- Email: [email protected]
A novel glutamate-dependent acid resistance among strains belonging to the Proteeae tribe of Enterobacteriaceae
FEMS Microbiology Letters 237 (2004) 303–309 www.fems-microbiology.org
A novel glutamate-dependent acid resistance among strains belonging to the Proteeae tribe of Enterobacteriaceae Geun woo Park, Francisco Diez-Gonzalez
*
Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Avenue, St. Paul, MN 55108, USA Received 25 February 2004; received in revised form 18 May 2004; accepted 28 June 2004 First published online 8 July 2004
Abstract Morganella, Providencia and Proteus strains were capable of surviving pH 2.0 for 1 h if glutamate was present. These strains did not have glutamic acid decarboxylase activity and the gadAB genes were not detected in any of these bacteria. When exposed to pH 2.0 acid shocks, the survival rate of these bacteria was significantly increased with glutamate concentrations as low as 0.3 mM in the acid media. Escherichia coli cells incubated at pH 3.4 consumed four times more glutamate and produced at least 7-fold more c-amino butyric acid than Morganella, Providencia and Proteus strains. These results indicate that strains belonging to the Proteeae tribe might have novel glutamate dependent acid-resistance mechanisms. Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Acid Resistance; Glutamate Decarboxylase; Morganella; Providencia; Proteus; Escherichia coli
1. Introduction The ability of bacteria to tolerate low pH is a very important trait to survive in a variety of environmental niches [1]. Bacteria that typically colonize the large intestine of mammals have a greater chance of success if they possess metabolic mechanisms that allow them to survive the acidic pH of the gastric stomach. A variety of mechanisms of resistance to low pH have been identified in enterobacteria, but most of these systems have only been studied in Escherichia coli and in related pathogenic species [2,3]. The acid-resistance response of E. coli appears to have multiple metabolic components and some of them have been thoroughly characterized. According to Richard and Foster [4], E. coli has three distinct acid-resistance mechanisms that are fully ex*
Corresponding author. Tel.: +1-612-624-9756; fax: +1-612-6255272. E-mail address: [email protected] (F. Diez-Gonzalez).
pressed at stationary phase: an oxidative, a glutamatedependent and an arginine-dependent systems. The glutamate-dependent system confers protection to pH 2.0, requires exogenous glutamate and its major metabolic component are the glutamic acid decarboxylase (GAD) enzymes that converts glutamate to c-amino butyric acid (GABA) [5]. The arginine-dependent acid-resistance system requires exogenous arginine that is decarboxylated via the arginine decarboxylase enzyme [5]. The combination of these acid-resistance systems may allow pathogenic E. coli to survive in the intestine and cause infection. Besides the glutamate and arginine decarboxylases, E. coli can also produce a lysine decarboxylase capable of conferring acid resistance [6]. Lysine decarboxylation was recognized as one of the major mechanisms of acid resistance in Salmonella, but this microorganism did not appear to have glutamate or arginine decarboxylases [7]. Other bacteria appear to have alternative mechanisms of acid resistance. The Salmonella acid tolerance
0378-1097/$22.00 Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsle.2004.06.050
304
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
response (ATR) includes 50 acid-shock proteins that might be regulated by RpoS, Fur and PhoP [8]. Yersinia, Helicobacter and Morganella have very potent ureases that protect these bacteria from acidic pH by rapidly producing ammonia [9,10]. Some lactic acid bacteria are capable of surviving acid environments if arginine is present, and this tolerance is due to the presence of an arginine deaminase that releases ammonia to increase pH [11]. This diversity of bacterial responses to low pH suggests that there might be other mechanisms of acid resistance yet to be identified. The GAD-mediated acid resistance is not only found in E. coli. Enzymes similar to the ones encoded by the gadA and gadB genes of E. coli have been found in related bacteria such as Shigella [12]. The presence of GAD activity has also been detected in other Gram-negative bacteria (Providencia alcalifaciens, Bacteroides fragilis) and in a variety of Gram-positive organisms that included Clostridium perfringens, Listeria monocytogenes and Lactobacillus lactis [13–15]. While the GAD-mediated acid-resistance system has been identified in different bacteria, there is no previous report of a glutamate-dependent mechanism that does not involve glutamate decarboxylation. The present report was undertaken with the purpose of detecting GAD-mediated acid resistance among enterobacteria and identifying alternative glutamate-dependent acid tolerance responses.
2. Materials and methods 2.1. Strains and culture conditions Enterobacteriaceae strains used in this study and their sources are listed in Table 1. All of these strains were cultivated in trypic soy broth (TSB) at 37 °C for 24-h.
2.2. Acid-resistance assay The strains were assayed for their amino acid-dependent acid resistance as follows: overnight cultures grown in TSB at 37 °C were diluted 100-fold into different acid-shock media (distilled water containing 10 g L1 Casamino acids, or 1 g L1 of an amino acid previously adjusted with HCl to final pH values between 2.0). The concentration of the individual amino acid solutions was 10-fold less than the Casamino acid one to reflect the fact that they are mostly present in this product at less than 10% of the total amino acid composition. After acid-shock treatments for 1 h at 37 °C, the viable cells counts were determined by the MPN method in triplicate in 96-well plates using TSB broth. All experiments were performed at least twice and the results were averaged by using Thomas approximation [16]. The percent survival was defined as the CFU per milliliter after acid shock divided by the CFU per milliliter a similar cell suspension in neutral media, and based on Thomas approximation, the confidence limits at 95% was derived1=2by multiplying or dividing mean value by 100:55ðlogðaÞ=nÞ (a: dilution factor and n, the number of total wells in the MPN). 2.3. Detection of gadAB genes The gadAB genes were detected using a PCR protocol as follows: primers (forward primer 5 0 -ACCTGCGTTGCGTAAATA-3 0 and reverse primer 5 0 -GGGCGGAGAAGTTGATG-3 0 ) were designed to amplify a 670-bp DNA fragment from the high-homology region of gadA and gadB as described by McDaniels et al. [17]. DNA templates were prepared by boiling 0.5 mL bacterial cell suspensions for 10 min and centrifuging (10,000g, 5 min) of cell debris. PCR was performed
Table 1 Bacterial strains used in this study Strains
Number of strains
Source
Escherichia coli Enterobacter cloacae Enterobacter sakazakii Enterobacter aerogenes Hafnia alvei Klebsiella spp. Kluyvera spp. Serratia spp. Yersinia enterocolitica Citrobacter spp. Proteus spp. Morganella morganii Providencia spp.
