Microbial Inactivation Kinetics in Soymilk during Continuous Flow High-Pressure Throttling

M:FoodMicrobiology&

Safety

JFS M: Food Microbiology and Safety

Microbial Inactivation Kinetics in Soymilk duringContinuous Flow High-Pressure ThrottlingVIJENDRA SHARMA, RAKESH K. SINGH, AND ROMEO T. TOLEDO

ABSTRACT: The thermal resistance of Clostridium sporogenes PA 3679 ATCC 7955 was determined in soymilk (pH 7)and 0.1% peptone water (pH 7) by the capillary tube method. In the continuous flow high-pressure throttling, thetemperature of soymilk increased due to instantaneous pressure release and the additional heat was supplied bya heat exchanger to achieve the set temperature. The soymilk was immediately cooled after a short preset holdtime to below 40 ◦C. A significant increase in the heat resistance was observed in C. sporogenes spores when heatedin soymilk in comparison to 0.1% peptone water. The D121-value for spores in soymilk was approximately 3-foldshigher than peptone water. The z-value was also much higher in soymilk as compared to that in 0.1% peptone water.Continuous flow high-pressure throttling (HPT) from 207 or 276 MPa to atmospheric pressure reduced the micro-bial populations in inoculated soymilk up to 6 log cycles when the holding times were 10.4, 15.6, and 20.8 s andthe process temperatures were 85, 121, 133, and 145 ◦C, respectively. The sporicidal effect increased as the operat-ing pressure, time, and temperature were increased. More injured spores were found at 207 MPa than at 276 MPa,indicating that lower pressure caused cell injury whereas high pressure caused cell death.

Keywords: high-pressure throttling, hold time, injured spores, soymilk, sporicidal effects

Introduction

High hydrostatic pressure is an emerging technique to producefresh-like, minimally processed and shelf-stable foods. Many

other alternative nonthermal food processing techniques are beingstudied and used, but high pressure has been most promising be-cause of its minimal processing effect on the quality. In 1899, the 1stexperiment involving high pressure was reported by Hite, in whichinactivation of microorganisms in milk by high pressure was stud-ied (Hoover 1993). Specially designed vessels are currently used forhigh hydrostatic pressure processing and usually pressure between100 and 1000 MPa is used. The most important attribute of highhydrostatic pressure is uniform and instantaneous pressure distri-bution throughout the food system irrespective of size, shape, andfood composition (Smelt 1998).

Different species and strains of microorganisms respond differ-ently to pressure and thermal inactivation. In addition, the pH,growth conditions, and suspending media also affect resistance toinactivation (Patterson 1999). Gram-positive bacteria are most re-sistant to pressure followed by Gram-negative and yeast and moldsare least sensitive to pressure (Smelt 1998). Bacillus species can bereduced by 1 log after pressure treatment of 300 to 350 MPa and10 ◦C for 20 min whereas complete inactivation of molds and yeastwas observed at pressure treatment between 300 and 350 MPa and10 ◦C for 20 min (Arroyo and others 1997). Stewart and others (1997)observed an additional 3 log reduction in cell counts of Listeriamonocytogenes scott CA at pH 4 as compared to that at pH 6 whenpressurized at 353 MPa, 45 ◦C for 10 min. Lysis of large cells wasreported by Hoover and others (1989) due to mechanical disrup-tion of stressed cell wall after pressure treatment of 20 to 40 MPa.Wouters and others (1998) reported loss of membrane functional-ity as a result of high-pressure treatment in L. plantarum.

MS 20080783 Submitted 10/6/2008, Accepted 4/16/2009. Authors are withthe Univ. of Georgia, Dept. of Food Science and Technology, Athens, GA30602, U.S.A. Direct inquiries to author Singh (E-mail: [email protected]).

Continuous flow high-pressure throttling (CFHPT) has been de-veloped at The Univ. of Georgia, Athens, Ga. (Toledo and MoormanJr. 2000). In CFHPT, the fluid food at high pressure is passed througha throttling valve with a high velocity and increased pressure. As thefood passes through a small orifice, the potential energy changes toheat energy and instantaneous rise in the temperature occurs as thefeed discharges from throttling valve. Thus, as a result of instanta-neous pressure reduction and shear on the fluid the microbial cellwall disrupts thereby killing or injuring microbial cells, whereas inHHP, there is no effect of shear as the product remains stationarythroughout the processing.

CFHPT processing includes simultaneous homogenization andsterilization which can benefit physical properties, texture, and sta-bility of sterilized product during storage. The use of CFHPT is lim-ited to fluid foods that contain very small size of suspended solids.

Microbial reduction of up to 2.14 logs in Bacillus stearother-mophillus and 5.12 log reduction in Bacillus megaterium was ob-served after high-pressure throttling treatment with a hold timeof 15 s at 135 ◦C (Areekul 2003). Also Moorman (1997) reported2 to 4 log reductions in native microbial population in skim milkthrough CFHPT process from 310 MPa to atmospheric pressurewith a high-pressure dwell of only 0.3 s and the author also reportedinactivation of P. putida inoculated in skim milk from an initialpopulation of approximately 108 colony forming units (CFU)/mLto undetectable levels.

