International Journal of Food Microbiology 131 (2009) 218–223

Contents lists available at ScienceDirect

International Journal of Food Microbiology

j ourna l homepage: www.e lsev ie r.com/ locate / i j foodmicro

Factors affecting the efficacy of pressure inactivation of Escherichia coli O157:H7 onalfalfa seeds and seed viability

Hudaa Neetoo, Mu Ye, Haiqiang Chen ⁎Department of Animal and Food Sciences, University of Delaware, Newark, DE 19716-2150, USA

⁎ Corresponding author. Tel.: +1 302 831 1045; fax: +E-mail address: haiqiang@udel.edu (H. Chen).

0168-1605/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.ijfoodmicro.2009.02.028

a b s t r a c t

a r t i c l e i n f o

Article history:Received 7 December 2008Received in revised form 19 February 2009Accepted 28 February 2009

Keywords:PressureAlfalfaSeedsEscherichia coli O157:H7Germination

The application of high hydrostatic pressure technology as a seed decontamination technology wasevaluated. Alfalfa seeds inoculated with approximately 105 CFU/g of Escherichia coli O157:H7 were subjectedto oscillatory pressure treatments at 600 MPa and 20 °C for up to five cycles with a holding time of 2 min/cycle. However, oscillatory pressurization was not able to eliminate E. coli O157:H7. The application ofpressure treatment at 600 MPa for 2 min at 20 °C in the presence of chemicals such as calcium hypochlorite,calcium hydroxide, lactic acid or sodium acid sulfate was subsequently investigated and it was demonstratedthat this “multiple hurdle” approach was unable to decontaminate alfalfa seeds. Soaking seeds prior topressure treatment was found to play a critical role on enhancing the pressure inactivation of E. coli O157:H7;seeds soaked in water for 60 min followed by treatment at 600 MPa for 2 min at 20 °C were decontaminatedand had a germination rate of 91% which was 4% lower than that of the untreated seeds (not statisticallysignificant, P>0.05). It was further demonstrated that a process involving soaking seeds in water for≥10 min followed by treatment at 600 MPa for 15 min at 20 °C was equally effective with respect to E. coliO157:H7 elimination and viability retention of the seeds.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Numerous outbreaks of food-borne diseases linked to consump-tion of fresh fruits and vegetables clearly point to the fact that thesecommodities constitute a serious hazard to public health. Seed sproutsare regarded by consumers as a health food because they are low in fatand calories, high in fiber and are rich in anti-cholesterolemic andanti-carcinogenic constituents (Kurtzweil, 1999). However, sproutedseeds continue to be implicated in outbreaks of food-borne illness andhence are considered a significant food safety risk (Anonymous, 1999;Kumar et al., 2006). Since 1995, there have been three outbreaksassociated with the consumption of alfalfa and clover sproutscontaminated with E. coli O157 in the United States (Fett, 2002;Ferguson et al., 2005). In the majority of the illness outbreaks, theseeds used for sprout production were proven to be the most likelysource of human pathogens (Anonymous, 1999).

Alfalfa sprouts are typically eaten raw and therefore interventiontreatments must be either applied to the seeds or to the germinatedsprouts. Since sprout-directed treatments could be more problematicbecause of the delicate nature of the sprouts, more research attentionhas been focused on alfalfa seed decontamination as opposed totreatmenton alfalfa sprouts (Langet al., 2000). In response to the urgentneed to curtail sprout-related outbreaks and to protect public safety, the

1 302 831 2822.

ll rights reserved.

U.S. Food and Drug Administration and the State of California arestrongly recommending the treatment of alfalfa seedswith a solution of20,000 ppm free chlorine from calcium hypochlorite. This treatment iscurrently considered the “benchmark” and is also recommended by theNational Advisory Committee on Microbiological Criteria for Foods(Anonymous, 1999). However, it has been shown that this method maynot be robust enough for seed decontamination. It reduces, but does noteliminate pathogens on seeds and consequently outbreaks continue tooccur (Brooks et al., 2001; Proctor et al., 2001). As a result, extensiveresearch is currently ongoing to investigate the ability of variousphysical, chemical andbiological preventive strategies to decontaminateseeds from foodborne pathogens.

High hydrostatic pressure (HHP) processing is a non-thermal foodpreservation technology. One of the most important effects of HHPprocessing is its ability to destroy or reduces the number of foodbornepathogens with minimum impact on food sensory and functionalqualities. Previous researchers investigating the application of HHP(Ariefdjohan et al., 2004; Wuytack et al., 2003) to decontaminatealfalfa and mung bean seeds have reported varying degrees of successdue to the inability of high pressure to eliminate pathogenicmicroorganisms from seeds with minimal impact on the seedsgerminability. Since these seeds are destined for sprouting, retentionof their seed viability is critical to ensure that pressure-treated seedscan consistently meet the yield anticipated by sprouters. A previousstudy in our laboratory demonstrated that continuous high pressuretreatment at 600 MPa for 20 min at 20 °C achieved a ~5 log reductionbut could not eliminate E. coli O157:H7 (105 CFU/g) on alfalfa seeds

219H. Neetoo et al. / International Journal of Food Microbiology 131 (2009) 218–223

(Neetoo et al., 2008). Therefore the overall aim of this study was todevelop a HHP process to decontaminate alfalfa seeds inoculated witha ~5 log CFU/g of E. coli O157:H7 with minimal adverse impact on theseeds germination potential. The specific objectives were to determinethe effect of (1) oscillatory HHP treatments, (2) the combined effect ofHHP with antimicrobial agents and (3) the application of HHP on pre-soaked seeds for varying holding times on seed decontamination andviability.

