ImernationalJoumal of Food Microbiology, 16 (It}92) 323-335 © 1992 EI.sevicrScience Publishers B.V. All rights reserved 0168-1605/q2/$05.00

3!'~

FOOD 1HI537

Minimal water activity levels for growth and survival of Listeria monocytogenes and Listeria innocua D.A. Nolan, D.C. Chamblin and J.A. Troller The Pr~wter& Gambh,Co., WiningHill Techni,'al Career, ('in(.mnati, 0ti. Ub'.A

(Received 13 February 1992;accepted 15 July 1'492)

Following initial range-finding experiments, total count determinations were used to determine minimal water activity (a w) levels for the grtr~lh and survival of Listeria monocylogt,n£'~ (L,m.) agd L. iltnoct~u (L.L). Media containing three different humectants; NaCI, sucruse or glycerol were u~d to determine minimal a w levels for growJh in the alxwc media which were 0.92, 0.92 and 0.%') respectively. The growth minima for L.i. were similar, or slightly higher than for L.m. in these media, Survival rates were generally lower in NaCI-adjustcd media than in systems adjaMcd with sucrose or glyccro|. Survival of L.m. and L.i. in these experiments was similar. Key words: Growth: Survival: Listeria m~m¢~vtogenes',l,isteria inmx'aa; Water activity level

Introduction Listeria monocytogene~ ( L . m . ) is k n o w n to he relatively resistant to a variety o f potentially inhibitory factors in its e n v i r o n m e n t . T h i s ability to g r o w o r survive in c o n d i t i o n s t h a t inhibit m a n y othe~ p a t h o g e n s is, to a significant degree, responsible for the i m p o r t a n c e of this o r g a n i s m as a f o o d - b o r n e p a t h o g e n , T h e characteristics o f the n o n - p a t h o g e n i c species, L. innocua (L.L), o n the o t h e r h a n d , are poorly d e f i n e d with p e r h a p s the exception o f those biochemical tests which distinguish this o r g a n i s m f r o m L . m . Listeria monocytogenes is generally r e g a r d e d as a p s y c h r o t r o p h , a n d r e c e n t studies by Lovett et al. 0 9 8 8 ) have definitively d e m o n s t r a t e d t h a t r e c o m m e n d e d g u i d e l i n e s for the pasteurization o f milk (71.7°C for 15 s) are adequate to assure the a b s e n c e o f viable cells f r o m milk. Listeria monocytogenes is c a p a b l e o f g r o w t h a n d survival o v e r a r a t h e r b r o a d r a n g e o f p H levels. G e n e r a l l y , the o r g a n i s m is r e p o r t e d by L o v e r (1988) to exhibit diminishing viability at p H levels less t h a n 4.5 a n d g r e a t e r t h a n 9.4, O t h e r s have s h o w n this o r g a n i s m to he relatively sensitive to chlorine (Brackett, 1987), sorbic

Correspondence address: D.A. Nolan, The Procter & Gamble Co., Winton Hill Technical Center, Cincinnati, OH 45224, USA.