Clostridium sporogenes PA 3679 ATCC 7955 is a heat resistant,Gram-positive, spore forming organism. In the commercial steril-ization process of low acid high moisture foods C. sporogenes PA3679 has been widely used as a surrogate organism for Clostridiumbotulinum since both have similar physiological properties and areassociated with the spoilage of canned food products where thecans swell or explode. The heat resistance of spores of C. sporo-genes PA 3679 has been found to be equal or greater than that ofC. botulinum (Koutchma and others 2005). The heat resistance ofC. sporogenes is very well documented in the literature. This studywas conducted in soymilk which is a low-acid food and therefore

M268 JOURNAL OF FOOD SCIENCE—Vol. 74, Nr. 6, 2009 C© 2009 Institute of Food Technologists R©doi: 10.1111/j.1750-3841.2009.01201.xFurther reproduction without permission is prohibited

M:Fo

odMi

crob

iolog

y&

Safet

y

Microbial inactivation in pressurized soymilk . . .

very susceptible to pathogenic microbial growth like some Bacil-lus species and Clostridium species. No information is available onD-value and z-value of C. sporogenes in soymilk. Therefore informa-tion was needed on the thermal death kinetics of this organism tofacilitate the adoption of processing parameters needed to ensureadequate spore inactivation in soymilk. We calculated the D- andz-value of C. sporogenes in both soymilk and 0.1% sterile peptonewater.

The overall objectives of this study were to determine heat resis-tance of C. sporogenes in soymilk, to quantify the kinetic parametersfor inactivation kinetics as affected by various parameters like pres-sure, temperature, and hold time. These parameters in differentcombinations were determined. The extent of spore injury duringcontinuous flow high-pressure throttling treatment of inoculatedsoymilk was also evaluated.

Materials and Methods

Soybeans were collected from the Georgia Seed DevelopmentCommission, 2420 South Milledge Ave., Athens, Ga. 30605,

U.S.A. The soybeans that were used for this study were Benningvariety, Group VII cultivar soybean [Glycine max (L.) Merrill] har-vested in 2005 from Davisboro, Ga., U.S.A. To minimize the changesin composition, soybeans were stored in closed polyethylene bagsat 4 ◦C and 20% RH in the dark throughout the experiments un-til it was processed into soymilk. Deionized water (DW) was usedthroughout the experiments to prepare soymilk.

Preparation of soymilkThe equipment used for soymilk preparation was made of either

stainless steel or plastic depending on the availability. The soymilkwas prepared according to the method developed by (Sivanandanand others 2008). The only change made in the process was thatthe comminution was done only in a Megatron (Model MTK 5000Q,Kinematica Inc., Cincinnati, Ohio, U.S.A.) at 624 × g for 15 min thatwas described as the best process to further reduce the particle size(Sivanandan 2007).

Preparation of sporesSpores from C. sporogenes PA 3679 ATCC 7955 were inoculated in

soymilk and processed with a continuous flow high-pressure throt-tling system to determine the lethality. To ensure healthy cultures,C. sporogenes was grown in reinforced clostridium medium (RCM,Difco Laboratories, Div. of Becton Dickinson and Co., Sparks, Md.,

Figure 1 ---Schematicdiagram of theCFHPT systemshowing theflow directionof fluid food.

U.S.A.) and incubated for 24 h at 37 ◦C during 3 initial transfers, inanaerobic jars with an atmosphere containing 5% to 10% CO2 ob-tained by the application of gas packs (BD BBLTM GasPakTM Anaer-obic System Envelopes, Difco Laboratories). Following initial trans-fers C. sporogenes vegetative cells were grown in RCM for 24 h at37 ◦C before sporulation. Spores were prepared by distributing ac-tively growing culture into a medium comprising 3% trypticasesoy broth (TSB, Difco Laboratories), 0.1% yeast extract (YE, DifcoLaboratories), and 1% ammonium sulfate (J.T. Baker, Phillipsburg,N.J., U.S.A.) (Kalchayanand and others 2004). The medium wasincubated for 14 d at 35 ◦C. Spore suspensions were centrifuged(CentrificTM Centrifuge Model 225, Fisher Scientific, Pittsburgh,Pa., U.S.A.) at 10000 × g for 20 min, rinsed with sterile distilledwater and centrifuged again as before. The process was repeated3 times. Spores were collected and stored at 4 ◦C for further use.Microscopic examinations were performed to confirm that the sus-pension consisted primarily of C. sporogenes spores. Prior to inoc-ulating the spores in soymilk, spores were heat shocked at 80 ◦Cfor 15 min in 100 mL of soymilk immediately prior to sample in-oculation. Enumeration of C. sporogenes spores were carried outin anaerobic jars as before, on brain heart infusion agar (BHIA,Difco Laboratories) for 48 h at 37 ◦C. Soymilk samples yieldingzero colonies on enumeration, that is, samples with no detectableCFU (detection limit ≤ 1 CFU/mL) were subjected to enrichment(48 h at 37 ◦C) in RCM and visually checked for turbidity. To detect1 CFU/mL, 1 mL of the spore culture after the CFHPT treatmentwas enumerated in duplicate on BHIA plates.

