!64:8/ $03,80*7 4- !#$&%##! #741(8,7 03 ( %()46(846; &,+092 ......OUO_ ]YOaS` dS_S W[QbPOaSR Oa +(i9 b[aWY aVS Q\Y\[WS` aVOa T\_ZSR dS_S YO_US S[\bUV a\ Q\b[a3 aVW` b`bOYYf \QQb__SR

$Page 1: !64:8/ $03,80*7 4- !#$&%##! #741(8,7 03 ( %()46(846; &,+092 ......OUO_ ]YOaS` dS_S W[QbPOaSR Oa +(i9 b[aWY aVS Q\Y\[WS` aVOa T\_ZSR dS_S YO_US S[\bUV a\ Q\b[a3 aVW` b`bOYYf \QQb__SR$
964

Journal of Food Protection Vol 61 No8 1998 Pages 964-968Copyrightcopy International Association of Milk Food and Environmental Sanitarians

Growth Kinetics of Salmonella Isolates in a Laboratory Mediumas Affected by Isolate and Holding Temperaturet

THOMAS P OSCAR

United States Department of Agriculture Agricultural Research Service Eastern Regional Research Center Microbial Food Safety Research Unit1124 Trigg Hall University of Maryland Eastern Shore Princess Anne Maryland 21853 USA

MS 97-176 Received 30 July 1997Accepted 29 October 1997

ABSTRACT

Salmonella isolates were surveyed for their growth kinetics in a laboratory medium for the purpose of identifying isolatessuitable for modeling experiments In addition the effect of holding stationary phase Salmonella cultures at different temperatureson their subsequent growth kinetics was evaluated for the purpose of developing a protocol to prevent the need for midnightsampling in modeling experiments In Experiment 1 16 isolates of Salmonella 2 from the American Type Culture Collection(ATCC) and 14 from broiler operations were surveyed for their growth kinetics in brain heart infusion (BHI) broth at 40degC Lagtime (P = 0005) and growth rate (P = 0022) were affected by identity of the isolate Lag time ranged from 073 to 138 hwhereas growth rate ranged from 078 to 094 10glOCFUmlh Overall isolate Sl (Salmonella infantis from ATCC) was thefastest growing In Experiment 2 4 isolates of Salmonella 1 from ATCC and 3 from broiler operations were used to determinewhether holding temperature influences subsequent growth kinetics Salmonella isolates were grown to stationary phase at 3rCin BHI and then held for 24 h at 5 22 or 37degC before dilution and reinitiation of growth in BHI at 3rC Holding temperature didnot alter or interact with identity of the isolate to alter subsequent growth kinetics From the latter finding a protocol was devisedin which a dual-flask system is used to prevent the need for midnight sampling in modeling experiments Similar to the resultsobtained in Experiment 1 identity of the isolate had only minor effects on growth kinetics in Experiment 2 indicating that allisolates examined were suitable for modeling experiments

Development of mathematical models that predictgrowth of Salmonella strains associated with poultry prod-ucts requires extensive kinetic data on responses of theSalmonella strains to combinations of environmental andfood formulation factors Collection of such data in food islabor intensive thus making it difficult to obtain sufficientdata for development of accurate predictive models (10)Consequently like other modelers we plan on developingpredictive models for Salmonella strains using extensivekinetic data collected in studies with a laboratory mediumand then validating the models by comparison of theirpredictions to less extensive kinetic data collected in studieswith poultry foods

A problem encountered when developing predictivemodels for Salmonella growth is the problem of which strainto use as there are over 2000 serotypes and many morestrains of Salmonella to choose from (8) Surprisingly thereis limited published information about the growth kinetics ofSalmonella strains to help with this decision Nonethelessthe available data suggest that there are only minor differ-ences in growth rates (4) and the minimum (18 19) andmaximum (7) temperatures at which different strains ofSalmonella will initiate growth Although these data suggestthat growth kinetics of Salmonella strains are not highly

Author for correspondence Tel 410-651-6062 Fax 410-651-6568E-mail toscarumes-birdumdedu

t Mention of a brand or firm name does not constitute an endorsement bythe US Department of Agriculture over others of a similar nature notmentioned

variable a systematic study of the effect of identity of strainon growth kinetics of Salmonella strains has not beenreported Consequently the current study was undertaken tosurvey isolates of Salmonella from broiler operations fortheir growth kinetics in a laboratory medium for the purposeof identifying strains that would be suitable for modelingexperiments

