Acid adaptation and food poisoning microorganisms

ELSFXIER International Journal of Food Microbiology 28 (1995) 245-254 he marionol Joumal ofFoodMicrobidogy Acid adaptation and food poisoning microorganisms Colin Hill a, * , Brid O’Driscoll a, Ian Booth b a National Food Biotechnology Centre and Department of Microbiology University College Cork, Cork, Ireland ’ Department of Molecular and Cell Biology, Marischal College, University of Aberdeen, Aberdeen, UK Keywords: Acid tolerance response; Pathogen; Food 1. Introduction The incidence of foodbome disease is increasing yearly, despite the growing body of information regarding the most common foodbome pathogens. One of the most fruitful research themes of recent years has been the discovery that pathogenic organisms are not limited to cardinal ranges of temperature, pH and water activity, but can adapt to survive at values outside of those given in textbooks. This communication will discuss the ability of pathogenic organisms to adapt to and tolerate low pH. This is an important topic which should be of concern to both food processors and clinicians, since organic acids are frequently used in foods to inhibit the growth and survival of undesirable organisms, and acid is also one of the barriers employed by the body to defend itself against microbial attack. Organic acids are intrinsic to some foods or they may be deliberately added as preservatives or can be present as a consequence of microbial fermentation processes. The extent of the preservative or protective effect will depend on the nature of the acid and the final pH of the food, among other environmental factors. If the intracellular pH of an organism was in equilibrium with the external pH, microbes would be unable to function outside of a very narrow pH spectrum close to neutrality. However, many pathogenic bacteria will only cease growth below pH 4.5, while lower pH values will cause cell death through the disruption of the pH homeostasis. Bacteria can only survive in acidic environments because of their ability to regulate their cytoplasmic pH (pH,), a process primarily driven by the controlled movement of cations across the membrane. However, this ability to * Corresponding author. Tel. + 353 (21) 902397; Fax + 353 (21) 903101; e-mail [email protected] 0168-1605/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDI 0168-1605(95)00060-7 246 C. Hill et al. /ht. J. Food Microbiology 28 (1995) 245-254 maintain the pHi close to neutrality can be overwhelmed at low extracellular pH values, leading to the death of the cell. The acid tolerance response (ATR) describes the phenomenon in which bacteria which have been exposed to mildly acidic conditions have been shown to acquire the ability to survive normally lethal pH values. The physiological changes involved in this response are not fully understood, but it is increasingly apparent that this adaptive phenomenon could have important consequences for the survival of spoilage and pathogenic bacteria in acid foods. Since the low pH of the stomach and the initial drop in pH experienced in the phagosome are important body defence mechanisms, the ability to tolerate low pH could have profound implications for the virulence of invasive foodborne pathogens. Much of the work in this area has focused on Escherichia coli (Goodson and Rowbury, 1989) and Salmonella (Foster, 1993b), and more recently on Listeria monocytogenes @roll and Patchett, 1992) and Aeromonas hydrophila (Karem et al., 1994). 2. Physiological basis of pH homeostasis Before discussing adaptive tolerance it is useful to briefly consider the physio- logical means by which bacteria control pH,, i.e. preserve the pH homeostasis (Booth, 1985). Most foodborne pathogenic bacteria are neutrophiles (optimum pH 6-7). The major source of acid-related stress likely to be encountered by these bacteria in food systems is the presence of lipid-permeable weak acids. Although the cell membrane is intrinsically impermeable to protons, undissociated weak acids pass through the membrane with relative ease and may dissociate to liberate protons in the cytoplasm. Consequently if the transmembrane pH gradient is 2 units or more (i.e. pH, of 7.0, external pH 5.0) then the internal concentration of the weak acid can be more than lOO-fold greater than the external concentration. For example, if the external acetate concentration is 1 mM, the internal concentra- tion would theoretically be of the order of 100 mM, if the cell could maintain a pHi of 7.0. This would result in the release of approximately 100 mM protons into the cytoplasm. In practice, this release of protons would outstrip the buffering capacity of the cell and in consequence the cytoplasmic pH would fall leading to growth inhibition and ultimately to death. pH homeostasis describes the ability of an organism to maintain the cytoplasmic pH at a value close to neutrality despite fluctuations in the external pH. For example, in Escherichia coli the cytoplasmic pH (pHi) changes by less than 0.1 unit per pH unit change in external pH in the range of external pH 4.5-7.9. Homeosta- sis is achieved in bacterial cells by a combination of passive and active mechanisms. 