CFHPT treatmentSoymilk was prepared according to the method developed by

Sivanandan and others (2008) containing all solids in the soybeans.The schematic diagram of the CFHPT system is shown in Figure 1.The CFHPT system consisted of a feed pump that is used to main-tain a constant pressure to the fluid feed to the intensifier, a dualpressure intensifier piston that works alternately to take in fluidfood product while the other discharges the high-pressure fluid, aheat exchanger to increase the temperature of the pressurized fluidproduct before throttling, a throttling valve to immediately drop thepressure that further increases the temperature of the fluid prod-uct, a back pressure valve to avoid flashing of the fluid food, a hold-ing tube to give the desired residence time to the fluid food, and atubular heat exchanger to immediately cool the product using icewater as the coolant after it leaves the holding tube. The process

Vol. 74, Nr. 6, 2009—JOURNAL OF FOOD SCIENCE M269

M:FoodMicrobiology&

Safety


included 2 levels of pressure treatment. The comminuted suspen-sion was pressurized at 207 and 276 MPa using CFHPT system.

Spores were heat shocked at 80 ◦C for 15 min in 500 mL ofsoymilk immediately before sample inoculation. The comminutedsuspension of the soymilk was pressurized at 207 and 276 MPa us-ing 2 separate intensifier pistons acting alternately. Piston move-ment was synchronized by a microprocessor which also controlsthe opening and closing of the intake and discharge valves of fluidentering and leaving the intensifier (Stansted Fluid Power Ltd.,Stansted, Essex, U.K.). The pistons were driven by a hydraulic pump(Model nG7900, Stansted Fluid Power Ltd.). The pressure generatedwas read from the pressure gauge located in the CFHPT system.The resulting fluid high pressure was generated when a pre-set hy-draulic fluid pressure drives the piston while a small opening ofthe throttling valve restricts flow out of the system. The throttlingvalve was a micrometering valve (Model 60VRMM4882, AutoclaveEngineers, Fluid Components, Erie, Pa., U.S.A.).

The inlet temperature of the soymilk fed into the CFHPT sys-tem was kept at 32 ± 2 ◦C. For all the treatments, a tubular heatexchanger was installed between the pressure intensifier and thethrottling valve to heat the soymilk to 53, 64, and 76 ◦C after pres-surization to get final exit temperature after depressurization of thesoymilk to 121, 133, and 145 ◦C, respectively, for 276 MPa. Simi-larly, for 207 MPa the soymilk was heated to 70, 82, and 94 ◦C toget final exit temperature of 121, 133, and 145 ◦C, respectively. Allthe exit temperatures fluctuated by ± 4 ◦C. When the tempera-ture of the soymilk leaving the tubular heat exchanger was 32 ±2 ◦C the final exit temperature was 85 ◦C at 207 MPa and 102 ◦C at276 MPa. K-type thermocouples (Everett, Wash., U.S.A.) were con-nected at the outlet of steam heated tubular heat exchanger and atthe end of holding tube that was located after the throttling valveto record the temperature of the soymilk after it leaves the tubu-lar heat exchanger and holding tube, respectively. The outputs ofthe thermocouples were recorded on a Fluke Hydra Data Bucket(Everett, Wash., U.S.A.). After throttling, a minimum back pressureof 0.4 MPa was applied to avoid flashing of vapors at the outlet byraising the boiling point of the fluid. The minimum back pressurevaried for each applied pressure as the adiabatic temperature risevaried with applied pressure. The minimum back pressure was cal-culated from the saturated steam table using the saturated temper-ature of the water at applied pressure (Toledo 2007). The soymilkremained at the elevated temperature after depressurization fordesired residence time in the holding tube between the throttlingvalve and back pressure valve. Thus the adiabatic temperature riseas the fluid passes through the throttling valve is responsible forboth microbial inactivation and desirable physical effects on thefluid product. A thermocouple was used to measure the elevatedtemperature at the end of holding tube. By using another heat ex-changer coil completely immersed in ice bath the temperature ofthe sterilized soymilk was lowered to 4 ◦C to avoid additional heat.The soymilk samples were collected in presterilized glass bottles.Volumetric flow rates of 0.75, 1.0, and 1.5 L/min with correspond-ing calculated hold times of 10.4, 15.6, and 20.8, respectively, wereused. Immediately after processing, samples were kept in a coolerat 4 ◦C and analyzed.

Determination of heat resistance of C. sporogenesThe pasteurized soymilk was inoculated with C. sporogenes

spores to yield an inoculum level of approximately 108 spores/mL.The inoculated soymilk was heat shocked as described previously,for determination of viable spore counts and to facilitate filling intocapillary tubes. The thermal treatment was determined by capillarytube method (Stumbo 1973).

Heating the inoculated soymilk (80 ◦C for 15 min) was done toactivate the spores for germination and to kill the vegetative cells ifpresent in soymilk. It is essential to heat shock the spore suspensionto break the dormancy of spores and for rapid germination (Stumbo1973).

Fifty microliters of inoculated soymilk were filled in the capil-lary tube (1.5 mm inside dia, 1.8 outside dia and 100 mm length,Kimax-51, Fisher Scientific Inc.) ensuring that there was no bubbleformation in the capillary tube. The open ends were flame sealed.The capillary tubes were then completely immersed in a constanttemperature oil bath (Isotemp Model 1013S, Fisher Scientific Inc.)at the desired temperatures ± 0.5 ◦C. Antifreeze (ValuecraftTM,Best Parts, Inc., Memphis, Tenn., U.S.A.) was used as the heatingmedium. After a designated heating time duplicate tubes were re-moved and immediately immersed in cooling water (4 ± 1 ◦C). Todetermine the come up time duplicate controls were used and thistime was included in thermal process determination. The capillarytubes were first washed in soap solution, rinsed with sterile deion-ized water and immersed in 70% ethanol and air dried. The controls(unheated capillary tubes) were cooled in ice water as the heatedsamples, but were heat shocked at 80 ◦C for 15 min to activate sporegermination before enumeration of viable spores. Spores were alsosuspended in sterile 0.1% peptone and subjected to thermal treat-ments at 98.9, 110, and 121 ◦C for different time combinations. Theviable counts were determined by plating on Brain Heart InfusionAgar and incubating for 48 h at 37 ◦C.