A second problem encountered when conducting model-ing experiments is the need to collect kinetic data in themiddle of the night Stagger-starting of two cultures pergrowth condition at 12-h intervals has potential for eliminat-ing the need for midnight sampling However one mustestablish that the Salmonella culture used to inoculate thesecond flask has similar growth kinetics to that used toinoculate the first flask Considering that previous growthconditions have been found to alter subsequent growthkinetics of other foodbome pathogens (3 13) the secondobjective of this study was to determine the effect of holdingtemperature on the subsequent growth kinetics of Salmo-nella cultures with the goal of devising a dual-flask systemthat would eliminate the need for midnight sampling inmodeling experiments

MATERIALS AND METHODS

Isolates Table 1 lists the sources and serotypes of the isolatesof Salmonella used in this study Isolates Sl and S2 were fromATCC (American Type Culture Collection Rockville Md) whereasthe other isolates were from broiler operations The broiler isolateswere serotyped by the Salmonella Reference Center at the Univer-sity of Pennsylvania (Kennet Square Pa) Stock cultures of all

J Food Prot Vol 61 No8 GROWTH OF SALMONELLA ISOLATES 965

Y = Baseline + Increase

LU = X - (lag time)

TABLE 1 Sources and serotypes of Salmonella used to measuregrowth kinetics in laboratory medium

isolates were maintained at -70degC in brain heart infusion (BHI)broth (pH 6) that contained 15 glycerol

Starter cultures Starter cultures of each isolate were initi-ated by inoculating 50 rnl of BHI (pH 6) with 50 III of theresuspended stock culture which had a Salmonella viable cellconcentration of 92 IOglOCFUml Starter cultures with an initialSalmonella cell level of 62 IOglOCFUml were incubated for 24 hat 30degC in Experiment 1 and for 24 h at 37degC in Experiment 2 toobtain stationary phase cells Starter cultures were incubated in250-ml Erlenmeyer flasks that were sealed with foam plugs andshaken at 150 orbits per minute (opm) In Experiment 1 startercultures were used at 24 h of incubation to inoculate growthcultures whereas in Experiment 2 starter cultures were held anadditional 24 h at 5 22 or 37degC before being used to inoculategrowth cultures

Growth cultures Growth cultures in both experiments wereinoculated to an initial level of 42 IOglO CFUml and wereincubated under aerobic conditions in 250-ml Erlenmeyer flasksthat contained 50 ml of BHI (pH 6) shaken at 150 opm InExperiment 1 growth cultures were incubated at 40degC whereas inExperiment 2 they were incubated at 37degC

Viable counts At 005 1 15 2 345 6 7 and 8 h afterinoculation the viable count of Salmonella cells in the growthcultures was determined A sample of the growth culture waswithdrawn diluted in peptone water and spiral plated onto BHIagar The BHI agar plates were incubated at 30degC until the coloniesthat formed were large enough to count this usually occurredbetween 18 and 24 h of incubation Colonies were counted using aProtos Colony Counter (Synoptics Ltd Cambridge UK)

Curve fitting Growth curves of viable counts (Y IOglOCFUml) versus time (X) were iteratively fit using GraphPadPRIZM (GraphPad Software San Diego Calif) to a two-phaselinear growth model

A Ar- - A- A A- _ABAB

AB AB AB ABr- - - - r-

ABC ABC- -

BC- ~ ~

100ltcDE

075Cl-

050

025

000

150

125

where the viable count Y was equal to Baseline (initial viablecount) plus Increase (increase of viable count) In tum Increasewas equal to zero if the sample time X was less than or equal to thelag time otherwise Increase was equal to the growth rate times LU(the sample time minus the lag time)

Statistical analysis Effects of identity of the isolate andholding temperature on growth characteristics (ie lag time andgrowth rate) were evaluated by analysis of variance using theStatistical Analysis System (24) The model for Experiment 1contained a term for identity of the isolate whereas the model forExperiment 2 contained terms for identity of the isolate holdingtemperature and the interaction of identity of the isolate withholding temperature When a significant F test was encounteredindividual means were compared using Duncans multiple rangetest Three to five replicate growth curves were obtained for eachisolate in Experiment 1 whereas two or three replicate growthcurves were obtained for each combination of isolate and holdingtemperature in Experiment 2