2. Materials and methods

2.1. Effect of oscillatory pressurization on the inactivation of E. coli O157:H7 on alfalfa seeds

2.1.1. Bacterial strainsE. coli O157:H7 strains 1730, 250, 251, Cider and J58, adapted to

grow in tryptic soy broth plus 0.6% yeast extract (Difco Laboratories,Sparks, MD, USA) supplemented with 50 µg/ml of nalidixic acid(Fisher Scientific, Hampton, NH, USA) (TSBYE-N) were used. The five-strain cocktail of E. coli O157:H7 was prepared as described by Neetooet al. (2008).

2.1.2. Inoculation of seedsThe cocktail (10 ml) was mixed with 100 ml of sterile 0.1% pep-

tone water (Fisher). Unscarified and undamaged alfalfa seeds (100 g),obtained from International Specialty Supply (Cookeville, Tenn.,USA), were added to the cell suspension (21±1 °C) and gentlystirred for 5 min. The seeds were separated from the cell suspensionby pouring the mixture over a double layer of cheesecloth supportedby a wire screen and dried inside a bio-safety hood at roomtemperature (21±1 °C) for 24 h to a final aw of 0.622. Dried seedswith an approximate inoculation level of 105 CFU/g of E. coli O157:H7were placed in sterile pouches and stored at 4 °C until needed.

2.1.3. Pressure treatmentTwo g of inoculated seeds and 3 ml of sterile DI water were placed

in a 3-mil thick pouch (Nylon/Polyethylene, Koch Supplies, KansasCity, MO). To avoid leakage during pressure-treatment, each samplepouch was placed in a larger pouch of an 8-mil thick PVC plastic(McMaster–Carr, Elmhurst, IL) and heat-sealed. HHP treatment ofsamples was carried out using a high-pressure unit with temperaturecontrol (Model Avure PT-1, Avure Technologies, Kent, WA). All theexperiments were conducted at 20 °C (initial seed sample tempera-ture prior to pressure treatment) using water as a hydrostaticmedium. The temperature of the water-bath surrounding the pressurechamber during pressurization was monitored through K-typethermocouples. The temperature and pressure data were recordedevery 2 s (DASYTEC USA, Bedford, NH). The pressure-come-up ratewas approximately 22 MPa/s. The pressure-release was <4 s.Pressurization time reported in this study did not include the pressurecome-up or release times. Oscillatory pressurization was performedby alternating atmospheric pressure (0.1 MPa) and high pressure(600 MPa) with 1, 2, 3, 4 or 5 cycles and a holding time of 2 min/cycle.

2.1.4. Microbiological analysisPouches containing treated seeds were cut open aseptically. The

seedmixture consisting of the seeds and immersionwater was pouredinto a stomacher bag to which 8 ml of sterile 0.1% peptone water wasadded and subsequently stomached for 2min at 260 rpm (Seward 400Stomacher, Seward Medical Co., London, U.K.). The seed slurry wasserially diluted in sterile 0.1% peptone and surface-plated in duplicateon tryptic soy agar with 0.6% yeast extract (Difco Laboratories, Sparks,MD, USA) supplemented with nalidixic acid to a final concentration of50 µg/ml (TSAYE-N). TSAYE-N plates were incubated at 35 °C for3 days. Presumptive colonies of E. coli O157:H7 formed on the plateswere enumerated. One or two colonies were picked at an initial stage

in the study and confirmed for E. coliO157:H7 identification using eithera BAX® system for screening E. coli O157:H7 PCR assay (Qualicon-DuPont, Wilmington, DE, USA) or Rapid E. coli O157:H7 Test Methods(Strategic Diagnostics Inc., Newark, DE, USA).