324 acid (EI-Shenawy and Marth, 1988a) and benzoic acid (E1-Shenawy and Marth, 1988b). The effect of reduced water activity (aw) levels on the growth of L.i. has not been determined except in broader studies to determine various growth characteristics of the genus Listeria in genei'al. Outbreaks of listeriosis involving preserved, cured meat products and various cheeses have suggested that this organism is capable of growth at elevated NaCl concentrations. Seeliger (1961) showed that L.m. was capable of growth at 10% (w/w) NaCI and survived for one year in a solution containing 16% (w/w) NaCI. Shahamat et al. (1980), on the other hand, examined the survival of this organism in bacteriological media adjusted to various aw levels with NaCI and reported that viability was lost within 10 days at 13% (w/w) NaCI (37°C). A solution with a concentration of 13% (w/w) NaCI has an a w level of approximately 0.91 in pure water at 25°C and, assuming that the a w level of the medium without added humectant was approximately 0.997, we can predict that the aw level of the adjusted salt solution was essentially unchanged by the medium components. Higher concentrations (to 30.5% NaCI) reduced survival times to about 5 days at 37°C, however lower temperatures appeared to ameliorate the effect of NaCI. Using gradient NaCi plates, McClure (1989) noted that growth appeared to be prevented by 8.8% (w/w) NaCI (0.94 a w) at 20°C, however increasing the incubation temperature to 25°C decreased the minimal a w level (increased the salt concentration) at which growth occurred to 9.9% (w/w) NaCI. Further increases in temperature to 30 and 35°C increased the minimal sodium chloride concentration tolerated for growth to greater than 9.9% (w/w). These determinations were in laboratory media; data obtained with natural substrates, such as saline cabbage juice, (Connor et al., 1986) suggest that minimal a w levels for survival and growth in these materials are somewhat higher. Papageorgiou and Marth (1989) evaluated the effect of two concentrations of NaCI on L.m. in skim milk and ,#hey and noted that growth of both strains occurred at 6% (w/w) NaCI (0.96 aw) and survival was observed for at least 132 days at 12% (w/w) NaCI (0.92 aw, 4°C). Based on these data, the minimal salt concentrations for growth of L. monocytogenes appears to be reasonably well-defined, however growth levels in substrates poised at various a w levels with other humectants, such as glycols and sugars, are not. Petran and Zottola (1989) obtained such data, however, their a w values were determined with an electric hygrometer which is reported by Troller and Christian (1978) to be generally inaccurate when measuring solutions containing glycols. The present work was initiated to compare a w minima for growth and survival of the pathogen, L.m. and the non-pathogen L.i., when exposed to a variety of humectants in the growth medium.

Materials and Methods

Organisms and media Cultures of L. monocytogenes Scott A. (serotype 4b) and L. innocua CI-94, were obtained from the Food and Drug Administration (Cincinnati, OH), subeultured

325 TABLE I Water activityof glycerol,sucroseand glycerolsolutions

% w/w

a,

Sucr~ % w/w

aN

Glycerol % w/w

aw

10.0 I0.5 I 1.0

0.934 0.929

43.0 48.0

0.952 0,940

29.0 30.0

0.920 0.915

0.924 0.92[ 0.917 0,913 0.910

49.0 50.0 51.0 52.0 53.0 54.0 55.0

0.938 0.935 0.932 0.928 0.925

31.0 32.0 33.0 34.0 35.0

0.911 0,908 0.q~4 0,900 0.897

0.920

36,0

0.893

NaCI

I 1.5 12.0 12.5 13.0

{}.915

Data from Chirife and Resnik(1984), Dallyn(1978)and Groverand Nicol(1940). and confirmed biochemically. These biochemical tests include: catalase; hemolysis; Triple Sugar Iron (TSI); urea (Christensen's Agar); motility (SIM); Methyl RedVoges Proskauer (MR-VP); Camp test; nih'ate reduction; esculin hydrolysis; dextrose, maltose, mannitol, rhamnos¢ and xylose utilization. The cultures were maintained in Tryptic Soy Broth with 0.6% yeast extract (TSB-YE) 10% ( w / w ) glycerin added at - 60~C. Cultures were prepared for inoculation by transferring a portion of the thawed contents of a vial into a tube of TSB-YE and incubating for 24 h at 37°C. Cultures were streaked onlo the surface of blood agar plates and incubated for 24 h at 37°C. Isolated colonies were picked for transfer to TSB-YE and incubated for 24 h at 3"PC. A transfer was then made into TSB-YE broth and incubated for 7 h at 37°C. This growth medium was diluted to obtain the counts required and used to inoculate aw-adjusted media which previously had been filter sterilized. Initial counts were confirmed culturally. Adjustment o f a w

Solutions of TSB-YE at various aw levels were prepared by adding the amounts of solutes shown in Figs. 1-6 and Table 1 to the medium. All solutes were ACS grade and were removed from freshly opened containers. TSB-YE in its normal formulation contains 5 g of NaCI per liter which was measured to be 0.997 a w. To compensate for this amount of humcctant, the a , levels shown in the figures were adjusted to reflect the product of the aw values of the test solution and the medium. This correction is derived from the Ross equation (Ross, 1975) and was applied equally to all solutions. Calculated a , levels are presented to the third place despite the fact that electric hygrometers and literature data (obtained via vapor pressure measurements) seldomly possess this level of sensitivity. In these studies data are interpolated from literature values and roflect the addition of very small increments (usually 0.5%) of hnmectants in an attempt to define more accurately the specific a w level for minimal growth under the stated experimental conditions. The pH levels of the media were not changed as a result of addition of humectant.