RESULTS

831 838 822 85 830814 852 817 844 826 820 833 87 82 862 81

Isolate

FIGURE 1 Effect of identity of the isolate on lag time ofSalmonella isolates in laboratory medium incubated at 40degC(Experiment 1) Each bar is the mean of three to five replicate lagtime determinations Analysis of variance indicated that identity ofthe isolate affected lag time (P = 0005) Bars having the sameletter in common are not significantly different at P lt 005 asdetermined by Duncan smultiple range test

In Experiment 1 16 isolates of Salmonella were sur-veyed for their growth kinetics in laboratory mediumincubated at 40degC Analysis of variance indicated thatidentity of the isolate affected lag time (P = 0005) andgrowth rate (P = 0022) Lag time ranged from 073 h forisolate S I to 138 h for isolates S31 and S38 (Figure 1)Growth rate ranged from 078 10glO CFUmlh for isolateS62 to 094 10glO CFUmlh per hour for isolate S44 (Figure2) Overall the maximum difference between isolates was90 for lag time and 20 for growth rate with most isolatesshowing similar lag times and growth rates

In an effort to identify fast- and slow-growing isolates inExperiment 1 the observed lag time (Figure 1) and growthrate (Figure 2) for each isolate were used to calculate thetime needed for an increase of 0 1 2 or 3 log cycles inSalmonella numbers The calculated time for a given logincrease was plotted against the corresponding log increaseas shown in Figure 3 The most striking revelation from this

if X (lag time)

if Xgt (lag time)

ATCC 51740ATCC 14028Broiler cecaBroiler cecaBroiler cecaBroiler feedBroiler cecaBroiler feedBroiler cecaBroiler cecaBroiler house litter swabBroiler processing plant sludgeBroiler processing plant sludgeBroiler house litter swabBroiler feedBroiler carcass rinseBroiler feedBroiler house beetles

Source

= (growth rate) X LU

Serotype

S infantisS typhimuriumS worthingtonS mbandakaS binzaS typhimuriumS senftenbergS binzaS brandenburgS senftenbergS senftenbergS thompsonS schwarzengrundS indianaS senftenbergS simsburyS worthingtonS typhimurium

Increase = 0

SlS2S5S7SlOSl4Sl6Sl7S20S22S26S30S31S33S38S44S52S62

Isolate

966 OSCAR J Food Prot Vol 61 No8

TABLE 2 Effect of isolate and holding temperature on growthkinetics of Salmonella in laboratory mediuma

S2 114 098 116 I09B 086 086 090 087ASIO 157 146 152 152A 084 082 085 084BS16 124 129 132 128AB081 084 084 083BS62 111 112 101 108B 086 085 085 085ABMain 126A 121A 125A O84A085A 086Aeffect

10

09

ltE 08Es 07LL

~ 06

i 05Tsect 04c 03(5

02

01

00

A-CA-C AS AB AB ~~ -B-D A-D A-D~ ~~ - ~ -

- -

~CD - -~

862 82 826 85 833 852 87 820 814 31 81 817 822 830 838 844

Isolate

Isolate

Lag time h

Holding temperature

Main5degC 22degC 37degC effect

Growth rate log CFUmllh

Holding temperature


FIGURE 2 Effect of identity of the isolate on growth rate ofSalmonella isolates in laboratory medium incubated at 40degC(Experiment 1) Each bar is the mean of three to five replicategrowth rate determinations Analysis of variance indicated thatidentity of the isolate affected growth rate (P = 0022) Barshaving the same letter in common are not significantly different atP lt005 as determined by Duncan smultiple range test

plot was that isolate Sl (Salmonella infantis from ATCC)was clearly the fastest growing (ie had the shortest time fora given log increase in numbers) In general most isolatesexhibited similar rates of growth as indicated by theclustering oflines in Figure 3