2.2. Combined effect of HHP and antimicrobial treatments on inactivationof E. coli O157:H7 on alfalfa seeds

The combined application of HHP and antimicrobial treatments onseed decontamination was investigated. Inoculated seeds were either(a) soaked in antimicrobial solutions for varying durations and thensubjected to a 2 min-pressure treatment or (b) directly pressure-treated in antimicrobial solutions (without prior soaking) for extendedexposure times. In the first approach, 2 g of inoculated seeds weresoaked in 20 ml of DI-water (control), calcium hypochlorite (AcrosOrganics, N.J.) solutions providing approximately 200, 1000, 2000, or20,000 ppm of free chlorine as determined using chlorine test strips(Fisher), 1% calcium hydroxide (Reheis Inc., Midlothian, TX) with orwithout 1% Tween 80 (Fisher), lactic acid (1, 2 or 5%) (Fisher), orsodium acid sulfate (0.05, 0.10 or 0.20%) (Jones-Hamilton, Walbridge,OH) for 5, 10 or 15 min at room temperature (21±1 °C). The con-centration unit was defined as %w/v. The calcium hypochloritesolutions were prepared in a sterile 0.05 M potassium phosphatebuffer solution with pH of 6.8. After soaking, the excess water orchemical solutions was decanted and the seeds were placed into asterile plastic pouch followed by the addition of 3 ml of the respectivetreatment solutions and treated at 600 MPa for 2 min at 20 °C. In thesecond approach, 2 g of the inoculated seeds were placed into sterilepouches and pressure-treated at 600MPa and 20 °Cwhilst submergedin 3 ml of various antimicrobial solutions for varying holding times (5,10 or 15 min). The antimicrobial solutions included 2000 and20,000 ppm calcium hypochlorite, 1% calcium hydroxide with orwithout 1% Tween 80 and 1, 2 and 5% lactic acid. Pouches containingtreated seedswere cut open aseptically. Seed sampleswere rinsedwith100 ml of sterile DI-water to remove any residues of the chemicalsolutions and thenpoured into a stomacher bag towhich 8ml of sterile0.1% peptonewater was added and subsequently stomached for 2min.The seed slurry was directly enriched in 90 ml of TSBYE-N andincubated for 48 h at 35 °C to allow resuscitation of sub-lethally injuredcells. Samples were streaked onto Sorbitol MacConkey agar (DifcoLaboratories, Sparks, MD, U.S.A.) plates supplemented with 50 µg/mlof nalidixic acid. After 24 h incubation, presence of growth exhibitingmorphological and biochemical characteristics typical of E. coli O157:H7 were determined by visually inspecting the plates. Colonies wereconfirmed to be E. coli O157:H7 using Rapid E. coli O157:H7 TestMethods (Strategic Diagnostics Inc., Newark, DE, USA).

2.3. Effect of soaking time in water prior to pressure treatment oninactivation of E. coli O157:H7 on alfalfa seeds and seed viability

Two g of inoculated seeds was soaked in sterile DI-water for 0(without soaking), 30, 60, 90, 120, 180, 240 or 300 min prior to beingtreated at 600 MPa for 2 min at 20 °C. Seeds were soaked either in alimiting volume (3 ml) or in a large excess (20 ml) of water todetermine whether the soaking time as well as the volume of wateravailable were critical factors in the pressure inactivation of E. coli O157:H7 on seeds. For seeds soaked in 3 ml of water, the 2 g inoculated seedsand 3mlwaterwere directly placed into a pouch and soaked for varyingdurations before being subjected to the pressure treatment. For seedssoaked in 20 ml of water, the excess water was subsequently decantedand seeds were placed into a pouch in the presence of 3 ml of freshsterile DI water and pressure-treated. Samples were then microbiolo-gically assayed post-treatment as described previously. Soaked controlswere additionally set up by immersing inoculated seeds in 20 ml ofsterile DI water for 30, 60, 90, 120, 180, 240 or 300 min beforedetermining the remaining populations of E. coli O157:H7 on the seeds.

Fig. 1. Effect of oscillatory pressure treatment on inactivation of E. coli O157:H7inoculated on alfalfa seeds at a level of 1.4×105 CFU/g. Initial sample temperature was20 °C. Pressure cycled between atmospheric pressure and 600 MPa (2-min hold time/cycle at 600 MPa). Data are the means of three replicates. Error bars represent ±1standard deviation.

220 H. Neetoo et al. / International Journal of Food Microbiology 131 (2009) 218–223

To determine the effect of soaking duration prior to pressuretreatment on the seeds germination potential, 2 g of un-inoculatedseeds were soaked in 20 ml of water for 0, 30, 60, 90, 120 or 180 min.Thewater was then decanted and the seeds introduced into a pouch towhich 3 ml of DI-water was added. The samples were treated at600 MPa for 2 min at 20 °C. After HHP, pressure-treated seeds anduntreated seeds (control) were soaked in DI-water for a total soakingtime of 3 h (including the soaking times before and after the pressuretreatment). One hundred seeds were drawn from the soaked seedsand spread evenly on pieces of wet paper towels on a plastic rack,which in turn was placed into a water-filled bucket to provide a moistenvironment for the seeds. The water level was maintained below theseeds' level. The bucket was kept at room temperature (21±1 °C) for8 days (as suggested by the seeds provider) and misted daily. Thebucket was covered loosely with a piece of plastic film to allowexchange of air between the inside and outside of the bucket. The seedswere visually evaluated for sprouting 3–8 days after setting up thegermination system.

To determine the rate of water absorption by seeds, 2 g of un-inoculated seedswere immersed in 20mlofwater at room temperature.Soaked seeds were weighed and their weight gain recorded at regulartime intervals for up to 19 h.

2.4. Effect of reduced soaking time and extended pressure holding timeon the inactivation of E. coli O157:H7 and seed viability

Two grams of inoculated seeds were soaked in 20ml of water for 10or 15 min. The water was later decanted and seeds mixed with 3 ml ofsterile DI-water, packaged and treated at 600 MPa and 20 °C for 5, 10 or15 min. Samples were then microbiologically assayed as describedpreviously and enriched for the detection of survivors. For thedetermination of seed germination, 2 g of un-inoculated seeds weresoaked for 10 or 15min and treated at 600 MPa for 15min at 20 °C. Onehundred seeds were drawn from the samples and assayed for ger-mination as described previously.

2.5. Statistical analysis

All experiments were replicated at least three times. Whereappropriate, statistical analyses were conducted using Minitab®

Release 15 (Minitab Inc., University Park, PA, USA). One-way analysisof variance (ANOVA) and Tukey's one-way multiple comparisonswere used to determine differences in the populations of E. coli O157:H7 recovered on treated alfalfa seeds as well as differences in thegermination percentage of seeds. Significant differences were con-sidered at the 95% confidence level (P<0.05).

