] and in principle they might be usedfor a 2-stage process for the bioconversion of glycerol intoVFAs followed by PHA production
Thus in recent years the glycerol glut problem hasled to several studies on the conversion of crude glycerolHowever valorization of crude glycerol derived from second-generation (2G) biodiesel has been scarcely investigated andto our knowledge bioconversion of crude glycerol from theprocessing of animal fat derived biodiesel has been reportedonly by Sarma and colleagues [16] so far On the other handproduction of 2G biodiesel is expected to increase in the nearfuture due to incentives Europe for instance has proposedsubsidies for the production of biofuels produced fromwaste feedstocks (ie ldquomultiple accounting mechanismrdquoRenewable Energy Directive 200928EC) thus leading toan enhanced production of crude glycerol derived from 2Gbiodiesel
Nevertheless the use of such a substrate containinghigh amounts of contaminants such as soaps and long chainfatty acids (LCFA) salts ashes and methanol can stronglyinterfere with or even inhibit the microbial growth andconversion efficiency especially in the case of pure strains[17 18] In fact crude glycerol derived from complex wastematerials such as meat processing and restaurant wasteis considered to have even more impurities (very highamount of sulfur and LCFA very low pH etc) than thecrude glycerol derived from pure substrates [16] For thisreason most studies working with pure strains focus onthe use of purified glycerol This allows for higher substrateconversion efficiency but significantly increases processingcosts [19] A very important step to reduce costs relatedto the conversion of glycerol would therefore be to usecrude glycerol directly without previous pretreatment Thismight be achieved by using selectedmixedmicrobial cultures(MMCs) Since sterile cultivation enables an easy way ofcontrolling microbial growth and product formation mostindustrial biotechnological processes today are based on asingle microbial strain Nonetheless there are many caseswhere the utilization of mixed cultures andor coculturesappears to be advantageous over a singlemicroorganism [20]
The ability of the selected MMC to create synergisticeffects can help degrading complex substrates with differentgrades of impurities also in nonsterile conditions MMCcan thus utilize a wide variety of complex substrates richin nutrients but also potentially inhibiting effluents This isparticularly advantageous if industrial waste feedstock con-taining compounds of undefined composition are used [21]In fact unlike monocultures MMCs show a complementarymetabolism and are able to utilize different carbon sourcesFor this reason they are considered by several authors to beof special interest in the fermentative processes [5 22 23]representing a promising alternative approach [5] in some
Table 1 Crude glycerol characteristics
Content Typical valuesRaw glycerine 75Fat 10Methanol lt1Sulphur 1-2Moisture 10Ash 5Density 12ndash13 KgLpH 15
cases even showing better performances than pure strains[24] Therefore a new promising direction in environmentalbiotechnology is to apply the principles of ecobiotechnologyand adaptive laboratory evolution to develop a mixed micro-bial population selected to achieve a higher production yieldand which would have unique metabolic capacities [25] atlower operational costs [6 26]
The objective of this study is the selection and adaptationof MMC able to ferment crude glycerol generated fromanimal fat derived biodiesel and produce building-blocksand green chemicals Various adaptation strategies havebeen investigated for the enrichment of suitable and stableMMCs trying to overcome inhibition problems and enhancesubstrate degradation efficiency as well as production ofsoluble fermentation products
2 Material and Methods
21 Choice of Crude Glycerol Unless differently stated non-pretreated crude glycerol provided by Daka Biodiesel (Den-mark) obtained from the transesterification of butcherywaste (based on animal fat categories 1 and 2 according tothe EU regulation number 10692009 and 1422011) was usedThe main characteristics of this type of crude glycerol arereported in Table 1
22 Experimental Plan The enrichment and selection wereperformed in small batches through repeated transfers of dif-ferent inocula in order to compare their performances Eachexperiment was performed in triplicate Activated sludgeand anaerobic sludge were used as inoculum source Thelatter underwent heat-shock treatment and the fermentationperformance was compared to the nonpretreated sludgeHeat-shock allows selecting for spore forming bacteria (typ-ically Gram-positive bacteria such as Clostridia which areabundant in anaerobic sludge and are well-known in darkfermentation processes) while getting rid of methanogensActivated sludge instead is mainly made of enterobacteriatypically nonspore forming bacteria which would be inhib-ited by the heat-shock Enterobacteria are considered to bean important component in dark fermentation processesand the heat shock would lead to a reduction of additionalfermentation pathways [27] Moreover the activated sludge isnot anaerobic and does not favor the growth ofmethanogens
BioMed Research International 3
Fed-batch
Starting inoculum Anaerobicsludge sludge
Activated
HeatshockPretreatment No No
Medium BA MM BA MM BA MM
Transfer KC EF KC EF KC EF KC EF KC EF KC EF
Figure 1 Transfer scheme for the selection and enrichment in batchconditions KC = Kinetic Control EF = End of Fermentation MM=Minimal Medium BA = BA medium
and thus the heat-shock treatment would not be necessary orbeneficial
Two different growth media were used for the enrich-ment containing 10 gL glycerol a medium rich in tracemetals vitamins and growth factors (BA) and a MinimalMedium (MM) which does not include yeast extract tryp-tone vitamin or mineral solutions
Transfers 10 inoculum was used in 125mL vials of 40mLworking volume The experiments were performed to com-pare the efficiency of two enrichment strategies (a) KineticControl (KC) and (b) non-Kinetic Control in which theinoculum was transferred only at the End of Fermentation(EF)
Kinetic Control Transfers occurred during the (late) expo-nential growth phase in rapid successions (after 21 h fermen-tation)
End of Fermentation The transfers occurred after 72 h whenno more fermentation gases were produced A scheme ofthe experimental inoculum transfers is presented in Figure 1In addition fed-batch experiments (400mL working volumein 1 L serum bottle) and enrichment on hexane-pretreatedcrude glycerol were also performed using anaerobic sludgeas starting inoculum
Liquid and gas samples were collected on a regular basis
221 Microorganisms Storage and Activation MMCs ob-tained during the exponential growth phase were stored inthe freezer at minus18∘C and periodically refreshed Prior to usethe frozen mixed culture was transferred to the refrigeratorat 4∘C for 2 hours and then for an additional hour at roomtemperature before being inoculated Activation was per-formed in the same conditions as the respective enrichmentand 10 vv inoculum was transferred into fresh medium
after 21 hours (in case of Kinetic Control experiments) or 72hours (in case of End of Fermentation experiments)
222 Batch Experiments 125mL serum vials were used forbatch experimentation to enrich the (activated or anaero-bic) sludge through repeated transfers into fresh mediumaccording to the transfer scheme shown in Figure 1 36mLgrowth medium (either MM or BA medium) containingaround 10 gL glycerol was flushed for 5 minutes with amixture of 80N
2and 20CO
2 in order to obtain anaerobic
conditions prior to inoculation and incubated at 37∘C usingan orbital shaker at 150 rpm Gas and liquid samples werecollected before transferring 10 vv of fermentation broth(representing the new inoculum) into fresh medium Alltransfer steps were performed in triplicate
223 Hexane Pretreatment of Crude Glycerol Enrichmentof (heat-shock treated) anaerobic sludge was also performed(in the same batch conditions described in Section 222)using hexane-pretreated crude glycerol The extraction stepwas applied in order to reduce the concentration of lipidsand (long chain) fatty acids present in the crude glycerol(coming from fat derived biodiesel) and evaluate its potentialinhibitory effect on the microbial growth Hexane pretreat-ment was performed as described by Anand and Saxena [28]and the batch transfers were performed with Kinetic Control(every 21 h)
224 Fed-Batch Experiments Repeated fed-batch culturewas used for the enrichment of heat-shock treated anaerobicsludge in a 1 L serum vial with 300mL work solutioncontaining 90 anaerobic sludge and 10 BA medium witharound 10 gL (nonpretreated) glycerol The serum vial wasflushed for 15 minutes with a mixture of 80 N
2and 20
CO2 in order to obtain anaerobic conditions and incubated
at 37∘C and 150 rpm Every day an aliquot of around 30mLwas collected and substituted with an equivalent amount offresh BA medium containing 10 gL glycerol Gas and liquidsamples were collected prior to this operation
23 Media Composition
231 Minimal Medium Minimal Medium (MM) is a verysimple growth medium containing per litre of distilledwater 10 g glycerol 34 g K
2HPO4sdot3H2O 13 g KH
2PO4 2 g
(NH4)2SO4 02 g MgSO
4sdot7H2O 20mg CaCl
2sdot2H2O and
5mg FeSO4sdot7H2O [29]
For cultivation 36mL of medium was dispensed into125mL serum bottles and sealed with butyl rubber stoppersSubsequently the medium was flushed with a mixture ofnitrogen and CO
2(80 20 vv) for 5 minutes and inoculated
with 4mL inoculum (10 vv inoculum) before being incu-bated at 37∘C with continuous stirring (150 rpm) Initial pHwas 7
232 Rich Medium A complete synthetic medium for an-aerobes (referred to as BA medium [30]) which con-tains salts vitamins and trace elements beside pH buffers
4 BioMed Research International
and reducing agents was also used The medium wasprepared from the following stock solutions (contain-ing per litre of distilled water) (A) 100 g NH
4Cl 10 g
NaCl 10 g MgCl2sdot6H2O and 5 g CaCl
2sdot2H2O (B) 200 g
K2HPO4sdot3H2O (C) trace metal and selenite solution 2 g
FeCl2sdot4H2O 005 g H
3BO3 005 g ZnCl
2 0038 g CuCl
2sdot2
H2O 005 g MnCl
2sdot4H2O 005 g (NH
4)6Mo7O24sdot4H2O
005 g AlCl3 005 g CoCl
2sdot6H2O 0092 g NiCl
2sdot6H2O 05 g
ethylenediaminetetraacetate 1mL concentrated HCl and01 g Na
2SeO3sdot5H2O (D) 52 g NaHCO
3 and (E) vitamin
mixture according to Wolin et al [31]974mL of redistilled water was added to the following
stock solutions A 10mL B 2mL C 1mL D 50mL and E1mL [30]
24 Inocula Activated sludgewas collected from thewastew-ater treatment plant of Daka Biodiesel Denmark as it wasanticipated that it should be already enriched inmicrobes ableto use glycerol and lipid substances as carbon source
Anaerobic sludge was obtained from the LundtofteWastewater Treatment plant (Denmark) and supplementedwith the effluent of a lab-scale anaerobic digester (5050 vv)treating swine manure