In Experiment 2 four isolates of Salmonella were usedto determine whether previous holding temperature influ-ences subsequent growth kinetics After growth to stationaryphase (24 h at 37degC) Salmonella cultures were held for 24 hat 5 22 or 37degC before dilution and reinitiation of growth inlaboratory medium at 37degC Holding temperature did notalter (P = 090 for lag time and 022 for growth rate) orinteract with identity of the isolate to alter (P = 098 for lagtime and 062 for growth rate) subsequent growth kinetics(Table 2) Again minor differences in growth kinetics wereseen between isolates Most notable the lag time of isolateS 10 was 40 longer than the lag time of isolates S2 and S62A plot of time for a given log increase in numbers versus the

5

a Salmonella were grown to stationary phase (Le 24 h at 37degC) inbrain heart infusion (BBI) broth (pH 6) and then held at 5 22 or3rC before measurement of growth kinetics in BBI pH 6incubated at 3rc Holding temperature did not alter or interactwith isolate to alter growth kinetics However isolate had asignificant effect on lag time and growth rate Means having thesame letter in common within the same main effect are notsignificantly different at P lt 005 as determined by Duncansmultiple range test

log increase calculated from lag time and growth raterevealed that isolates S2 and S62 grew at a similar rate butthey grew faster than isolate S 16 which grew slightly fasterthan isolate S 10 (Figure 4)

Subsequent to Experiment 2 my colleagues and Iadopted a protocol in which the starter culture is diluted by afactor of 10-4 in buffered peptone water and the dilutedculture is then used to inoculate two flasks one at 8 am(ie flask A) and one at 8 pm (ie flask B) for each growthcondition being modeled The diluted culture used toinoculate flask A is stored at 4degC for 12 h before it is used toinoculate flask B Although this protocol differs from thatformally tested in Experiment 2 growth curves obtained bysampling flasks A and B usually yielded kinetic data that fellin line with each other An example is illustrated in Figure 5Here the data points enclosed in boxes were collected fromflask B whereas the data points not enclosed in boxes were

6

4

c 3

ti=

2

oo

Log increase

2

--+-81 ~826~82 ~830~85 ~831--0--87 ~833-814-838~8170S44~820 bull 852-822 bull 862

3

5

4

cE 3i=

2-0-82-0-810-gt- 816-0-862

00 2 3

Log increase

FIGURE 3 Effect of identity of the isolate on time for a given logincrease in Salmonella viable cell numbers in laboratory mediumincubated at 40deg C (Experiment 1)

FIGURE 4 Effect of identity of the isolate on time for a given logincrease in Salmonella viable cell numbers in laboratory mediumincubated at 37degC (Experiment 2)


DISCUSSION

s- 8- g4plusmn-o------r5-~--~10-----1~5--~---20

Time h

collected from flask A Figure 5 also shows how well thetwo-phase linear equation fits the Salmonella growth dataThe r2 values a measure of goodness of fit for all 92 curvefits in this study ranged from 09364 to 09993 with a mean plusmnstandard error of the mean of 09902 plusmn 00012

FIGURE 5 An example of a growth curve obtained using adual-flask system Viable count data indicated by circles enclosedin boxes were collected from flask B which was inoculated withisolate 82 S typhimurium at 8 pm whereas other viable countdata (indicated by circles with no boxes) were collected from flaskA which was inoculated at 8 am S typhimurium was grown inbrain heart infusion broth adjusted to pH 65 and incubated at22SC

(pH 6) at 40degC Although most isolates exhibited similargrowth kinetics identity of the isolate was found to influ-ence growth of Salmonella isolates The maximum differ-ence between isolates was 90 for lag time and 20 forgrowth rate

The importance of assessing strain differences under avariety of growth conditions for the predictive microbiolo-gist is to identify strains that can be used to developpredictive models that are accurate and err on the side offood safety An appropriate strain is one that grows slightlyfaster than other strains pertinent to the situation beingmodeled In the present study isolate S1 was found to growslightly faster than other isolates and thus would be a goodchoice for model development This of course assumes thatisolate S1 grows slightly faster under other conditions nottested but included in development of a predictive modelAgain one may want to verify the latter assumption before afinal decision on use of a particular strain for modeling ismade A safe conclusion based on the observed similarity ofgrowth kinetics in this study is that all isolates of Salmo-nella examined are suitable for modeling experiments iean empirical choice is acceptable

