Production of Recombinant ProteinsUsing Multiple-Copy Gene Integration inHigh-Cell-Density Fermentations ofRalstonia eutropha

Sriram Srinivasan, Gavin C. Barnard, Tillman U. Gerngross

Thayer School of Engineering, Dartmouth College, 8000 Cummings Hall,Hanover, New Hampshire 03755; telephone: 603-646-3161; fax:603-646-2277; e-mail: tillman.gerngross@dartmouth.edu

Received 11 February 2003; accepted 6 May 2003

Published online 25 June 2003 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bit.10756

Abstract: We have previously reported the developmentof a novel protein expression system based on Ralstoniaeutropha. In this study we report on the influence of genecopynumber on recombinant protein expression in R. eu-tropha. We compare recombinant gene stability and ex-pression levels of chromosomal integration with a plas-mid-based expression system. Single, double, and triplecopies of a gene encoding organophosphohydrolase(OPH), an enzyme prone to inclusion-body formation inE. coli, were integrated into the R. eutropha chromo-some. A linear increase between the concentration ofsoluble, active OPH and gene copynumber was found.Using a triple-copy integrant, we were able to produceapproximately 4.3 g/L of OPH in a high-cell-density fer-mentation. This represents the highest titer reported todate for this enzyme, and is approximately 30 timesgreater than expression levels reported in E. coli. © 2003Wiley Periodicals, Inc. Biotechnol Bioeng 84: 114–120, 2003.Keywords: protein expression; gene dosage; high celldensity; fermentation

INTRODUCTION

High-cell-density fermentations have been used extensivelyfor increasing the productivity and titer of recombinant pro-teins (Chen et al., 1992; Cutayar and Poillon, 1989; Knorreet al., 1991; Panda et al., 1999). Protein production in fer-mentation processes depends on several factors, includingpromoter strength (Wilms et al., 2001), induction time (Wuet al., 2001), gene dosage, and stability of expression con-struct (Grabherr et al., 2002). Various studies have beenconducted to maximize the recombinant protein yieldfor Escherichia coli-based fermentations (Chae et al.,2000; Grabherr et al., 2002; Portner and Markl, 1998).These have resulted in tremendous improvements in titerand in amount of soluble protein. However, high-cell-density fermentations of E. coli suffer from various draw-

backs, including the formation of organic acids (Akesson etal., 1999; Kleman and Strohl, 1994), proteolysis (Han et al.,1999), and inclusion-body formation (Han et al., 1999;Makrides, 1996).

To overcome the shortcomings of E. coli fermentations, anovel high-cell-density fermentation based on Ralstoniaeutropha has been developed in our laboratory (Srinivasanet al., 2002). This system has been shown to producehigh levels of soluble recombinant protein in a simple, scal-able, high-cell-density fermentation. Using this system, >1g/L of soluble recombinant organophosphohydrolase (OPH)can be produced by a single chromosomal copy of OPHunder the control of an inducible phosphate-responsive pro-moter.

In prokaryotes, high-level protein expression has beenachieved by modulating gene dosage through the use ofmedium- or high-copynumber plasmids. However, an in-crease in plasmid copynumber usually increases the meta-bolic burden on the cell. For example, a 200-copy, 5-kbplasmid represents approximately 1 Mb of DNA, constitut-ing 20% of the bacterial genome. The effect is compoundedin a high-cell-density culture and results in unstable andheterogeneous fermentation. This reduces the protein yieldand results in inclusion-body formation and proteolysis(Lee, 1996). In lower eukaryotes, especially Pichia pastoris,high levels of protein expression have been achieved bymultiple random integration of the gene of interest. Studieshave shown that an increase in gene dosage results in anincrease in recombinant protein production up to a copy-number of 15 (Werten et al., 1999).

To better understand the effect of gene dosage and itsimpact on fermentation characteristics, we constructedstrains containing single, double, and triple copies of theexpression construct integrated into the chromosome. Wealso constructed a strain containing a multicopy plasmidwith the expression construct. In this study, we report theeffect of multiple copies on protein expression and stabilityof the constructs in the genome.