15 5 12 3 3 9 1 3 1 6 6 5 4
ATCC (2), FML (11) and USACM (2) ATCC (2) and FML (3) ATCC (1) and FDA (11) ATCC (1), DLMP (1) and FML (1) FML (3) FML (9) FML (1) DLMP (2) and FML(1) FML (1) ATCC (1), DLMP (2) and FML(3) ATCC (2) and FML(4) DLMP (2) and FML (3) FML (4)
ATCC, American Type Culture Collection. DLMP, Department of Laboratory Medicine and Pathology, University of Minnesota Medical School, MN, USA. FDA, Food and Drug Administration, MD, USA. FML, Food Microbiology Lab (Dr. Sita Tatini), University of Minnesota, MN. USAM, Dr. John W. Foster, University of Southern Alabama College of Medicine, AL.
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
by addition of 2 lL of the boiled bacterial cell suspensions to a reaction mixture containing 1.5 mM MgCl2 and 5 lL of 10-fold concentrated polymerase synthesis buffer, 200 lM (each) of deoxynucleoside triphosphate, and 0.5 U of Taq polymerase (Promega, Madison,WI). Reaction mixtures were first heated at 94 °C for 5 min and amplification was done using a thermo cycler (Stratagene, Inc., La Jolla, CA) programmed for 35 cycles of 1 min each at 94, 58 and 72 °C, followed by a final 7 min extension. 2.4. Glutamic acid decarboxylase assay The enzymatic assay buffer (pH 3.4), containing (per liter) 1 g glutamate, 90 g sodium chloride, 0.005 g bromocresol green and 5 mL Triton X-100 for 4 h at 37 °C were prepared as described by Rice and co-workers [18]. Cultures were harvested by centrifugation (10,000g, 5 min), cell pellets were re-suspended in 1 mL enzymatic assay buffer, and incubated at 37 °C for 4 h. The GAD activities were qualitatively determined by observing the color change from yellow to green. 2.5. Enzymatic analysis The production of GABA and the consumption of glutamate were measured by using enzymatic assays after incubation of cells using the same buffer as described for the GAD assay, except that no bromocresol blue was added. After incubation of cell suspensions at 37 °C, the reactions were terminated by boiling for 5 min. The supernatants were collected by centrifugation (10,000g, 5 min) and the enzymatic assays were performed immediately. The GABA concentration reaction mixture assay was done in 0.9 mL volumes that contained: 0.3 M Tris–HCl buffer (pH 8.9), 10 mM a-ketoglutarate, 2 mM 2-mercaptoenthol, 0.5 mM NADP and 200 mU L1 GABase (a mixture of 4-aminobutyrate: 2-oxoglutarate aminotransferase and succinate-semialdehyde:NAD oxidoreductase; (Sigma–Aldrich, Inc., St. Louis, MO). The reactions were initiated by adding 0.1 mL sample. After 30 min, the reactions were terminated by boiling for 5 min. The amount of GABA consumed was calculated from the amount of NAPDH formed as spectrophotometrically determined at 340 nm based on a standard curve [19]. The free glutamate in the supernatant was determined as follows: four stock solutions were prepared: solution I contained: 0.2 M TEA, 25 mM phosphate, 13.2 g L1 Triton X-100 (pH 8.6); solution II contained: 17 mM NAD, 1.2 g diaphorase (4.8 U mg1, Sigma–Aldrich, Inc.); solution III contained 1.19 mM iodonitrotetrazolium chloride (INT); and solution IV contained glutamate dehydrogenase (GIDH, 1200 kU L1, Sigma– Aldrich). The reaction mixture included 0.3 mL solution
305
I, 0.1 mL solution II, 0.1 mL solution III, 0.1 mL sample (in the control, 0.1 mL water was added), and 0.9 mL water. After 2-min incubations at room temperature, 0.015 mL solution IV was added into the mixture. After an additional 2 min, the final reactant (INT-formazan) was determined spectrophotometrically at 492 nm [20]. 2.6. Statistical analysis The data consisted of the mean value of at least two independent experiments, and in the case of survival rate calculations the variations were less than 50%. Significant differences between the treatments were determined using the StudentÕs t-test [21].
3. Results Exposure to pH 2.0 saline solution for 1 h reduced the viable count of all enterobacteria tested to less than 0.001%, with the exception of a Proteus vulgaris strain G16 that had a 1.3% survival rate (Table 2). When the pH 2.0 saline solution was supplemented with 10 g L1 Casamino acids the survival rate of as many as 49 bacterial strains did not increase, but in most E. coli, four Morganella morganii, three Providencia, one Proteus mirabilis, one Enterobacter cloacae, one Hafniae alvei, and one Citrobacter braakii strain, this number increased markedly to 1% or greater. Similar results were observed if the saline solution included 6 mM glutamate. Cultures of several of these acid-resistant enterobacteria that were subjected to acid shocks in pH 2.0 saline solution supplemented with 1 g L1 of one of 17 other amino acids had a survival rate of less than 0.001%, with the exception of most E. coli with arginine (>0.2%) and P. vulgaris strain G16 with all of them (1.3%). Most E. coli strains tested positive for the presence of gadAB genes and had detectable GAD activity (Table 3). Besides strain EF522 that was deficient in GAD production due to transposon mutagenesis, E. coli O157:H7 strain 43890 was the only acid sensitive E. coli that had no enzymatic activity, but yielded PCR products for gadAB. All other acid-sensitive enterobacteria had no genotypic or phenotypic evidence of the presence of GAD. In addition to E. coli, the only acid-resistant enterobacteria that tested positive for gadAB and had GAD activity were a H. alvei and a C. braakii strain. The remaining acid-resistant enterobacteria strains Enterobacter cloacae, M. morganii, Providencia and P. mirabilis had no detectable gadAB genes and GAD activity. The E. coli strain EF522 that had insertions in the gadA and gadB genes was sensitive to acid, produced approximately 0.4 mmol g protein1 GABA and increased the pH to 3.9 after incubation of cell suspensions with glutamate at pH 3.4 (Table 3). Wild-type strain EK227
306
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
Table 2 Acid resistance of enterobacteria as affected by acid-shock media Bacterial species or genera
Escherichia coli Enterobacter cloacae Enterobacter sakazakii Enterobacter aerogenes Hafniae alvei Klebsiella spp. Serratia spp. Citrobacter spp. Kluyvera spp. Yersinia enterocolitica Proteus mirabilis Proteus vulgaris Morganella morganii Providencia spp.
