Acid-Sensitive Enteric Pathogens Are Protected from Killing under Extremely Acidic Conditions of pH 2.5 when They Are Inoculated onto Certain Solid Food Sources

Abstract

Gastric acidity is recognized as the first line of defense against food-borne pathogens, and the ability of pathogens to resist this pH corresponds to their oral infective dose (ID). Naturally occurring and genetically engineered acid-sensitive enteric pathogens were examined for their ability to survive under acidic conditions of pH 2.5 for 2 h at 37°C when inoculated onto ground beef. Each of the strains displayed significantly high survival rates under these normally lethal conditions. The acid-sensitive pathogens Campylobacter jejuni and Vibrio cholerae, which were protected at lower levels from acid-induced killing by ground beef under these conditions, were sensitive to killing in acidified media at pH 5.0 but survived at pH 6.0. Salmonella inoculated onto the surface of preacidified ground beef could not survive if the pH on the surface of the beef was 2.61 or lower but was viable if the surface pH was 3.27. This implies that the pH of the microenvironment occupied by the bacteria on the surface of the food source is critical for their survival. Salmonella was also shown to be protected from killing when inoculated onto boiled egg white, a food source high in protein and low in fat. These results may explain why Salmonella species have a higher oral ID of approximately 10⁵ cells when administered under defined conditions but have been observed to cause disease at doses as low as 50 to 100 organisms when consumed as part of a contaminated food source. They may also help explain why some pathogens are associated primarily with food-borne modes of transmission rather than fecal-oral transmission.

The low pH of gastric secretions has long been recognized as the first line of defense against food-borne enteric pathogens (13, 28). The ability of enteric bacteria to resist killing by acid during transit through the stomach increases their likelihood of colonizing the intestines and causing an infection. The infective dose (ID) of different enteric pathogens corresponds to their relative abilities to resist killing by acid (23). The ID of Vibrio cholerae, nontyphi Salmonella species, and Shigella flexneri are approximately 10⁹, 10⁵, and 10², respectively (5). These doses correspond with the relative levels of acid resistance of these pathogens, with V. cholerae being the least resistant and S. flexneri being the most resistant (23).

Enteric microorganisms have evolved several mechanisms for handling acid stress. Escherichia coli O157:H7 and S. flexneri, which both have a low ID, can survive extreme acid conditions of pH 2.5 or less for a number of hours in vitro (2, 4, 15, 23, 34, 35). This acid resistance is induced in stationary phase or under starvation conditions and is dependent upon the alternate sigma factor, ς^s, encoded by rpoS (7, 31, 34, 35). ς^s regulates a set of genes which may enable the cell to transport glutamate from the acidified media to the interior of the cell, where it is converted by glutamate decarboxylase to γ-amino butyric acid. This basic amine presumably provides a buffering effect and maintains the internal pH homeostasis of the cell (35). S. flexneri is commonly transmitted by person-to-person spread via the fecal-oral route and by waterborne infections, but it can also be transmitted via a contaminated food source. Although E. coli O157:H7 is usually a food-borne pathogen, person-to-person spread can also occur and is a significant cause of secondary infections (3). The acid resistance of E. coli O157:H7 is not surprising, since most E. coli strains, as members of the normal fecal flora, are very acid resistant (15).

S. typhimurium possess two stationary-phase acid tolerance response systems, one that is acid induced and ς^s independent and one that is unresponsive to pH but ς^s dependent (21). This inducible acid tolerance system is important to the virulence of the organism (12) but does not allow Salmonella to survive at the extremes of pH (1.5 to 2.5) that are tolerated by the stationary-phase acid resistance phenotype of S. flexneri and E. coli (15). Yersinia enterocolitica can also tolerate a pH less than 1.5 in vitro. This tolerance is dependent upon a cytoplasmic urease induced in the stationary phase and during cytoplasmic acidification (10, 36).