Correspondence to: T. U. GerngrossContact grant sponsors: ARO; DOJ; NISTContract grant numbers: DAAD19-00-1-0180; 2000-DT-CX-K001(S-

1); 60NANB1D0064

© 2003 Wiley Periodicals, Inc.

MATERIALS AND METHODS

Strains, Plasmids, and Media

All E. coli and R. eutropha strains and plasmids used in thisstudy are described in Table I. All pKNOCK series of vec-tors were transformed in E. coli S-17, whereas other vectorswere manipulated in E. coli TOP10. E. coli strains weregrown in Luria–Bertani (LB) media. R. eutropha strainswere cultivated in tryptic soy broth (TSB) media (DifcoLaboratories, Detroit, MI) or PCT media (Srinivasan et al.,2002). Antibiotics were added to growth media to the fol-lowing final concentrations: for R. eutropha, gentamicin (10�g/mL) and chloramphenicol (50 �g/mL); and for E. coli,ampicillin (100 �g/mL), gentamicin (10 �g/mL), and chlor-amphenicol (50 �g/mL) all antibiotics were obtained fromSigma Co. (St. Louis, MO).

DNA Preparation, Manipulation, and SouthernBlot Hybridization

Preparation (Miniprep Kit, Qiagen, Inc., Valencia, CA) andmanipulation of DNA, genomic DNA preparation (Geno-

mic-tip 500, Qiagen), transformation, and hybridization(North2South direct HRP labeling and detection kit; Pierce,Rockford, IL) were carried out using standard procedures(Sambrook and Russell, 2001) or the manufacturer’s in-structions. Sequencing was performed at the Molecular Bi-ology Core Facility at Dartmouth College. PCR productswere cloned into pCR2.1-TOPO vector (Invitrogen Corp.,Carlsbad, CA) and then cut with the appropriate restrictionenzymes.

Construction of Plasmids

A list of primers used in the study is shown in Table I. A100-bp transcription terminator was amplified by PCR fromthe vector pEZSeq (Lucigen Corp., Middleton, WI) usingthe primers TERMplus and TERMminus. The construct wasthen cloned into pCR2.1-TOPO yielding pTOPO-Terminator. This vector was then cut with the restrictionenzymes BglII/PstI and cloned into the BamHI/PstI frag-ment of vector pUCPPCm, generating pUCPPT (containsthe PPT expression cassette: GA24 promoter-OPH openreading frame-transcriptional terminator).

Three separate PCRs were performed to amplify the

Table I. Strains, plasmids, and primers used in this study.

Strain or plasmid Description Reference or source

R. eutropha strainsNCIMB 40124 Wild-type NCIMBGS5 R. eutropha strain containing plasmid pKTPPCm Srinivasan et al. (2002)SS17 Single-copy integrant of the OPH expression cassette derived from NCIMB40124 and pKNOCK-PPT1 This studySS18 Double-copy integrant of the OPH expression cassette derived from NCIMB40124 and pKNOCK-PPT2 This studySS19 Triple-copy integrant of the OPH expression cassette derived from NCIMB40124 and pKNOCK-PPT3 This study

E. coli strainsTOP10 Strain for ligation/cloning of genes InvitrogenS-17 Host strain for pKNOCK series of plasmids Delorenzo et al. (1990)

PlasmidspCR2.1-TOPO High-copynumber plasmid for cloning PCR products InvitrogenpEZSeq High-copynumber plasmid with strong transcription terminators LucigenpKNOCK-Cm Plasmid for recombination conferring chloramphenicol resistance Alexeyev (1999)pKNOCK-PPT1 pKNOCK-Cm plasmid containing one copy of the OPH expression cassette (PPT) This studypKNOCK-PPT2 pKNOCK-Cm plasmid containing two copies of the OPH expression cassette (PPT) This studypKNOCK-PPT3 pKNOCK-Cm plasmid containing three copies of the OPH expression cassette (PPT) This studypKTPPCm phaPp-OPH translational fusion and CAT gene cloned into plasmid pKT230 Srinivasan et al. (2002)pTOPO-PPTa pCR2.1-TOPO plasmid containing PPT expression cassette with SacII/PstI cloning sites This studypTOPO-PPTb pCR2.1-TOPO plasmid containing PPT expression cassette with PstI/HindIII cloning sites This studypTOPO-PPTc pCR2.1-TOPO plasmid containing PPT expression cassette with HindIII/XhoI cloning sites This studypTOPO-PPT2 pTOPO-PPTc plasmid containing PPT expression cassette from pTOPO-PPTb This studypTOPO-Terminator pCR2.1-TOPO plasmid containing transcription terminator from pEZSeq This studypUC19 High copynumber plasmid used for cloning, confers ampicillin resistance New England BiolabspUCPPCm phaPp-OPH translational fusion and CAT gene cloned into pUC19 Srinivasan et al. (2002)pUCPPT pUC19 vector containing the phaPp-OPH-terminator (PPT) expression cassette This studyPrimersTERMplus 5�-GAGAGATCTTAATAAATTTAAATCATACCTGACCT-3�

TERMminus 5�-GAGCTGCAGATTTCATTATGGTGAAAGTTGGAA-3�

mulSPEplus 5�-GCCACTAGTATCGCGCAGCATGCTGTACTTGCGC-3�

mulPSTminus 5�-GCCCTGCAGATTTCATTATGGTGAAAGTTGGAAC-3�

mulPSTplus 5�-GCCCTGCAGATCGCGCAGCATGCTGTACTTGCGC-3�

mulHINDminus 5�-GCCAAGCTTATTTCATTATGGTGAAAGTTGGAAC-3�

mulHINDplus 5�-GCCAAGCTTATCGCGCAGCATGCTGTACTTGCGC-3�

mulXHOminus 5�-GCCCTCGAGATTTCATTATGGTGAAAGTTGGAAC-3�

SRINIVASAN ET AL.: PRODUCTION OF RECOMBINANT PROTEINS IN R. EUTROPHA 115

1600-bp product containing the expression cassettewith the following pairs of primers: (a) mulSPEplus andmulPSTminus; (b) mulPSTplus and mulHINDminus; and(c) mulHINDplus and mulXHOminus, using pUCPPT asthe template. The products were then cloned into pCR2.1-TOPO vector generating pTOPO-PPTa, pTOPO-PPTb, andpTOPO-PPTc. The insert from pTOPO-PPTc was cut withHindIII and cloned into the pTOPO-PPTb vector, digestedwith HindIII, and dephosphorylated with calf intestinalphosphatase (CIP), generating the pTOPO-PPT2 vector.pTOPO-PPT

2

contains two copies of the expression cassette(PPT).

The pTOPO-PPTa vector was cut with SacII/PstI and thePPT fragment was cloned into the SacII/PstI fragment ofvector pKNOCK-Cm, generating pKNOCK-PPT1. Simi-larly, TOPO-PPT2 was cut with PstI/XhoI and cloned intopKNOCK-Cm and pKNOCK-PPT1 cut with the same en-zymes, generating the vectors pKNOCK-PPT2 andpKNOCK-PPT3. Thus, three vectors, pKNOCK-PPT1,pKNOCK-PPT2, and pKNOCK-PPT3, containing one, two,and three copies of the PPT expression cassette, respec-tively, were generated. These vectors were then transformedinto R. eutropha by biparental mating as described bySrinivasan et al. (2002). Eight colonies from each transfor-mation were selected and tested for the presence of OPHactivity. For each copynumber, one positive strain was se-lected, designated SS17, SS18, or SS19, containing one,two, or three copies of the expression cassette, respectively.(All restriction enzymes and modifying enzymes were ob-tained from New England Biolabs, Beverly, MA.)

Fermentation and Enzyme Assay

Fermentations with R. eutropha strains were performed ac-cording to a previously described method (Srinivasan et al.,2002). Fed-batch fermentations were carried out in a 3-Lfermentor (Applikon, Foster City, CA) at 30°C, with a 1.5-L/min airflow, 630-rpm initial stirrer speed, and an initialworking volume of 1 L (1× PCT media with 5% [v/v] in-oculum). An Applikon programmable logic controller(ADI1030) was used for maintaining temperature at 30°Cand pH 6.8. Media pH was maintained with aqueous am-monia (28%). Dissolved oxygen concentration was main-tained at 30% by controlling the stirrer speed up to a maxi-mum speed of 1250 rpm, at which point the dissolved oxy-gen concentration dropped below 30% and became afunction of the glucose feed rate. Fermentation was carriedout in a batch mode until the initial glucose (20 g/L) wasconsumed.