2.1. Passive homeostasis The very low permeability of the membrane to protons and other ions plays a major role in preventing large changes in pH, as the pH of the environment is varied. Thus, protons present in the environment cannot simply traverse the cell C. Hill et al. /ht. J. Food Microbiology 28 (1995) 245-254 247 membrane and reduce the internal pH. This lack of permeability can only be overcome by undissociated lipid-permeable weak acids, or if the cell is treated with an ionophore. A second major factor preventing substantial disruption of pH, is the high buffering capacity of the cell, which results both from the protein content of the cytoplasm and from cytoplasmic pools of glutamate and polyamines. Maximum buffering capacity is found at the extremities of the pH range. There is little evidence that microorganisms can specifically alter their cytoplasmic constitution with the objective of increasing the buffering capacity. However, some mutants of Salmonella typhimurium that possess enhanced survival at acid pH replace part of the glutamate pool with citrate and isocitrate (pK 6.4) due to mutations that affect isocitrate dehydrogenase. The enhanced buffering capacity of the cytoplasm in the neutral range is believed to be an important aspect of the survival of these mutants (Foster and Hall, 1991). 2.2. Active pH homeostasis Active pH homeostasis depends primarily on the potassium ion and proton circuits. In addition, the Na+ circuit is very important in the alkaliphiles (pH optimum 8-9) but plays a more limited role in acidophilic and neutrophilic bacteria. The potassium ion and proton circuits are important for all bacterial species. A cell at low external pH must extrude those protons entering the cell associated with weak acids. However, the translocation of protons across the membrane generates a membrane potential that limits further proton extrusion. Large scale proton movement can only take place if the membrane potential is dissipated through the movement of cations into the cell. Potassium ion entry fulfils this purpose and leads to the generation of a transmembrane pH gradient (Kroll and Booth, 1981). Over the course of each doubling time bacteria must accumulate sufficient potassium to maintain a constant concentration of cations as the cell grows in volume and subsequently divides. Associated with this potassium uptake will be net proton extrusion, which can cause the cytoplasmic pH to become more alkaline. The situation is a little more complicated with alkaliphiles, bacteria which require a lower pH, than the external pH, but that specific instance will not be considered here. 3. Bacterial adaptation to acid stress Any change in environmental conditions from the optima towards extreme conditions imposes a stress on an organism. The degree of change will determine whether the organism is killed, ceases growth or simply grows at a reduced rate. The majority of bacteria have the capacity to withstand small changes in an environmental parameter and can adapt over the time scale of minutes, hours or days. Larger changes away from the optimal values can cause the induction of more elaborate stress responses. These adaptive strategies are generally directed 248 C. Hill et al. /ht. J. Food ~Uicrobiology 28 (1995) 245-254 0 20 40 60 80 100 Time/min Fig. 1. Induced acid tolerance in Livteria morwcytogenes LOB. The open circles represent those cells transferred directly from pH 7.0 to pH 3.5. The closed circles represent the cells which have been induced at pH 5.0 for 90 min. towards survival rather than growth. An example of a typical acid adaptation is presented in Fig. 1. In this instance the ability of L. monocytogenes LO28 to survive at pH 3.5 is dramatically enhanced by prior induction for 90 min at pH 5. No growth is observed for either culture, but the adapted cells are much more likely to survive the exposure to low pH. This discussion is solely concerned with the ability of organisms to adapt to changes in pH, but it is important to note that this adaptive phenomenon cannot be viewed in isolation. For example, a number of laboratories have reported that induction of acid tolerance can also induce other stress responses, including resistance to thermal stress, osmotic stress, and to the action of surface-active agents such as crystal violet (Farber and Pagotto, 1992; Leyer and Johnson, 1993). Equally, other stresses can invoke acid tolerance. This most probably reflects two facets of stress adaptation; firstly, there are a limited number of basic systems for eliciting gene expression and therefore, cross-induc- tion of systems that respond to the same signals will occur. It is also possible that those proteins induced by one stress would be capable of protecting cells against other