Number of strains
1 14 4 1 12 3 2 1 9 3 5 1 1 1 4 1 1 1 4 1 3
Average acid resistance (%) No addition
Casamino acid (10 g L1)
Glutamate (1 g L1)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 1.29 <0.001 <0.001 <0.001 <0.001
<0.001 18.0 <0.001 1.00 <0.001 <0.001 <0.001 10 <0.001 <0.01 <0.001 17.79 <0.001 <0.001 <0.001 1.80 2.30 <0.001 8.97 <0.001 39.02
<0.001 11.7 <0.001 6.31 <0.001 <0.001 <0.001 1 <0.001 <0.01 <0.001 70.40 <0.001 <0.001 <0.001 1.08 1.29 <0.001 32.80 <0.001 6.92
Table 3 Determination of gadAB genes, GAD and GABA production in enterobacteria Strains (acid sensitive/resistant)
gadAB PCR assay
GAD assay
GABA levels mmol g protein1
Final pH
Escherichia coli EK227 (R) Escherichia coli EF522 (S) Escherichia coli O157 ATCC 43895 (R) Escherichia coli O157 ATCC 43890 (S) Enterobacter cloacae ATCC 700323 (S) Enterobacter cloacae ATCC 13047 (S) Enterobacter cloacae G07 (R) Enterobacter sakazakii ATCC 57329 (S) Enterobacter sakazakii G25 (S) Enterobacter aerogenes ATCC 13048 (S) Hafniae alvei G18 (R) Citrobacter braakii G15 (R) Proteus mirabilis G08 (R) Proteus vulgaris G16 (S) Morganella morganii G35 (R) Morganella morganii G57 (R) Morganella morganii G62 (R) Proteus stuartii G22 (R) Proteus stuartii G23 (S) Proteus stuartii G70 (R) Proteus rettgeri G37 (R)
+ ND + + – – – – – – + + – – – – – – – – –
+ – + – – – – – – – + + – – – – – – – – –
2.15 ± 0.12 0.35 ± 0.01 1.82 ± 0.08 0.36 ± 0.00 0.18 ± 0.01 0.17 ± 0.01 0.65 ± 0.15 0.20 ± 0.01 0.76 ± 0.20 0.15 ± 0.01 ND 1.43 ± 0.02 0.23 ± 0.02 0.31 ± 0.03 0.25 ± 0.01 0.67 ± 0.02 0.20 ± 0.04 0.43 ± 0.03 0.78 ± 0.09 0.19 ± 0.02 0.35 ± 0.03
5.6 ± 0.06 3.95 ± 0.02 5.66 ± 0.02 4.01 ± 0.02 3.94 ± 0.04 3.80 ± 0.01 4.46 ± 0.21 3.97 ± 0.02 4.66 ± 0.54 3.85 ± 0.04 ND 5.74 ± 0.07 3.99 ± 0.01 3.88 ± 0.02 3.80 ± 0.07 3.96 ± 0.13 3.86 ± 0.08 3.79 ± 0.02 4.41 ± 0.39 3.77 ± 0.058 3.99 ± 0.07
ND, not determined; R, acid-resistant strain that had a survival rate of 1% or greater after acid shock at pH 2.0 for 1 h; S, acid-sensitive strain that had a survival rate of 0.01% or less after acid shock at pH 2.0 for 1 h.
and most acid-resistant E. coli cells produced 5-fold more GABA than EF522 (2.1 vs. 0.35 mmol mg protein1) and increased the final pH to 5.6. C. braakii strain G15 produced approximately 1.4 mmol g protein1. The amount of GABA produced by other enterobacteria ranged from 0.17 to 0.78 mmol g protein1 and there was no statistically significant difference in the con-
centration produced by those strains that were resistant to pH 2.0 in the presence of glutamate as compared to those that were not. The final incubation media pH of strains that produced less than 0.8 mmol g protein1 GABA was always less than 4.5. E. coli strains EK227 and 43895 cultures were capable of consuming 5.6 mM glutamate when incubated
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
at pH 3.4 after 4 h with an initial concentration of 5.9 mM (Table 4). E. coli strain EF522 only consumed 10% of the original glutamate concentration under the same conditions. The amount of glutamate consumed by P. mirabilis G08 and M. morganii G62 was approximately 23% of the initial concentration and the rest of the acid-resistant enterobacteria consumed less than 20% of that number. When the concentration of glutamate in the pH 2.0acid-shock media was decreased to 37 lM or less the survival rate of all acid-resistant strains was less than 0.01% (Table 5). The survival rate of E. coli, C. braakii and H. alvei strains that were positive for GAD was maximal at glutamate concentrations between 74 and 148 lM. The survival rate of the Providencia spp. strains and M. morganii G35 strain was only maximal if the glutamate concentrations were greater than 296 lM. The remainder of the glutamate-dependent acid-resistant strains reached their maximum survival rate at 148 lM. If cells suspensions of any of the glutamate-dependent acid-resistant strains containing approximately 106 cells L1 were subjected to an acid shock in the
Table 4 Extent of glutamate consumption by cultures of enterobacteria when incubated at pH 3.4 with an initial glutamate concentration of 5.9 mM Strains
Escherichia coli EK227 Escherichia coli EF522 Escherichia coli O157 43895 Escherichia coli O157 43890 Proteus mirabilis G08 Morganella morganii G35 Morganella morganii G57 Morganella morganii G62 Providencia spp. G22 Providencia rettgeri G37
Consumed amount of glutamate mM
mmol (g protein)1
5.60 0.54 5.57 0.87 1.30 0.98 0.68 1.39 0.77 1.00
2.60 0.25 2.59 0.40 0.61 0.46 0.32 0.65 0.36 0.47
307
presence of 0.6 lM glutamate and 1.5 mM of L -transpyrrolidine-2,4-dicarboxylic acid (L -PDC), no survivors were detected.