Following the batch phase, a feed containing glucose,phosphoric acid, CoCl2 � 7H2O, and MgSO4 (600 g/L, 220mM, 0.15 mM, and 19 mM, respectively) was used to pro-mote linear growth until a biomass concentration of ap-proximately 100 g/L dry cell weight was achieved. Theglucose feed rate was initially maintained at 11.7 g/h and,after 18 h of feeding, was increased to 14.0 g/h. Inductionwas initiated by changing the feed solution to glucose (600

g/L) containing 15 mM CoCl2. The second glucose feedsolution contained no phosphate, which led to a gradualdepletion in the medium, which in turn induced OPH ex-pression. Off-gas was analyzed in real time for O2 and CO2

concentrations with a CO2/O2 analyzer (Model 3750, Illi-nois Instruments, Ingleside, IL).

Fermentation with the plasmid-bearing strain was per-formed with a supply of chloramphenicol (80 mg/L in theinitial media and 400 mg/L in the feed) to maintain selectionpressure during the fermentation. Plasmid stability wasmeasured by plating dilutions of the samples on Luria –Ber-tani (LB) plates with and without chloramphenicol. Theplates were incubated at 30°C for 2 days. The stability wasmeasured as the ratio of the number of colonies in the platecontaining chloramphenicol to that of the number of colo-nies in the antibiotic-free plate. Cell viability was also mea-sured by the number of colonies in the antibiotic-free plate.

OPH enzyme assays were performed according to themethod of Serdar and Gibson (1985), with modifications(Srinivasan et al., 2002).

Western Blots

Whole cell pellets were resuspended in water to a finalconcentration of 0.2 g (dry cell weight)/L. The cell suspen-sion (20 �L) was added to 20 �L of 2× sodium dodecyl-sulfate (SDS) lysis buffer (100 mM Tris-HCl [pH 6.8], 4%SDS, 0.2% bromophenol blue, 20% glycerol), and boiledfor 10 min. Samples (15 �L) were then loaded on a Tris-HCl SDS–12% polyacrylamide gel (Bio-Rad, Hercules,CA) and resolved at 100 V. The proteins were then trans-ferred from the gel to a nitrocellulose membrane (0.2 �m,Schleicher–Schuell, Keene, NH) for 1 h at 100 V/350 mA.The membrane was then blocked in a TBST solution buffer(50 mM Tris base, 188 mM sodium chloride, 0.05% Tween-20 [pH 7.5]) containing 3% bovine serum albumin (BSA;Sigma Co.) and incubated for 2 h at room temperature. Themembrane was then exposed to a crude primary rabbit anti-OPH serum (generously supplied by Dr. Janet Grimsley,Texas A&M University) diluted in the TBST buffer. Afterrepeated washing with TBST buffer, the membrane wasexposed to a secondary goat anti-rabbit antibody conjugatedto horseradish peroxidase (Pierce Co., Rockford, IL), di-luted 1:5000 in TBST buffer. After repeated washing withTBST buffer, color development was achieved in a Tris-buffered saline (TBS) buffer containing 500 mg/L diami-nobenzinidine (DAB; Pierce) and 100 �L/L of a 30% H2O2

solution (Fisher, Fair Lawn, NJ).

Quantification of Polyhydroxybutyrate inR. eutropha Cells

Polyhydroxybutyrate (PHB) was quantified by the sulfuricacid–high-performance liquid chromatography (HPLC)method of Karr et al. (1983), with modifications (York etal., 2001).