326

Growth experiments Media adjusted to specified a w levels with the humectants stated were inoculated in duplicate with 0.1 ml of the actively growing cultures. Throughout this work, aw levels of NaCI solutions were obtained from the data of Chirifc and Rcsnik (1984), glycerine solutions from the data of Grover and Nicol (1940), and sucrose solutions from the data of Dallyn (1978). With the exception of the glycerol values, the solutions adjusted according to literature values were determined to be in agreement with hygrometric measurements. Concentrations of humcetants used to obtain specified levels of a w are shown in Table 1. As noted above, glycerol poses special problems when measured with many electric hygrometers and, for this reason, reference data relating to this humectant was accepted without analytical confirmation. It was determined experimentally that the inocula did not alter the aw level of the tube contents. The initial concentrations of cells in the test media containing humcctants wcrc approximately 103, 10 "~or l0 T cclls/ml. Where appropriate, cultures wcrc incubated at 2I°C for at least 22 days before being declared negative with regard to the initiation of growth. The development of visually interpreted turbidity, as described by Cole c t a l . (1990) was considered to bc evidence of growth in preliminary range-finding experiments. Growth determinations reported herein were made using plate counts on Tryptic Soy Agar (TSA) with 0.6% yeast extract. Duplicate plates were incubated at 21°C for 48 h and counted. Earlier experiments using visual density observations were found to bc unsatisfactory for determinations other than range finding because of plasmolysis which occurred as a result of turgor loss when cultures were exposed to humectant solutions. This shrinking of individual cells is a result of osmotic imbalance between the medium and cell contents and is reported by Booth et al. (1988) to be injurious or lethal to cells depending on the extent of imbalance, type of organism, and other environmental factors. Survival experiments Survival of L.m. and L.i. was determined by counting test cultures adjusted to aw levels below minimums required for growth in TSA with 0.6% yeast extract.

Results and Discussion

For the purposes of these experiments, growth was defined as a one log increase in viable bacteria count within 22 to 24 days, the entire experimental period examined. Determination of growth craves is essential in interpreting the effects of aw on the two strains used in these experiments, however growth rates also are calculated and are included in this report. The g~owth characteristics of L.m. in NaCI are shown in Fig. 1. These data ~how ~h~t growth ceases at an a w level between 0.924 and 0.921. The generation time at 0.924 is approximately 62 rain (Table II). The minimal a w level for growth of L.i. (Fig. 2) is very slightly higher; between 0.929 and 0.924. Comparison (not

327

o

I

2

4

6

8

10 12 DAYS

14

t6

18

20

22

Fig. 1. Growthtff L. mont~'ytogenes in NaCI~lutions: o. 0.910: o, 0.913: ~, 0.917; A, 0.921; n. 0.924; I1, 0.929(Tuptic St~jBrothwlth0.6% yeast extract: 2[°C).

shown) of the influence of inoculation levels on growth showed that inoculations to obtain 10 3 and l0 s L . m . / m l resulted in growth at lower aw levels than when the initial count was 10~/ml. A similar effect occurred with L.L in NaCI-adjustcd media. Estimations of survival obtained from these data show that L.L seems to tolerate low a , levels slightly better than L.m. Although not shown, survival of L.m. is similar at all three inoculation levels. Sodium chloride has been used to preserve foods for millenia. Dairy products and preserved, processed meat products, of course, are commonly preserved with this humectant, therefore the fate of Listeria species in systems containing NaC! is highly relevant. The fact that a,~ levels of many cheese products known to support the growth of L.m. are in the a , ranges (Marcos et al., 1981) discussed in this

328 T A B L E II Effect of u w level on $r(c~vth rates o f L. monocylogcnes a n d L. innocua

Generation time (rain) a w

NaC[

S ucrosc

Glycerol

0.910 0.t)13 0.917 0.921 (}.924 0.929 0,934 0,915 0.920 0.925 0.928 0.932 0.935

0.893 0.897 0.904 0.908 0.911 0.920

L. monocytogenes

L.