The previous incubation temperature has been shown toinfluence subsequent thermal inactivation kinetics of Salmo-nella strains Ng et al (21) found that Salmonella strainsgrown at 44degC were more heat tolerant than Salmonellastrains grown at lower temperatures (15 to 35degC) Otherinvestigators (6 17) have also observed that increasing thegrowth temperature enhances subsequent thermal resistanceof Salmonella strains Previous temperature also alterssubsequent growth kinetics offoodbome pathogens Mackeyand Derrick (15 16) demonstrated that the lag time ofSalmonella growth increases when strains are previouslyexposed to low temperatures (- 5 to - 20degC) or hightemperatures (50 to 57degC) Buchanan and Klawitter (3) forListeria monocytogenes and Hudson (13) for Aeromonashydrophila found that growing these pathogens to stationaryphase at different temperatures altered their subsequent lagtime but not growth rate In contrast I found that holdingstationary phase Salmonella cultures at nongrowth tempera-ture (5degC) and growth temperatures (22 and 37degC) for 24 hdid not alter their subsequent growth kinetics These resultssuggest that it is possible to prevent effects of preconditionson subsequent growth of Salmonella strains by imposingpreconditions that prevent cell division and death

Based on the latter supposition my colleagues and Idevised a protocol to prevent the need for midnight samplingin modeling experiments In this protocol we grow aSalmonella culture to stationary phase dilute it and then useit to inoculate two flasks one at 8 am and one at 8 pm Thediluted Salmonella culture used to inoculate the first flask isheld at 4degC for 12 h before being used to inoculate thesecond flask By inoculating flasks at 12-h intervals we areable to achieve 24-h coverage of sampling between 8 amand 8 pm without the need to come into the laboratoryduring the middle of the night We have used this protocol tosuccessfully develop a predictive model for combinedeffects of previous pH (55 to 85) temperature (15 to 40degC)

[Q]

Isect]

Isecti

bullbull 1Qi

0 bullmiddotbullmiddotplt

Q bullbullbullbullbullbull

6

9

Thermal death time studies indicate that although mostSalmonella strains exhibit similar rates of inactivation thereare a number of heat-tolerant Salmonella strains (1 21 2526) Notably Salmonella senftenberg 775W is as much as20-fold more heat tolerant than other serotypes of Salmo-nella (21 22) and other strains of S senftenberg (5 14)However at low water activity (12) or in exposure to dryheat conditions (11 23) heat resistance of S senftenberg775W is similar to that of other Salmonella strains Thusstrain identity and environmental conditions interact toinfluence thermal inactivation kinetics of Salmonella strains

The influence of strain and environmental conditions ongrowth and nonthermal inactivation kinetics of Salmonellastrains is less well defined The available data suggest thatthere are only minor differences in growth kinetics (datafrom references 4 and 20 and this study) and minimum (1819) and maximum (7) temperatures at which strains ofSalmonella grow In addition environmental factors appearto interact with strain to influence growth kinetics ofSalmonella strains (2 9) However unlike heat tolerancethere do not appear to be strains of Salmonella with unusualgrowth characteristics such as the extreme heat resistance ofS senftenberg 775W

In the present study a systematic investigation wasconducted of the effect of identity of the isolate on growthkinetics of Salmonella isolates in a laboratory medium undertwo sets of environmental conditions In Experiment 1 twoATCC isolates and 14 broiler isolates of Salmonella weresurveyed for their growth kinetics in a laboratory medium

E~ 8Uo

ig 7(I)rosectt1l 60

2iE 5JZ


and pH (5 to 7) on growth of isolate S2 in laboratory medium(unpublished data)

In summary 16 broiler isolates of Salmonella and twoATCC isolates were surveyed for their growth kinetics in alaboratory medium under two sets of environmental condi-tions In general growth kinetics were similar betweenisolates indicating that all of them would be suitable formodeling experiments However the limited number ofisolates tested and growth conditions investigated justifyfurther research defining the variability of growth kineticsamong isolates of Salmonella In addition an experimentwas conducted that led to development of a two-flask systemto prevent the need for midnight sampling in modelingexperiments It was demonstrated that one can inoculate twoflasks at 12-h intervals and obtain similar growth kinetics ineach flask by taking advantage of the inability of Salmonellastrains to grow at refrigerated temperatures (18 19)