116 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 84, NO. 1, OCTOBER 5, 2003

RESULTS AND DISCUSSION

High-Cell-Density Fermentation With Strain GS5

A high-cell-density fed-batch fermentation was carried outwith the R. eutropha strain GS5 containing the plasmidpKTPPCm. This plasmid is a derivative of pKT230, whichhas been shown to have approximately ten copies per cell inR. eutropha (Jackson and Srienc, 1999). The stability of theplasmid, cell viability, and OPH expression levels weremonitored during the course of the fermentation (data notshown). During the batch phase approximately 50% of thecells lost the plasmid and only 10% of the cells containedthe plasmid at the end of the fermentation (as measured byresistance to chloramphenicol). Cell viability also droppedsignificantly during the induction phase. However, OPHexpression levels increased in the induction phase andreached a maximum specific activity of approximately 240U/mg. This represents a 50% increase of protein concentra-tion relative to a single chromosomal copy of the OPH gene.We hypothesize that, although the majority of the cells hadlost the plasmid, the subpopulation that maintained the plas-mid compensated for the loss by expressing a significantamount of OPH. Hence, single-copy gene integration doesnot saturate the inherent capability of recombinant proteinproduction in R. eutropha. To further analyze the extent oftranscriptional limitation, we constructed strains containingmultiple gene-expression constructs integrated into thechromosome.

Construction of Multiple-CopyGene-Expression Strains

Multiple gene integrations for recombinant protein produc-tion have been used in yeast to increase protein expressionlevels. However, this approach has not been widely used inprokaryotes, mostly due to the availability of multiple-copynumber plasmids. We constructed single-, double-, and

triple-copy integrations of the OPH expression cassette in R.eutropha to evaluate protein expression levels. An expres-sion cassette, PPT, was made that contained the open read-ing frame of OPH under the control of the GA24 promoterand a strong transcriptional terminator. Because the copieswere integrated in tandem, the transcriptional terminatorsequence was used to reduce the synthesis of polycistronicmessages. The expression cassette was cloned into the in-tegration vector pKNOCK-Cm, generating three separatevectors containing one, two, or three copies of the OPHgene. R. eutropha wild-type strain was then transformedwith these vectors and colonies were selected on a minimalmedia plate containing chloramphenicol and gentamicin.One colony for each copynumber (SS17, SS18, and SS19)was then selected and Southern blots were performed toconfirm the copynumber and region of integration (data notshown). These strains were then evaluated for the OPHexpression in a high-cell-density fermentation.

High-Cell-Density Fermentation WithMultiple-Copy Integrants

High-cell-density fermentations (Srinivasan et al., 2002)were performed with strains SS17, SS18, and SS19. A typi-cal fermentation involves a batch phase, a growth phase,and an induction phase. During the batch phase, the initialsupply of glucose (20 g/L) is consumed. This was followedby a fed-batch growth phase, during which a biomass con-centration of approximately 100 g/L (dry cell weight) isachieved. The PHB concentration during this phase remainsat <5% of the total biomass (Fig. 1). Induction of the OPHgene is accomplished by limiting the phosphate concentra-tion in the feed. A typical fermentation (SS19, the straincontaining three copies of PPT) is shown in Figure 1. In thisfermentation, the final concentration in biomass was 130.6g/L (107.3 g/L real biomass and 23.3 g/L PHB). This indi-cates that gene dosage does not negatively affect the fer-mentation capability of R. eutropha.

Figure 1. A typical high-cell-density fermentation of R. eutropha. (A) Dry cell weight, real biomass, PHB, O2 uptake rate (OUR), and CO2 evolutionrate (CER) profile for high-cell-density fermentation with strain SS19 (R. eutropha strain with three copies of the expression construct). The arrow indicatesthe time of induction. Symbols: (�) total dry cell weight (g/L); (�) real biomass (g/L); (�) PHB (g/L); (�) oxygen uptake rate [OUR, mmol/(L � h)];(X) carbon dioxide evolution rate [CER, mmol/(L � h)]. (B) Phosphate concentration (g/L) in the reactor and the specific activity of organophosphohy-drolase produced (U/mg). Symbols: (�) phosphate concentration; (�) specific activity.