environmental challenges. For example, chaperonins may be induced which would prevent the denaturation of vital enzymes by low pH or by thermal denaturation. Secondly, severe stress will reduce the growth rate and this could possibly activate a number of global responses. Initially, it is presumed that the cell possesses relatively low activity of the ‘stress response system’. Imposition of stress leads to a higher level of the stress response system relative to that of the initial state. However, if this initial response cannot re-establish homeostasis, novel gene expression may continue and further genes may be expressed. Generally as a stress persists the genes expressed are for pathways that are increasingly costly solutions to the cells’ problems. In stress stimulons a hierarchy is generally found in which C. Hill et al. /ht. J. Food Microbiology 28 (1995) 245-254 249 Time/hours Fig. 2. Effect of growth phase on resistance to pH 3.5. The growth curve is represented by the open circles, while the % survivors after exposure to pH 3.5 for 60 min is indicated by the closed circles. genes encoding the most beneficial solution are expressed when the stress is mild, but the genetic systems for more drastic remedies are activated as the severity of the stress increases. It is also noteworthy that the ability to tolerate stress is growth phase depen- dent. This can be illustrated once again using L. monocytogenes LO28 as a model system (Fig. 2). Cells grown overnight (> 16 h) at pH 7.0 are tolerant to challenge at pH 3.5, although the pH of the culture remains at pH 7.0 during this extended growth period. Upon inoculation into fresh broth growth is initiated within 60 min. The acid tolerance is rapidly lost in early logarithmic growth, until a maximum sensitivity is observed during mid-logarithmic growth. As the culture proceeds to the end of the exponential phase and into the stationary phase the tolerance gradually returns. A similar phenomenon has been noted in respect of other stresses and with other genera. 4. Genetic control of ATR Foster and his co-workers (Foster, 1991, 1993a; Foster and Hall, 1990, 1991; Foster et al., 1994) have examined acid tolerance in S. typhimurium, and have concluded that it involves a two stage process. These two stages have been described as the pre-acid shock (induced at an external pH of approximately pH 5.8) and post-acid shock (induced at or below an external pH 4.0) phases. During post-acid shock a number of acid-shock proteins (ASPS) are produced which are important for the subsequent survival of the organism. If a cell is transferred directly from pH 7.0 to pH 3.3 these ASPS are not synthesized and the cell will be killed. If, however, the pre-acid shock at pH 5.8 is performed then the cell will be 250 C. Hill et al. /ht. J. Food Microbiology 28 (1995) 245-254 able to maintain pH homeostasis long enough at pH 3.3 to allow synthesis of the ASPS. Thus, both phases are required for maximum protection against low pH. Two-dimensional protein analysis has revealed that during pre-acid shock the concentrations of 18 polypeptides are affected, 12 of which are induced and 6 repressed. These changes are manifested physiologically by an inducible pH homeostasis system which helps to maintain the pH, above pH 5.5 at low external pH ( < 4.0) and thus minimise denaturation of acid sensitive proteins. The induc- tion of these pre-acid shock proteins also allows the synthesis of the post-acid shock proteins to occur at low pH. However, the pre-acid shock stage can be substituted by a brief shock at pH 4.4 for 15 min, during which a transiently transcribed group of ASPS are detectable (Foster, 1993aLK@ cell is held at pH 4.4 for 30 min these proteins will have disappeared and the cells will no longer be able resist a subsequent challenge at pH 3.3. Acid shock results in the induction of at least 43 proteins, some or all of which contribute to survival in high acid conditions. A number of the genes encoding these proteins have been identified recently using Mud-Zuc operon fusion analysis to target pH-regulated genes (Foster et al., 1994). This study has demonstrated the presence of six regulatory circuits in S. typhimurium involving at least five regulatory loci, including fur (ferric uptake regulator), oxrG (oxidative stress regulator), ompR (Outer mem- brane porin regulator), eurAG and earC (external acid regulators). The role of the fir locus has been determined to be independent of iron, since the induction of the ATR is unaffected by iron availability (Foster and Hall, 1992). The fur mutants are acid-sensitive and it has been suggested that the fur gene product may be involved in the regulation of a number of acid-regulated genes. It has also proven possible to isolate acid-tolerant and acid-sensitive mutants of L. monocytogenes, both spontaneously and after mutagenesis with acridine orange. The acid tolerant mutants are fully tolerant throughout all stages of the growth cycle, and display extended survival in acid foods and higher virulence in mouse assays (see below). Genetic analysis of these interesting mutants is in progress. 