innoct//J

_ t

62 56 ND :

83 30

-

I b2 62 ND ND

105 89 6(')

-

-

I 17 59

360 78 25 ND

I no growlh 2 ND, not d o n e .

report suggests that precise control of this parameter can be an important factor in determining the safety of these foods. The minimal a w for growth of L.m. in medium adjusted with sucrose is between 0.925 and 0.920 aw (Fig. 3). Significant extensions in the lag phase took place where counts fell three log cycles over a period of 10 days before recovery and growth occurred. Because our definition of growth stipulated an increase of one log cycle of growth within the 22 day incubation period, a three log loss in count followed by a one log or greater, increase would, by our definition, be regarded as growth. An important conclusion to be drawn from the data in Fig. 4 is that growth, which might be undetected with a method sensitivity of, for example, 102 organisms/ml, could, eventually (after 10-12 days), take place. We conclude that systems poised at near sub-threshold growth levels with sucrose, will eventually support growth of L.m. Water activity levels below the growth threshold ( < 0.915 aw) do not support such growth within the time span (22 days) of these experiments. Whether or not growth will eventually occur below these levels is under investigation at this time. As is the case with NaCI, sucrose concer, trations inhibitory to L.i. are slightly higher than for L.m. The minimal aw level for growth in sucrose-containing systems was found to be between 0.928 and 0.925 aw. In situations in which NaCl

329

2

4

6

8

10 12 DAYS

14

16

18

20

22

Fig. 2. Growth of L. innoclm in NaCI ~lutions: o, 0.910; e, 0.913; z,, 0.917; A, 0.921; 13, 0.924; I1, 0.929; O, 0.934 (Ttyptic Soy Broth with 0.6% yeast extract; 2I~C).

or sucrose are the humcctants, an initial drop in counts occurs in which populations of inoculated L.i. decrease rapidly. This probably is the result of osmotic injury and death of portions of the population which occurs with a decrease in cell turgor following exposure to systems of high osmotic pressure (Epstein, 1986; Booth, 1988). Recovery from this condition occurs in gram-negative cells by the immediate accumulation of intracellular K + and its eventual replacement by compatible solutes. This recovery normally is very rapid unlike the recovery of total populations in the present study at threshhold aw levels, hence physiological adjustment of this type may not b¢ a factor in the observed recovery. A more likely explanation of this phenomenon is the possible selection of osmotically remediated populations that are capable of growth under conditions detrimental to the population as a whole. This hypothesis is the subject of current investigations.

330

7

6.

5q o

f

I

i

i

t

i

2

4

8

8

10

i

12 DAYS

i

t

i

i

i

14

16

18

20

22

Fig. 3. Growth of L. monoCylogenes in sucrose solutions: o. 0.915; e, 0.920; z~, 0.925; A, 0.928 (Tryplic Soy Broth with 0.6% yeast extract; 21"C).

Unlike the situation in sucrose-adjusted media, initial populations in media adjusted with glycerol do not exhibit early decreases in viable cells (Figs. 5 and 6). The most plausible explanation of this is that turgot loss is prevented as a result of the rapid influx of glycerol directly into the cytoplasm, lsoosmotic conditions thus created are proportional to the increase in extracellular concentrations of glycerol. Microorganisms generally are reported by Troller and Christian (1978) to be more tolerant of glycerol than either NaCI or sucrose. An important exception is Staphylococcus aureus which is reported by Christian and Stewart (1973) to grow more rapidly in the presence of NaCi than in either glycerol or sucrose. In the present experiments, the former situation occurred in which both L.m. and L.i. were able to grow at slightly lower aw levels in glycerol-containing media than in either sucrose or NaCI-adjusted systems.

331

7t

2

4

6

8

tO

12 DAYS

t4

t6

18

20

22

Fig. 4. Grtlwth of L. innc~,'lm in sucrose solutions: o , 0.915:. o, (|.920; z~, 0.925; A, 0.928; O, 0.932; I1, a.935 (Twptic Soy Broth with 0.bc/~ yeast extract: 2 I°C).