SRINIVASAN ET AL.: PRODUCTION OF RECOMBINANT PROTEINS IN R. EUTROPHA 117

Stability of Integration Constructs

Recombination of the integration locus during the course ofthe fermentation is of concern because of the large numberof generations (approximately 25 generations). Moreover,

direct repeats of DNA loci are known to be geneticallyunstable. Recombination can be prevented by knocking outrecombination genes like recA, recE, etc. (Amundsen et al.,2000; Hou and Hill, 2002). Because transformation in R.eutropha is achieved using homologous recombination,recA-deficient strains would not be suitable for cloning. Toevaluate the recombination frequency between the multiplecopies of OPH, a Southern blot was performed on samplescollected during the initial growth phase and late inductionphase (Fig. 2). The results show that multiple copies werestably integrated and recombination during the fermentationwas absent. We suspect that multiple-copy integrations canbe used successfully in other prokaryotic expression sys-tems to achieve stable protein production in high-cell-density fermentation.

Comparison of OPH Expression

High-level expression of soluble OPH in E. coli has beenunsuccessful due to the formation of inclusion bodies (Ser-dar et al., 1989; Wu et al., 2000). We investigated the effectof gene dosage to determine if soluble OPH expression canbe saturated by multiple chromosomal integrants in R. eu-tropha. Figure 3A shows the specific activity of OPH pro-duced during the induction phases of the three fermentationswith single, double, and triple copies. In addition to enzymeactivity assays, OPH expression levels can be observed as a36-kDa band in a Western blot, with samples from beforeand after induction for the three fermentations (Fig. 3B). Alinear increase in the amount of OPH with gene dosage wasobserved, whereas basal level expression remained low(maximum specific activities of OPH of 170, 312, and 487U/mg for one-, two-, and three-copy integrations, respec-tively). After 30 h of induction (68 h after inoculation) withSS19 (three copies per cell), the specific activity of theprotein was found to be 487 U/mg. The specific activity of

Figure 2. Southern blot analysis of R. eutropha strains SS17, SS18, andSS19. Genomic DNA from strains single-copy (SS17), double-copy(SS18), and triple-copy (SS19) integrants were extracted from samplescollected in the initial phase (approximately 50 g/L biomass) and afterinduction (approximately 125 g/L biomass), and digested with HindIII. A1-kb OPH open reading frame sequence (the SacI/BamHI fragment ofplasmid pUCPPCm) was used to probe the blot. (A) Results of Southernblot analysis of genomic DNA of strains probed with a horseradish per-oxidase-labeled OPH open reading frame. Lanes 1, 3, and 5 correspond togenomic DNA before induction and lanes 2, 4, and 6 correspond to geno-mic DNA after induction, for single-, double-, and triple-copy integrants.The sizes of the markers are indicated on the left. (B) Map of the genomicregion analyzed in strains SS17, SS18, and SS19, indicating the expressionconstruct (PPT) and the HindIII restriction enzyme sites. H, HindIII. Theopen reading frame of organophosphohydrolase (OPH) is also indicated.

Figure 3. Induction profile for fermentations with strains SS17, SS18, and SS19. (A) Specific activity profiles for the strains containing one (SS17), two(SS18), and three (SS19) chromosomal copies of the OPH expression cassette, PPT. Symbols: (�) SS17; (�) SS18; (�) SS19. (B) Western blot analysisof fermentation samples. Total cell pellets were diluted to approximately 0.2 g/L DCW and then boiled on sodium dodecylsulfate–polyacrylamide gelelectrophoresis (SDS-PAGE) lysis buffer and resolved on a 12% polyacrylamide gel. The proteins were transferred to a nitrocellulose membrane andblotting was done with anti-OPH antibody. The figure shows OPH expression levels before (approximately 100 g/L dry cell weight; lanes 2, 4, and 6) andafter (approximately 130 g/L dry cell weight; lanes 3, 5, and 7) induction for fermentations with strains SS17, SS18, and SS19, respectively. M, broad-rangemolecular weight marker.

118 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 84, NO. 1, OCTOBER 5, 2003

purified OPH (with cobalt as the divalent cation) has beenreported to be 8020 U/mg, using the paraoxon enzyme ac-tivity assay (Omburo, 1992). As previously noted, the pro-tein content of R. eutropha was found to be 68% of dry cellweight (Henderson and Jones, 1997). Thus, the specific ac-tivity of 487 U/mg corresponds to 6% of total cellular pro-tein, respectively. The product of 107.3 g/L biomass, 0.68 gprotein/g biomass, and 0.06 g OPH/g protein yields an OPHconcentration of 4.3 g/L. This is approximately 30 timesgreater than the levels produced in E. coli and 3 timesgreater than the previously published data in R. eutropha(Srinivasan et al., 2002).