Thermal resistance of C. sporogenes spores was determined us-ing thermal death time capillary tube method procedures de-scribed by Stumbo (1973) and kinetic data were further calculatedas discussed by Toledo (2007). The destruction of C. sporogenesspores was modeled following the 1st-order reaction kinetics:

Log(N/No) = −t/D

where N is the viable number of spores after heating, No is the num-ber of viable spores before heating, t is the heating time, and D is thedecimal death time (D-value is the time needed to destroy 90% ofthe microorganisms). D-values were calculated as the negative re-ciprocal of the slopes of the survival curves against time. In additionto D-value determinations, z-values (the temperature in Celsius re-quired to bring about a 10-fold change in D-value) were determinedby calculating the negative inverse slope of the log10 D comparedwith temperature plot.

Enumeration of microorganismsSoymilk samples were stored at (4 ◦C) and enumerated within

5 h. Prior to enumeration, the soymilk samples and the control inglass bottles were heat shocked at 80 ◦C for 15 min (to heat shockthe ungerminated spores and to inactivate the vegetative cells).Spores were enumerated by spread plating in duplicate onto brainheart infusion agar (BHIA, Difco Laboratories) for 48 h at 37 ◦C.Further investigation was carried out to confirm if complete inac-tivation had been achieved. For this purpose 1 mL of each soymilksample yielding zero enumeration results, that is, samples with lessthan detectable CFU (detection limit ≤ 1 CFU/mL) were subjectedto enrichment (48 h at 37 ◦C) in RCM. Then a loopful of RCM wasstreaked out on brain heart infusion agar, which was incubatedat 37 ◦C for 48 h. To enumerate the injured spores, the thin agarlayer (TAL) method was used to recover the injured spores (Kangand Fung 2000). To prepare TAL plates, selective medium (brainheart infusion agar) was poured in the plate and left to solidify, andthen selective medium was overlaid with 14 mL of melted trypticsoy agar (7 mL and after solidification another 7 mL). Appropriate

M270 JOURNAL OF FOOD SCIENCE—Vol. 74, Nr. 6, 2009

M:Fo

odMi

crob

iolog

y&

Safet

y


decimal dilutions were made in 0.1% peptone (BactoTM peptone,Difco Laboratories). The colony forming units in plates with 25 to250 colonies were enumerated and the averages of duplicate platesin the samples were presented.

Statistical analysisAnalysis of variance using GLM procedure of SAS

R©software (ver-

sion 9.1, SAS Inst., Cary, N.C., U.S.A.) was done for the log of theinactivation (N/No, where N is the number of surviving cells andNo is the number of initial cells). Level of significance was set forP < 0.05. For each factor (temperature, time, pressure) differencesamong means were also determined to determine the significantdifference between treatments. Further, Tukey test was conductedto find that which levels were significantly different at each factor.

Results and Discussion

Heat resistance of C. sporogenesA larger population of C. sporogenes (approximately 2 logs

CFU/mL) was obtained when the spores were heat shocked as com-pared to no heat shocking (data not presented). This was in accor-dance with Alcock (1984) who observed a D-value of 0.6 min whenthe spores were not heat shocked and the thermal resistance in-creased to D-value of 1 min when the spores were heat shocked for10 min at 100 ◦C.

Survivor curves were obtained by plotting the number of sur-vivors against the heating time which also included 1 to 2 s of come-up time. The D-value for C. sporogenes in 0.1% peptone water waslower than that observed for C. sporogenes in soymilk for all the 3different temperatures. The calculated D121-values were 1.47 and0.53 min for soymilk and 0.1% peptone, respectively. The D-valuesof C. sporogenes in 0.1% peptone and soymilk at various temper-atures are shown in Table 1. The z-value for 0.1% peptone waterwas also much lower than that in soymilk (Table 1). For instance,the z-value for 0.1% peptone and soymilk were 14 and 16.7 ◦C, re-spectively. (Figure 2). The survival curves for C. sporogenes in both

Table 1 --- Decimal reduction time (D-value, min) and z-value and R2-value for C. sporogenes in soymilk and peptonewater.

Soymilk Peptone water

Temperature (◦C) D-value (min) R2-value z-value (◦C) D-value (min) R2-value z-value (◦C)

99 30.9 0.97 20.0 0.89110 14.6 0.99 16.7 4.4 0.90 14.0121 1.5 0.99 0.54 0.93

Figure 2 --- Plot showing z-value for C.sporogenes in 0.1% peptone waterand soymilk.