5. Industrial significance of the ATR response Perhaps the most well known example of the industrial significance of ATR in a foodbome pathogen was detailed in a paper by Leyer and Johnson (19921, in which it was shown that acid-induced SulmoneZlu was capable of prolonged survival in cheese as compared with the non-induced parent. The Salmonella strains used in the study were acid-induced by adjusting the pH of the growth medium to pH 5.8 with HCI for one to two doublings. Induced and non-induced cells were surface inoculated onto Cheddar (pH 5.2), Swiss (pH 5.6) and mozzarella (pH 5.3) cheeses which were subsequently stored at 5°C. Under aerobic storage conditions it could be demonstrated that the non-induced Salmonella were rapidly inactivated and were completely absent from the Cheddar cheese within 35 days. However, while there was an approximately 99% reduction in the initial load, the adapted cells were detectable after prolonged storage (74 days). Similar results were obtained C. Hill et al. /ht. J. Food Microbiology 28 (1995) 245-254 2.51 0 1 2 3 4 5 6 Time/days Fig. 3. Survival of Listeria monocytogenes LOB and an acid tolerant mutant in yoghurt (pH 4.3) incubated at 4°C. LO28 and the mutant are represented by closed and open circles, respectively. with mozzarella cheese, but the results were less dramatic with Swiss (perhaps due to the presence of additional inhibitors in this cheese type). The same authors also demonstrated that acid-adapted Salmonella strains can survive during a dairy fermentation better than a non-induced parent strain. Milk was fermented to a final pH 4.27 using a combination of Streptococcus ther- mophilus and Lactobacillw helveticus. S. typhimurium was added when the fermen- tation had reached pH 5.1 and the survival was monitored for 5 h. A reduction was observed for both induced and non-induced strains, but after 5 h a lOOO-fold difference in cell numbers was observed, suggesting that acid-induced cells are more adapted to survive during a dairy fermentation. We have also observed that acid tolerant strains of L. morwcytogenes are better adapted to survive in fermented dairy products. L. monocytogenes 22 is an acid tolerant mutant of strain LO28 isolated after prolonged exposure to pH 3.5 in the absence of induction. The mutant and non-induced LO28 were inoculated into yoghurt (pH 4.3) and stored at 4°C. The results are presented in Fig. 3. It can be seen that the mutant displayed a higher survival rate than the parental strain. While growth does not occur with either induced or mutant strains in the experiments described above, it remains to be seen what role if any the induction of acid tolerance, or the selection of acid tolerant mutants, might play in the transmission of foodborne illness. For those bacteria with a low infectious dose, survival of a small number of bacteria may be sufficient to cause disease. Altema- tively, pathogenic bacteria may survive until such a time as conditions change in a manner to favour growth. For example, the contaminated food may be used as an ingredient for the manufacture of other foods, or in the preparation of a meal, or 252 C. Hill et al. /ht. J. Food Microbiology 28 (1995) 245-254 Table 1 Virulence of Listen% monocytogenes LO28 and an acid tolerant mutant Inoculum Strain CFU/spleen at 3 days (average of 4 mice/assay) 1x105 LO28 2.21 x 104 Mutant 8.25x 106a 1x104 LO28 2.12x 104 Mutant 3.55 x 105 a Two of the four mice died within 3 days. the pH might rise due to growth of other organisms (as is the case in surface ripened cheeses and in fermented meats). 6. A role for ATR in the virulence of L. morwcytogenes Many authors have speculated that the ATR may play a role in virulence (Foster, 1993b), particularly with invasive organisms such as Lbteriu and Sulmonelfu, which will encounter at least two acidic barriers en route to a successful infection; the acidic environment of the stomach, and the low pH encountered after ingestion by macrophages. We investigated this possibility using acid tolerant mutants of L. monocytogenes M28. The spontaneous acid tolerant mutants were recovered after prolonged exposure to low pH without prior induc- tion. These mutants exhibited tolerance to low pH at all stages of the growth cycle, and in the absence of further induction. The virulence of one such acid tolerant mutant was assessed using a mouse model. Two groups of four mice were injected intraperitoneally with either the parent or mutant strains in separate experiments, using two different levels. After three days the mice were sacrificed and the number of L. monocytogenes in the spleen was estimated by direct plating on listeria selective agar. Significantly higher counts were recorded from the spleens of mice infected with the acid tolerant mutant as compared to the parental strains at both inoculum