ACKNOWLEDGMENTS

The author would like to thank Jaci Wheatley and Pat Shannon ofARS and Marianita Albano Octavia Pickett and Lisa Collier of UMES fortheir excellent assistance on this project In addition the author appreciatesthe critical review of this research provided by Donald W Thayer andRobert L Buchanan of ARS

REFERENCES

I Baird-Parker A C M Boothroyd and E Jones 1970 The effect ofwater activity on the heat resistance of heat sensitive and heat resistantstrains of salmonellae J Appl Bacteriol 33515-522

2 Banwart G J and J C Ayres 1957 The effect of pH on the growthof Salmonella and functional properties of liquid egg white FoodTechnol II 244-246

3 Buchanan R L and L A Klawitter 1991 Effect of temperaturehistory on the growth of Listeria monocytogenes Scott A at refrigera-tion temperature Int J Food Microbiol 12235-246

4 Christian J H B and W J Scott 1953 Water relations ofsalmonellae at 30degC Aust J BioI Sci 6565-573

5 Davidson C M M Boothroyd and D L Georgala 1966 Thermalresistance of Salmonella senftenberg Nature (London) 212 1060-1061

6 Dega C A J M Goepfert and C H Amundson 1972 Heatresistance of salmonellae in concentrated milk Appl Microbial23415-420

7 Elliott R P andP K Heiniger 1965 Improved temperature-gradientincubator and the maximal growth temperature and heat resistance ofSalmonella Appl Microbiol 1373-76

8 Farber J M 1986 Predictive modeling of food deterioration andsafety p 57-70 In M D Pierson and N J Stern (ed) Foodborne

microorganisms and their toxins developing methodology MarcelDekker Inc New York

9 Genigeorgis C A Nychas and C Loullis 1977 Interaction ofSalmonella with food environments p 269-275 In Proceedings ofthe 7th International Symposium of the World Association of Veteri-nary Food Hygienists Frankfurt

10 Gibson A M N Bratchell and T A Roberts 1988 Predictingmicrobial growth growth responses of salmonellae in a laboratorymedium as affected by pH sodium chloride and storage temperatureInt J Food Microbiol 6 155-178

II Goepfert J M and R A Biggie 1968 Heat resistance of Salmonellatyphimurium and Salmonella senftenberg 775W in milk chocolateAppl Microbiol 161939-1940

12 Goepfert J M J K Iskander and C H Amundson 1970 Relationof the heat resistance of salmonellae to the water activity of theenvironment Appl Microbiol 19429-433

13 Hudson J A 1993 Effect of pre-incubation temperatureon the lagtime of Aero monas hydrophila Lett Appl Microbiol 16274-276

14 Liu T S V L Carlson and G H Snoeyenbos 1968 Thermalresistance of smooth and rough derivatives of Salmonella senftenberg775W Poult Sci 471155-1162

15 Mackey B M and C M Derrick 1982 The effect of sublethal injuryby heating freezing drying and gamma-radiation on the duration ofthe lag phase of Salmonella typhimurium J Appl Bacteriol 53243-251

16 Mackey B M and C M Derrick 1984 Conductance measurementsof the lag phase of injured Salmonella typhimurium J ApplBacteriol 57299-308

17 Mackey B M and C M Derrick 1986 Elevation of the heatresistance of Salmonella typhimurium by sublethal heat shock JAppl Bacteriol 61 389-393

18 Matches J R and J Liston 1968 Low temperature growth ofsalmonella J Food Sci 33641-645

19 Matches J R and J Liston 1972 Effect of pH on low temperaturegrowth of salmonella J Milk Food Technol 3549-52

20 Matches J R and J Liston 1972 Effects of incubation temperatureon the salt tolerance of Salmonella J Milk Food Technol 3539-44

21 Ng H H G Bayne and J A Baribaldi 1969 Heat resistance ofSalmonella the uniqueness of Salmonella senftenberg 775W ApplMicrobiol 1778-82

22 Osborne W W R P Straka and H Lineweaver 1954 Heatresistance of strains of Salmonella in liquid whole egg egg yolk andegg white Food Res 19451-463

23 Riemann H 1968 Effect of water activity on the heat resistance ofSalmonella in dry materials Appl Microbiol 161621-1622