0.1% peptone and soymilk are shown in Figure 3. The D121-valuefor C. sporogenes spores can be as high 3 min (Cameron and oth-ers 1980) but in general the D121-value ranges between 0.1 and1.5 min (Stumbo 1973). We calculated the D-value in 0.1% pep-tone and soymilk so direct comparison with the published data isnot sensible as various factors like bacterial strain, environmentalconditions including media and temperature, the heating mediumand properties and the composition of recovery medium, incu-bation conditions, influences the heat resistance and recovery ofheat-treated bacterial spores affect the inactivation (Stumbo 1973;Cameron and others 1980). The D- and z-values calculated werein similar range to those found by some researchers and are dif-ferent from others. This could be due to the techniques used inthe thermal resistance determination, sample heating, and recov-ery medium and different enumeration methods employed. Signif-icant differences in D- and z-values were observed depending onthe media, type of heat, and sporulation (Augustin and Pflug 1967).The highest D-values obtained by these researchers were in beef in-fusion or pea infusion when moist heat was used. They observeda D-value of 1.4 min when grown in beef heart and the recoverymedium was beef infusion. Our results were similar to those ob-tained by these researchers; however, we used different media forsporulation, enumeration and calculated the D-value in soymilk.The D121- and D110-values in phosphate buffer were reported as 1.3and 10.9 min, respectively, by Da Vi and Zottola (1978). The resultspresented in this study were slightly higher in soymilk and slightlylower in 0.1% peptone, which could be due to the reason that thecomponents of soymilk (protein, fat, and sugars) had some protec-tive effect on the thermal resistance of spores. The calculated D121-value of 1.46 min in soymilk and 0.50 min in 0.1% peptone couldbe due to the types of sugars, fatty acids present in soymilk as thepH and aw was in the same range. The z-value obtained for soymilkwas also higher than that obtained in peptone. The z-value also de-pends on the nature of heat treatment. The z-value obtained for C.sporogenes subjected to moist heat was approximately twice thatof obtained with dry heat (Augustin and Pflug 1967). Stumbo andothers (1950) reported z-value of C. sporogenes that ranged from


M:FoodMicrobiology&

Safety


16.6 to 20.5 ◦F. Thermal resistance also varies with different buffersolutions at same pH and some researchers have reported higherD-value in buffer than in food substrate (Tsuji and others 1960).Cameron and others (1980) observed a lower D-value of 2.6 min at121 ◦C and higher z-value of 14 ◦C in phosphate buffer than in peapuree when the pH was 7 in both cases. We obtained a lower D- andz-value in 0.1% peptone than in soymilk.

Microbicidal effects of continuous flow high-pressurethrottling on C. sporogenes in soymilk

The log reduction of C. sporogenes in soymilk increased as thehold time, pressure, and temperature were increased. At 207 MPa,log reductions of 0.4 to 5.6 CFU/mL were observed at differentexit temperatures (85, 121, 133, and 145 ◦C) and different resi-dence times in the holding tube (10.4, 15.6, and 20.8 s), respectively(Figure 4). Similarly at 276 MPa and the residence time of 10.4, 15.6,and 20.8 s, log reduction increased from 0.85 to 5.8 CFU/mL at dif-ferent exit temperatures (102, 121, 133, and 145 ◦C) (Figure 5). Meanlog reductions achieved at different combinations of pressure, holdtime, and exit temperatures are shown in Table 2. Almost completeinactivation (5.84 log CFU/mL) was achieved when the exit tem-perature was 145 ◦C for both 207 and 276 MPa. It was observed thatat 276 MPa increasing the temperature of the soymilk entering thethrottling valve from 53 to 64 ◦C using a tubular heat exchanger, in-creased the microbial reduction from 2.05 to 3.32 log CFU/mL andfurther increasing the temperature to 76 ◦C almost caused com-

Figure 3 --- Survival curve for C.sporogenes at 121 ◦C in peptone andsoymilk.

¿

Figure 4 --- Log reduction for C.sporogenes in soymilk at 207 MPa at(85, 121, 133, and 145 ◦C).

plete inactivation reducing the counts by 5.84 log CFU/mL. Simi-larly at 207 MPa increasing the temperature of the soymilk enteringthe throttling valve from 70 to 82 ◦C using a tubular heat exchangerincreased the reduction from 1.04 to 3.14 log CFU/mL and furtherincreasing the temperature to 94 ◦C caused 5.64 log CFU/mL reduc-tions. Feijoo and others (1997) also observed that increasing inlettemperature and operating pressure had positive sporicidal effectsin Bacillus spores. Similar results were observed by Areekul (2003)who observed 2.37 to 5.26 log reductions in B. megaterium after in-creasing the temperature of the fluid entering the throttling valvefrom 75 to 85 ◦C. Not much research is available on HPT and mostof the works on HPH were conducted on vegetative cells. These re-sults were much more effective over microbicidal effects of pres-sure reported previously (Feijoo and others 1997; Thiebaud andothers 2003). We found that the D-value decreased by increasingthe pressure, and the D-value calculated at 121 ◦C at 207 MPa was12.39 s and D121 at 276 MPa was 9.50 s. The z-value calculated at 207MPa was also much higher than that at 276 MPa. The z-value at 207MPa was 71 and 61 ◦C for 276 MPa (Figure 6). The mean D-value ateach temperature and pressure and z-value are shown in Table 3.Rovere and others (1996) observed a D-value of 41.7 s at 800 MPaand 108 ◦C and similarly Koutchma and others (2005) observed aD-value of 49 s at 108 ◦C and 800 MPa in C. sporogenes whereaswe observed a D-value of 14.4 s at 276 MPa and 102 ◦C. Pressurehad a significant role in the inactivation of C. sporogenes spores,as the D-value calculated at 121 ◦C by the capillary tube method