Despite the amount of information derived from studies using in vitro acid resistance assays, the minimum number of ingested Salmonella organisms necessary to produce clinical symptoms in humans remains a controversial issue. Earlier work on experimental human salmonellosis involving volunteers from a penal institution showed that ingestion of greater than or equal to 10⁵ CFU of Salmonella meleagridis or Salmonella anatum was required to produce illness (24). Results with other Salmonella species found similar requirements for infection (18, 25, 26). In contrast, the infectivity of Salmonella in pharmaceutical preparations and foods including pancreatin, oral vaccine, water, hamburger, milk chocolate, and cheddar cheese was found to be less than 10³ CFU (5). There have been other reports of ID of less than 100 CFU of Salmonella eastbourne (9) or 50 CFU of Salmonella napoli (16) in chocolate and of 100 to 500 CFU of Salmonella heidelberg (11) or 1 to 6 CFU of S. typhimurium (8) in cheddar cheese (8). These results suggest that Salmonella species can cause infection at a much lower ID if they are ingested with a food source, which implies that food may provide protection against the acidity of the stomach for this pathogen.

The acidity of the human stomach is dependent on physiological variables that include previous food intake. Under fasting conditions, the median of the luminal pH in healthy volunteers is around 2.0, ranging from 1.5 to 5.5 (33). Ingestion of a meal characteristic of a Western diet produces an immediate rise in the median gastric pH to about 6.0 (20, 32). Reduction of gastric acidity has been associated with an increase in the survival rates of some food-borne pathogens (28) and with a lowering of the ID (6, 30).

In this study we have shown that pathogens classified as acid sensitive in an in vitro acid resistance assay can survive under the same acidic conditions if inoculated onto certain food sources. To further test this hypothesis, we used a genetically engineered acid-sensitive rpoS deletion mutant of Shigella flexneri as an acid-sensitive control.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The strains of pathogenic enteric bacteria and their isogenic derivatives used in this study are described in Table 1. The acid-resistant S. flexneri M25-8A and its acid-sensitive isogenic rpoS mutant W422 were from our laboratory collection and have been described previously (35). E. coli O157:H7 strain PS2 is a naturally occurring rpoS mutant which is restored to its acid-resistant state when complemented by pPS4.4 (containing rpoS) (34). All strains used in this study were transformed with pACYC184 (Cm^r) as a selectable marker unless stated otherwise. This has no effect on the viability of the organisms or their ability to resist killing by acid. Campylobacter jejuni 81116 GRK1 contains a kanamycin cartridge inserted in the flaA gene and was obtained from C. Grant. Strains were grown in 5 ml of Luria-Bertani broth (LB) with chloramphenicol (30 μg/ml) at 37°C for 24 h with shaking. C. jejuni was grown for 48 h at 37°C in 5% CO₂ on LB plates containing 5% sheep erythrocytes and kanamycin (30 μg/ml).

TABLE 1.

Bacterial strains used in this study

Strain	Relevant genotype	Source or reference
S. flexneri M25-8A	Cured of large plasmid	29
S. flexneri W422	rpoS deletion mutant of M25-8A	35
S. typhimurium SL1344	Wild type	17
S. typhi ISP 1820	Wild type	19
E. coli O157:H7 PS2 (rpoS)	Natural isolate with rpoS mutation	34
E. coli O157:H7 PS2 (rpoS+)	rpoS+, contains pPS4.4 (pACYC184 with 4.4-kb ClaI fragment with rpoS, Cmr)	34
C. jejuni 81116 GRK1	flaA::Kmr	Chris Grant
V. cholerae 14035	Wild type	ATCCa

^aATCC, American Type Culture Collection. 

Acid resistance assays.