4. Discussion We were able to identify bacterial strains capable of surviving pH 2.0 with the addition of amino acids. These bacteria were: fourteen Escherichia coli strains, one Enterobacter, one Hafnia, one Citrobacter, four Morganella, three Providencia, and one Proteus strains. Compared to other amino acids only glutamate enabled these bacteria to survive extreme acid shocks. When these strains were subjected to an acid shock in the presence of casamino acids (Table 2), their survival rate ranged from 1 to 39% which was slightly higher than with glutamate alone. This difference could have been due to a potential enhancement of the glutamate-dependent acid resistance when other amino acids were present. Since GAD has been one of most important amino acid-dependent acid-resistance elements identified in enterobacteria, we conducted PCR assays targeting the gadAB genes. Our results indicated that among acidresistant strains, E. coli, Hafnia alvei and Citrobacter braakii had the gadAB genes and this result was confirmed by the detection of GAD activity (Table 3). Because these bacteria produced GABA, they could utilize glutamate for acid resistance. Among enterobacteria, the GAD-mediated system has been identified in E. coli, Shigella and Providencia [5,12,13], but this is the first report in which H. alvei and C. braakii strains have been observed to possess this stress response mechanism. The presence of GAD has been suggested as a specific marker for detection of E. coli [18], but our findings suggest that this enzyme activity is also found among other enterobacteria. It should be noted that E. coli O157:H7 strain ATCC 43890 was the only strain that had the gadAB genes present, but had no GAD
Table 5 Effect of glutamate concentration on the survival rate of enteric bacteria Strains
Survival rate (%) with glutamate (lM) 18
37
74
148
296
592
Escherichia coli EK227 Escherichia coli O157:H7 ATCC 43895 Enterobacter cloacae G07 Citrobacter brakii G15 Hafniae alvei G18 Proteus mirabilis G08 Morganella morganii G35 Morganella morganii G57 Morganella morganii G62 Providencia stuartii G22 Providencia rettgeri G37
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.002 <0.001 <0.001 <0.001
0.3 25.7 0.002 100.0 0.4 0.056 <0.001 17.8 1.00 <0.001 <0.001
17.8 25.7 0.018 25.7 1.8 0.1 0.1 38.9 4.6 0.4 0.6
10 25.7 0.1 45.8 0.4 0.6 1.0 100.0 2.6 56.2 21.9
25.7 0.1 45.8 1.8 1.0 0.4 38.9 2.6 56.2 38.9 25.7
308
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
activity. The explanation for this unique trait could be due to unidentified potential defects in gene expression and/or protein synthesis. These GAD-positive strains were capable of excreting significant levels of GABA, the product of glutamate decarboxylation, and this production appeared to be linked to the disappearance of glutamate from the media. Among recent studies that have investigated the glutamate-mediated acid resistance, very few have reported the production of GABA and the consumption of glutamate as important factors. De Biase et al. [22] indicated that acid-resistant E. coli could produce as much as 1.3 mM GABA during an acid shock, but the glutamate consumption was not measured. The link of GABA production rate with an increased acid resistance remains to be determined, but in our study these parameters were useful to identify non-GAD glutamate-dependent acid-resistance systems. Among acid-resistance strains, gad genes and glutamic acid decarboxylation were not detected in strains belonging to the Proteeae Tribe of Enterobacteriaceae (Morganella, Providencia, and Proteus) and in an Enterobacter isolate. However, glutamate addition to the acid media significantly increased their survival after a pH-2.0 acid shock. The glutamate supplementation stimulated the survival rate of these bacteria to similar levels observed with GAD-positive E. coli and their response to glutamate concentrations was comparable to E. coli. Because the Morganella, Providencia and Proteus strains did not consume significant amounts of glutamate when subjected to acid shocks, it appeared that their acid-resistance system was metabolically different from that of E. coli. Waterman and Small [23] reported that the acid resistance of E. coli and Shigella could be completely inhibited if the glutamate-containing acid-shock media was supplemented with L -trans-pyrrolidine-2,4-dicarboxylic acid (L -PDC), an inhibitor of glutamate transporters. Similar to E. coli, the acid resistance of Morganella, Providencia, Proteus and Enterobacter was sensitive to the addition of L -PDC. The glutamate-dependent systems of these strains, however, did not appear to metabolize glutamate at the same rate as E. coli (90% vs. less than 25%) and they produced significantly less GABA. These comparisons suggest that Morganella, Providencia, and Proteus possess glutamate-dependent acid resistance, but markedly different from the gad-mediated acid resistance identified in E. coli. E. coli utilizes glutamate for osmotic as well as acid resistance. Uptake of potassium glutamate enable bacteria cells to resist high osmotic pressure by converting glutamate into proline, an osmoprotectant solute [24]. Ogahara et al. [25] suggested that the gad system is induced during an osmotic shock to maintain the charge balance during the rapid uptake of K+ by converting
glutamate into GABA. Based on these reports it appears that E. coli relies on glutamate for both hyperosmotic as well as acid tolerance, but the potential link between the glutamante-dependent acid resistance of the Proteeae group strains and their osmotolerance remains to be determined. Proteus, Morganella, and Providencia share some phenotypic properties rarely encountered in other enterobacteria. The most intriguing of these properties is the ability to oxidatively deaminate some amino acids to keto acids, which may function in iron transport [26]. Similar to other enterobacteria, these Proteeae group also have a urease that can hydrolyze urea and cause alkalinization of the media. We confirmed that all of the Proteeae strains that we used in this research produced urease (data not shown). These bacteria have been implicated in urinary tract infection because release of ammonia can cause crystal deposition, stone formation, and cellular damage to the renal epithelium [27]. Despite that all these strains have urease activity, only Morganella can survive extreme acid as low as pH 2.0 by generating ammonia from urea [10]. Because of the unique physiological characteristics in Proteeae group, commercial detection systems rarely have difficulty in providing a correct identification of these organisms [28]. However, these unusual biochemical characteristics would imply the special metabolism in these groups. These alternative metabolic pathways may compensate the relatively limited ability to ferment diverse carbon sources. A potentially new acid resistance relying on glutamate might be uniquely present in the Proteeae. Glutamate dependent acid resistance that have newly identified in Proteeae tribe will contribute to broadening the understanding of these bacteria itself. Further research will be undertaken to identify this mechanisms.
Acknowledgements Funding for this project was provided by the Minnesota Agricultural Experiment Station. The authors thank Dr. John W. Foster (University of South Alabama College of Medicine) for providing E. coli EK227 and EF522 strains, Dr. Patricia Ferrieri for providing enterobacteria, and Dr. Kun-Ho Seo for providing Enterobacter sakazakii.
References [1] Gorden, J. and Small, P.L.C. (1993) Acid resistance in enteric bacteria. Infect. Immun. 61, 364–367. [2] Small, P., Blankenhorn, D., Welty, D., Zinser, E. and Slonczewski, J.L. (1994) Acid and base resistance in Escherichia coli and Shigella flexneri: role of rpoS and growth pH. J. Bacteriol. 176, 1729–1737.