M:Fo

odMi

crob

iolog

y&

Safet

y


Figure 5 --- Log reduction for C.sporogenes in soymilk at 276 MPa at(102, 121, 133, and 145 ◦C).

in soymilk was much higher than the D-value calculated at 121 ◦Cwhen pressure was also applied, that is at 207 MPa or at 276 MPa.Similarly the z-value at 121 ◦C calculated in capillary tube methodwas 16.7 ◦C whereas at 207 and 276 MPa the z-value calculated was71 and 61 ◦C, respectively (Table 1 and 3). The lower D-value ob-served in this study indicated that the multiple effects of shear andother mechanical forces were involved in increasing inactivationon C. sporogenes spores. In this study, we observed that the effectof pressure and hold time were significantly different from eachother in the inactivation of C. sporogenes spores. Temperature wasthe most significantly effective in spore inactivation (P < 0.0001).The average mean in the inactivation of C. sporogenes inoculatedin soymilk for each temperature and hold times were significantlydifferent (P < 0.0001 and P < 0.01, respectively) from each other.

Table 2 --- Mean log reductions [log10 (N/No) (CFU/mL)] forC. sporogenes in BHIA (brain heart infusion agar, se-lective medium) and TAL (thin agar layer, nonselectivemedium + selective medium) after plating on day 0.

BHIA TALPressure Temperature Time mean mean(MPa) (◦C) (s) log N/No log N/No

207 85 10.4 −0.45 −0.25207 85 15.6 −0.64 −0.47207 85 20.8 −0.82 −0.68207 121 10.4 −1.04 −0.88207 121 15.6 −1.22 −1.07207 121 20.8 −1.61 −1.41207 133 10.4 −2.39 −2.19207 133 15.6 −2.62 −2.50207 133 20.8 −3.11 −2.95207 145 10.4 −4.37 −4.37207 145 15.6 −4.95 −4.95207 145 20.8 −5.47 −5.47276 102 10.4 −0.82 −0.58276 102 15.6 −1.15 −0.93276 102 20.8 −1.34 −1.11276 121 10.4 −1.42 −1.24276 121 15.6 −1.71 −1.47276 121 20.8 −1.97 −1.82276 133 10.4 −2.65 −2.49276 133 15.6 −3.23 −3.02276 133 20.8 −4.11 −4.05276 145 10.4 −5.27 −5.27276 145 15.6 −5.79 −5.79276 145 20.8 −5.79 −5.79Note: Shelf-life studies for soymilk were discontinued after observing counts onday 0.

No significant inactivation of C. sporogenes spores occurredwhen treated at 600 MPa for 30 min at 20 ◦C (Mills and others1998). However, we observed a high reduction at lower pressuresand high temperature due to the combination of instantaneoustemperature rise, impacts of pressure, high shear, high turbu-lence, and exposure to hydrodynamic cavitations that usually con-tributes to the microbial inactivation by disrupting cell membraneintegrity.

In general Gram-positive bacteria are more pressure resistantthan Gram-negative microorganisms. Gram-negative microorgan-isms are normally inactivated with 300 to 400 MPa at 25 ◦C for10 min to achieve inactivation, while Gram-positive microorgan-isms are inactivated with treatments of 500 to 600 MPa at 25 ◦C for10 min and the bacterial spores especially of Clostridium species,are relatively resistant to it (Hoover and others 1989; Smelt 1998). Inmicroorganisms, the cell membrane is supposed to be the 1st andforemost site of damage caused by pressure (Patterson 2005). Phys-ical damage to cell membrane occurs as a result of pressure, such asleakage of ATP from cell membrane (Patterson 2005). The HPT pro-cess has an affect on the morphological characteristics of microbialcells (Kheadr and others 2002). In HPT the death of a cell occurs asa result of sudden pressure drop that causes the shear, and instan-taneous temperature rise and high turbulence after the throttlingvalve.

Injured sporesThis study was also conducted to assure presence/absence of

injured spores after the CFHPT treatment. Injured spores can re-pair which could affect the microbiological quality of foodstuffsand safety especially in low acid canned foods (Bozoglu andothers 2004). In this study, we observed that injured spores werepresent after the CFHPT treatment. We observed that CFHPT didnot completely inactivate the microorganisms. It is very impor-tant to be able to detect undamaged as well as injured, dormantspores in food as the repair and proliferation of injured or dormantspores may lead to food safety problems or product quality lossor both (Ray 1989). To enumerate the injured spores, we followedthe method developed by Kang and Fung (2000) (TAL method), inwhich the nonselective (TSA) media was overlaid on selective me-dia (BHIA). The number of injured spores was determined by sub-tracting cell counts enumerated on BHIA from cell counts observedin TAL. Shelf life of treated soymilk was studied up to 10 d and if in-jured spores were observed after plating on day 0, the study for shelflife was discontinued.


M:FoodMicrobiology&

Safety


Figure 6 --- Plot showing z-value for C.sporogenes at 207 and 276 MPa.