As is the case with other solutes, there was little difference observed between a , minima for growth of either of the two strains tested when glycerol was the humectant. The minimum a , level for growth of L.m. was 0.911 with no growth occurring following 22 days of incubation at 0.908 aw. Sharp decreases in growth rates immediatcly after inoculation were not obselved in media adjusted with glycerol. Growth rate data are summarized in Table II, however as noted above, growth curves present a more explicit description of the growth sequences in each medium. The minimal aw for growth of L.L in media adjusted with glycerol was between 0.904 and 0.897. Based on these investigations, inhibition as a result of aw limitation in most foods in which L.m. is a potential threat will occur at slightly above 0.92 a . . Our experiments (not reported) show that the presence of 0.6% yeast extract increases

332

o 8

2

4

6

8

10

12 DAYS

14

16

18

20

22

Fig. 5. Growth of L. monocytogenes in glycerol solutions: o, 0.897; e, 0.900; zx, 0.904; A, 0.908; O , 0.911 (Tryptic Soy Broth with 0.6% yeast extract; 2I°C).

the tolerance of both strains to all three humcctants when compared to TSB without yeast extract. It could be expected that similar sources of nutrients, as might occur in food products, could similarly reduce the effect of a . levels on growth. The reason for this stimulation has not been determined, however, yeast extract is known to contain a variety of amino acids and related compounds some of which have been reported (Miller et al., 1991) to simulate the growth of gram-positive organisms in the presence of solutes. Unlike other humectants, glycerol has immediate intracellular access as a result of direct transmittal through the cell membrane. According to Prior and Kenyon (1980), glycerol does not interfere, to any appreciable extent, with major metabolic enzymes hence it acts as an efficient compatible solute, exerting its stabilizing activity immediately upon exposure.

333

o g

>=

I

3-

2.

.

2

.

.

4

.

.

6

,

8

tO

t DAYS

,

1

,

!

,

,

,

,

1

20

22

24

Fig. 6 Growth of L. imuwua in glycerol: o , 0.893; a, 0.897; z,, 0.904; &, 0.908; f-l, 0.911; II, 0.920 (T~ptic Soy Broth with 11.6N yeast extract: 21°C).

The present experiments show that, while not as osmotolcrant as S. aureus, L.m. is one of the most resistant foodborne pathogens to deleterious osmotic conditions and is exceeded in resistance to conditions of high osmolality only by S. a/~re/Js,

Survival rates of both species were approximately the same. Upon exposure to humectants at stated levels, survival rates were lowest in media containing NaCI. These data (Figs. 1-6), however, were obtained at a w levels in the bacterial growth range whereas lower a , ranges are reported by Uzelac and Stille (1977) to increase survival, it is especially noteworthy that the survival curve of L.m. in sucrose solutions drops characteristically with time for as long as 10 days whereupon it begins to climb as growth eventually occurs.