The heat-shock pretreatment was obtained by heatingthe anaerobic sludge mixture for 15 minutes at 90∘C whileflushing with the N
2-CO2mixture
25 Analytical Methods Detection and quantification ofglycerol ethanol 13-propanediol and lactic acid wereobtained with a HPLC equipped with a refractive index andAminex HPX-87H column (BioRad) at 60∘C A solutionof 4mM H
2SO4was used as an eluent at a flow rate of
06mLminSamples for HPLC analysis were centrifuged at
10000 rpm for 10min filtered through a 045 120583mmembranefilter and finally acidified with a 10 ww solution of H
2SO4
For the quantification of volatile fatty acids (VFAs)filtered samples were acidified with H
3PO4(30 120583L of 17
H3PO4was added in 1mL of sample) and analyzed on a gas
chromatograph (PerkinElmer Clarus 400) equipped with aflame ionization detector and a capillary column (AgilentHP-FFAP 30m long 053mm inner diameter) The oven wasprogrammed to start with 105∘C (for 3minutes) followed by aramp that reaches 130∘Cat a rate of 8∘Cmin and subsequently230∘C (held for 3min) at a rate of 45∘Cmin Nitrogen wasused as the carrier gas at 13mLmin the injector temperaturewas set at 240∘C and the detector at 230∘C
The total volume of gas production was measured using awater displacement system [32]
Hydrogen content in the produced gas was measuredwith a gas chromatograph (SRI GC model 310) equippedwith a thermal conductivity detector and a packed column(Porapak-Q length 6 ft and inner diameter 21mm) Thevolume of H
2produced in sealed vials during glycerol fer-
mentation tests was calculated by the mass balance equation[33]
Multivariate data analysis was performed usingUnscram-bler X 101 software (by Camo) A Principal Component
Analysis (PCA) [34] was chosen as a tool to explore the bigdatamatrix obtained from themain fermentation parametersmonitored during the enrichments
26 Next Generation Sequencing DNA was extracted fromthe pellets of 5mL crude samples using the PowerSoil DNAIsolation Kit (MoBio) according to the standard procedureSequencing amplicon librarieswere generated byPCR follow-ing the ldquo16S Metagenomic Sequencing Library PreparationPreparing 16S Ribosomal RNA Gene Amplicons for theIllumina MiSeq Systemrdquo protocol (Illumina part number15044223 rev B) Internal parts of the 16S ribosomal RNA(rRNA) gene covering variable regions V3 and V4 werePCR-amplified with the KAPA HiFi HotStart ReadyMix(KAPA Biosystems) and the primers 51015840-TCGTCGGCAGC-GTCAGATGTGTATAAGAGACAGCCTACGGGNGG-CWGCAG-31015840 and 51015840-GTCTCGTGGGCTCGGAGATG-TGTATAAGAGACAGGACTACHVGGGTATCTAATCC-31015840 and purifiedwith theAgencourt AMPure XP kit (BeckmanCoulter Genomics) The Nextera XT Index Kit was used toadd sequencing adapters and multiplexing indices PooledDNA libraries were sequenced on a MiSeq sequencer (Illu-mina) using theMiSeq Reagent Kit v3 in the 2sdot300 bp paired-end mode
Sequencing reads were demultiplexed trimmed andOTU-classified using the Metagenomics Workflow of theMiSeq Reporter Software v23 (Illumina) This workflowuses an Illumina proprietary classification algorithm andan Illumina-curated version of the Greengenes 135 (May2013) taxonomy database which covers 3 kingdoms 33 phyla74 classes 148 orders 321 families 1086 genera and 6466species
Due to the relatively high number of unclassified readsfound at the species level comparisons between samples arepresented at the genus level while comparisons at the speciesfamily order class and phylum level are available as supp-lementary information (in Supplementary Material availableonline at httpdxdoiorg1011552015932934) Sequencingreads have been deposited to the sequence read archive ofNCBI under the Bioprojects PRJNA285034 (httpwwwncbinlmnihgovbioproject285034) and PRJNA284929 (httpwwwncbinlmnihgovbioproject284929)
3 Results and Discussion
31 Enrichment in Batch Conditions
311 Activated Sludge Based on the experimental scheme(Figure 1) 12 different selection conditionswere tested in trip-licate The enrichment using activated sludge showed goodresults in terms of substrate degradation and it continuedunhindered for several transfers with no evident inhibition(due to the use of crude glycerol) This actually indicatedthe possibility to increase the substrate concentration infuture studiesThe best results obtained in terms of substratedegradation efficiency (practically reaching 100) and biogasproduction were observed with MM-KC This experimentalcondition led to the highest ethanol production converting
BioMed Research International 5
about 10 gL glycerol in 21 h (maximum yield = 46 gg) witha concomitant 13 PD yield of approximately 3 gg After 16transfers however the distribution of the main metaboliteschanged with 13 PD becoming the dominant one andshowing an increase in butyrate during the last transfers
MM-EF also showed a high substrate degradation effi-ciency and (with exception of transfers 5ndash7) the mainmetabolites were represented by 13 PD and butyrate Thiscondition performed the best butyrate production with amaximum yield of 33 gg (from 85 gL glycerol in 72 hfermentation) together with 13 PD yield of 47 gg
The use of BAmedium (experiments 3 and 4) seemed notto favor solventogenesis pathway (almost no ethanol produc-tion was observed) while 13 PD was still by far the mainmetabolite (with an average production of 367plusmn056 gL and399plusmn074 gL for KC and EF resp) followed by butyrate andacetate Also in this case the End of Fermentation seemed tofavor butyrate production with a yield reaching up to 299 gg(from 77 gL glycerol in 72 h fermentation) in BA-EF
Hydrogen in the biogas was rather modest in allexperiments reaching in most cases around 20
The distribution ofmainmetabolites and substrate degra-dation () observed during the enrichment process withactivated sludge are shown in Figures 2(a) and 2(b)
Principal Component Analysis based on the completedatamatrix of 240 samples with 11 variables showed clear dif-ferences between the tested enrichment strategies (Figure 3)with EF closer related to butyrate (especially MM-EF) andBA-KC closer related to acetate In general the first PrincipalComponent (PC) showed an increase of ethanol and hydro-gen moving towards the right while the second PC showedan increase of butyrate productionmoving upwardsThe firstPC roughly separated EF and KC (with the exception ofMM-EF) while the second PC separated MM from BA
Furthermore a comparison of the correlation loadingsobtained with the data of the four enrichment conditions(MM-KC MM-EF BA-KC and BA-EF) separately showedthat only in the case of BA butyric acid was related to H
2
production (Figure 4) as would be expected from a directglycerol conversion into butyrate In fact glycerol conversionto butyric acid has a theoretically yield of 2molmol [35]
Interestingly in the case of MM butyrate production wasnegatively correlated with lactic and acetic acid and alsowith hydrogen in MM-EF while it was positively correlatedwith hydrogen production when using BA medium thusimplying a secondary fermentation (sensu Agler et al [21])(a butyrate production which does not come directly fromglycerol conversion)
There might be several possible pathways leading tobutyrate production through the conversion of lactate andacetate [36] besides the above-mentioned conversion ofglycerol Some examples are provided in
Lactate+ 04Acetate+ 07H+
997888rarr 07Butyrate+ 06H2 +CO2 + 04H2O
ΔG = minus1839
(1)
Lactate+Acetate+H+
997888rarr Butyrate+ 08H2 + 14CO2 + 06H2O
ΔG = minus594
(2)
2Lactate+H+ 997888rarr Butyrate+ 2H2+ 2CO2
ΔG = minus641(3)
It is also worth noting that Zhu and Yang [37] observed ametabolic shift from butyrate formation to lactate and acetateat pH lt 63 associated with decreased activities of phos-photransbutyrylase and NAD-independent lactate dehydro-genase and increased activities of phosphotransacetylase andlactate dehydrogenase Our batch experiments were operatedwithout pH control starting at pH 7 and typically endingat around 48 due to glycerol acidification Therefore it islikely that such a metabolic shift was also involved in ourfermentation tests
312 Anaerobic Sludge Differently from activated sludge theenrichment of anaerobic sludge in batch conditions showed aclear inhibition regardless of the selection strategy (BA andMMgrowthmedium EF or KC transfers)The inhibitionwaspresumably related to the high concentration of LCFA andthe negative interaction with the cell membranes of Gram-positive anaerobic bacteria of the anaerobic sludge ratherthan product inhibition In fact even after centrifuging theinoculum washing away the supernatant and resuspendingthe pellet into freshmedium (thus washing away extracellularsoluble metabolites) no recovery of the fermentation wasachieved Addition of specific elements such as yeast extractor vitamin andmineral solution did not have any effect either
The distribution of main metabolites and fraction of H2
(in the headspace) detected during the enrichment processwith anaerobic sludge are shown in Figure 5 The use ofMM (without nutrient supplements) led to inactivation afteronly 1 transfer while BA reached 6-7 transfers before beinginhibited (Figure 5(a)) Nonpretreated sludge (Figure 5(b))showed a high production of propionic acid while withheat-shock treated sludge (Figure 5(c)) butyric acid was thedominant metabolite The latter condition was chosen for analternative selection strategy using fed-batch conditions
313 Hexane-Pretreated Glycerol Tests As mentioned aboveheat-shock treated (HS) inoculum was chosen for furtherexperimentation The possible inhibiting effect of LCFA andldquolipidic compoundsrdquo was evaluated in the following test Thehypothesis was that the animal fat derived crude glycerolwould contain inhibiting amounts of LCFA which mightnegatively interfere with the membrane of Gram-positivebacteria of the anaerobic sludge Activated sludge was notincluded in this test since it did not show any inhibition
Nonextracted crude glycerol showed an organic carboncontent expressed as chemical oxygen demand (COD) of1309 plusmn 32 g CODL while the extracted crude glycerol was1172 plusmn 12 g CODL thus suggesting that approximately 137 gCODL of ldquolipidic compoundsrdquo was removed (which would
6 BioMed Research International
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(2) MM-EF gas products
0005101520253035404550
(gL
)
(2) MM-EF liquid products
0005101520253035404550
(gL
)
(1) MM-KC liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(1) MM-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(a)
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(4) BA-EF gas products
0005101520253035404550
(gL
)
(3) BA-KC liquid products
0005101520253035404550
(gL
)
(4) BA-EF liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(3) BA-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(b)
Figure 2 Fermentation products monitored during the enrichment of activated sludge in batch conditions through repeated transfers usingMM (a) and BA (b) medium (1) MM-KC = Minimal Medium with Kinetic Control (21 h) (2) MM-EF = Minimal Medium with End ofFermentation (72 h) (3) BA-KC = Basal Medium with Kinetic Control (21 h) (4) BA-EF = Basal Medium with End of Fermentation (72 h)