The ability of enteric bacteria to survive killing by acid when inoculated onto a food source was tested by using ground beef, boiled egg white, and boiled rice. Ground beef (9% fat) was purchased at a local supermarket. Samples (0.1 g) of the ground beef were weighed and inoculated with 10 μl of an overnight (o/n) culture of the appropriate bacteria diluted 10⁻¹ (approximately 10⁶ CFU) in phosphate-buffered saline (PBS), and the bacteria were allowed to dry on the surface of the beef for 10 min at room temperature. For inoculations with C. jejuni, the bacteria were harvested from a blood plate and resuspended in 1 ml of PBS (pH 7.4) prior to dilution. The inoculated beef was then placed into 10 ml of LB acidified to pH 2.5 with HCl and incubated at 37°C with gentle shaking (100 rpm) for 2 h in a 20-ml plastic tube (Sterilin). Following incubation, the acidified medium was decanted from the beef fragments and its pH was measured. Surviving bacteria were recovered by extracting the ground beef with 10 ml of PBS followed by vigorous vortexing. Appropriate aliquots were taken from the resuspended samples, plated onto MacConkey lactose agar containing chloramphenicol (30 μg/ml), and incubated o/n at 37°C unless stated otherwise. Surviving bacteria were enumerated and expressed as a percentage of the original inoculum exposed to acid challenge. An uninoculated ground-beef control was placed in acidified LB under identical conditions and examined for the presence of any contaminating bacteria. Ground-beef samples inoculated with S. typhimurium or S. flexneri were placed in 10 ml of PBS under identical conditions to those used for acid challenge as a control to determine the percent cell recovery obtained from untreated ground beef. This indicates the efficiency of the extraction procedure. Samples taken from ground beef inoculated with C. jejuni were plated onto LB agar containing 5% sheep erythrocytes and kanamycin and incubated as described above. Samples taken from ground beef inoculated with V. cholerae were plated onto LB containing chloramphenicol.

For experiments involving the direct acidification of ground beef, 0.1 g of the beef was acidified with HCl to pH 1.0. Ground beef was also acidified by being soaked in LB (pH 2.5) for 1 h or o/n where stated and then placed in 10 ml of fresh LB (pH 2.5) for 10 min before being inoculated with bacteria. The pH at the surface of the acidified ground beef was determined after treatment. Acidified ground beef was inoculated with bacteria for 2 h at room temperature, and the surviving bacteria were recovered by extraction with 10 ml of PBS as described above. In some experiments, 0.05 g of boiled rice grains and 0.1 g of boiled egg white were used as a food source instead of ground beef and were inoculated with bacteria in the same manner as ground beef. In vitro acid resistance assays of freely suspended cells were performed as described previously (34) with LB acidified with HCl to pH 2.5, 4.0, 5.0, and 6.0, and the results are given as the percentage of the number of bacteria surviving compared with the original inoculum exposed. Results are expressed as the mean of duplicate samples from a typical experiment. Larger amounts of uninoculated ground beef were used in some experiments under identical conditions, and the pH of the acidified media was measured after incubation, as described above.

RESULTS

Survival of enteric pathogens inoculated onto ground beef under acidic conditions.

Our studies show that all the enteric pathogens tested were able to survive at pH 2.5 when inoculated onto ground beef (Table 2) even though many of these isolates could not survive when assayed in acidified LB at the same pH (Table 2). Even sensitive rpoS mutants of S. flexneri and E. coli O157:H7, as well as S. typhimurium, S. typhi, V. cholerae, and C. jejuni, were able to survive when inoculated onto ground beef (Table 2). Measurement of the pH of the acidified LB used to challenge each pathogen revealed that the addition of ground beef did not alter the pH of the medium (Table 2). This is in contrast to results obtained with larger amounts of ground beef in which the pH of the medium was raised significantly (see Table 4). A control ground beef sample which was not inoculated was treated at pH 2.5 under identical conditions and did not harbor any contaminating chloramphenicol- or kanamycin-resistant bacteria (Table 2). The percent survival rates were significant for all the acid-sensitive strains tested, compared to their ability to survive as freely suspended cells in acidified LB (Table 2). The survival (recovery) rates of control samples where S. typhimurium and S. flexneri were inoculated onto ground beef but challenged in PBS instead of acidified LB (Table 2) were measured to gauge the efficiency of the extraction procedure and were found to be 100.00 and 66.52%, respectively.