We observed more counts in TAL in comparison to BHIA, indi-cating the presence of both healthy and injured spores. The injuredspore counts were more at 207 MPa than at 276 MPa. The counts inTAL decreased as the treatment temperature or the holding timewas increased. It was noticed that the TAL counts for both pres-sures were significantly different from each other. Hold times werealso significantly different from each other in the inactivation of C.sporogenes spores inoculated in soymilk. Temperature was the mosteffective in spore inactivation (P < 0.0001). The average means forthe inactivation of C. sporogenes spores for each temperature andhold times were significantly different from each other. The resultsindicate that an increasing number of spores were destroyed as a re-sult of pressure treatment and the destruction increased with bothtemperature and hold time. The BHIA counts (healthy spores) de-creased more rapidly than TAL counts. This indicates that morespores get injured than are immediately killed. Finally, at higherpressure and temperature, the counts in both TAL and BHIA wereundetectable indicating that no spores were present. In most of thetreatments, injured spores were able to resuscitate after plating onday 0 (Table 2). Only at few treatments, for example, at higher tem-perature (133, 145 ◦C) spores took longer time to repair indicat-ing that at higher temperatures spores were severely injured andshowed counts on day 5 and day 10 for treatments at 207 and 276MPa, respectively (Table 4). These results were in accordance withthe results presented by other researchers (Yuste and others 2003;Ariefdjohan and others 2004). The results also indicate that theCFHPT system caused sublethal injury to spores instead of com-pletely eliminating them.

Table 3 --- Mean D-values and z-value calculated from plot-ting [log10(N/No)] compared with time for C. sporogenes insoymilk.

Pressure Temperature D-value z-value(MPa) (◦C) (s) (◦C)

207 85 24.7207 121 12.4 71207 133 6.0207 145 3.3276 102 14.4276 121 9.5 61276 133 4.8276 145 2.9

Table 4 --- Mean log reductions [log10(N/No) (CFU/mL)] forC. sporogenes in BHIA (brain heart infusion agar, se-lective medium) and TAL (thin agar layer, nonselectivemedium + selective medium) after plating on day 5 andday 10.

TAL TALBHIA day 5 day 10

Pressure Temperature Time mean mean mean(MPa) (◦C) (s) log N/No log N/No log N/No

207 133 15.6 −2.88 −2.81 NA207 133 20.8 −3.09 −3.02 NA207 145 10.4 −4.37 −4.16 NA207 145 15.6 −4.95 −4.95 −4.88207 145 20.8 −5.47 −5.47 −5.47276 133 10.4 −2.57 −2.52 NA276 133 15.6 −3.41 −3.31 −3.31276 133 20.8 −4.11 −4.11 −4.06276 145 10.4 −5.27 −5.27 −5.27276 145 15.6 −5.79 −5.79 −5.79276 145 20.8 −5.79 −5.79 −5.79NA = not applicable.

Conclusions

The thermal resistance of C. sporogenes was found to be greaterin soymilk in comparison to 0.1% peptone water. The D121-

value for C. sporogenes in soymilk was 3-folds higher than that in0.1% peptone water. When pressurized soymilk was passed througha steam heated tubular heat exchanger prior to depressurizationsignificant 5.6 and 5.8 log reductions occurred as a result of pres-sure treatment at 207 and 276 MPa, respectively. The lethal effect in-creased as the operating pressure, temperature, and the residencetime in the holding tube was increased. The injured spores werealso present (as the counts in TAL were greater in comparison tothose in BHIA) after passing the soymilk through CFHPT system.The high pressure and temperature treatment cause the injuredspores to undetectable level.

High hydrostatic pressure is an effective technique to satisfyconsumer demand for fresh-like, minimally processed shelf-stableproduct. Many other alternative nonthermal food processing tech-niques are being studied and used but among all, the high-pressureprocessing has been the most promising. CFHPT system commer-cially pasteurizes the soymilk in seconds without any viable mi-croorganisms. Time and temperature are key factors in inactivatingmicroorganisms even under high pressure.


M:Fo

odMi

crob

iolog

y&

Safet

y


AcknowledgmentsThe authors would like to thank Natl. Research Initiative (grant nr2005-35503-15374, USDA Cooperative State Research, Educationand Extension Service NCGP program) for their financial support.Soybeans used in this study were supplied by Georgia Seed Com-mission, 2420 South Milledge Ave., Athens, Ga., U.S.A.

ReferencesAlcock S. 1984. Elevation of heat resistance of Clostridium sporogenes following heat

shock. Food Microbiol 1(1):39–47.Areekul V. 2003. High pressure sterilization of honey: physiochemical changes, sen-

sory attributes and shelf life [PhD diss]. Athens, Ga.: Univ. of Georgia. 147 p.Ariefdjohan MW, Nelson PE, Singh RK, Bhunia AK, Balasubramaniam VM, Singh N.

2004. Efficacy of high hydrostatic pressure treatment in reducing Escherichia coliO157 and Listeria monocytogenes in alfalfa seeds. J Food Sci 69(5):M117–20.

Arroyo G, Sanz PD, Prestamo G. 1997. Effect of high pressure on the reduction of mi-crobial populations in vegetables. J Appl Microbiol 82(6):735–42.

Augustin JA, Pflug IJ. 1967. Recovery patterns of spores of putrefactive anaerobe 3679in various subculture media after heat treatment. Appl Microbiol 15(2):266–76.