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
BioMed Research International 3
Fed-batch
Starting inoculum Anaerobicsludge sludge
Activated
HeatshockPretreatment No No
Medium BA MM BA MM BA MM
Transfer KC EF KC EF KC EF KC EF KC EF KC EF
Figure 1 Transfer scheme for the selection and enrichment in batchconditions KC = Kinetic Control EF = End of Fermentation MM=Minimal Medium BA = BA medium
and thus the heat-shock treatment would not be necessary orbeneficial
Two different growth media were used for the enrich-ment containing 10 gL glycerol a medium rich in tracemetals vitamins and growth factors (BA) and a MinimalMedium (MM) which does not include yeast extract tryp-tone vitamin or mineral solutions
Transfers 10 inoculum was used in 125mL vials of 40mLworking volume The experiments were performed to com-pare the efficiency of two enrichment strategies (a) KineticControl (KC) and (b) non-Kinetic Control in which theinoculum was transferred only at the End of Fermentation(EF)
Kinetic Control Transfers occurred during the (late) expo-nential growth phase in rapid successions (after 21 h fermen-tation)
End of Fermentation The transfers occurred after 72 h whenno more fermentation gases were produced A scheme ofthe experimental inoculum transfers is presented in Figure 1In addition fed-batch experiments (400mL working volumein 1 L serum bottle) and enrichment on hexane-pretreatedcrude glycerol were also performed using anaerobic sludgeas starting inoculum
Liquid and gas samples were collected on a regular basis
221 Microorganisms Storage and Activation MMCs ob-tained during the exponential growth phase were stored inthe freezer at minus18∘C and periodically refreshed Prior to usethe frozen mixed culture was transferred to the refrigeratorat 4∘C for 2 hours and then for an additional hour at roomtemperature before being inoculated Activation was per-formed in the same conditions as the respective enrichmentand 10 vv inoculum was transferred into fresh medium
after 21 hours (in case of Kinetic Control experiments) or 72hours (in case of End of Fermentation experiments)
222 Batch Experiments 125mL serum vials were used forbatch experimentation to enrich the (activated or anaero-bic) sludge through repeated transfers into fresh mediumaccording to the transfer scheme shown in Figure 1 36mLgrowth medium (either MM or BA medium) containingaround 10 gL glycerol was flushed for 5 minutes with amixture of 80N
2and 20CO
2 in order to obtain anaerobic
conditions prior to inoculation and incubated at 37∘C usingan orbital shaker at 150 rpm Gas and liquid samples werecollected before transferring 10 vv of fermentation broth(representing the new inoculum) into fresh medium Alltransfer steps were performed in triplicate
223 Hexane Pretreatment of Crude Glycerol Enrichmentof (heat-shock treated) anaerobic sludge was also performed(in the same batch conditions described in Section 222)using hexane-pretreated crude glycerol The extraction stepwas applied in order to reduce the concentration of lipidsand (long chain) fatty acids present in the crude glycerol(coming from fat derived biodiesel) and evaluate its potentialinhibitory effect on the microbial growth Hexane pretreat-ment was performed as described by Anand and Saxena [28]and the batch transfers were performed with Kinetic Control(every 21 h)
224 Fed-Batch Experiments Repeated fed-batch culturewas used for the enrichment of heat-shock treated anaerobicsludge in a 1 L serum vial with 300mL work solutioncontaining 90 anaerobic sludge and 10 BA medium witharound 10 gL (nonpretreated) glycerol The serum vial wasflushed for 15 minutes with a mixture of 80 N
2and 20
CO2 in order to obtain anaerobic conditions and incubated
at 37∘C and 150 rpm Every day an aliquot of around 30mLwas collected and substituted with an equivalent amount offresh BA medium containing 10 gL glycerol Gas and liquidsamples were collected prior to this operation
23 Media Composition
231 Minimal Medium Minimal Medium (MM) is a verysimple growth medium containing per litre of distilledwater 10 g glycerol 34 g K
2HPO4sdot3H2O 13 g KH
2PO4 2 g
(NH4)2SO4 02 g MgSO
4sdot7H2O 20mg CaCl
2sdot2H2O and
5mg FeSO4sdot7H2O [29]
For cultivation 36mL of medium was dispensed into125mL serum bottles and sealed with butyl rubber stoppersSubsequently the medium was flushed with a mixture ofnitrogen and CO
2(80 20 vv) for 5 minutes and inoculated
with 4mL inoculum (10 vv inoculum) before being incu-bated at 37∘C with continuous stirring (150 rpm) Initial pHwas 7
232 Rich Medium A complete synthetic medium for an-aerobes (referred to as BA medium [30]) which con-tains salts vitamins and trace elements beside pH buffers
4 BioMed Research International
and reducing agents was also used The medium wasprepared from the following stock solutions (contain-ing per litre of distilled water) (A) 100 g NH
4Cl 10 g
NaCl 10 g MgCl2sdot6H2O and 5 g CaCl
2sdot2H2O (B) 200 g
K2HPO4sdot3H2O (C) trace metal and selenite solution 2 g
FeCl2sdot4H2O 005 g H
3BO3 005 g ZnCl
2 0038 g CuCl
2sdot2
H2O 005 g MnCl
2sdot4H2O 005 g (NH
4)6Mo7O24sdot4H2O
005 g AlCl3 005 g CoCl
2sdot6H2O 0092 g NiCl
2sdot6H2O 05 g
ethylenediaminetetraacetate 1mL concentrated HCl and01 g Na
2SeO3sdot5H2O (D) 52 g NaHCO
3 and (E) vitamin
mixture according to Wolin et al [31]974mL of redistilled water was added to the following
stock solutions A 10mL B 2mL C 1mL D 50mL and E1mL [30]
24 Inocula Activated sludgewas collected from thewastew-ater treatment plant of Daka Biodiesel Denmark as it wasanticipated that it should be already enriched inmicrobes ableto use glycerol and lipid substances as carbon source
Anaerobic sludge was obtained from the LundtofteWastewater Treatment plant (Denmark) and supplementedwith the effluent of a lab-scale anaerobic digester (5050 vv)treating swine manure
The heat-shock pretreatment was obtained by heatingthe anaerobic sludge mixture for 15 minutes at 90∘C whileflushing with the N
2-CO2mixture
25 Analytical Methods Detection and quantification ofglycerol ethanol 13-propanediol and lactic acid wereobtained with a HPLC equipped with a refractive index andAminex HPX-87H column (BioRad) at 60∘C A solutionof 4mM H
2SO4was used as an eluent at a flow rate of
06mLminSamples for HPLC analysis were centrifuged at
10000 rpm for 10min filtered through a 045 120583mmembranefilter and finally acidified with a 10 ww solution of H
2SO4
For the quantification of volatile fatty acids (VFAs)filtered samples were acidified with H
3PO4(30 120583L of 17
H3PO4was added in 1mL of sample) and analyzed on a gas
chromatograph (PerkinElmer Clarus 400) equipped with aflame ionization detector and a capillary column (AgilentHP-FFAP 30m long 053mm inner diameter) The oven wasprogrammed to start with 105∘C (for 3minutes) followed by aramp that reaches 130∘Cat a rate of 8∘Cmin and subsequently230∘C (held for 3min) at a rate of 45∘Cmin Nitrogen wasused as the carrier gas at 13mLmin the injector temperaturewas set at 240∘C and the detector at 230∘C
The total volume of gas production was measured using awater displacement system [32]
Hydrogen content in the produced gas was measuredwith a gas chromatograph (SRI GC model 310) equippedwith a thermal conductivity detector and a packed column(Porapak-Q length 6 ft and inner diameter 21mm) Thevolume of H
2produced in sealed vials during glycerol fer-
mentation tests was calculated by the mass balance equation[33]
Multivariate data analysis was performed usingUnscram-bler X 101 software (by Camo) A Principal Component
Analysis (PCA) [34] was chosen as a tool to explore the bigdatamatrix obtained from themain fermentation parametersmonitored during the enrichments
26 Next Generation Sequencing DNA was extracted fromthe pellets of 5mL crude samples using the PowerSoil DNAIsolation Kit (MoBio) according to the standard procedureSequencing amplicon librarieswere generated byPCR follow-ing the ldquo16S Metagenomic Sequencing Library PreparationPreparing 16S Ribosomal RNA Gene Amplicons for theIllumina MiSeq Systemrdquo protocol (Illumina part number15044223 rev B) Internal parts of the 16S ribosomal RNA(rRNA) gene covering variable regions V3 and V4 werePCR-amplified with the KAPA HiFi HotStart ReadyMix(KAPA Biosystems) and the primers 51015840-TCGTCGGCAGC-GTCAGATGTGTATAAGAGACAGCCTACGGGNGG-CWGCAG-31015840 and 51015840-GTCTCGTGGGCTCGGAGATG-TGTATAAGAGACAGGACTACHVGGGTATCTAATCC-31015840 and purifiedwith theAgencourt AMPure XP kit (BeckmanCoulter Genomics) The Nextera XT Index Kit was used toadd sequencing adapters and multiplexing indices PooledDNA libraries were sequenced on a MiSeq sequencer (Illu-mina) using theMiSeq Reagent Kit v3 in the 2sdot300 bp paired-end mode
Sequencing reads were demultiplexed trimmed andOTU-classified using the Metagenomics Workflow of theMiSeq Reporter Software v23 (Illumina) This workflowuses an Illumina proprietary classification algorithm andan Illumina-curated version of the Greengenes 135 (May2013) taxonomy database which covers 3 kingdoms 33 phyla74 classes 148 orders 321 families 1086 genera and 6466species
Due to the relatively high number of unclassified readsfound at the species level comparisons between samples arepresented at the genus level while comparisons at the speciesfamily order class and phylum level are available as supp-lementary information (in Supplementary Material availableonline at httpdxdoiorg1011552015932934) Sequencingreads have been deposited to the sequence read archive ofNCBI under the Bioprojects PRJNA285034 (httpwwwncbinlmnihgovbioproject285034) and PRJNA284929 (httpwwwncbinlmnihgovbioproject284929)
3 Results and Discussion
31 Enrichment in Batch Conditions
311 Activated Sludge Based on the experimental scheme(Figure 1) 12 different selection conditionswere tested in trip-licate The enrichment using activated sludge showed goodresults in terms of substrate degradation and it continuedunhindered for several transfers with no evident inhibition(due to the use of crude glycerol) This actually indicatedthe possibility to increase the substrate concentration infuture studiesThe best results obtained in terms of substratedegradation efficiency (practically reaching 100) and biogasproduction were observed with MM-KC This experimentalcondition led to the highest ethanol production converting
BioMed Research International 5
about 10 gL glycerol in 21 h (maximum yield = 46 gg) witha concomitant 13 PD yield of approximately 3 gg After 16transfers however the distribution of the main metaboliteschanged with 13 PD becoming the dominant one andshowing an increase in butyrate during the last transfers
MM-EF also showed a high substrate degradation effi-ciency and (with exception of transfers 5ndash7) the mainmetabolites were represented by 13 PD and butyrate Thiscondition performed the best butyrate production with amaximum yield of 33 gg (from 85 gL glycerol in 72 hfermentation) together with 13 PD yield of 47 gg