CONCLUSIONS

In our earlier work (Srinivasan et al., 2002), we reported thedevelopment of a simple fermentation process for producinghigh levels of recombinant proteins in R. eutropha. Weshowed that plasmids are unstable in R. eutropha, even inthe presence of antibiotic. We overcame this difficulty byintegrating multiple copies of our gene of interest into thechromosome and thereby increasing the gene dosage whileeliminating instability. In E. coli, translational fusions weremade with OPH to increase the specific activity of the ex-pressed protein and to reduce inclusion-body formation (Wuet al. 2001).

Our study with R. eutropha has shown that a linear in-crease in OPH production can be achieved by integratingmultiple copies of the gene of interest into the chromosome.Moreover, we have shown that the basal level of inductionis not increased substantially, and that multiple copies re-main stable in the chromosome. It should be noted that noinclusion bodies were seen, even with such high-level ex-pression of the gene of interest. The levels of proteinachieved with three-copy integrant (4.3 g/L) is the highestreported for this protein. We also believe that gene dosagecan be further increased by integrating more copies, but itmay increase basal expression levels.

The authors thank Dr. Mikhail Alexeyev and Dr. Janet Grimsleyfor providing us with the plasmids and the antibodies for theproject. We also thank Mary Kay Brown for continued support.

References

Akesson M, Karlsson EN, Hagander P, Axelsson JP, Tocaj A. 1999. On-line detection of acetate formation in Escherichia coli cultures usingdissolved oxygen responses to feed transients. Biotechnol Bioeng 64:590–598.

Alexeyev MF. 1999. The pKNOCK series of broad-host-range mobilizablesuicide vectors for gene knockout and targeted DNA insertion into thechromosome of gram-negative bacteria. Biotechniques 26:824–828.

Amundsen SK, Taylor AF, Smith GR. 2000. The RecD subunit of theEscherichia coli RecBCD enzyme inhibits RecA loading, homologousrecombination, and DNA repair. Proc Natl Acad Sci USA 97:7399–7404.

Chae HJ, Delisa MP, Cha HJ, Weigand WA, Rao G, Bentley WE. 2000.Framework for online optimization of recombinant protein expressionin high-cell-density Escherichia coli cultures using GFP-fusion moni-toring. Biotechnol Bioeng 69:275–285.

Chen HC, Hwang CF, Mou DG. 1992. High-density Escherichia coli cul-tivation process for hyperexpression of recombinant porcine growthhormone. Enzyme Microb Technol 14:321–326.

Cutayar JM, Poillon D. 1989. High-cell-density culture of Escherichia coliin a fed batch system with dissolved oxygen as substrate feed indica-tor. Biotechnol Lett 11:155–160.

Delorenzo V, Herrero M, Jakubzik U, Timmis KN. 1990. Mini-Tn5 trans-poson derivatives for insertion mutagenesis, promoter probing, andchromosomal insertion of cloned DNA in gram-negative eubacteria. JBacteriol 172:6568–6572.

Grabherr R, Nilsson E, Striedner G, Bayer K. 2002. Stabilizing plasmidcopy number to improve recombinant protein production. BiotechnolBioeng 77:142–147.

Han KG, Lee SS, Kang CW. 1999. Soluble expression of cloned phageK11 RNA polymerase gene in Escherichia coli at a low temperature.Prot Expr Purif 16:103–108.

Henderson RA, Jones CW. 1997. Physiology of poly-3-hydroxybutyrate(PHB) production by Alcaligenes eutrophus growing in continuousculture. Microbiology UK 143:2361–2371.

Hou R, Hill TM. 2002. Loss of RecA function affects the ability of Esch-erichia coli to maintain recombinant plasmids containing a Ter site.Plasmid 47:36–50.