3. Results and discussion

3.1. Effect of oscillatory pressurization on the inactivation of E. coli O157:H7 on alfalfa seeds

Oscillatory pressure treatment was investigated to determinewhether it could be used to enhance pressure inactivation sincecontinuous high pressure treatment at 600 MPa for 20 min at 20 °Ccould not eliminate E. coli O157:H7 (105 CFU/g) on alfalfa seeds(Neetoo et al., 2008). Results for the oscillatory HHP treatments aresummarized in Fig. 1. The degree of pressure inactivationwas found tovary as a function of the number of cycles applied, achieving amaximum reduction of 3.7 logs after 5 cycles for a total holding timeof 10min at 600MPa. Contrary to findings garnered by other research-ers, oscillatory pressure treatments did not confer a significantadvantage in the inactivation of E. coli O157:H7. Palou et al. (1998a,b)evaluated the inactivation of Zygosaccharomyces bailii by oscillatorypressure and continuous pressure treatments and found that oscilla-tory pressure treatments increased pressure inactivation. Aleman et al.

(1996) also reported greater reductions of Saccharomyces cerevisiaewhen oscillatory pressure treatments were applied. Enhancement inoscillatory pressure inactivation has also been reported for spores(Hayakawa et al., 1998). It is thought that the effectiveness ofoscillatory HHP treatments can be attributed to spore burst whichcan be promoted by increased spore wall permeability at highpressures (Palou et al., 1998a,b). However, our results are comparableto those reported by Kingsley et al. (2006) for the pressure inactivationof hepatitis A virus (HAV). Oscillatory high pressure processing for2, 4, 6, and 8 cycles at 400 MPa did not considerably enhance theinactivation of HAV as compared with continuous high pressureapplication.

3.2. Combined effect of HHP and antimicrobial treatments on inactivationof E. coli O157:H7 on alfalfa seeds

Since oscillatory pressurization with as many as 5 cycles could noteliminate 105 CFU/g of E. coli O157:H7, a “multiple hurdle approach”was investigated whereby the simultaneous application of antimicro-bials and HHP were evaluated on inoculated alfalfa seeds. Previousresearch has shown that washing or soaking in antimicrobial agentscan reduce the microbial load in seeds (Jaquette et al., 1996; Piernasand Guiraud, 1997). The drawback associated with these chemicaltreatments however is that high antimicrobial concentrations areusually required to demonstrate any appreciable effect and thetreatments cannot eliminate pathogens. In this experiment, theconcerted application of HHP and chemical treatments was investi-gated. Calcium hydroxide was included in this study as an alternativeto chlorinated sanitizers while lactic acid and sodium acid sulfatewerestudied because of their acidulating property, which was thought toenhance pressure inactivation of bacterial pathogens. Since we wereinterested in determining whether the application of HHP inconjunction with chemical treatments would be able to achieve100% lethality or not, samples were analyzed qualitatively forpresence/absence of survivors after enrichment in non-selectivemedia. Soaking seeds in the different chemical solutions at low,medium or high concentrations prior to pressure treatment did notachieve 100% kill of the pathogen as survivors were consistentlydetected across all treatments in all three trials (Table 1). Since the 2-min pressure holding time did not bring about elimination of E. coliO157:H7, seeds were then pressure-treated with selected antimicro-bials consisting of calcium hypochlorite, calcium hydroxide or lacticacid with an extended pressure holding time of up to 15 min.Regardless of the nature or the level of antimicrobials used, survivorswere still detected post-enrichment (Table 2).

Table 1Effect of HHP (600 MPa for 2 min at 20 °C) in conjunction with antibacterial agents oninactivation of E. coli O157:H7 on alfalfa seeds inoculated at a level of 2.0×105 CFU/g.

Fluid Concentration Soaking time (min)

0 5 10 15

Water 0 3/3 3/3 3/3 3/3Ca(OCL)2 200 ppm 3/3 3/3 3/3 3/3Ca(OCL)2 1000 ppm 3/3 3/3 3/3 3/3Ca(OCL)2 2000 ppm 3/3 3/3 3/3 3/3Ca(OCL)2 20,000 ppm 3/3 3/3 3/3 3/3Ca (OH)2 1% 3/3 3/3 3/3 3/3Ca (OH)2 1%+1% Tween 80 3/3 3/3 3/3 3/3Lactic acid 1% 3/3 3/3 3/3 3/3Lactic acid 2% 3/3 3/3 3/3 3/3Lactic acid 5% 3/3 3/3 3/3 3/3Sodium acid sulfate 0.05% 3/3 3/3 3/3 3/3Sodium acid sulfate 0.10% 3/3 3/3 3/3 3/3Sodium acid sulfate 0.20% 3/3 3/3 3/3 3/3

Numbers separated by a “/” represent the number of samples testing positive afterenrichment out of a total of three trials.

221H. Neetoo et al. / International Journal of Food Microbiology 131 (2009) 218–223

It is believed that a small population of E. coli O157:H7 wasparticularly resistant to pressure treatments. Indeed, other authorshave shown that vegetative bacteria such as E. coli O157:H7 can bevery refractory to high pressure with pronounced “tails” on survivalcurves (Chen, 2007). These pressure-resistant cells might be sub-lethally injured during a pressure treatment and with subsequentappropriate conditions conducive to growth, these cells can recover(Earnshaw et al., 1995). Hence sub-lethal damage and subsequentrecovery can present a significant problem for alfalfa sprouts that areeaten raw given that these recalcitrant pathogenic cells can multiplyduring sprouting.