24 SAS Institute 1990 SASSTAT users guide version 6 4th ed vol 2SAS Institute Inc Cary NC

25 Solowey M R R Sutton and E J Calesnick 1948 Heat resistanceof Salmonella organisms isolated from spray-dried whole-egg pow-der Food Technol 29-14

26 Winter A RG F Stewart V H McFarlane and M Solowey 1946Pasteurization of liquid egg products III Destruction of salmonella inliquid whole egg Am J Public Health 36451-460


Y = Baseline + Increase

LU = X - (lag time)

TABLE 1 Sources and serotypes of Salmonella used to measuregrowth kinetics in laboratory medium

isolates were maintained at -70degC in brain heart infusion (BHI)broth (pH 6) that contained 15 glycerol

Starter cultures Starter cultures of each isolate were initi-ated by inoculating 50 rnl of BHI (pH 6) with 50 III of theresuspended stock culture which had a Salmonella viable cellconcentration of 92 IOglOCFUml Starter cultures with an initialSalmonella cell level of 62 IOglOCFUml were incubated for 24 hat 30degC in Experiment 1 and for 24 h at 37degC in Experiment 2 toobtain stationary phase cells Starter cultures were incubated in250-ml Erlenmeyer flasks that were sealed with foam plugs andshaken at 150 orbits per minute (opm) In Experiment 1 startercultures were used at 24 h of incubation to inoculate growthcultures whereas in Experiment 2 starter cultures were held anadditional 24 h at 5 22 or 37degC before being used to inoculategrowth cultures

Growth cultures Growth cultures in both experiments wereinoculated to an initial level of 42 IOglO CFUml and wereincubated under aerobic conditions in 250-ml Erlenmeyer flasksthat contained 50 ml of BHI (pH 6) shaken at 150 opm InExperiment 1 growth cultures were incubated at 40degC whereas inExperiment 2 they were incubated at 37degC

Viable counts At 005 1 15 2 345 6 7 and 8 h afterinoculation the viable count of Salmonella cells in the growthcultures was determined A sample of the growth culture waswithdrawn diluted in peptone water and spiral plated onto BHIagar The BHI agar plates were incubated at 30degC until the coloniesthat formed were large enough to count this usually occurredbetween 18 and 24 h of incubation Colonies were counted using aProtos Colony Counter (Synoptics Ltd Cambridge UK)

Curve fitting Growth curves of viable counts (Y IOglOCFUml) versus time (X) were iteratively fit using GraphPadPRIZM (GraphPad Software San Diego Calif) to a two-phaselinear growth model

A Ar- - A- A A- _ABAB

AB AB AB ABr- - - - r-

ABC ABC- -

BC- ~ ~

100ltcDE

075Cl-

050

025

000

150

125

where the viable count Y was equal to Baseline (initial viablecount) plus Increase (increase of viable count) In tum Increasewas equal to zero if the sample time X was less than or equal to thelag time otherwise Increase was equal to the growth rate times LU(the sample time minus the lag time)

Statistical analysis Effects of identity of the isolate andholding temperature on growth characteristics (ie lag time andgrowth rate) were evaluated by analysis of variance using theStatistical Analysis System (24) The model for Experiment 1contained a term for identity of the isolate whereas the model forExperiment 2 contained terms for identity of the isolate holdingtemperature and the interaction of identity of the isolate withholding temperature When a significant F test was encounteredindividual means were compared using Duncans multiple rangetest Three to five replicate growth curves were obtained for eachisolate in Experiment 1 whereas two or three replicate growthcurves were obtained for each combination of isolate and holdingtemperature in Experiment 2

RESULTS

831 838 822 85 830814 852 817 844 826 820 833 87 82 862 81

Isolate

FIGURE 1 Effect of identity of the isolate on lag time ofSalmonella isolates in laboratory medium incubated at 40degC(Experiment 1) Each bar is the mean of three to five replicate lagtime determinations Analysis of variance indicated that identity ofthe isolate affected lag time (P = 0005) Bars having the sameletter in common are not significantly different at P lt 005 asdetermined by Duncan smultiple range test