The use of BAmedium (experiments 3 and 4) seemed notto favor solventogenesis pathway (almost no ethanol produc-tion was observed) while 13 PD was still by far the mainmetabolite (with an average production of 367plusmn056 gL and399plusmn074 gL for KC and EF resp) followed by butyrate andacetate Also in this case the End of Fermentation seemed tofavor butyrate production with a yield reaching up to 299 gg(from 77 gL glycerol in 72 h fermentation) in BA-EF
Hydrogen in the biogas was rather modest in allexperiments reaching in most cases around 20
The distribution ofmainmetabolites and substrate degra-dation () observed during the enrichment process withactivated sludge are shown in Figures 2(a) and 2(b)
Principal Component Analysis based on the completedatamatrix of 240 samples with 11 variables showed clear dif-ferences between the tested enrichment strategies (Figure 3)with EF closer related to butyrate (especially MM-EF) andBA-KC closer related to acetate In general the first PrincipalComponent (PC) showed an increase of ethanol and hydro-gen moving towards the right while the second PC showedan increase of butyrate productionmoving upwardsThe firstPC roughly separated EF and KC (with the exception ofMM-EF) while the second PC separated MM from BA
Furthermore a comparison of the correlation loadingsobtained with the data of the four enrichment conditions(MM-KC MM-EF BA-KC and BA-EF) separately showedthat only in the case of BA butyric acid was related to H
2
production (Figure 4) as would be expected from a directglycerol conversion into butyrate In fact glycerol conversionto butyric acid has a theoretically yield of 2molmol [35]
Interestingly in the case of MM butyrate production wasnegatively correlated with lactic and acetic acid and alsowith hydrogen in MM-EF while it was positively correlatedwith hydrogen production when using BA medium thusimplying a secondary fermentation (sensu Agler et al [21])(a butyrate production which does not come directly fromglycerol conversion)
There might be several possible pathways leading tobutyrate production through the conversion of lactate andacetate [36] besides the above-mentioned conversion ofglycerol Some examples are provided in
Lactate+ 04Acetate+ 07H+
997888rarr 07Butyrate+ 06H2 +CO2 + 04H2O
ΔG = minus1839
(1)
Lactate+Acetate+H+
997888rarr Butyrate+ 08H2 + 14CO2 + 06H2O
ΔG = minus594
(2)
2Lactate+H+ 997888rarr Butyrate+ 2H2+ 2CO2
ΔG = minus641(3)
It is also worth noting that Zhu and Yang [37] observed ametabolic shift from butyrate formation to lactate and acetateat pH lt 63 associated with decreased activities of phos-photransbutyrylase and NAD-independent lactate dehydro-genase and increased activities of phosphotransacetylase andlactate dehydrogenase Our batch experiments were operatedwithout pH control starting at pH 7 and typically endingat around 48 due to glycerol acidification Therefore it islikely that such a metabolic shift was also involved in ourfermentation tests
312 Anaerobic Sludge Differently from activated sludge theenrichment of anaerobic sludge in batch conditions showed aclear inhibition regardless of the selection strategy (BA andMMgrowthmedium EF or KC transfers)The inhibitionwaspresumably related to the high concentration of LCFA andthe negative interaction with the cell membranes of Gram-positive anaerobic bacteria of the anaerobic sludge ratherthan product inhibition In fact even after centrifuging theinoculum washing away the supernatant and resuspendingthe pellet into freshmedium (thus washing away extracellularsoluble metabolites) no recovery of the fermentation wasachieved Addition of specific elements such as yeast extractor vitamin andmineral solution did not have any effect either
The distribution of main metabolites and fraction of H2
(in the headspace) detected during the enrichment processwith anaerobic sludge are shown in Figure 5 The use ofMM (without nutrient supplements) led to inactivation afteronly 1 transfer while BA reached 6-7 transfers before beinginhibited (Figure 5(a)) Nonpretreated sludge (Figure 5(b))showed a high production of propionic acid while withheat-shock treated sludge (Figure 5(c)) butyric acid was thedominant metabolite The latter condition was chosen for analternative selection strategy using fed-batch conditions
313 Hexane-Pretreated Glycerol Tests As mentioned aboveheat-shock treated (HS) inoculum was chosen for furtherexperimentation The possible inhibiting effect of LCFA andldquolipidic compoundsrdquo was evaluated in the following test Thehypothesis was that the animal fat derived crude glycerolwould contain inhibiting amounts of LCFA which mightnegatively interfere with the membrane of Gram-positivebacteria of the anaerobic sludge Activated sludge was notincluded in this test since it did not show any inhibition
Nonextracted crude glycerol showed an organic carboncontent expressed as chemical oxygen demand (COD) of1309 plusmn 32 g CODL while the extracted crude glycerol was1172 plusmn 12 g CODL thus suggesting that approximately 137 gCODL of ldquolipidic compoundsrdquo was removed (which would
6 BioMed Research International
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(2) MM-EF gas products
0005101520253035404550
(gL
)
(2) MM-EF liquid products
0005101520253035404550
(gL
)
(1) MM-KC liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(1) MM-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(a)
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(4) BA-EF gas products
0005101520253035404550
(gL
)
(3) BA-KC liquid products
0005101520253035404550
(gL
)
(4) BA-EF liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(3) BA-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(b)
Figure 2 Fermentation products monitored during the enrichment of activated sludge in batch conditions through repeated transfers usingMM (a) and BA (b) medium (1) MM-KC = Minimal Medium with Kinetic Control (21 h) (2) MM-EF = Minimal Medium with End ofFermentation (72 h) (3) BA-KC = Basal Medium with Kinetic Control (21 h) (4) BA-EF = Basal Medium with End of Fermentation (72 h)
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
4 BioMed Research International
and reducing agents was also used The medium wasprepared from the following stock solutions (contain-ing per litre of distilled water) (A) 100 g NH
4Cl 10 g
NaCl 10 g MgCl2sdot6H2O and 5 g CaCl
2sdot2H2O (B) 200 g
K2HPO4sdot3H2O (C) trace metal and selenite solution 2 g
FeCl2sdot4H2O 005 g H
3BO3 005 g ZnCl
2 0038 g CuCl
2sdot2
H2O 005 g MnCl
2sdot4H2O 005 g (NH
4)6Mo7O24sdot4H2O
005 g AlCl3 005 g CoCl
2sdot6H2O 0092 g NiCl
2sdot6H2O 05 g
ethylenediaminetetraacetate 1mL concentrated HCl and01 g Na
2SeO3sdot5H2O (D) 52 g NaHCO
3 and (E) vitamin
mixture according to Wolin et al [31]974mL of redistilled water was added to the following
stock solutions A 10mL B 2mL C 1mL D 50mL and E1mL [30]
24 Inocula Activated sludgewas collected from thewastew-ater treatment plant of Daka Biodiesel Denmark as it wasanticipated that it should be already enriched inmicrobes ableto use glycerol and lipid substances as carbon source
Anaerobic sludge was obtained from the LundtofteWastewater Treatment plant (Denmark) and supplementedwith the effluent of a lab-scale anaerobic digester (5050 vv)treating swine manure
The heat-shock pretreatment was obtained by heatingthe anaerobic sludge mixture for 15 minutes at 90∘C whileflushing with the N
2-CO2mixture
25 Analytical Methods Detection and quantification ofglycerol ethanol 13-propanediol and lactic acid wereobtained with a HPLC equipped with a refractive index andAminex HPX-87H column (BioRad) at 60∘C A solutionof 4mM H
2SO4was used as an eluent at a flow rate of
06mLminSamples for HPLC analysis were centrifuged at
10000 rpm for 10min filtered through a 045 120583mmembranefilter and finally acidified with a 10 ww solution of H
2SO4
For the quantification of volatile fatty acids (VFAs)filtered samples were acidified with H
3PO4(30 120583L of 17
H3PO4was added in 1mL of sample) and analyzed on a gas
chromatograph (PerkinElmer Clarus 400) equipped with aflame ionization detector and a capillary column (AgilentHP-FFAP 30m long 053mm inner diameter) The oven wasprogrammed to start with 105∘C (for 3minutes) followed by aramp that reaches 130∘Cat a rate of 8∘Cmin and subsequently230∘C (held for 3min) at a rate of 45∘Cmin Nitrogen wasused as the carrier gas at 13mLmin the injector temperaturewas set at 240∘C and the detector at 230∘C
The total volume of gas production was measured using awater displacement system [32]
Hydrogen content in the produced gas was measuredwith a gas chromatograph (SRI GC model 310) equippedwith a thermal conductivity detector and a packed column(Porapak-Q length 6 ft and inner diameter 21mm) Thevolume of H
2produced in sealed vials during glycerol fer-
mentation tests was calculated by the mass balance equation[33]
Multivariate data analysis was performed usingUnscram-bler X 101 software (by Camo) A Principal Component
Analysis (PCA) [34] was chosen as a tool to explore the bigdatamatrix obtained from themain fermentation parametersmonitored during the enrichments
26 Next Generation Sequencing DNA was extracted fromthe pellets of 5mL crude samples using the PowerSoil DNAIsolation Kit (MoBio) according to the standard procedureSequencing amplicon librarieswere generated byPCR follow-ing the ldquo16S Metagenomic Sequencing Library PreparationPreparing 16S Ribosomal RNA Gene Amplicons for theIllumina MiSeq Systemrdquo protocol (Illumina part number15044223 rev B) Internal parts of the 16S ribosomal RNA(rRNA) gene covering variable regions V3 and V4 werePCR-amplified with the KAPA HiFi HotStart ReadyMix(KAPA Biosystems) and the primers 51015840-TCGTCGGCAGC-GTCAGATGTGTATAAGAGACAGCCTACGGGNGG-CWGCAG-31015840 and 51015840-GTCTCGTGGGCTCGGAGATG-TGTATAAGAGACAGGACTACHVGGGTATCTAATCC-31015840 and purifiedwith theAgencourt AMPure XP kit (BeckmanCoulter Genomics) The Nextera XT Index Kit was used toadd sequencing adapters and multiplexing indices PooledDNA libraries were sequenced on a MiSeq sequencer (Illu-mina) using theMiSeq Reagent Kit v3 in the 2sdot300 bp paired-end mode
Sequencing reads were demultiplexed trimmed andOTU-classified using the Metagenomics Workflow of theMiSeq Reporter Software v23 (Illumina) This workflowuses an Illumina proprietary classification algorithm andan Illumina-curated version of the Greengenes 135 (May2013) taxonomy database which covers 3 kingdoms 33 phyla74 classes 148 orders 321 families 1086 genera and 6466species
Due to the relatively high number of unclassified readsfound at the species level comparisons between samples arepresented at the genus level while comparisons at the speciesfamily order class and phylum level are available as supp-lementary information (in Supplementary Material availableonline at httpdxdoiorg1011552015932934) Sequencingreads have been deposited to the sequence read archive ofNCBI under the Bioprojects PRJNA285034 (httpwwwncbinlmnihgovbioproject285034) and PRJNA284929 (httpwwwncbinlmnihgovbioproject284929)