Past research by other authors has demonstrated that chemicalsanitizers as stand-alone treatments are not effective at eliminatingE. coli O157:H7 from alfalfa seeds (Beuchat and Scouten, 2002;Taormina and Beuchat, 1999; Weissinger and Beuchat, 2000). It wasthus hypothesized in this study that the simultaneous application ofantibacterial agents and HHP could provide a good opportunity for thechemicals to capitalize on the widespread cellular and biochemicaldamages induced by pressure treatment. The fact that pressurizationof the seeds in the presence of very high concentration of thesecompounds could not eliminate E. coli O157:H7 lends strong credenceto the explanation that these pathogens are lodged in deep cracks orcrevices on the seed coat or even internalized. As a result, thesechemicals cannot permeate the seed coat to target those internalizedcells within the time-frame investigated in our study.

Microscopy of mung bean seeds has shown that although the seedsurface is relatively smooth, the stem scar is porous, allowing bacteriato penetrate into the seed. Microscopic examination of alfalfa seedshas revealed similarities in the seed coat, both being smooth but withareas capable of harboring pathogens, thus protecting them from

Table 2Effect of extended HHP holding times (600 MPa and 20 °C) in conjunction withantibacterial agents on inactivation of E. coli O157:H7 on alfalfa seeds inoculated at alevel of 1.2×105 CFU/g.

Fluid Concentration Pressure holding time (min)

5 10 15

Water 0 3/3 3/3 3/3Ca(OCL)2 2000 ppm 3/3 3/3 3/3Ca(OCL)2 20000 ppm 3/3 3/3 3/3Ca (OH)2 1% 3/3 3/3 3/3Ca (OH)2 1%+1% Tween 80 3/3 3/3 3/3Lactic acid 1% 3/3 3/3 3/3Lactic acid 2% 3/3 3/3 3/3Lactic acid 5% 3/3 3/3 3/3

Numbers separated by a “/” represent the number of samples testing positive afterenrichment out of a total of three trials.

aqueous sanitizers (Delaquis et al., 1999). In addition to its topo-graphic complexity, the surface of alfalfa seeds is known to be coveredwith a waxy cuticle (cutin) and it is unlikely that aqueous solutionscan effectively wet these surfaces. Given the hydrophobic nature ofalfalfa seeds, we hypothesized that addition of a surfactant such asTween 80 as an adjunct to 1% calcium hydroxide could enhancepermeation of the surfactant or the sanitizer into the seeds. However,contrary to findings reported by Weissinger and Beuchat (2000), wefound that the application of surfactant during pressure treatment didnot enhance the antimicrobial activity of calcium hydroxide and wasnot effective in achieving complete lethality.

3.2.1. Effect of soaking time and immersion volume in water prior topressure treatment on inactivation of E. coli O157:H7 on alfalfa seeds

Since the combined application of antimicrobial treatments withHHP was unsuccessful in achieving elimination of E. coli O157:H7,modification of the HHP strategy to decontaminate seeds was thensought in an attempt to enhance the microbial safety of alfalfa seeds.Seeds soaked in sterile DI water for up to 300 min (soaked controls)did not undergo any significant reduction in the counts of E. coli O157:H7 with final populations ranging from 4.7–5.3 log CFU/g. Table 3shows that treatment at 600MPawithout prior soaking brought abouta 2.7-log reduction. Pre-soaking of the seeds in water was accom-panied by a greater extent of inactivation; in fact a direct relationshipbetween the degree of inactivation and soaking time were observed.When seeds were soaked in a limited volume of 3 ml of water,pressure inactivation of the vegetative cells was strongly dependenton the soaking time. This was likely due to the fact that longer soakingtimes allowed water to permeate deeper into the cracks and crevicesof seeds raising their local water activity and hence enhancingpressure inactivation of cells trapped in these spaces. Seeds soaked for120–180 min had no detectable counts although survivors were stilldetected post-enrichment. However, soaking for ≥240 min followedby pressure treatment achieved elimination of the 5 log CFU/g initialload. Seeds were also soaked in a large excess of water (20 ml) suchthat the volume of water for soaking would not be a limiting factor.The same trend was observed i.e., the longer the soaking time, thegreater the degree of pressure inactivation. After the seeds weresoaked for ≥60 min followed by pressure treatment, elimination ofE. coli O157:H7 was consistently achieved. Hence it can be concludedthat both the soaking time and water volume are critical factors inensuring the efficacy of HHP to decontaminate alfalfa seeds.

In the light of these results, it can be inferred that in the presence ofexcess soaking water, there was uninhibited penetration of water intothe seeds such that an hour of imbibitionwas adequate to allowwaterto access the deepest cracks or crevices of the seed coat. In thepresence of limiting volume of water, water uptake was slower;delaying the consequentmovement of water into the seed coat spaces.The results from this soaking study reinforce the findings reported inSection 3.2 suggesting that the poor efficacy of HHPwith antimicrobialagents might be attributed to the sub-surface location of the pathogenon the seed. It is thought the pre-soaking step is critical in raising the

Table 3Effect of soaking time prior to treatment at 600MPa for 2 min at 20 °C on inactivation ofE. coli O157:H7 on alfalfa seeds inoculated at a level of 2.5×105 CFU/g.