TABLE 2.

Survival of pathogens inoculated on ground beef and challenged in acidified LB^a

Pathogen	% Survivalb	% Acid resistancec	pH of LB after treatment
S. flexneri M25-8A(pACYC184)d	95.25	28.03	2.50
S. flexneri W422(pACYC184)	41.40	<0.001	2.50
S. typhimurium SL1344(pACYC184)	62.86	<0.001	2.50
S. typhi ISP 1820(pACYC184)	10.63	<0.001	2.50
E. coli O157:H7 PS2 (rpoS)(pACYC184)	43.55	<0.009	2.50
E. coli O157:H7 PS2 (rpoS+)(pPS4.4)e	79.06	103.70	2.50
C. jejuni 81116 GRK1 (flaA::Kmr)	1.99	<0.001	2.50
V. cholerae 14035(pACYC184)	0.20	<0.001	2.50

^aExperiments were performed in duplicate, and numbers are means. The inoculum for each experiment was 9 × 10⁵ to 2 × 10⁶ CFU. LB was originally pH 2.5. Ground beef was examined for the presence of contaminating antibiotic-resistant bacteria, and no Cm^r or Km^r bacteria were found. 

^bPercent survival is expressed as 100 × CFU of bacteria recovered from ground beef after incubation in LB (pH 2.5) for 2 h/CFU of inoculum. 

^cPercent acid resistance is expressed as 100 × CFU of bacteria recovered from acidified LB (pH 2.5) at 2 h/CFU of inoculum. 

^dpACYC184, Cm^r. 

^epPS4.4 harbors the rpoS gene and Cm^r. 

TABLE 4.

Effect on pH of 10 ml of acidified LB and on S. typhimurium SL1344(pACYC184) survival of the addition of increasing amounts of ground beef

Amt (g) of ground beef	pH of LB after treatmenta	% Survivalb
0.1	2.50	32.07
0.5	2.70	83.47
1.0	2.83	86.37
2.0	2.98	82.93

^aThe pH of the medium was examined after a 2-h incubation at 37°C with increasing amounts of ground beef incubated in 10 ml of LB (original pH 2.5). 

^bGround beef was inoculated with approximately 10⁶ CFU of S. typhimurium, and the number of survivors was determined as a percentage of the original inoculum exposed. 

Acid-sensitive strains were tested in acidified LB for their ability to survive at pH 4.0, 5.0, and 6.0. We found that the acid-sensitive rpoS mutants of S. flexneri and E. coli and all the Salmonella species could survive at pH 4.0; however, C. jejuni survival at this pH 4.0 and 5.0 was low and V. cholerae was unable to survive at either pH 4 or 5 (Table 3). At pH 6.0, both species showed significant survival rates. These studies suggest that the pH of the media to which these enteric pathogens are exposed is a critical factor in determining their survival.

TABLE 3.

Survival of acid-sensitive pathogens at pH 4.0, 5.0, and 6.0 in acidified LB

Pathogen	% Acid resistancea at:
	pH 4.0	pH 5.0	pH 6.0
S. typhimurium SL1344	91.52	NDb	ND
S. flexneri W422 ΔrpoS	38.06	ND	ND
S. typhi ISP 1820	59.60	ND	ND
E. coli O157:H7 PS2	25.76	ND	ND
C. jejuni 81116 GRK1	0.12	0.19	52.76
V. cholerae 14305	<0.002	<0.002	25.67

^aPercent acid resistance is expressed as 100 × CFU recovered after a 2-h exposure to acidified LB/CFU of inoculum. 

^bND, not done. 

Effect on the pH of LB caused by increasing the inoculation size of ground beef.