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309 [3] Foster, J.W. (2000) Microbial responses to acid stress In: Bacterial Stress Responses (Storz, G., Hengge-Aronis, R., Eds.), pp. 99–115. ASM Press, Washington, DC. [4] Richard, H.T. and Foster, J.W. (2003) Acid resistance in Escherichia coli. Adv. Appl. Microbiol. 52, 167–186. [5] Lin, J., Lee, I.S., Frey, J., Slonczewski, J.L. and Foster, J.W. (1995) Comparative analysis of extreme acid survival in Salmonella typhimurium, Shigella flexneri, and Escherichia coli. J. Bacteriol. 177, 4097–4104. [6] Meng, S.Y. and Bennett, G.N. (1992) Nucleotide sequence of the Escherichia coli cad operon: a system for neutralization of low extracellular pH. J. Bacteriol. 174, 2659–2669. [7] Park, Y.K., Bearson, B., Bang, S.H., Bang, I.S. and Foster, J.W. (1996) Internal pH crisis, lysine decarboxylase and the acid tolerance response of Salmonella typhimurium. Mol. Microbiol. 20, 605–611. [8] Bearson, S., Bearson, B. and Foster, J.W. (1997) Acid stress responses in enterobacteria. FEMS Microbiol. Lett. 147, 173–180. [9] Matin, A., Zychlinsky, E., Keyhan, M. and Sachs, G. (1996) Capacity of Helicobacter pylori to generate ionic gradients at low pH is similar to that of bacteria which grow under strongly acidic conditions. Infect. Immun. 64, 1434–1436. [10] Young, G.M., Amid, D. and Miller, V.L. (1996) A bifunctional urease enhances survival of pathogenic Yersinia enterocolitica and Morganella morganii at low pH. J. Bacteriol. 178, 6487–6495. [11] Marquis, R.E. and Bender, G.R. (1987) Arginine deaminase system and bacterial adaptation to acid environments. Appl. Environ. Microbiol. 53, 198–200. [12] Waterman, S.R. and Small, P.L. (2003) Identification of the promoter regions and sigma (s)-dependent regulation of the gadA and gadBC genes associated with glutamate-dependent acid resistance in Shigella flexneri. FEMS Microbiol. Lett. 225, 155– 160. [13] Freir, P.A., Graves, M.H. and Kocka, F.E. (1976) A rapid glutamic acid decarboxylase test for identification of bacteria. Ann. Clin. Lab. Sci. 6, 537–539. [14] Cotter, P.D., OÕReilly, K. and Hill, C. (2001) Role of the glutamate decarboxylase acid resistance system in the survival of Listeria monocytogenes LO28 in low pH foods. J. Food Prot. 64, 1362–1368. [15] Sanders, J.W. and Leenhouts, K. (1998) A chloride-inducible acid resistance mechanism in Lactococcus lactis and its regulation. Mol. Microbiol. 27, 299–310.
309
[16] Swanson, K.M.J., Petran, R.L. and Hanlin, J.H. (2001) Culture Methods for Enumeration of Microorganisms. Microbiological Examination of Foods. American Public Health Association, Washington, DC. [17] McDaniels, A.E., Rice, E.W., Reyes, A.L., Johnson, C.H., Haugland, R.A. and Stelma Jr., G.N. (1996) Confirmational identification of Escherichia coli, a comparison of genotypic and phenotypic assays for glutamate decarboxylase and b-D -glucuronidase. Appl. Environ. Microbiol. 62, 3350–3354. [18] Rice, E.W., Johnson, C.H., Dunnigan, M.E. and Reasoner, D.J. (1993) Rapid glutamate decorboxylase assay for detection of Escherichia coli. Appl. Environ. Microbiol. 59, 4347–4349. [19] Higuchi, T., Hayashi, H. and Abe, K. (1997) Exchange of glutamate and c-aminobutyrate in a Lactobacillus strain. J. Bacteriol. 179, 3362–3364. [20] Beutler, H.-O. (1983) L -glutamate, colorimeteric method with glutamate dehydrogenase and diaphorase In: Metabolites 3: Lipids, Amino Acids and Related Compounds (Grabl, M., Ed.), pp. 369–376. VCH Verlagsgesellschaft mbH, Weinheim. [21] Moore, D.S. and McCabe, G.P. (1999) Introduction to the Practice of Statistics. W.H. Freeman and Co., New York, USA. [22] De Biase, D., Tramonti, A., Bossa, F. and Visca, P. (1999) The response to stationary-phase stress conditions in Escherichia coli: role and regulation of the glutamic acid decarboxylase system. Mol. Microbiol. 32, 1198–1211. [23] Waterman, S.R. and Small, P.L. (2003) The glutamate-dependent acid resistance of Escherichia coli and Shigella flexneri is inhibited in vitro by L -trans-pyrrolidine-2,4-dicarboxylic acid. FEMS Microbiol Lett. 224, 119–125. [24] Booth, I.R. and Higgins, C.F. (1990) Enteric bacteria and osmotic stress: intracellular potassium glutamate as a secondary signal of osmotic stress. FEMS Microbiol. Rev. 75, 239–246. [25] Ogahara, T., Ohno, T., Takayama, M., Igarashi, K. and Kobayashi, H. (1995) Accumulation of glutamate by osmotically stressed Escherichia coli is dependent on pH. J. Bacteriol. 177, 5987–5990. [26] Ewing, W.H. (1986) Edwards and EwingÕs Identification of Enterobacteriaceae, 4th Ed.. Elsevier Inc., New York. [27] Mobley, H.L.T. and Warren, J.W. (1987) Urease-positive bacteriuria and obstruction of long-term urinary catheters. J. Clin. Microbiol. 25, 560–564. [28] Janda, J.M. and Abbott, S.L. (1998) The Enterobacteria. Lippincott-Raven Publishers, Philadelphia, PA.