Bozoglu F, Alpas H, Kaletunc G. 2004. Injury recovery of foodborne pathogens in highhydrostatic pressure treated milk during storage. FEMS Immunol Med Microbiol40(3):243–7.

Cameron MS, Leonard SJ, Barrett EL. 1980. Effect of moderately acidic pH on heatresistance of Clostridium sporogenes spores in phosphate buffer and in bufferedpea puree. Appl Environ Microbiol 39(5):943–9.

Da Vi RT, Zottola EA. 1978. Thermal inactivation of Clostridium sporogenes PA 3679and Bacillus stearothermophilus 1518 in low-acid home canned foods. J Food Sci43(6):1738–40.

Feijoo SC, Hayes WW, Watson CE, Martin JH. 1997. Effects of microfluidizer R© tech-nology on Bacillus licheniformis spores in ice cream mix. J Dairy Sci 80(9):2184–7.

Hoover DG. 1993. Pressure effects on biological systems. Food Technol (Chicago)47(6):150–5.

Hoover DG, Metrick C, Papineau AM, Farkas DF, Knorr D. 1989. Biological effects ofhigh hydrostatic pressure on food microorganisms. Food Technol 43(3):99–107.

Kalchayanand N, Dunne CP, Sikes A, Ray B. 2004. Germination induction and inacti-vation of Clostridium spores at medium-range hydrostatic pressure treatment. In-nov Food Sci Emerg Technol 5(3):277–83.

Kang DH, Fung DYC. 2000. Application of thin agar layer method for recovery of in-jured Salmonella typhimurium. Int J Food Microbiol 54(1–2):127–32.

Kheadr EE, Vachon JF, Paquin P, Fliss I. 2002. Effect of dynamic high pressure on mi-crobiological, rheological and microstructural quality of cheddar cheese. Int DairyJ 12(5):435–46.

Koutchma T, Guo B, Patazca E, Parisi B. 2005. High pressure–high temperature steril-ization: from kinetic analysis to process verification. J Food Process Engr 28(6):610–29.

Mills G, Earnshaw R, Patterson MF. 1998. Effects of high hydrostatic pressure onClostridium sporogenes spores. Lett Appl Microbiol 26:227–30.

Moorman JE Jr. 1997. Microbicidal and rheological effects of high-pressure throttling[MS thesis]. Athens, Ga.: Univ. of Georgia. 74 p.

Patterson M. 1999. High-pressure treatment of foods. Encyclopedia Food Microbiol1059–65.

Patterson MF. 2005. A Review. Microbiology of pressure-treated foods. J Appl Micro-biol 98(6):1400–9.

Ray B. 1989. Enumeration of injured indicator bacteria from foods. In: Ray B, editor.Injured index and pathogenic bacteria. Boca Raton, Fla.: CRC Press Inc. p 9–54.

Rovere P, Maggi A, Scaramuzza N, Gola S, Miglioli L, Carpi G, Dall’Aglio G. 1996. High-pressure heat treatments: evaluation of the sterilizing effect and of thermal dam-age. Ind Conserv 71(4):473–83.

Sivanandan L. 2007. Characterization of soymilk produced by continuous flow highpressure throttling process. [PhD diss]. Athens, Ga.: Univ. of Georgia. 187 p.

Sivanandan L, Toledo RT, Singh RK. 2008. Effect of continuous flow high-pressurethrottling on rheological and ultrastructural properties of soy milk. J Food Sci73(6):E288–96.

Smelt J. 1998. Recent advances in the microbiology of high pressure processing. TFood Sci Technol 9(4):152–8.

Stewart CM, Jewett FF, Dunne CP, Hoover DG. 1997. Effect of concurrent high hydro-static pressure, acidity and heat on the injury and destruction of Listeria monocy-togenes. J Food Saf 17(1):23–6.

Stumbo CR. 1973. Thermobacteriology in food processing. New York: Academic Press.Stumbo CR, Murphy JR, Cochran J. 1950. Nature of thermal death time curves for PA

3679 and Clostridium botulinum. Food Technol 4:321–6.Thiebaud M, Dumay E, Picart L, Guiraud JP, Cheftel JC. 2003. High-pressure homog-

enization of raw bovine milk. Effects on fat globule size distribution and microbialinactivation. Int Dairy J 13(6):427–39.

Toledo RT. 2007. Fundamentals of food process engineering. 3rd ed. New York:Springer. 579 p.

Toledo RT, Moorman JE Jr, inventors; Univ. of Georgia Research Foundation Inc.(Athens, Ga.), assignee. 2000 Sept 19. Microbial inactivation by high-pressure throt-tling. U.S. patent 6,120,732.

Tsuji K, El-Bisi HM, Esselsen WB. 1960. Thermal resistance of Clostridium sporogenes(PA 3679) at the intermediate pH range. Food Technol 14:47.

Wouters PC, Glaasker E, Smelt J. 1998. Effects of high pressure on inactivation kinet-ics and events related to proton efflux in Lactobacillus plantarum. Appl EnvironMicrobiol 64(2):509–14.

Yuste J, Capellas M, Pla R, Llorens S, Fung DYC, Mor-Mur M. 2003. Use of conven-tional media and thin agar layer method for recovery of foodborne pathogens frompressure-treated poultry products. J Food Sci 68(7):2321–4.


Documents

Microbial Inactivation Kinetics in Soymilk during Continuous Flow High-Pressure Throttling