To determine if the addition of ground beef was having a neutralizing effect on the pH of the acidified medium used for challenge, we examined the pH of the medium after incubation with different amounts of ground beef under identical conditions of acidity (Table 4). We also examined the effect of different amounts of ground beef on the survival of S. typhimurium. We observed that the larger the amount of ground beef challenged under acidic conditions, the greater the increase in the pH of the medium. This rise in pH was correlated with an increase in the survival of S. typhimurium. This demonstrates that ground beef can raise the pH of acidified media in vitro.

Survival of S. typhimurium inoculated onto the surface of preacidified ground beef.

To demonstrate if there was a protective effect against pH observed at the surface level of the ground beef, 0.1 g of ground beef was acidified before being inoculated with S. typhimurium. Ground beef acidified directly with HCl to a pH of 1.0 was shown to be unable to support the survival of S. typhimurium (Table 5). In other experiments, ground beef was also acidified by being soaked in LB (pH 2.5) for either 1 h or o/n and then placed in fresh LB (pH 2.5) for 10 min. Following this acidification, the ground beef was removed from the medium, its surface pH was measured, and was inoculated with S. typhimurium and incubated at room temperature for 2 h. The ground beef soaked for 1 h had a surface pH of 3.27 and provided an environment that was demonstrated to support the survival of S. typhimurium (Table 5). S. typhimurium has been demonstrated previously to be able to tolerate conditions of this pH (22). In contrast, the ground beef soaked o/n had a surface pH of 2.61 and did not support the survival of S. typhimurium. However, the ground beef had raised the pH of the LB to 2.66 as a result of the o/n soaking. These results demonstrate that even 0.1 g of ground beef has an alkalinating effect on the acidified LB over time and that if the surface pH of the ground beef is below the threshold for S. typhimurium survival, it cannot support this acid-sensitive pathogen.

TABLE 5.

Survival of S. typhimurium SL1344(pACYC184) inoculated onto the surface of preacidified ground beef

Ground beef acidificationa	pH on surface of beefb	% Survivalc
HCl	1.0	<0.01
LB pH 2.5 (1 h)	3.27	103.7
LB pH 2.5 (16 h)	2.61	<0.01

^aGround beef (0.1 g) was acidified by the direct addition of HCl or by being soaked in 10 ml of LB (pH 2.5) for the time indicated; it was then placed in 10 ml of fresh LB (pH 2.5) for 10 min. 

^bThe pH on the surface of the ground beef was determined immediately after the acidification treatment and before bacterial inoculation. 

^cAcidified ground beef was inoculated with approximately 10⁶ CFU of S. typhimurium and incubated for 2 h at room temperature. Surviving bacteria were enumerated by vigorous vortexing in 10 ml of PBS and are expressed as a percentage of the original inoculum exposed. 

Survival of serial dilutions of S. typhimurium inoculated onto ground beef under acidic conditions.

To determine if the inoculum size had any effect on the bacterial survival rate, serial dilutions of S. typhimurium were inoculated onto ground beef and challenged as described above. The survival rates obtained for all inoculations ranging from 10² to 10⁵ CFU were similar (Table 6) despite variations in the inoculum size.

TABLE 6.

Survival of S. typhimurium SL1344(pACYC184) inoculated onto ground beef in various inocula and challenged in acidified LB^a

Inoculum	No. of surviving bacteria (CFU)	% Survival
1.77 × 105	5.00 × 104	28.18
1.77 × 104	3.70 × 103	20.86
1.77 × 103	3.67 × 102	20.67
1.77 × 102	3.33 × 101	18.77

^aExperiments were performed in duplicate, and the results are means. 

Survival of S. typhimurium inoculated onto boiled rice and boiled egg white under acidic conditions.

S. typhimurium was inoculated onto boiled rice under the same conditions as those described for the ground-beef challenge to examine whether a food source high in carbohydrate would have a protective effect against an acidic environment. S. typhimurium inoculated onto boiled rice was not protected against low-pH challenge, in contrast to the protective effect observed with ground beef (Table 7). Rice inoculated with S. typhimurium and incubated under identical conditions in PBS (pH 7.4) was used as a control to determine the percent recovery.