3 Results and Discussion
31 Enrichment in Batch Conditions
311 Activated Sludge Based on the experimental scheme(Figure 1) 12 different selection conditionswere tested in trip-licate The enrichment using activated sludge showed goodresults in terms of substrate degradation and it continuedunhindered for several transfers with no evident inhibition(due to the use of crude glycerol) This actually indicatedthe possibility to increase the substrate concentration infuture studiesThe best results obtained in terms of substratedegradation efficiency (practically reaching 100) and biogasproduction were observed with MM-KC This experimentalcondition led to the highest ethanol production converting
BioMed Research International 5
about 10 gL glycerol in 21 h (maximum yield = 46 gg) witha concomitant 13 PD yield of approximately 3 gg After 16transfers however the distribution of the main metaboliteschanged with 13 PD becoming the dominant one andshowing an increase in butyrate during the last transfers
MM-EF also showed a high substrate degradation effi-ciency and (with exception of transfers 5ndash7) the mainmetabolites were represented by 13 PD and butyrate Thiscondition performed the best butyrate production with amaximum yield of 33 gg (from 85 gL glycerol in 72 hfermentation) together with 13 PD yield of 47 gg
The use of BAmedium (experiments 3 and 4) seemed notto favor solventogenesis pathway (almost no ethanol produc-tion was observed) while 13 PD was still by far the mainmetabolite (with an average production of 367plusmn056 gL and399plusmn074 gL for KC and EF resp) followed by butyrate andacetate Also in this case the End of Fermentation seemed tofavor butyrate production with a yield reaching up to 299 gg(from 77 gL glycerol in 72 h fermentation) in BA-EF
Hydrogen in the biogas was rather modest in allexperiments reaching in most cases around 20
The distribution ofmainmetabolites and substrate degra-dation () observed during the enrichment process withactivated sludge are shown in Figures 2(a) and 2(b)
Principal Component Analysis based on the completedatamatrix of 240 samples with 11 variables showed clear dif-ferences between the tested enrichment strategies (Figure 3)with EF closer related to butyrate (especially MM-EF) andBA-KC closer related to acetate In general the first PrincipalComponent (PC) showed an increase of ethanol and hydro-gen moving towards the right while the second PC showedan increase of butyrate productionmoving upwardsThe firstPC roughly separated EF and KC (with the exception ofMM-EF) while the second PC separated MM from BA
Furthermore a comparison of the correlation loadingsobtained with the data of the four enrichment conditions(MM-KC MM-EF BA-KC and BA-EF) separately showedthat only in the case of BA butyric acid was related to H
2
production (Figure 4) as would be expected from a directglycerol conversion into butyrate In fact glycerol conversionto butyric acid has a theoretically yield of 2molmol [35]
Interestingly in the case of MM butyrate production wasnegatively correlated with lactic and acetic acid and alsowith hydrogen in MM-EF while it was positively correlatedwith hydrogen production when using BA medium thusimplying a secondary fermentation (sensu Agler et al [21])(a butyrate production which does not come directly fromglycerol conversion)
There might be several possible pathways leading tobutyrate production through the conversion of lactate andacetate [36] besides the above-mentioned conversion ofglycerol Some examples are provided in
Lactate+ 04Acetate+ 07H+
997888rarr 07Butyrate+ 06H2 +CO2 + 04H2O
ΔG = minus1839
(1)
Lactate+Acetate+H+
997888rarr Butyrate+ 08H2 + 14CO2 + 06H2O
ΔG = minus594
(2)
2Lactate+H+ 997888rarr Butyrate+ 2H2+ 2CO2
ΔG = minus641(3)
It is also worth noting that Zhu and Yang [37] observed ametabolic shift from butyrate formation to lactate and acetateat pH lt 63 associated with decreased activities of phos-photransbutyrylase and NAD-independent lactate dehydro-genase and increased activities of phosphotransacetylase andlactate dehydrogenase Our batch experiments were operatedwithout pH control starting at pH 7 and typically endingat around 48 due to glycerol acidification Therefore it islikely that such a metabolic shift was also involved in ourfermentation tests
312 Anaerobic Sludge Differently from activated sludge theenrichment of anaerobic sludge in batch conditions showed aclear inhibition regardless of the selection strategy (BA andMMgrowthmedium EF or KC transfers)The inhibitionwaspresumably related to the high concentration of LCFA andthe negative interaction with the cell membranes of Gram-positive anaerobic bacteria of the anaerobic sludge ratherthan product inhibition In fact even after centrifuging theinoculum washing away the supernatant and resuspendingthe pellet into freshmedium (thus washing away extracellularsoluble metabolites) no recovery of the fermentation wasachieved Addition of specific elements such as yeast extractor vitamin andmineral solution did not have any effect either
The distribution of main metabolites and fraction of H2
(in the headspace) detected during the enrichment processwith anaerobic sludge are shown in Figure 5 The use ofMM (without nutrient supplements) led to inactivation afteronly 1 transfer while BA reached 6-7 transfers before beinginhibited (Figure 5(a)) Nonpretreated sludge (Figure 5(b))showed a high production of propionic acid while withheat-shock treated sludge (Figure 5(c)) butyric acid was thedominant metabolite The latter condition was chosen for analternative selection strategy using fed-batch conditions
313 Hexane-Pretreated Glycerol Tests As mentioned aboveheat-shock treated (HS) inoculum was chosen for furtherexperimentation The possible inhibiting effect of LCFA andldquolipidic compoundsrdquo was evaluated in the following test Thehypothesis was that the animal fat derived crude glycerolwould contain inhibiting amounts of LCFA which mightnegatively interfere with the membrane of Gram-positivebacteria of the anaerobic sludge Activated sludge was notincluded in this test since it did not show any inhibition
Nonextracted crude glycerol showed an organic carboncontent expressed as chemical oxygen demand (COD) of1309 plusmn 32 g CODL while the extracted crude glycerol was1172 plusmn 12 g CODL thus suggesting that approximately 137 gCODL of ldquolipidic compoundsrdquo was removed (which would
6 BioMed Research International
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(2) MM-EF gas products
0005101520253035404550
(gL
)
(2) MM-EF liquid products
0005101520253035404550
(gL
)
(1) MM-KC liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(1) MM-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(a)
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(4) BA-EF gas products
0005101520253035404550
(gL
)
(3) BA-KC liquid products
0005101520253035404550
(gL
)
(4) BA-EF liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(3) BA-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(b)
Figure 2 Fermentation products monitored during the enrichment of activated sludge in batch conditions through repeated transfers usingMM (a) and BA (b) medium (1) MM-KC = Minimal Medium with Kinetic Control (21 h) (2) MM-EF = Minimal Medium with End ofFermentation (72 h) (3) BA-KC = Basal Medium with Kinetic Control (21 h) (4) BA-EF = Basal Medium with End of Fermentation (72 h)
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
BioMed Research International 5
about 10 gL glycerol in 21 h (maximum yield = 46 gg) witha concomitant 13 PD yield of approximately 3 gg After 16transfers however the distribution of the main metaboliteschanged with 13 PD becoming the dominant one andshowing an increase in butyrate during the last transfers
MM-EF also showed a high substrate degradation effi-ciency and (with exception of transfers 5ndash7) the mainmetabolites were represented by 13 PD and butyrate Thiscondition performed the best butyrate production with amaximum yield of 33 gg (from 85 gL glycerol in 72 hfermentation) together with 13 PD yield of 47 gg
The use of BAmedium (experiments 3 and 4) seemed notto favor solventogenesis pathway (almost no ethanol produc-tion was observed) while 13 PD was still by far the mainmetabolite (with an average production of 367plusmn056 gL and399plusmn074 gL for KC and EF resp) followed by butyrate andacetate Also in this case the End of Fermentation seemed tofavor butyrate production with a yield reaching up to 299 gg(from 77 gL glycerol in 72 h fermentation) in BA-EF
Hydrogen in the biogas was rather modest in allexperiments reaching in most cases around 20
The distribution ofmainmetabolites and substrate degra-dation () observed during the enrichment process withactivated sludge are shown in Figures 2(a) and 2(b)
Principal Component Analysis based on the completedatamatrix of 240 samples with 11 variables showed clear dif-ferences between the tested enrichment strategies (Figure 3)with EF closer related to butyrate (especially MM-EF) andBA-KC closer related to acetate In general the first PrincipalComponent (PC) showed an increase of ethanol and hydro-gen moving towards the right while the second PC showedan increase of butyrate productionmoving upwardsThe firstPC roughly separated EF and KC (with the exception ofMM-EF) while the second PC separated MM from BA
Furthermore a comparison of the correlation loadingsobtained with the data of the four enrichment conditions(MM-KC MM-EF BA-KC and BA-EF) separately showedthat only in the case of BA butyric acid was related to H
2
production (Figure 4) as would be expected from a directglycerol conversion into butyrate In fact glycerol conversionto butyric acid has a theoretically yield of 2molmol [35]
Interestingly in the case of MM butyrate production wasnegatively correlated with lactic and acetic acid and alsowith hydrogen in MM-EF while it was positively correlatedwith hydrogen production when using BA medium thusimplying a secondary fermentation (sensu Agler et al [21])(a butyrate production which does not come directly fromglycerol conversion)
There might be several possible pathways leading tobutyrate production through the conversion of lactate andacetate [36] besides the above-mentioned conversion ofglycerol Some examples are provided in
Lactate+ 04Acetate+ 07H+
997888rarr 07Butyrate+ 06H2 +CO2 + 04H2O
ΔG = minus1839
(1)
Lactate+Acetate+H+
997888rarr Butyrate+ 08H2 + 14CO2 + 06H2O
ΔG = minus594
(2)
2Lactate+H+ 997888rarr Butyrate+ 2H2+ 2CO2
ΔG = minus641(3)
It is also worth noting that Zhu and Yang [37] observed ametabolic shift from butyrate formation to lactate and acetateat pH lt 63 associated with decreased activities of phos-photransbutyrylase and NAD-independent lactate dehydro-genase and increased activities of phosphotransacetylase andlactate dehydrogenase Our batch experiments were operatedwithout pH control starting at pH 7 and typically endingat around 48 due to glycerol acidification Therefore it islikely that such a metabolic shift was also involved in ourfermentation tests