3.'M

References Braekett, R.E. (1987) Sodium benzoate. J. Food Prot. 50. 999-101L';, 1008. Booth, I.R. Cairney J., Snlherland L. and Higgins C.F. (1988) Enteric bacteria and osmotic stress: an integrated homeostatic system. J. Appl. Bacteriol. Symp. Suppl. 355-495. Chirife, J. and Resnik S.L (1984) Unsalurated solutions of sodium chloride as reference sources of water activity at various temperatures. J. Focal Sci. 49, 1486-1488. Christian, J.H.B. and Stewart. B.J, (1973) Survival of Staphyfococcu,r aurcus and Salmonella newport in dried foods, as influenced by water activity and oxygen. Microbiol. Safely Food Proc. Int. Syrup. Food Microbiol.. 8lb. 1972. pp. 107-119, Christian, J.H.B. and Waltho J.A, (1962) Solute concentrations within cells of halophilic and nonhalophilic bacleria. Biochim. Biophys. Acta 65, 506-508. Cole, M.B., Jones M.V. Ilolyoak C. (1990) The effect of p|t, sail concentration and 4t.mperature on the survival and growth of LLvtetia monocymgenes. J. Appl. Bacleriol. 69, f~3-72. Conner, D.E., Brackett R.E. and Ikuchat L.R. 0986) Effect of temperature, sodium chloride and pH on growth of Listeriu monocyrogenes in cabbage juice. AppL Environ. Microbiol. 52, 59-63. Dallyn, It. (1978,) The effect of substrate water activity on the growth of certain xerophilie fungi. Ph.D. Thesis, CNAA Polytechnic of Ihe South Bank, ~mdon. EI-Shenawy, M.A. and Marth E.H. (1988a) Sodium benzoate inhibits growth of or inactivates Listeria moma'yiogenes. J. Food Prof. 51,525-539. EI-Shenawy, M.A. and Marth E,H. (1988b) Inhibition and inactivation of Lister'ia monoc,ytogenes by ~rbic acid. J. Food Prol. 51,842-847. Epstein, W. (1986) Osmoregulation by potassium translxlrt in Escheriehia coil FEMS Microbiol. Revs, 39: 73-78, Gifford, H., Lorenz, K, and Solos. J. (1991) Analysis of bakery prt~ucls for presence of Listeria mom~'ytt~gem, s. Lehenson-Wiss. Technol. 24. 476-477. Grover, D.W. a~id Nicol J.M. (1940) The vapor pressure of glycerine solutions at 21rC. J. Soc. Chem. (London) 59. 175-177. Lovett, J., Bradshaw, J.G., Francis. D.W., Crawford, R.G.. Donnclly, C.W., Murlby. G.K. and Wesley, I.V. (1988) Efficacy of high temperature short lime pasteurization for inactivation of Liswria motltic'ytogent, s in milk. Abstracts 75th Meeting Int. Ass¢~. Milk F~v..I Environ. Sanitarians. Tampa, FL. Lovett, J. (1989) Lisleria monocytogenes. In: M.P. D~rylc (Ed.), Fotxiborne Bacterial Pathogens, Marcel Dekker, Inc., New York, pp. 290-291. Marcos, A., Alcala M., Leon F. Fernandez.Salguerro J. and Estaban M.A, (1981) Water activity and chemical composition of cheeses. J. Dairy Sci. 64:622-626 McLure, P.J., Roberts, T.A. and Otto Agura, P. (1989) Comparison of the c[fcc:~ ~f ~tuJium chloride, pH and temperature on the grt~vlh of Listeria monocytogene.~ on gradient plates and in liquid medium. Lett. Appl. Microbiol. 9, 95-99. Miller, KJ., Zeit, S.C. and Bat, J.H. (1991) Glycine betaiuc and proline are the principal eompatible solutes of Staphyloc'~u'cus aureus. Curr. Microbiol. 23, 131-137. Pelran, R.L. and Zt)ttola E.A. (1989) A study of factors affecting grt~,vth and rec~wery of Lisreria mom~'ytogenes Scott A. J. Food Sci, 54, 458-460. Papagcorgiou D.K. and Marth E.H. (1989) Behavior of L£s'wria monewytogenes al 4 and 22°C in whey and skim milk containing 6 or 12% sodium chlori~ , J. F o ~ Plot. 52, 625-630. Prior, B.A. and Kenyon, C.P. (1980) water relations of glucose: catabolizing enzymes in Pseudomonas fluorescens. J. Appl. Bacteriol. 48, 211-225. Ross, K.D. (1975) Estimation of water activity in intermediate moisture fotvJs. Fix)d TcchnoL 29, 26-30. Seeliger, tI.P.R. (1961) Listeriosis. llaffner, New York. Shahamat, M., Seaman A. and Woodbine M. (1980) Survival of Lis~eria monocytogenes in high salt concentrations. Zbl. Bakl. Hyg. I. Abt. Orig. A 246, 506-511.

335 Troller, J.A. (1975) Influence of water activity on growlh and entcrotoxin formation by Staphylococclls aureus in foods. J. Food Sci. 40, 802-804. Troller, J.A. and Christian, J.H.B. (1978). Water Activity and F~x)d. Academic Press, NY. Uzelae, G. and Stille B. (1977) Survival of bacteria of faecal origia in d~J foods, in relation to water activity. I1. Studies on Eschenchia colt and Streptococcus faecalis. Deutsch Lebensm.-Rundsch. 73. 325-329.