TABLE 7.

Survival of S. typhimurium SL1344(pACYC184) inoculated onto boiled rice or boiled egg white and challenged in acidified LB^a

Challenge substance and food source	Inoculum (CFU)	No. of surviving bacteria (CFU)	% Survival	pH of LB after treatment
LB (pH 2.5)
Rice	2.97 × 106	0	<0.01b	2.50
Egg white	9.40 × 105	3.07 × 105	32.66b	2.50
PBS (pH 7.4)d
Rice	1.72 × 106	9.40 × 105	54.57c
Egg white	9.40 × 105	1.72 × 106	100.00c

^aExperiments were performed in duplicate, and numbers are means. 

^bPercent survival of S. typhimurium inoculated onto the surface of 0.1 g of boiled rice on egg white and challenged with LB (pH 2.5). 

^cPercent survival of S. typhimurium inoculated onto the surface of 0.1 g of boiled rice or egg white and challenged with PBS (pH 7.4). 

^dPBS was used as a control. 

S. typhimurium was also inoculated onto boiled egg white to determine whether a food source high in protein and low in fat could offer protection against extreme acid conditions. S. typhimurium was protected from killing by low pH at a significant level when inoculated on egg white (Table 7), suggesting that this survival was not due to a protective effect of fat. The percent recovery of S. typhimurium inoculated on egg white was determined by using a similar PBS control.

DISCUSSION

The studies presented in this paper were undertaken to examine what effect food might have on the survival of acid-sensitive pathogens challenged under acidic conditions and perhaps to shed light on the contradictory evidence concerning the ID of Salmonella species. In clinical trials where defined inocula were fed to human volunteers, the ID was demonstrated to be at least 10⁵ bacteria (5). However, careful analysis of ID in outbreak studies has shown that individuals ingesting as few as 50 to 100 of these organisms have become ill (5, 8). Since we had previously shown that S. typhimurium cannot survive under extreme acidic conditions (below pH 3.0) (15), we examined whether food could provide a protective effect to acid-sensitive bacteria by facilitating their survival under extreme acidic conditions.

In vitro acid resistance assays have been used by a number of investigators to study the ability of pathogens to resist acidic conditions (10, 15, 22). Likewise, the ability of pathogens to survive in acidic food sources such as yogurt, fermented sausage, and apple cider has received similar attention (1, 14, 27). In the present study, we have brought the two fields together to determine if a solid-food source can contribute to the survival of a pathogen under extreme acidic conditions. Information gleaned from these studies may help explain why pathogens which have a high ID when studied in clinical trials may require a much lower dose to cause natural infections.

We have shown previously that the stationary-phase acid resistance phenotype of S. flexneri and E. coli O157:H7 is dependent upon the expression of rpoS. rpoS mutants of these species are completely sensitive to killing under extreme acidic conditions of pH 2.5 (34, 35). Other pathogens, such as Salmonella, behave similarly to rpoS mutants of Shigella in this in vitro assay. In this study, we have shown that acid-sensitive rpoS mutants of Shigella and E. coli, as well as naturally acid-sensitive pathogens such as Salmonella, survive under these extreme low-pH conditions in vitro when inoculated onto the surface of certain food sources. This suggests that in some cases the solid food source provides a protection against low pH.

It has been speculated by D’Aoust (8) that Salmonella outbreaks with a low ID are often associated with a food source with a high fat content such as chocolate and cheese. This led to the hypothesis that the fat content of contaminated foods may play a significant role in human salmonellosis. The rationale behind this hypothesis is that organisms trapped in hydrophobic lipid moieties may readily survive the acidic conditions of the stomach and pass into the intestinal tract. The precise role played by fat in protecting bacteria from killing by low pH has yet to be determined and may be important in the protection offered by other food sources not examined here. If organisms are trapped in hydrophilic lipid moieties, it would be expected that the protective effect afforded by ground beef would have been equal for all the pathogens that were tested. This was not the case, since C. jejuni and V. cholerae were not as well protected as the other pathogens. We observed, however, that if S. typhimurium was inoculated onto boiled egg white, which is low in fat, it was protected from killing by acid.