312 Anaerobic Sludge Differently from activated sludge theenrichment of anaerobic sludge in batch conditions showed aclear inhibition regardless of the selection strategy (BA andMMgrowthmedium EF or KC transfers)The inhibitionwaspresumably related to the high concentration of LCFA andthe negative interaction with the cell membranes of Gram-positive anaerobic bacteria of the anaerobic sludge ratherthan product inhibition In fact even after centrifuging theinoculum washing away the supernatant and resuspendingthe pellet into freshmedium (thus washing away extracellularsoluble metabolites) no recovery of the fermentation wasachieved Addition of specific elements such as yeast extractor vitamin andmineral solution did not have any effect either
The distribution of main metabolites and fraction of H2
(in the headspace) detected during the enrichment processwith anaerobic sludge are shown in Figure 5 The use ofMM (without nutrient supplements) led to inactivation afteronly 1 transfer while BA reached 6-7 transfers before beinginhibited (Figure 5(a)) Nonpretreated sludge (Figure 5(b))showed a high production of propionic acid while withheat-shock treated sludge (Figure 5(c)) butyric acid was thedominant metabolite The latter condition was chosen for analternative selection strategy using fed-batch conditions
313 Hexane-Pretreated Glycerol Tests As mentioned aboveheat-shock treated (HS) inoculum was chosen for furtherexperimentation The possible inhibiting effect of LCFA andldquolipidic compoundsrdquo was evaluated in the following test Thehypothesis was that the animal fat derived crude glycerolwould contain inhibiting amounts of LCFA which mightnegatively interfere with the membrane of Gram-positivebacteria of the anaerobic sludge Activated sludge was notincluded in this test since it did not show any inhibition
Nonextracted crude glycerol showed an organic carboncontent expressed as chemical oxygen demand (COD) of1309 plusmn 32 g CODL while the extracted crude glycerol was1172 plusmn 12 g CODL thus suggesting that approximately 137 gCODL of ldquolipidic compoundsrdquo was removed (which would
6 BioMed Research International
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(2) MM-EF gas products
0005101520253035404550
(gL
)
(2) MM-EF liquid products
0005101520253035404550
(gL
)
(1) MM-KC liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(1) MM-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(a)
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(4) BA-EF gas products
0005101520253035404550
(gL
)
(3) BA-KC liquid products
0005101520253035404550
(gL
)
(4) BA-EF liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(3) BA-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(b)
Figure 2 Fermentation products monitored during the enrichment of activated sludge in batch conditions through repeated transfers usingMM (a) and BA (b) medium (1) MM-KC = Minimal Medium with Kinetic Control (21 h) (2) MM-EF = Minimal Medium with End ofFermentation (72 h) (3) BA-KC = Basal Medium with Kinetic Control (21 h) (4) BA-EF = Basal Medium with End of Fermentation (72 h)
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
6 BioMed Research International
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(2) MM-EF gas products
0005101520253035404550
(gL
)
(2) MM-EF liquid products
0005101520253035404550
(gL
)
(1) MM-KC liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(1) MM-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(a)
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(4) BA-EF gas products
0005101520253035404550
(gL
)
(3) BA-KC liquid products
0005101520253035404550
(gL
)
(4) BA-EF liquid products
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
(3) BA-KC gas products
AcetatePropionateButyrate
EthanolLactate13-Propanediol
Degradation ()
Biogas (mL)H2 ()
(b)
Figure 2 Fermentation products monitored during the enrichment of activated sludge in batch conditions through repeated transfers usingMM (a) and BA (b) medium (1) MM-KC = Minimal Medium with Kinetic Control (21 h) (2) MM-EF = Minimal Medium with End ofFermentation (72 h) (3) BA-KC = Basal Medium with Kinetic Control (21 h) (4) BA-EF = Basal Medium with End of Fermentation (72 h)
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
BioMed Research International 7
PC-1 (41)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(26
)
minus1
minus08
minus06
minus04
minus02
0
02
04
06
08
1Correlation loadings (X)
Biogas
Acetate
Butyrate
Ethanol
Lactate
Succinate
13 PD
PC-1 (41)minus4 minus3 minus2 minus1 0 1 2 3 4 5 6 7
PC-2
(26
)
minus4
minus3
minus2
minus1
0
1
2
3
4
Scores
MM_KCMM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KC
BA_EF
BA_EFMM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EFBA_KC
BA_KCBA_KCBA_EFBA_EF
BA_EF
MM_KCMM_KCMM_KCMM_EF
MM_EFMM_EFBA_KCBA_KCBA_KC
BA_EFBA_EF
BA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KC
BA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EF
BA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EF
MM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EF
MM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EF
MM_KCMM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KC
BA_KC
BA_EFBA_EFBA_EF
MM_KC
MM_KC
MM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EF
BA_EFBA_EF
MM_KCMM_KCMM_KC
MM_EFMM_EFMM_EF
BA_KCBA_KCBA_KC
BA_EFBA_EFBA_EFH2 ()
Figure 3 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) and thedistribution of the samples (score plot) during the experiments with activated sludge MM-EF (in red) MM-KC (in blue) BA-EF (in grey)and BA-KC (in green) The first two components explain together about 67 of the total variability
PC-1 (53)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(18
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
DegradationFinal pHBiogas
Acetate
Butyrate
Ethanol
Lactate
Succinate13 PD
PC-1 (55)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(25
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pHBiogas
Acetate
Propionate
Butyrate
EthanolLactate
Succinate
13 PD
PC-1 (37)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(24
)
minus1minus08minus06minus04minus02
002040608
1
Correlation loadings (X)
Degradation
Final pH
BiogasAcetate
PropionateButyrateEthanol
Lactate
13 PD
PC-1 (46)minus1 minus08 minus06 minus04 minus02 0 02 04 06 08 1
PC-2
(28
)
minus1minus08minus06minus04minus02
002040608
1Correlation loadings (X)
Degradation
Final pH
Biogas
Acetate
Butyrate
13 PD
MM-KC BA-KC
MM-EF BA-EF
H2 ()
H2 ()H2 ()
H2 ()
Figure 4 Principal Component Analysis showing the distribution of the main fermentation parameters (correlation loading plot) of the fourexperimental conditions (namely MM-EF MM-KC BA-EF and BA-KC) separately during the experiments with activated sludge The firsttwo components explain together more than 60 of the total variability in all cases
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
8 BioMed Research International
0
(a)
(b)
5
10
15
20
25
30
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
Nonpretreated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
05
1015202530
T1 T2 T3 T4 T5 T6 T7 T8 T9Transfers
HS treated anaerobic sludge
MM-KCMM-EF
BA-KCBA-EF
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers nonpretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers nonpretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers nonpretreated
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2
(gL
)
MM-EF transfers HS pretreated
002040608
112141618
T1 T2
(gL
)
MM-KC transfers HS pretreated
H2
()
H2
()
Figure 5 Continued
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
BioMed Research International 9
(c)
AcetatePropionate
ButyrateValerate
AcetatePropionate
ButyrateValerate
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)
BA-EF transfers HS pretreated
002040608
112141618
T1 T2 T3 T4 T5 T6 T7 T8 T9
(gL
)BA-KC transfers HS pretreated
Figure 5 Results of batch transfers during the enrichment of anaerobic sludge showing H2 (a) in the headspace soluble metabolites fromnonpretreated anaerobic sludge (b) and soluble metabolites from heat-shock treated anaerobic sludge (c) MM-KC =Minimal Medium withKinetic Control (21 h) MM-EF = Minimal Medium with End of Fermentation (72 h) BA-KC = Basal Medium with Kinetic Control (21 h)BA-EF = Basal Medium with End of Fermentation (72 h)
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10Transfers
Enrichment of anaerobic sludge gaseous products
Biogas (mL)H2 ()
(a)
02468
1012
0 2 4 6 8 10
(gL
)
Transfers
Enrichment of anaerobic sludge soluble compounds
Glycerol consumed13-Propanediol
Butyrate
(b)
Figure 6 Results from the batch transfers of anaerobic sludge using hexane-treated crude glycerol showing gas products (a) and glycerolconsumption together with the main soluble metabolites (b)
0
1
2
3
4
5
6
1 3 5 7 9 11 13 15 17 19 21 23 25
(gL
)
Feeds
Anaerobic sludge fed-batch
AcetatePropionateButyrate
Ethanol13-Propanediol
Figure 7 Distribution of main soluble metabolites observed during the fed-batch enrichment process with heat-shock treated anaerobicsludge
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
10 BioMed Research International
approximately correspond to 347 gL of oleic acid a typicalLCFA known for its inhibiting effect)
As can be seen in Figure 6 repeated transfers in batchconditions with the hexane-treated crude glycerol led to highsubstrate degradation efficiency and the MMC was neverinactivated showing glycerol fermentation performancescomparable with those obtained with activated sludge Thisimplied that indeed the inactivation of anaerobic sludgedepended on the high LCFA content of the 2G crude glycerol
However since the aim of this study was the selectionof MMC that can grow on nonpretreated crude glycerolthe possibility to achieve enrichment and adaptation tests ofanaerobic sludge using fed-batch conditionswas investigated
32 Enrichment in Fed-Batch Conditions As can be seen inFigure 7 the fed-batch operations allowed effective overcom-ing of crude glycerol inhibitionwith anaerobic sludge leadingto a good substrate conversion into mainly 13 PD ethanoland butyrate (after about 14 feedings) However the reactorstarted to develop a community of sulfate reducing bacteria(SRB) that inhibited fermentation after roughly 7 feedingsFor this reason the sludge underwent a second heat-shocktreatment (at 10 feedings) to allow further glycerol fermen-tation Nonetheless H
2S production occurred again after
21 feedings Probably continuous mode fermentation withshort hydraulic retention time (HRT) would thus representa suitable approach for successful adaptationenrichment ofanaerobic sludge to untreated crude glycerol (possibly help-ing to rinse out slower growing SRB) For this reason ongoingwork is now focusing on identification of the operatingparameters for maintaining a stable MMC in continuousmode and statistical optimization of key parameters for greenchemicals production Since activated sludgewas successfullyenriched in batch conditions there was no need to performfed-batch tests with this inoculum