Water(ml)

Soaking time prior to HHP (min)

0 30 60 90 120 180 240 300

3 2.7±0.8a 1.5±0.1b 1.1±0.4b 1.4±0.1b 1/3 1/3 0/3 0/320 2.7±0.8 1/3 0/3 0/3 0/3 0/3 0/3 0/3

Data are the means of log survivors (CFU/g)±1 S.D. Numbers separated by a “/”represent the number of samples testing positive after enrichment out of a total of threetrials (The counts were below the detection limit by direct plating (0.8 log CFU/g) in thethree trials).Data in the same row followed by the same superscripted letter are not significantlydifferent (P>0.05).

Fig. 2. The % fresh weight gain of imbibing alfalfa seeds. “% fresh weight gain” wascalculated as the difference between the final and the initial weight at defined timeintervals as a percentage of the initial weight. Data are the means of three replicates.Error bars represent ±1 standard deviation.

222 H. Neetoo et al. / International Journal of Food Microbiology 131 (2009) 218–223

water activity of the seeds which is known to have a bearing on thepressure inactivation of vegetative bacterial cells. It is well documen-ted that low water activity levels protect microorganisms againstpressure treatment (Oxen and Knorr, 1993; Palou et al., 1997; Kingsleyand Chen, 2008).

3.2.2. Effect of soaking time in water prior to pressure treatment on theseed viability

To determine the effect of pre-soaking time on seed viability, thegermination rate of seeds soaked for varying durations before HHPwas determined (Table 4). Soaking for ≤60 min followed by HHP stillallowed seeds to retain their viability to a large extent. The process ofsoaking seeds in 20 ml of water for 60 min followed by pressuretreatment for 2 min achieved a final germination yield of 91% whichwas 4% lower than the control. This difference in seed germinationwas not statistically significant (P>0.05). However with prolongedsoaking of >60 min, the effect of pressure treatment on the seeds'germinability became more severe. It is not clear what mechanismsare responsible for the progressively lower germination yields but it ispossible that denaturation of molecules such as enzymes might haveoccurred during HHP following their activation during prolongedwater imbibition. It is also highly likely that during soaking, seedsimbibe water and as a result of the increased moisture content, theseeds advance into an active physiological state whereby metabolicchanges associated with the initial stages of germination start to takeplace. The effect of high pressure on the delicate early germinativestage of the seeds most probably evoked negative changes in themicrostructure of the seeds. Blaszczak et al. (2007) compared themicro-structural and biochemical changes undergone in raw andgerminated chickpea seeds after pressure treatment. They demon-strated the occurrence of more pronounced damage to the structuralintegrity (physical damage) and protoplasmic contents of cells(biochemical damage) after germinated chickpea seeds were pres-sure-treated compared to raw chickpea seeds.

In order to understand the developmental changes undergone inseeds during soaking, the % fresh weight gain profile of alfalfa seedswas determined as shown in Fig. 2. Seeds are known to be quiescentand can be stored for months without harm, but once supplied withwater, they become hydrated again and embark on a different stage ofactivity. This stage results in outgrowth of the root and later the shootat the expense of the reserve materials. The imbibition profile wasthus studied in order to determine the kinetics of this new pattern ofdevelopment and to understand how HHP possibly interfered withthis developmental process. The imbibition curvewas found to exhibita biphasic pattern of water uptake characterized by a rapid increase infresh weight during the first 3 h of imbibition (Phase I) followed by amuch slower rate of water uptake lasting for approximately 15 h(Phase II). Dry seeds by virtue of their low water potentials set up avery high water potential gradient as soon as they become in contactwith water resulting in imbibition. This helps to explain the rapiduptake of water shown in Fig. 2 which occurred at a fairly uniform rate

Table 4Effect of soaking time prior to treatment at 600 MPa for 2 min at 20 °C on thegermination rate of alfalfa seeds.

Days ofgermination

control Soaking time prior to HHP (min)

0 30 60 90 120 180

3 87±4a 80±5ac 75±6ad 80±6ae 71±7a 60±12bcde 40±8b

4 89±3a 87±6a 79±7a 85±6a 75±8a 67±11ac 47±10bc

5 91±4a 91±7a 81±7a 88±6a 78±8a 73±10ac 52±11bc

6 93±3a 93±7a 82±7a 89±6a 80±9a 76±11ac 55±1bc

7 95±3a 95±6a 84±7a 90±6a 81±9a 79±11ac 58±13bc

8 95±3a 95±6a 86±7ac 91±5a 84±9ad 82±13ae 62±13bcde

Data are the means of % germination ±1 S.D.Data for the same day of germination followed by the same superscripted letter are notsignificantly different (P>0.05).

within the first 3 h. Imbibition then slackened off leading to a slowerrate of water uptake and plateaued out until no further fresh weightgain could be observed. This point was coincided by the onset ofsprouting marked by the protrusion of the radicle from the seed,which occurred 18 h after soaking.