We have demonstrated that increasing the amount of ground beef challenged in the assay raised the pH of the medium and increased the survival rate of S. typhimurium. By raising the pH of the medium in vitro, we confirmed that the pH of the acidified medium is a critical factor in determining the survival of each enteric pathogen. The importance of pH was also confirmed by the ability of Salmonella to survive when inoculated onto the surface of preacidified ground beef if the surface pH was high enough to be tolerated by this pathogen. However, if the pH of the ground beef was acidified so that its pH at the surface remained lethal to S. typhimurium, no survival was observed. These results imply that the survival of acid-sensitive bacteria on the surface of ground beef is most probably the result of the ability of ground beef to raise the pH of the acidified medium at the microenvironment occupied by the bacteria.

The number of bacteria consumed in contaminated food can vary considerably, and so we looked at the effect of inoculum size on the survival of S. typhimurium in our ground-beef assay. We found that the inoculum size had no significant effect on the survival rates of S. typhimurium in ground beef. This experiment also demonstrated that a very low inoculum of S. typhimurium can survive extreme acidity when present on the surface of certain solid food sources and may explain the low ID observed in some food-borne outbreaks involving Salmonella species.

At present it is not clear exactly how food protects bacteria from acidic conditions. The fat content of food may still provide a significant barrier against acidic conditions, but other food sources with lower fat content can still offer significant protection. This could be tested in the future by examining the protective effect of a solid food which has an extremely high fat content and is low in protein. It should be noted, however, that foods high in fat may not maintain their consistency under these conditions of temperature and pH (indeed, fatty oils were observed at the surface of the acidified media following the ground-beef challenge). Carbohydrate does not seem to offer protection, since Salmonella did not survive when inoculated onto the surface of rice. It has been demonstrated in vitro that the acid resistance phenotype of Shigella flexneri and E. coli is dependent upon the presence of amino acids in the acidified media (22, 23, 35). Acid resistance is expressed under these conditions by supplementing the acidified media with extremely small quantities of amino acids. It is possible that the protein content of some solid foods may be the essential component responsible for the protective effect observed in the experiments described here.

In this study, we have demonstrated that bacteria can survive acidic conditions in vitro when inoculated onto the surface of certain solid food sources whereas the same level of acidity is lethal to the inoculum in an acidified broth environment. The protective effect of some solid foods may be the result of at least partially raising the pH of the acidified medium at the microenvironment occupied by the bacteria on the surface of the food source. This is consistent with evidence that the ID of some food-borne pathogens is lowered when gastric acidity is reduced. Studies with the acid-sensitive food-borne pathogen Listeria monocytogenes have shown that the ID of this organism in the Sprague-Dawley rat model can be lowered by raising the pH of the stomach to 5.5 to 6.0 with cimetidine (30). Rats that were inoculated with a lower ID exhibited the same degree of severity of bacterial colonization in specific organs and tissue as did rats that received a higher ID, suggesting that once the pH barrier of the stomach has been breached, the number of surviving bacteria reaching the intestines does not affect the severity of the disease. This implies that the ability to survive the acidic conditions of the stomach is the principal factor determining the ID of a specific enteric pathogen.

In summary, these studies may help resolve the controversy surrounding the ID of pathogenic Salmonella species. In volunteer studies, the inoculum was administered in a liquid form, which may offer little protection against stomach acidity. Food-borne outbreaks characterized by a low ID are often associated with consumption of bacteria on solid food. This environment may protect bacteria from the lethal effects of stomach acidity.