33 Molecular Characterization of the MMC during theEnrichment Process The development of the MMC wasmonitored by sequencing amplicons of the V3 and V4 vari-able regions of the 16S rRNA gene Operational taxonomicunits (OTUs) were then assigned from each sequencing readand used as a measure of the microbial diversity of eachsample The copy number of the 16S rRNA gene varies from1 to 15 depending on the species and the OTUs are thereforeonly providing an estimate of the true microbial diversityThe copy number is varying but is relatively high in thetaxa Firmicutes and Gammaproteobacteria with a mean of58 plusmn 28 copies while it is lower for Bacteroidetes (35 plusmn 15)Betaproteobacteria (33 plusmn 16) Actinobacteria (31 plusmn 17)and Spirochaetes (24 plusmn 10) [38] Overall the Firmicutes andGammaproteobacteria are overestimated in the analysis andthe cell-count may for some genera be sim5ndash10-fold lower thanthe OTU count
331 Activated Sludge Experiments In all these samplesthere was a dominance of bacteria belonging to the phylumFirmicutes in particular from the classes Clostridia and
Bacilli and of the classGammaproteobacteria (Figures S1ndashS6)
MM-KC The enrichment was characterized by a strongdecrease of the genera Clostridium and Lactobacillus bothFirmicutes and an increase ofKlebsiella andEscherichia bothGammaproteobacteria (Table 2 Figure S2) In particular thejoint increase of the latter two probably favored an enhancedethanol production (T10 and T13) while the dominanceof Klebsiella alone (T18) was associated with a metabolicshift towards 13 PD (see Figure 2(a)) These results are ingood agreement with previous observations with enrichedactivated sludge selected with Kinetic Control [39]
MM-EFThe distribution of themain genera observed duringthese tests showed a sequence of dominance shifts goingfrom Escherichia to Klebsiella and finally to Clostridium andEscherichia The ethanol peak observed in T6 is associatedwith the dominance of Escherichia (around 55) whilethe subsequent increase of Klebsiella (reaching almost 70)shifted towards 13 PD production (T8 52 gL 13 PD and noethanol production) Moreover the stability of the commu-nity from T8 to T15 is also reflected in the distribution of themain metabolites (see Figure 2(a)) The higher butyric acidproduction observed after T7might be related to the increaseof the genus Clostridium which includes several butyric acidproducing species
BA-KC Interestingly a clear increase in biodiversity could beobserved during the enrichment of BA-KC with an initialdominance of Clostridium (86) and a sharp decrease overtime leading to less than 8This decrease is associated witha concomitant increase of other genera such as Escherichia(reaching 34) Lactobacillus (13) and a number of unclas-sified genera (approximately 14 in total primarily fromthe classes Gammaproteobacteria and Clostridia Figure S5)followed by Serratia andKlebsiella (10) Higher butyric acidwas observed in T1 and T12 in the presence of at least 70of Clostridium while an increased acetic acid production wasobserved in T18
BA-EF In general this enrichment was characterized bya dominance of Clostridium with a decrease towards thelast transfers A decrease of acetic acid and concomitantincrease in butyric acid could be observed comparing thesamples T7 and T11 which were associated with a decreaseof the genus Slackia (typically producing acetic acid andlactic and formic acid [40]) and an increase in ClostridiumA very sharp decrease of butyric acid (together with anincrease in acetic acid and ethanol) could be observed inT15 which was associated with a decrease in Clostridiumand a concomitant increase of unclassified genera primarilybelonging to the phylum Proteobacteria and in particular theclass Gammaproteobacteria (Figures S5 and S6)
332 Anaerobic Sludge Experiments This subparagraphreports the results of MMC taxonomical characterization forthe anaerobic sludge enriched on hexane-pretreated crude
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
BioMed Research International 11
Table2Metagenom
iccla
ssificatio
nof
theMMCat
thegenu
slevel
Results
ofbatchtransfe
rsd
uringtheenric
hmento
factivated
sludgeexpressedas
fractio
n(
)MM-KC=Minim
alMedium
with
Kinetic
Con
trol(21h)M
M-EF=Minim
alMedium
with
Endof
Ferm
entatio
n(72h
)BA
-KC=Ba
salM
edium
with
Kinetic
Con
trol(21h)B
A-EF
=Ba
salM
edium
with
End
ofFerm
entatio
n(72h
)T0
ndashT20
=transfe
rnum
bersN
D=Not
detectedG
eneraa
ppearin
gatfre
quencies
below1
inallsam
ples
wereo
mitted
GEN
ERA
MM-KC
MM-EF
BA-KC
BA-EF
T1T3
T7T10
T13
T18
T0T6
T7T8
T15
T20
T1T12
T18
T0T7
T11
T15
T20
Clostridium
513
808
370
124
288
181
284
142
121
401
432
839
864
674
792
679
603
737
320
454
Klebsiella
074
042
474
289
191
654
280
013
667
015
007
003
030
912
918
002
003
019
661
003
Escherich
ia054
846
060
335
287
105
099
542
705
310
316
005
074
516
344
005
011
084
309
428
Uncla
ssified
647
283
831
814
115
888
138
184
105
135
109
123
291
656
142
183
341
593
572
403
Lactobacillus
297
007
011
001
002
001
039
353
078
079
474
143
605
472
133
001
168
148
218
412
Slackia
007lt001lt001lt001
001lt001
001
704
121
516
041
003
lt001
361
055
105
181
261
021
101
Serratia
001
206
045
997
572
059
064
802
268
436
435
001
028
105
103
001
001
014
173
135
Enterobacter
001
158
093
386
372
139
066
367
178
227
250lt001
008
055
261
lt001lt001
007
074
025
Alkaliphilus
029
001
002lt001lt001
001
366
001lt001
001
001lt001
014lt001
001
065lt001
002
ND
ND
Tolumonas
001
022
174
119
070
273
143
004
266
003
003
ND
006
048
063
lt001lt001
001
001lt001
Negativ
icoccus
002lt001
001lt001lt001lt001
001lt001
000lt001lt001lt001
001
014
256
006
002
004lt001
ND
Blautia
020
005
057
004
001
004
092
007
002
001
001lt001
010
001
029
003
001
001
ND
ND
Ruminococcus
ND
ND
NDlt001
ND
ND
002
013
217
014
013
ND
lt001lt001lt001
005
024
ND
ND
ND
Erwinia
lt001
211
006
010
006
007
003
005
046
003
004
ND
lt001
004
025
NDlt001lt001
005lt001
Methylotenera
091
005lt001lt001
NDlt001
089
NDlt001
ND
NDlt001
010
NDlt001
197
ND
ND
ND
ND
Geobacillus
014
002
005
003lt001
005
144lt001
002lt001lt001lt001
005lt001
003
003lt001lt001
ND
ND
Pseudomonas
073
012
006
018
004
001
116
004
002
004
003
001
002
001
001
001lt001
002
001
027
Weis
sella
108
002lt001lt001lt001lt001
055lt001lt001lt001lt001
ND
021lt001
001
041lt001
001
ND
ND
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
12 BioMed Research International
Table 3 Metagenomic classification of the MMC at the genuslevel for the anaerobic sludge enriched on hexane-pretreated crudeglycerol in batch tests (HT) and with the untreated crude glycerolin fed-batch expressed as fraction () T0ndashT11 = transfer numbersND =Not detected Genera appearing at frequencies below 1 in allsamples were omitted
GeneraHT FED-BATCH
T0 T9 T11
Blautia 024 004 508Clostridium 301 466 162Unclassified 315 645 989Klebsiella 001 288 002Escherichia 006 103 lt001Enterococcus 002 027 619Alkaliphilus 564 006 088Soehngenia lt001 ND 352Serratia 001 267 004Pedobacter 238 002 008Enterobacter 002 221 001Propionispora 199 001 003Treponema 142 001 003Peptoniphilus 007 002 135Flavobacterium 133 003 054Sedimentibacter 033 lt001 126
glycerol in batch tests (HT) and with the untreated crudeglycerol in fed-batch (Figures S7ndashS12) Anaerobic sludgegrown on untreated glycerol underwent quick inhibition andwas thus not analyzed
The main difference that can be observed between thebatch and fed-batch conditions was the dominant presenceof Blautia (up to 50) in the latter (Table 3) The fed-batchcommunity was also characterized by the genus Clostridiumin addition to a number of unclassified genera primarily ofthe phylumFirmicutes Dominant genera in batch conditions(HT) at T0 were Clostridium and unclassified genera (botharound 30) with an increase of Clostridium (reachingmore than 45) and Klebsiella (almost 30) in T9 It isworth noting that T0 was a highly diverse sample withmultiple genera having abundances in the range of 01ndash09explaining why the total fraction only reached about 75(see Figure S8) The unclassified genera found in T0 mainlybelonged to the phyla Proteobacteria (in particular to theclass Deltaproteobacteria) and Firmicutes (especially to theclass Clostridia) (Figures S11 and S12)
A total of 19 genera belonging to SRB were retrievedin the different anaerobic sludge samples even thoughalways at a very low (far below the cut-off set at 1)Initial sludge (HS T0) contained 18 different genera (mainlyDesulfovibrio andDesulfofrigus) accounting for 119 whichdecreased to 10 genera (00023) in T9 This suggests thatthe Kinetic Control was effective in enriching faster grow-ing (glycerol consuming) bacteria such as Clostridium andKlebsiella species over SRB In fed-batch conditions instead
the absence of a Kinetic Control allowed the growth of SRBThus even though a second heat-shock treatment (T11) wasable to decrease SRB from initial 19 genera to 16 (accountingfor 059) this was probably sufficient to allow SRB togrow in the following weeks of fed-batch experimentation aswitnessed by the H
2S production observed in the fed-batch
reactor (which turned black and was characterized by thetypical strong H
2S smell) The most abundant genus found
in T11 was Desulfotomaculum (mainly with the species Dhalophilum) Desulfotomaculum comprises endospore form-ing Gram-positive bacteria Desulfotomaculum spp are ableto grow autotrophically (using H
2CO2) and produce sulfide
and acetate Besides H2as electron donor they are able
to utilize alcohols and organic acids which were likely toaccumulate in the fed-batch system Besides sulfate reductionthey may also use various other sulfur compounds [41]
4 Conclusions
The selection and adaptation of activated sludge inoculumthrough successive transfers in batch conditions were per-formed successfully and continued unhindered for severalmonths The best results showed a substrate degradationefficiency of almost 100 (about 10 gL) and different dom-inant metabolic products were obtained depending on theselection strategy (mainly 13 PD ethanol or butyrate) Inparticular the strategy of Kinetic Control coupled withMinimalMedium (MM-KC) led to a maximum ethanol yieldof 46 gL together with a 13 PD yield of around 3 ggwith complete substrate degradation within 21 h The Endof Fermentation coupled with Minimal Medium (MM-EF)showed a degradation efficiency of around 90ndash95 with amaximum butyric acid yield of 33 gg (from 85 gL glycerolin 72 h fermentation) together with a 13 PD yield of 47 ggTests with the rich BA medium showed a general lower sub-strate degradation efficiency but were also characterized bya high 13 PD and butyric acid production Multivariate dataanalysis showed clear differences between different strategiesand further suggested that only in the case of BAmedium thebutyric acid was directly produced from glycerol In additionEnd of Fermentation enrichment seemed to favor butyricacid production On the other hand anaerobic sludge (bothheat pretreated and not) exhibited inactivation after a fewtransfers in batch conditions probably due to the presenceof high concentration of lipidic compounds Fed-batch modeturned out to be a valid alternative adaptation strategyovercoming inhibition problems related to crude glycerolcomposition but was also associated with H