As mentioned previously, we observed that an hour of soakingbefore HHP treatment was optimal and soaking beyond an hourbrought about a gradual decrease in the seeds viability after pressuretreatment. Reconciling the germination rate data shown in Table 4with the imbibition profile of Fig. 2, we attribute this critical time-dependence to the fact that imbibing seeds may not swell uniformlythroughout their tissues. In pea seeds for example, it has been shownthat the testa is completely wetted at a relatively early stage ofimbibition (Houben, 1966). It is likely that during the first hour ofsoaking, the testa swells to the extent of leaving a space between theseed coat and the embryo. This space, probably water-filled, acts as aprotective cushion for the seed embryo during HHP. More prolongedsoaking however may cause water to permeate all the way into theembryo which also swells, eventually occupying the whole volumeinside the testa causing the embryo to stretch the slightly elastic testa.This has been demonstrated to happen in pea seeds after 4–5 h ofsoaking (Houben, 1966). Hence controlling the length of soaking ishighly critical in order to ensure complete decontamination of seedswhile maintaining a high germination yield.

3.3. Effect of reduced soaking time and extended pressure holding time at600 MPa on inactivation of E. coli O157: H7 and on the germination rateof seeds

Since previous findings showed that the process of soaking seedsfor an hour followed by HHP treatment was able to eliminate path-ogens at a slight expense on the germination rate, we thus inves-tigated whether seeds could be decontaminated by shortening thesoaking time whilst extending the pressure holding time in order to

Table 5Effect of reduced soaking time and extended pressure holding time on inactivation ofE. coli O157: H7 on alfalfa seeds inoculated at a level of 2.5×105 CFU/g.

Holding time at600 MPa (min)

Soaking time prior to HHP (min)

10 15

5 3/3 2/310 3/3 3/315 0/3 0/3

Numbers represent the number of samples testing positive after enrichment out of atotal of three trials.The counts were below the detection limit by direct plating (0.8 log CFU/g) throughout.

Fig. 3. Effect of reduced soaking time followed by treatment at 600 MPa for 15 min at20 °C on the germination rate of un-inoculated seeds. Data are the means of threereplicates. Error bars represent ±1 standard deviation.

223H. Neetoo et al. / International Journal of Food Microbiology 131 (2009) 218–223

alleviate the structural damage of HHP on pre-soaked seeds. Resultsfor the inactivation and germination experiments with seeds soakedand pressure-treated for various time combinations are shown inTable 5 and Fig. 3, respectively. Elimination was consistently achievedwhen the soaking time was 10 or 15 min with a pressure exposuretime of 15 min. In addition, un-inoculated seeds treated under theseconditions still retained their viability to a large extent (Fig. 3). Theprocess of soaking seeds for 15 min followed by pressure treatment of15 min achieved a final germination yield of 89% which was 4% lowerthan the control. Differences in seed germination rate between thecontrol and either treatment on the same day of germinationwere notstatistically significant (P>0.05).

4. Conclusions

Results of this study demonstrate that the application of oscillatorypressurization produced little enhancement in pressure inactivationof E. coli O157:H7 on alfalfa seeds. Although the degree of pressureinactivation increased as a function of the number of cycles, oscillatoryHHP treatment could not achieve a >5 log reduction even after five 2-min cycles at 600 MPa. This work also provides evidence that pressuretreatments in conjunction with antimicrobials including sanitizers,acidulants or a combination of sanitizer and surfactant were notsufficient to eliminate E. coli O157:H7 on alfalfa seeds. However, whenseeds were soaked in water for 60 min and subsequently treated at600 MPa for 2 min, elimination of E. coli O157:H7 occurred at a slightexpense on the seed germinability. In addition, it was demonstratedthat seeds could be alternatively processed by a brief soaking of10min inwater followed by an extended pressure treatment of 15minto achieve the same goals with respect tomicrobial safety and viabilityof alfalfa seeds.

References

Anonymous, 1999. Microbiological safety evaluations and recommendations onsprouted seeds. International Journal of Food Microbiology 52, 123–153.

Aleman, G.D., Ting, E.Y., Mordre, S.C., Hawes, A.C.O., Walker, M., Farkas, D.F., Torres, J.A.,1996. Pulsed ultra high pressure treatments for pasteurization of pineapple juice.Journal of Food Science 61, 389–390.

Ariefdjohan, M.W., Nelson, P.E., Singh, R.K., Bhunia, A.K., Balasubramaniam, V.M., Singh, N.,2004. Efficacy of high hydrostatic pressure treatment in reducing Escherichia coliO157and Listeria monocytogenes in alfalfa seeds. Journal of Food Science 69, 117–126.

Beuchat, L.R., Scouten, A.J., 2002. Combined effects of water activity, temperature andchemical treatments on the survival of Salmonella and Escherichia coli O157:H7 onalfalfa seeds. Journal of Applied Microbiology 92, 382–395.

Blaszczak, W., Doblado, R., Frias, J., Vidal-Valverde, C., Sadowska, J., Fornal, J., 2007.Microstructural and biochemical changes in raw and germinated cowpea seedsupon high-pressure treatment. Food Research International 40, 415–423.

Brooks, J.T., Rowe, S.Y., Shillam, P., Heltzel, D.M., Hunter, S.B., Slutsker, L., Hoekstra, R.M.,Luby, S.P., 2001. Salmonella typhimurium infections transmitted by chlorine-pretreated clover sprout seeds. American Journal of Epidemiology 154, 1020–1028.

Chen, H., 2007. Use of linear, Weibull, and log–logistic functions to model pressureinactivation of seven foodborne pathogens in milk. Food Microbiology 24, 197–204.