Jackson JK, Srienc F. 1999. Effects of recombinant modulation of thephbCAB operon copy number on PHB synthesis rates in Ralstoniaeutropha. J Biotechnol 68:49–60.

Karr DB, Waters JK, Emerich DW. 1983. Analysis of poly-�-hydroxy-butyrate in Rhizobium japonicum bacteroids by ion exclusion highpressure liquid chromatography and UV detection. Appl Environ Mi-crobiol 46:1339–1344.

Kleman GL, Strohl WR. 1994. Acetate metabolism by Escherichia coli inhigh cell density fermentation. Appl Environ Microbiol 60:3952–3958.

Knorre WA, Deckwer WD, Korz D, Pohl HD, Riesenberg D, Ross A,Sanders E, Schulz V. 1991. High cell density fermentation of recom-binant Escherichia coli with computer-controlled optimal growth rate.Ann NY Acad Sci 646:300–306.

Lee SY. 1996. High cell-density culture of Escherichia coli. Trends Bio-technol 14:98–105.

Makrides SC. 1996. Strategies for achieving high-level expression of genesin Escherichia coli. Microbiol Rev 60:512–538.

Omburo GA, Kuo JM, Mullins LS, Raushel FM. 1992. Characterization ofthe zinc binding site of bacterial phosphotriesterase. J Biol Chem267:13278–13283.

Panda AK, Khan RH, Rao KB, Totey SM. 1999. Kinetics of inclusion bodyproduction in batch and high cell density fed-batch culture of Esch-erichia coli expressing ovine growth hormone. J Biotechnol 75:161–72.

Portner R, Markl H. 1998. Dialysis cultures. Appl Microbiol Biotechnol50:403–414.

Sambrook J, Russel D. 2001. Molecular cloning: A laboratory manual, 3rded. Cold Spring, NY: Cold Spring Harbor Laboratory Press.

Serdar CM, Gibson DT. 1985. Enzymatic hydrolysis of organophos-phates—cloning and expression of a parathion hydrolase gene fromPseudomonas diminuta. Bio/Technology 3:567–571.

Serdar CM, Murdock DC, Rohde MF. 1989. Parathion hydrolase genefrom Pseudomonas diminuta MG—subcloning, complete nucleotidesequence, and expression of the mature portion of the enzyme inEscherichia coli. Bio/Technology 7:1151–1155.

Srinivasan S, Barnard GC, Gerngross TU. 2002. A novel high cell densityprotein expression system based on Ralstonia eutropha. Appl EnvironMicrobiol 68:5925–5932.

Werten MW, van den Bosch TJ, Wind RD, Mooibroek H, de Wolf FA.

SRINIVASAN ET AL.: PRODUCTION OF RECOMBINANT PROTEINS IN R. EUTROPHA 119

1999. High yield secretion of recombinant gelatins by Pichia pastoris.Yeast 15:1087–1096.

Wilms B, Hauck A, Reuss M, Syldatk C, Mattes R, Siemann M, Alten-buchner J. 2001. High cell density fermentation for production ofL-N-carbamoylase using an expression system based on the Esch-erichia coli rhaBAD promoter. Biotechnol Bioeng 73:95–103.

Wu CF, Cha HJ, Rao G, Valdes JJ, Bentley WE. 2000. A green fluorescentprotein fusion strategy for monitoring the expression, cellular location,and separation of biologically active organophosphorus hydrolase.Appl Microbiol Biotechnol 54:78–83.

Wu CF, Valdes JJ, Bentley WE. 2001. Effects of in situ cobalt ion additionon the activity of a gfp-oph fusion protein: The fermentation kinetics.Biotechnol Progr 17:606–611.

Wu CF, Valdes JJ, Rao G, Bentley WE. 2001. Enhancement of organo-phosphorus hydrolase yield in Escherichia coli using multiple genefusions. Biotechnol Bioeng 75:100–103.

York GM, Junker BH, Stubbe JA, Sinskey AJ. 2001. Accumulation of thePhaP phasin of Ralstonia eutropha is dependent on production ofpolyhydroxybutyrate in cells. J Bacteriol 183:4217–4226.

120 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 84, NO. 1, OCTOBER 5, 2003