A novel glutamate-dependent acid resistance among strains belonging to the Proteeae tribe of Enterobacteriaceae Geun woo Park, Francisco Diez-Gonzalez
*
Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Avenue, St. Paul, MN 55108, USA Received 25 February 2004; received in revised form 18 May 2004; accepted 28 June 2004 First published online 8 July 2004
Abstract Morganella, Providencia and Proteus strains were capable of surviving pH 2.0 for 1 h if glutamate was present. These strains did not have glutamic acid decarboxylase activity and the gadAB genes were not detected in any of these bacteria. When exposed to pH 2.0 acid shocks, the survival rate of these bacteria was significantly increased with glutamate concentrations as low as 0.3 mM in the acid media. Escherichia coli cells incubated at pH 3.4 consumed four times more glutamate and produced at least 7-fold more c-amino butyric acid than Morganella, Providencia and Proteus strains. These results indicate that strains belonging to the Proteeae tribe might have novel glutamate dependent acid-resistance mechanisms. Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Acid Resistance; Glutamate Decarboxylase; Morganella; Providencia; Proteus; Escherichia coli
1. Introduction The ability of bacteria to tolerate low pH is a very important trait to survive in a variety of environmental niches [1]. Bacteria that typically colonize the large intestine of mammals have a greater chance of success if they possess metabolic mechanisms that allow them to survive the acidic pH of the gastric stomach. A variety of mechanisms of resistance to low pH have been identified in enterobacteria, but most of these systems have only been studied in Escherichia coli and in related pathogenic species [2,3]. The acid-resistance response of E. coli appears to have multiple metabolic components and some of them have been thoroughly characterized. According to Richard and Foster [4], E. coli has three distinct acid-resistance mechanisms that are fully ex*
Corresponding author. Tel.: +1-612-624-9756; fax: +1-612-6255272. E-mail address: [email protected] (F. Diez-Gonzalez).
pressed at stationary phase: an oxidative, a glutamatedependent and an arginine-dependent systems. The glutamate-dependent system confers protection to pH 2.0, requires exogenous glutamate and its major metabolic component are the glutamic acid decarboxylase (GAD) enzymes that converts glutamate to c-amino butyric acid (GABA) [5]. The arginine-dependent acid-resistance system requires exogenous arginine that is decarboxylated via the arginine decarboxylase enzyme [5]. The combination of these acid-resistance systems may allow pathogenic E. coli to survive in the intestine and cause infection. Besides the glutamate and arginine decarboxylases, E. coli can also produce a lysine decarboxylase capable of conferring acid resistance [6]. Lysine decarboxylation was recognized as one of the major mechanisms of acid resistance in Salmonella, but this microorganism did not appear to have glutamate or arginine decarboxylases [7]. Other bacteria appear to have alternative mechanisms of acid resistance. The Salmonella acid tolerance
0378-1097/$22.00 Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsle.2004.06.050
304
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
response (ATR) includes 50 acid-shock proteins that might be regulated by RpoS, Fur and PhoP [8]. Yersinia, Helicobacter and Morganella have very potent ureases that protect these bacteria from acidic pH by rapidly producing ammonia [9,10]. Some lactic acid bacteria are capable of surviving acid environments if arginine is present, and this tolerance is due to the presence of an arginine deaminase that releases ammonia to increase pH [11]. This diversity of bacterial responses to low pH suggests that there might be other mechanisms of acid resistance yet to be identified. The GAD-mediated acid resistance is not only found in E. coli. Enzymes similar to the ones encoded by the gadA and gadB genes of E. coli have been found in related bacteria such as Shigella [12]. The presence of GAD activity has also been detected in other Gram-negative bacteria (Providencia alcalifaciens, Bacteroides fragilis) and in a variety of Gram-positive organisms that included Clostridium perfringens, Listeria monocytogenes and Lactobacillus lactis [13–15]. While the GAD-mediated acid-resistance system has been identified in different bacteria, there is no previous report of a glutamate-dependent mechanism that does not involve glutamate decarboxylation. The present report was undertaken with the purpose of detecting GAD-mediated acid resistance among enterobacteria and identifying alternative glutamate-dependent acid tolerance responses.
2. Materials and methods 2.1. Strains and culture conditions Enterobacteriaceae strains used in this study and their sources are listed in Table 1. All of these strains were cultivated in trypic soy broth (TSB) at 37 °C for 24-h.
2.2. Acid-resistance assay The strains were assayed for their amino acid-dependent acid resistance as follows: overnight cultures grown in TSB at 37 °C were diluted 100-fold into different acid-shock media (distilled water containing 10 g L1 Casamino acids, or 1 g L1 of an amino acid previously adjusted with HCl to final pH values between 2.0). The concentration of the individual amino acid solutions was 10-fold less than the Casamino acid one to reflect the fact that they are mostly present in this product at less than 10% of the total amino acid composition. After acid-shock treatments for 1 h at 37 °C, the viable cells counts were determined by the MPN method in triplicate in 96-well plates using TSB broth. All experiments were performed at least twice and the results were averaged by using Thomas approximation [16]. The percent survival was defined as the CFU per milliliter after acid shock divided by the CFU per milliliter a similar cell suspension in neutral media, and based on Thomas approximation, the confidence limits at 95% was derived1=2by multiplying or dividing mean value by 100:55ðlogðaÞ=nÞ (a: dilution factor and n, the number of total wells in the MPN). 2.3. Detection of gadAB genes The gadAB genes were detected using a PCR protocol as follows: primers (forward primer 5 0 -ACCTGCGTTGCGTAAATA-3 0 and reverse primer 5 0 -GGGCGGAGAAGTTGATG-3 0 ) were designed to amplify a 670-bp DNA fragment from the high-homology region of gadA and gadB as described by McDaniels et al. [17]. DNA templates were prepared by boiling 0.5 mL bacterial cell suspensions for 10 min and centrifuging (10,000g, 5 min) of cell debris. PCR was performed
Table 1 Bacterial strains used in this study Strains
Number of strains
Source
Escherichia coli Enterobacter cloacae Enterobacter sakazakii Enterobacter aerogenes Hafnia alvei Klebsiella spp. Kluyvera spp. Serratia spp. Yersinia enterocolitica Citrobacter spp. Proteus spp. Morganella morganii Providencia spp.
15 5 12 3 3 9 1 3 1 6 6 5 4
ATCC (2), FML (11) and USACM (2) ATCC (2) and FML (3) ATCC (1) and FDA (11) ATCC (1), DLMP (1) and FML (1) FML (3) FML (9) FML (1) DLMP (2) and FML(1) FML (1) ATCC (1), DLMP (2) and FML(3) ATCC (2) and FML(4) DLMP (2) and FML (3) FML (4)
ATCC, American Type Culture Collection. DLMP, Department of Laboratory Medicine and Pathology, University of Minnesota Medical School, MN, USA. FDA, Food and Drug Administration, MD, USA. FML, Food Microbiology Lab (Dr. Sita Tatini), University of Minnesota, MN. USAM, Dr. John W. Foster, University of Southern Alabama College of Medicine, AL.