In Experiment 1 16 isolates of Salmonella were sur-veyed for their growth kinetics in laboratory mediumincubated at 40degC Analysis of variance indicated thatidentity of the isolate affected lag time (P = 0005) andgrowth rate (P = 0022) Lag time ranged from 073 h forisolate S I to 138 h for isolates S31 and S38 (Figure 1)Growth rate ranged from 078 10glO CFUmlh for isolateS62 to 094 10glO CFUmlh per hour for isolate S44 (Figure2) Overall the maximum difference between isolates was90 for lag time and 20 for growth rate with most isolatesshowing similar lag times and growth rates

In an effort to identify fast- and slow-growing isolates inExperiment 1 the observed lag time (Figure 1) and growthrate (Figure 2) for each isolate were used to calculate thetime needed for an increase of 0 1 2 or 3 log cycles inSalmonella numbers The calculated time for a given logincrease was plotted against the corresponding log increaseas shown in Figure 3 The most striking revelation from this

if X (lag time)

if Xgt (lag time)

ATCC 51740ATCC 14028Broiler cecaBroiler cecaBroiler cecaBroiler feedBroiler cecaBroiler feedBroiler cecaBroiler cecaBroiler house litter swabBroiler processing plant sludgeBroiler processing plant sludgeBroiler house litter swabBroiler feedBroiler carcass rinseBroiler feedBroiler house beetles

Source

= (growth rate) X LU

Serotype

S infantisS typhimuriumS worthingtonS mbandakaS binzaS typhimuriumS senftenbergS binzaS brandenburgS senftenbergS senftenbergS thompsonS schwarzengrundS indianaS senftenbergS simsburyS worthingtonS typhimurium

Increase = 0

SlS2S5S7SlOSl4Sl6Sl7S20S22S26S30S31S33S38S44S52S62

Isolate




10

09

ltE 08Es 07LL

~ 06

i 05Tsect 04c 03(5

02

01

00


- -

~CD - -~

862 82 826 85 833 852 87 820 814 31 81 817 822 830 838 844

Isolate

Isolate

Lag time h

Holding temperature



Holding temperature





5




6

4

c 3

ti=

2

oo

Log increase

2

--+-81 ~826~82 ~830~85 ~831--0--87 ~833-814-838~8170S44~820 bull 852-822 bull 862

3

5

4

cE 3i=

2-0-82-0-810-gt- 816-0-862

00 2 3

Log increase




DISCUSSION

s- 8- g4plusmn-o------r5-~--~10-----1~5--~---20

Time h







[Q]

Isect]

Isecti

bullbull 1Qi



6

9




E~ 8Uo

ig 7(I)rosectt1l 60

2iE 5JZ




ACKNOWLEDGMENTS


REFERENCES































10

09

ltE 08Es 07LL

~ 06

i 05Tsect 04c 03(5

02

01

00


- -

~CD - -~

862 82 826 85 833 852 87 820 814 31 81 817 822 830 838 844

Isolate

Isolate

Lag time h

Holding temperature



Holding temperature





5




6

4

c 3

ti=

2

oo

Log increase

2

--+-81 ~826~82 ~830~85 ~831--0--87 ~833-814-838~8170S44~820 bull 852-822 bull 862

3

5

4

cE 3i=

2-0-82-0-810-gt- 816-0-862

00 2 3

Log increase




DISCUSSION

s- 8- g4plusmn-o------r5-~--~10-----1~5--~---20

Time h







[Q]

Isect]

Isecti

bullbull 1Qi



6

9




E~ 8Uo

ig 7(I)rosectt1l 60

2iE 5JZ




ACKNOWLEDGMENTS


REFERENCES





























DISCUSSION

s- 8- g4plusmn-o------r5-~--~10-----1~5--~---20

Time h







[Q]

Isect]

Isecti

bullbull 1Qi



6

9




E~ 8Uo

ig 7(I)rosectt1l 60

2iE 5JZ




ACKNOWLEDGMENTS


REFERENCES































ACKNOWLEDGMENTS


REFERENCES




























Documents

!64:8/ $03,80*7 4- !#$&%##! #741(8,7 03 ( %()46(846; &,+092 ......OUO_ ]YOaS` dS_S W[QbPOaSR Oa +(i9 b[aWY aVS Q\Y\[WS` aVOa T\_ZSR dS_S YO_US S[\bUV a\ Q\b[a3 aVW` b`bOYYf \QQb__SR