Delaquis, P.J., Sholberg, P.L., Stanich, K., 1999. Disinfection of mung bean seed withgaseous acetic acid. Journal of Food Protection 62, 953–957.

Earnshaw, R.G., Appleyard, J., Hurst, R.M., 1995. Understanding physical inactivationprocesses: combined preservation opportunities using heat, ultrasound and pressure.International Journal of Food Microbiology 28, 197–219.

Ferguson, D.D., Scheftel, J., Cronquist, A., Smith, K., Woo-Ming, A., Anderson, E., Knutsen,J., De, A.K., Gershman, K., 2005. Temporally distinct Escherichia coli O157 outbreaksassociated with alfalfa sprouts linked to a common seed source—Colorado andMinnesota, 2003. Epidemiology and Infection 133, 439–447.

Fett, W.F., 2002. Reduction of Escherichia coli O157:H7 and Salmonella spp. on laboratory-inoculated mung bean seed by chlorine treatment. Journal of Food Protection 65,848–852.

Hayakawa, I., Furukawa, S., Midzunaga, A., Horiuchi, H., Nakashima, T., Fujio, Y., Yano, Y.,Ishikura, T., Sasaki, K., 1998. Mechanism of inactivation of heat-tolerant spores ofBacillus stearothermophilus IFO 12550 by rapid decompression. Journal of FoodScience 63, 371–374.

Houben, J., 1966. Imbibition des graines de Pisum Sativum 1ere partie Imbibition parTrempage (Imbibition of seeds of Pisum Sativum 1st stage of imbibition by soaking).Planta 71, 87–97.

Jaquette, C.B., Beuchat, L.R., Mahon, B.E., 1996. Efficacy of chlorine and heat treatment inkilling Salmonella stanley inoculated onto alfalfa seeds and growth and survival ofthe pathogen during sprouting and storage. Applied and Environmental Micro-biology 62, 2212–2215.

Kingsley, D.H., Chen, H., 2008. Aqueous matrix compositions and pH influence virusinactivation by high pressure processing. Journal of Food Protection 71, 1598–1603.

Kingsley, D.H., Guan, D., Hoover, D.G., Chen, H., 2006. Inactivation of hepatitis A virus byhigh-pressure processing: the role of temperature and pressure oscillation. Journalof Food Protection 69, 2454–2459.

Kumar, M., Hora, R., Kostrzynska, M., Waites, W.M., Warriner, K., 2006. Inactivation ofEscherichia coli O157:H7 and Salmonella on mung beans, alfalfa, and other seedtypes destined for sprout production by using an oxychloro-based sanitizer. Journalof Food Protection 69, 1571–1578.

Kurtzweil, P., 1999. Questions keep sprouting about sprouts. FDA Consumer 33, 18–22.Lang, M.M., Ingham, B.H., Ingham, S.C., 2000. Efficacy of novel organic acid and

hypochlorite treatments for eliminating Escherichia coli O157:H7 from alfalfa seedsprior to sprouting. International Journal of Food Microbiology 58, 73–82.

Neetoo, H., Ye, M., Chen, H., 2008. Potential application of high hydrostatic pressure toeliminate Escherichia coli O157:H7 on alfalfa sprouted seeds. International Journalof Food Microbiology 128, 348–353.

Oxen, P., Knorr, D., 1993. Baroprotective effects of high solute concentrations againstinactivation of Rhodotorula rubra. LWT Food Science & Technology 26, 220–223.

Palou, E., Lopez-Malo, A., Barbosa-Canovas, G.V., Welti-Chanes, J., Swanson, B.G., 1997.Effect of water activity on high hydrostatic pressure inhibition of Zygosaccharomycesbailii. Letters in Applied Microbiology 24, 417–420.

Palou, E., Lopez-Malo, A., Barbosa-Canovas, G.V., Welti-Chanes, J., Davidson, P.M.,Swanson, B.G., 1998a. Effect of oscillatory high hydrostatic pressure treatments onByssochlamys nivea ascospores suspended in fruit juice concentrates. Letters inApplied Microbiology 27, 375–378.

Palou, E., Lopez-Malo, A., Barbosa-Canovas, G.V., Welti-Chanes, J., Swanson, B.G., 1998b.Oscillatory high hydrostatic pressure inactivation of Zygosaccharomyces bailii.Journal of Food Protection 61, 1213–1215.

Piernas, V., Guiraud, J.P., 1997. Disinfection of rice seeds prior to sprouting. Journal ofFood Science 62, 611–615.

Proctor, M.E., Hamacher, M., Tortorello, M.L., Archer, J.R., Davis, J.P., 2001. Multistateoutbreak of Salmonella serovar Muenchen infections associated with alfalfa sproutsgrown from seeds pretreated with calcium hypochlorite. Journal Clinical Micro-biology 39, 3461–3465.

Taormina, P.J., Beuchat, L.R., 1999. Comparison of chemical treatments to eliminateenterohemorrhagicEscherichia coliO157:H7onalfalfa seeds. Journal of FoodProtection62, 318–324.

Weissinger, W.R., Beuchat, L.R., 2000. Comparison of aqueous chemical treatments toeliminate Salmonella on alfalfa seeds. Journal of Food Protection 63, 1475–1482.

Wuytack, E.Y., Michiels, C.W., Meersseman, K., Diels, A.M., 2003. Decontamination ofseeds for seed sprout production by high hydrostatic pressure. Journal of FoodProtection 66, 918–923.