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
by addition of 2 lL of the boiled bacterial cell suspensions to a reaction mixture containing 1.5 mM MgCl2 and 5 lL of 10-fold concentrated polymerase synthesis buffer, 200 lM (each) of deoxynucleoside triphosphate, and 0.5 U of Taq polymerase (Promega, Madison,WI). Reaction mixtures were first heated at 94 °C for 5 min and amplification was done using a thermo cycler (Stratagene, Inc., La Jolla, CA) programmed for 35 cycles of 1 min each at 94, 58 and 72 °C, followed by a final 7 min extension. 2.4. Glutamic acid decarboxylase assay The enzymatic assay buffer (pH 3.4), containing (per liter) 1 g glutamate, 90 g sodium chloride, 0.005 g bromocresol green and 5 mL Triton X-100 for 4 h at 37 °C were prepared as described by Rice and co-workers [18]. Cultures were harvested by centrifugation (10,000g, 5 min), cell pellets were re-suspended in 1 mL enzymatic assay buffer, and incubated at 37 °C for 4 h. The GAD activities were qualitatively determined by observing the color change from yellow to green. 2.5. Enzymatic analysis The production of GABA and the consumption of glutamate were measured by using enzymatic assays after incubation of cells using the same buffer as described for the GAD assay, except that no bromocresol blue was added. After incubation of cell suspensions at 37 °C, the reactions were terminated by boiling for 5 min. The supernatants were collected by centrifugation (10,000g, 5 min) and the enzymatic assays were performed immediately. The GABA concentration reaction mixture assay was done in 0.9 mL volumes that contained: 0.3 M Tris–HCl buffer (pH 8.9), 10 mM a-ketoglutarate, 2 mM 2-mercaptoenthol, 0.5 mM NADP and 200 mU L1 GABase (a mixture of 4-aminobutyrate: 2-oxoglutarate aminotransferase and succinate-semialdehyde:NAD oxidoreductase; (Sigma–Aldrich, Inc., St. Louis, MO). The reactions were initiated by adding 0.1 mL sample. After 30 min, the reactions were terminated by boiling for 5 min. The amount of GABA consumed was calculated from the amount of NAPDH formed as spectrophotometrically determined at 340 nm based on a standard curve [19]. The free glutamate in the supernatant was determined as follows: four stock solutions were prepared: solution I contained: 0.2 M TEA, 25 mM phosphate, 13.2 g L1 Triton X-100 (pH 8.6); solution II contained: 17 mM NAD, 1.2 g diaphorase (4.8 U mg1, Sigma–Aldrich, Inc.); solution III contained 1.19 mM iodonitrotetrazolium chloride (INT); and solution IV contained glutamate dehydrogenase (GIDH, 1200 kU L1, Sigma– Aldrich). The reaction mixture included 0.3 mL solution
305
I, 0.1 mL solution II, 0.1 mL solution III, 0.1 mL sample (in the control, 0.1 mL water was added), and 0.9 mL water. After 2-min incubations at room temperature, 0.015 mL solution IV was added into the mixture. After an additional 2 min, the final reactant (INT-formazan) was determined spectrophotometrically at 492 nm [20]. 2.6. Statistical analysis The data consisted of the mean value of at least two independent experiments, and in the case of survival rate calculations the variations were less than 50%. Significant differences between the treatments were determined using the StudentÕs t-test [21].
3. Results Exposure to pH 2.0 saline solution for 1 h reduced the viable count of all enterobacteria tested to less than 0.001%, with the exception of a Proteus vulgaris strain G16 that had a 1.3% survival rate (Table 2). When the pH 2.0 saline solution was supplemented with 10 g L1 Casamino acids the survival rate of as many as 49 bacterial strains did not increase, but in most E. coli, four Morganella morganii, three Providencia, one Proteus mirabilis, one Enterobacter cloacae, one Hafniae alvei, and one Citrobacter braakii strain, this number increased markedly to 1% or greater. Similar results were observed if the saline solution included 6 mM glutamate. Cultures of several of these acid-resistant enterobacteria that were subjected to acid shocks in pH 2.0 saline solution supplemented with 1 g L1 of one of 17 other amino acids had a survival rate of less than 0.001%, with the exception of most E. coli with arginine (>0.2%) and P. vulgaris strain G16 with all of them (1.3%). Most E. coli strains tested positive for the presence of gadAB genes and had detectable GAD activity (Table 3). Besides strain EF522 that was deficient in GAD production due to transposon mutagenesis, E. coli O157:H7 strain 43890 was the only acid sensitive E. coli that had no enzymatic activity, but yielded PCR products for gadAB. All other acid-sensitive enterobacteria had no genotypic or phenotypic evidence of the presence of GAD. In addition to E. coli, the only acid-resistant enterobacteria that tested positive for gadAB and had GAD activity were a H. alvei and a C. braakii strain. The remaining acid-resistant enterobacteria strains Enterobacter cloacae, M. morganii, Providencia and P. mirabilis had no detectable gadAB genes and GAD activity. The E. coli strain EF522 that had insertions in the gadA and gadB genes was sensitive to acid, produced approximately 0.4 mmol g protein1 GABA and increased the pH to 3.9 after incubation of cell suspensions with glutamate at pH 3.4 (Table 3). Wild-type strain EK227
306
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
Table 2 Acid resistance of enterobacteria as affected by acid-shock media Bacterial species or genera
Escherichia coli Enterobacter cloacae Enterobacter sakazakii Enterobacter aerogenes Hafniae alvei Klebsiella spp. Serratia spp. Citrobacter spp. Kluyvera spp. Yersinia enterocolitica Proteus mirabilis Proteus vulgaris Morganella morganii Providencia spp.