2S production
thus implying the use of continuousmode to better select andadapt anaerobic sludge to the conversion of animal fat derivedcrude glycerol After overcoming inhibition problems mainmetabolites produced were comparable with those obtainedwith activated sludge with a high 13 PD and butyric acidproduction
Next Generation Sequencing represented a useful toolto monitor the changes in microbial composition of MMCshighlighting the development of a glycerol consuming com-munity (with numerous strains belonging to the genera
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
BioMed Research International 13
ClostridiumKlebsiella and Escherichia) thus confirming theeffectiveness of the enrichment strategy
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Acknowledgment
The authors wish to thank the European Commission for thefinancial support of this work under FP7 Grant Agreementno 613667 (acronym GRAIL)
References
[1] M Ayoub and A Z Abdullah ldquoCritical review on the currentscenario and significance of crude glycerol resulting frombiodiesel industry towards more sustainable renewable energyindustryrdquo Renewable amp Sustainable Energy Reviews vol 16 no5 pp 2671ndash2686 2012
[2] C Varrone R Liberatore T Crescenzi G Izzo and A WangldquoThe valorization of glycerol economic assessment of aninnovative process for the bioconversion of crude glycerol intoethanol and hydrogenrdquo Applied Energy vol 105 pp 349ndash3572013
[3] N Kolesarova MHutan I Bodık andV Spalkova ldquoUtilizationof biodiesel by-products for biogas productionrdquo Journal ofBiomedicine and Biotechnology vol 2011 Article ID 126798 16pages 2011
[4] I Ntaikou C Valencia Peroni C Kourmentza et al ldquoMicrobialbio-based plastics from olive-mill wastewater generation andproperties of polyhydroxyalkanoates from mixed cultures in atwo-stage pilot scale systemrdquo Journal of Biotechnology vol 188pp 138ndash147 2014
[5] K Johnson Y Jiang R Kleerebezem G Muyzer and MC M van Loosdrecht ldquoEnrichment of a mixed bacterialculture with a high polyhydroxyalkanoate storage capacityrdquoBiomacromolecules vol 10 no 4 pp 670ndash676 2009
[6] P Kumar M Singh S Mehariya S K S Patel J-K Lee andV C Kalia ldquoEcobiotechnological approach for exploiting theabilities of Bacillus to produce co-polymer of polyhydroxyalka-noaterdquo Indian Journal of Microbiology vol 54 no 2 pp 151ndash1572014
[7] H Moralejo-Garate R Kleerebezem A Mosquera-Corraland M C M Van Loosdrecht ldquoImpact of oxygen limitationon glycerol-based biopolymer production by bacterial enrich-mentsrdquoWater Research vol 47 no 3 pp 1209ndash1217 2013
[8] A-P Zeng and H Biebl ldquoBulk chemicals from biotechnologythe case of 13-propanediol production and the new trendsrdquoAdvances in Biochemical EngineeringBiotechnology vol 74 pp239ndash259 2002
[9] J Hao R Lin Z Zheng H Liu and D Liu ldquoIsolation and char-acterization ofmicroorganisms able to produce 13-propanediolunder aerobic conditionsrdquo World Journal of Microbiology andBiotechnology vol 24 no 9 pp 1731ndash1740 2008
[10] G P da Silva M Mack and J Contiero ldquoGlycerol a promis-ing and abundant carbon source for industrial microbiologyrdquoBiotechnology Advances vol 27 no 1 pp 30ndash39 2009
[11] E K C Yu and J N Saddler ldquoBiomass conversion to butanediolby simultaneous saccharification and fermentationrdquo Trends inBiotechnology vol 3 no 4 pp 100ndash104 1985
[12] P Kumar R Sharma S Ray et al ldquoDark fermentative bio-conversion of glycerol to hydrogen by Bacillus thuringiensisrdquoBioresource Technology vol 182 pp 383ndash388 2015
[13] P Kumar S Mehariya S Ray A Mishra and V C KalialdquoBiodiesel industry waste a potential source of bioenergy andbiopolymersrdquo Indian Journal of Microbiology vol 55 pp 1ndash72014
[14] A Zhou J Du C Varrone Y Wang A Wang and W LiuldquoVFAs bioproduction from waste activated sludge by couplingpretreatments with Agaricus bisporus substrates conditioningrdquoProcess Biochemistry vol 49 no 2 pp 283ndash289 2014
[15] L Marang Y Jiang M C M van Loosdrecht and R Kleere-bezem ldquoButyrate as preferred substrate for polyhydroxybu-tyrate productionrdquo Bioresource Technology vol 142 pp 232ndash239 2013
[16] S J Sarma S K Brar Y Le Bihan G Buelna and C R SoccolldquoHydrogen production from meat processing and restaurantwaste derived crude glycerol by anaerobic fermentation andutilization of the spent brothrdquo Journal of Chemical Technologyand Biotechnology vol 88 no 12 pp 2264ndash2271 2013
[17] Z Chi D Pyle Z Wen C Frear and S Chen ldquoA laboratorystudy of producing docosahexaenoic acid from biodiesel-wasteglycerol by microalgal fermentationrdquo Process Biochemistry vol42 no 11 pp 1537ndash1545 2007
[18] S K Athalye R A Garcia and Z Wen ldquoUse of biodiesel-derived crude glycerol for producing eicosapentaenoic acid(EPA) by the fungus Pythium irregularerdquo Journal of Agriculturaland Food Chemistry vol 57 no 7 pp 2739ndash2744 2009
[19] W J Choi ldquoGlycerol-based biorefinery for fuels and chemicalsrdquoRecent Patents on Biotechnology vol 2 no 3 pp 173ndash180 2008
[20] J Bader E Mast-Gerlach M K Popovic R Bajpai andU Stahl ldquoRelevance of microbial coculture fermentations inbiotechnologyrdquo Journal of Applied Microbiology vol 109 no 2pp 371ndash387 2010
[21] M T Agler B A Wrenn S H Zinder and L T AngenentldquoWaste to bioproduct conversion with undefined mixed cul-tures the carboxylate platformrdquoTrends in Biotechnology vol 29no 2 pp 70ndash78 2011
[22] P A Selembo J M Perez W A Lloyd and B E LoganldquoEnhanced hydrogen and 13-propanediol production fromglycerol by fermentation using mixed culturesrdquo Biotechnologyand Bioengineering vol 104 no 6 pp 1098ndash1106 2009
[23] A Gadhe S S Sonawane andMN Varma ldquoKinetic analysis ofbiohydrogen production from complex dairy wastewater underoptimized conditionrdquo International Journal of Hydrogen Energyvol 39 no 3 pp 1306ndash1314 2014
[24] I Z Boboescu M Ilie V D Gherman et al ldquoRevealingthe factors influencing a fermentative biohydrogen productionprocess using industrial wastewater as fermentation substraterdquoBiotechnology for Biofuels vol 7 no 1 article 139 2014
[25] B S Saharan A Grewal and P Kumar ldquoBiotechnologicalproduction of polyhydroxyalkanoates a review on trends andlatest developmentsrdquo Chinese Journal of Biology vol 2014Article ID 802984 18 pages 2014
[26] J Wang W-W Li Z-B Yue and H-Q Yu ldquoCultivationof aerobic granules for polyhydroxybutyrate production fromwastewaterrdquo Bioresource Technology vol 159 pp 442ndash445 2014
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
14 BioMed Research International
[27] A Marone G Izzo L Mentuccia et al ldquoVegetable waste assubstrate and source of suitable microflora for bio-hydrogenproductionrdquo Renewable Energy vol 68 pp 6ndash13 2014
[28] P Anand and R K Saxena ldquoA comparative study of solvent-assisted pretreatment of biodiesel derived crude glycerol ongrowth and 13-propanediol production from Citrobacter fre-undiirdquo New Biotechnology vol 29 no 2 pp 199ndash205 2012
[29] F Barbirato C Camarasa-Claret J P Grivet and A BoriesldquoGlycerol fermentation by a new 13-propanediol-producingmicroorganism Enterobacter agglomeransrdquo Applied Microbiol-ogy and Biotechnology vol 43 no 5 pp 786ndash793 1995
[30] I Angelidaki S P Petersen and B K Ahring ldquoEffects of lipidson thermophilic anaerobic digestion and reduction of lipidinhibition upon addition of bentoniterdquo Applied Microbiologyand Biotechnology vol 33 no 4 pp 469ndash472 1990
[31] E A A Wolin M J J Wolin and R S S Wolfe ldquoFormationof methane by bacterial extractsrdquo The Journal of BiologicalChemistry vol 238 pp 2332ndash2286 1963
[32] V C Kalia S R Jain A Kumar and A P Joshi ldquoFermentationof biowaste to H
2
by Bacillus licheniformisrdquo World Journal ofMicrobiology and Biotechnology vol 10 no 2 pp 224ndash227 1994
[33] B E Logan S-E Oh I S Kim and S Van Ginkel ldquoBiologicalhydrogen production measured in batch anaerobic respirome-tersrdquo Environmental Science and Technology vol 36 no 11 pp2530ndash2535 2002
[34] J E Jackson A Userrsquos Guide to Principal Components Wiley2003
[35] B T Maru M Constanti A M Stchigel F Medina and JE Sueiras ldquoBiohydrogen production by dark fermentation ofglycerol using Enterobacter and Citrobacter Sprdquo BiotechnologyProgress vol 29 no 1 pp 31ndash38 2013
[36] A Marone G Massini C Patriarca A Signorini C Varroneand G Izzo ldquoHydrogen production from vegetable waste bybioaugmentation of indigenous fermentative communitiesrdquoInternational Journal of Hydrogen Energy vol 37 no 7 pp 5612ndash5622 2012
[37] Y Zhu and S-T Yang ldquoEffect of pH on metabolic pathwayshift in fermentation of xylose by Clostridium tyrobutyricumrdquoJournal of Biotechnology vol 110 no 2 pp 143ndash157 2004
[38] T Vetrovsky and P Baldrian ldquoThe variability of the 16S rRNAgene in bacterial genomes and its consequences for bacterialcommunity analysesrdquo PLoS ONE vol 8 no 2 Article IDe57923 2013
[39] C Varrone Bioconversion of crude glycerol into hydrogen andethanol by microbial mixed culture [PhD dissertation] HarbinInstitute of Technology Harbin China 2015
[40] F Nagai Y Watanabe and M Morotomi ldquoSlackia piriformissp nov and Collinsella tanakaei sp nov new members of thefamily Coriobacteriaceae isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol60 no 11 pp 2639ndash2646 2010
[41] T Aullo A Ranchou-Peyruse B Ollivier and M MagotldquoDesulfotomaculum spp and related gram-positive sulfate-reducing bacteria in deep subsurface environmentsrdquo Frontiersin Microbiology vol 4 article 362 2013
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anatomy Research International
PeptidesInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation httpwwwhindawicom
International Journal of
Volume 2014
Zoology
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Molecular Biology International
GenomicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioinformaticsAdvances in
Marine BiologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Signal TransductionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Evolutionary BiologyInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Biochemistry Research International
ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Genetics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in
Virolog y
Hindawi Publishing Corporationhttpwwwhindawicom
Nucleic AcidsJournal of
Volume 2014
Stem CellsInternational
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Enzyme Research
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
International Journal of
Microbiology