levels (Table 1). At the higher inoculum (lo5 cells/mouse) two of the four mice infected with the tolerant mutant died within the three days, whereas none of the mice infected with the parental strain showed overt symptoms of infection, thus confirming the enhanced virulence associated with the mutant. Since this mutant was spontaneous, and the nature of the underlying mutation remains unknown, it would be unwise to unequivocally implicate their acid toler- ance as the sole cause of the increased virulence. However, the initial results are exciting, and we are continuing our work in this area. 7. Conclusions Outside of the laboratory cells are forced to deal with hostile environments which differ in many respects from their controlled intracellular environments. The C. Hill et al. /Int. J. Food Microbiology 28 (1995) 245-254 253 ability of the cell to maintain the difference between internal and external conditions will dictate its persistence in the type of complex milieu found in foods. The objective of the food processor is to overwhelm the cells’ ability to withstand these hostile environments, while employing a minimum of energy wasting process- ing steps. Thus, the extent to which bacteria can withstand stress is an important factor to be considered when attempting to provide effective barriers (hurdles) to prevent spoilage or safety problems (Leistner and Gorris, 1995). This is particularly important when it is noted that imposition of one stress may lead to the induction of multiple stress responses. This is extremely significant in the light of the recent developments in minimal processing, in which low levels of different stresses are employed rather than a single intensive processing step. If these stresses are employed in sequence it is possible that a protective effect could be induced which would reduce the efficacy of later treatments. Pathogenic bacteria, particularly those which invade the body, must also be able to cope with the stresses imposed by body defence systems. These will include at least two acid-based defences in the stomach and after uptake by macrophages. Therefore, it is also possible, and preliminary evidence described in this communication would support the theory, that induction of tolerance would impact on the virulence of a pathogenic invasive organism. Much remains to be done in determining the extent of tolerance which can be developed by pathogenic and spoilage bacteria, and the underlying control of the stress response. Acknowledgements The authors thank the AAIR Concerted Action PL920630 for support. References Booth, I.R. (1985) Regulation of cytoplasmic pH in bacteria. Microbial. Rev. 49, 359-378. Kroll, R.G., and Booth, I.R. (1981) The role of potassium transport in the generation of a pH gradient in Escherichia coli. Biochem. J. 198, 691-698. Farber, J.M. and Pagotto, F. (1992) The effect of acid shock on the heat resistance of Listeria monocyfogenes. Lett. Appl. Microbial. 15, 197-201. Foster, J.W. (1991) Salmonella acid shock proteins are required for the adaptive acid tolerance response. J. Bacterial. 173, 6896-6902. Foster, J.W. (1993a) The acid tolerance response of Salmonella typhimurium involves transient synthesis of key acid shock proteins. J. Bacterial. 175, 1981-1987. Foster, J.W. (1993b) Beyond pH homeostasis: the acid tolerance response of samonellae. ASM News 58, 266-270. Foster, J.W. and Hall, H.K. (1990) Adaptive acidification tolerance response of Salmonella ty- phimurium. J. Bacterial. 172, 771-778. Foster, J.W. and Hall, H.K. (1991) Inducible pH homeostasis and the acid tolerance response of Salmonella typhimurium J. Bacterial. 173, 5129-5135. Foster, J.W. and Hall, H.K. (1992) Effect of Salmonella typhimurium ferric uptake regulator (fur) mutations on iron- and pH-regulated protein synthesis. J. Bacterial. 174, 4317-4323. 254 C. Hill et al. / Int. J. Food Microbiology 28 (1995) 245-254 Foster, J.W., Park, Y.K., Soo Bang, I., Karem, K., Betts, H., Hall, H.K. and Shaw, E. (1994) Regulatory circuits involved with pH-regulated gene expression in Salmonefla typhimurium. Microbiology 140, 341-352. Goodson, M. and Rowbury, R.J. (1989) Habituation to normally lethal acidity by prior growth of Escherichia coli at a sub-lethal acid pH value. Lett. Appl. Microbial. 8, 77-79. Karem, K.L., Foster, J.W. and Bej, AK. (1994) Adaptive acid tolerance response (ATR) in Aeromonas hydrophila. Microbiology 140, 1731-1736. Kroll, R.G. and Patchett, R.A. (1992) Induced acid tolerance in Listerio monocytogenes. Lett. Appl. Microbial. 14, 224-227. Leistner, L. and Gorris, L.G.M. (1995) Food preservation by hurdle technology. Trends Food Sci. Technol. 6, 41-46. Leyer, G.J. and Johnson, E.A. (1993) Acid adaptation induces cross-protection against environmental stresses in Salmonella typhimurium. Appl. Environ. Microbial. 59, 1842-1847. Leyer, G.J. and Johnson, E.A. (1992) Acid adaptation promotes survival of Sabnonelfo spp. in cheese. Appl. Environ. Microbial. 58, 2075-2080.