Number of strains
1 14 4 1 12 3 2 1 9 3 5 1 1 1 4 1 1 1 4 1 3
Average acid resistance (%) No addition
Casamino acid (10 g L1)
Glutamate (1 g L1)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 1.29 <0.001 <0.001 <0.001 <0.001
<0.001 18.0 <0.001 1.00 <0.001 <0.001 <0.001 10 <0.001 <0.01 <0.001 17.79 <0.001 <0.001 <0.001 1.80 2.30 <0.001 8.97 <0.001 39.02
<0.001 11.7 <0.001 6.31 <0.001 <0.001 <0.001 1 <0.001 <0.01 <0.001 70.40 <0.001 <0.001 <0.001 1.08 1.29 <0.001 32.80 <0.001 6.92
Table 3 Determination of gadAB genes, GAD and GABA production in enterobacteria Strains (acid sensitive/resistant)
gadAB PCR assay
GAD assay
GABA levels mmol g protein1
Final pH
Escherichia coli EK227 (R) Escherichia coli EF522 (S) Escherichia coli O157 ATCC 43895 (R) Escherichia coli O157 ATCC 43890 (S) Enterobacter cloacae ATCC 700323 (S) Enterobacter cloacae ATCC 13047 (S) Enterobacter cloacae G07 (R) Enterobacter sakazakii ATCC 57329 (S) Enterobacter sakazakii G25 (S) Enterobacter aerogenes ATCC 13048 (S) Hafniae alvei G18 (R) Citrobacter braakii G15 (R) Proteus mirabilis G08 (R) Proteus vulgaris G16 (S) Morganella morganii G35 (R) Morganella morganii G57 (R) Morganella morganii G62 (R) Proteus stuartii G22 (R) Proteus stuartii G23 (S) Proteus stuartii G70 (R) Proteus rettgeri G37 (R)
+ ND + + – – – – – – + + – – – – – – – – –
+ – + – – – – – – – + + – – – – – – – – –
2.15 ± 0.12 0.35 ± 0.01 1.82 ± 0.08 0.36 ± 0.00 0.18 ± 0.01 0.17 ± 0.01 0.65 ± 0.15 0.20 ± 0.01 0.76 ± 0.20 0.15 ± 0.01 ND 1.43 ± 0.02 0.23 ± 0.02 0.31 ± 0.03 0.25 ± 0.01 0.67 ± 0.02 0.20 ± 0.04 0.43 ± 0.03 0.78 ± 0.09 0.19 ± 0.02 0.35 ± 0.03
5.6 ± 0.06 3.95 ± 0.02 5.66 ± 0.02 4.01 ± 0.02 3.94 ± 0.04 3.80 ± 0.01 4.46 ± 0.21 3.97 ± 0.02 4.66 ± 0.54 3.85 ± 0.04 ND 5.74 ± 0.07 3.99 ± 0.01 3.88 ± 0.02 3.80 ± 0.07 3.96 ± 0.13 3.86 ± 0.08 3.79 ± 0.02 4.41 ± 0.39 3.77 ± 0.058 3.99 ± 0.07
ND, not determined; R, acid-resistant strain that had a survival rate of 1% or greater after acid shock at pH 2.0 for 1 h; S, acid-sensitive strain that had a survival rate of 0.01% or less after acid shock at pH 2.0 for 1 h.
and most acid-resistant E. coli cells produced 5-fold more GABA than EF522 (2.1 vs. 0.35 mmol mg protein1) and increased the final pH to 5.6. C. braakii strain G15 produced approximately 1.4 mmol g protein1. The amount of GABA produced by other enterobacteria ranged from 0.17 to 0.78 mmol g protein1 and there was no statistically significant difference in the con-
centration produced by those strains that were resistant to pH 2.0 in the presence of glutamate as compared to those that were not. The final incubation media pH of strains that produced less than 0.8 mmol g protein1 GABA was always less than 4.5. E. coli strains EK227 and 43895 cultures were capable of consuming 5.6 mM glutamate when incubated
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
at pH 3.4 after 4 h with an initial concentration of 5.9 mM (Table 4). E. coli strain EF522 only consumed 10% of the original glutamate concentration under the same conditions. The amount of glutamate consumed by P. mirabilis G08 and M. morganii G62 was approximately 23% of the initial concentration and the rest of the acid-resistant enterobacteria consumed less than 20% of that number. When the concentration of glutamate in the pH 2.0acid-shock media was decreased to 37 lM or less the survival rate of all acid-resistant strains was less than 0.01% (Table 5). The survival rate of E. coli, C. braakii and H. alvei strains that were positive for GAD was maximal at glutamate concentrations between 74 and 148 lM. The survival rate of the Providencia spp. strains and M. morganii G35 strain was only maximal if the glutamate concentrations were greater than 296 lM. The remainder of the glutamate-dependent acid-resistant strains reached their maximum survival rate at 148 lM. If cells suspensions of any of the glutamate-dependent acid-resistant strains containing approximately 106 cells L1 were subjected to an acid shock in the
Table 4 Extent of glutamate consumption by cultures of enterobacteria when incubated at pH 3.4 with an initial glutamate concentration of 5.9 mM Strains
Escherichia coli EK227 Escherichia coli EF522 Escherichia coli O157 43895 Escherichia coli O157 43890 Proteus mirabilis G08 Morganella morganii G35 Morganella morganii G57 Morganella morganii G62 Providencia spp. G22 Providencia rettgeri G37
Consumed amount of glutamate mM
mmol (g protein)1
5.60 0.54 5.57 0.87 1.30 0.98 0.68 1.39 0.77 1.00
2.60 0.25 2.59 0.40 0.61 0.46 0.32 0.65 0.36 0.47
307
presence of 0.6 lM glutamate and 1.5 mM of L -transpyrrolidine-2,4-dicarboxylic acid (L -PDC), no survivors were detected.
4. Discussion We were able to identify bacterial strains capable of surviving pH 2.0 with the addition of amino acids. These bacteria were: fourteen Escherichia coli strains, one Enterobacter, one Hafnia, one Citrobacter, four Morganella, three Providencia, and one Proteus strains. Compared to other amino acids only glutamate enabled these bacteria to survive extreme acid shocks. When these strains were subjected to an acid shock in the presence of casamino acids (Table 2), their survival rate ranged from 1 to 39% which was slightly higher than with glutamate alone. This difference could have been due to a potential enhancement of the glutamate-dependent acid resistance when other amino acids were present. Since GAD has been one of most important amino acid-dependent acid-resistance elements identified in enterobacteria, we conducted PCR assays targeting the gadAB genes. Our results indicated that among acidresistant strains, E. coli, Hafnia alvei and Citrobacter braakii had the gadAB genes and this result was confirmed by the detection of GAD activity (Table 3). Because these bacteria produced GABA, they could utilize glutamate for acid resistance. Among enterobacteria, the GAD-mediated system has been identified in E. coli, Shigella and Providencia [5,12,13], but this is the first report in which H. alvei and C. braakii strains have been observed to possess this stress response mechanism. The presence of GAD has been suggested as a specific marker for detection of E. coli [18], but our findings suggest that this enzyme activity is also found among other enterobacteria. It should be noted that E. coli O157:H7 strain ATCC 43890 was the only strain that had the gadAB genes present, but had no GAD
Table 5 Effect of glutamate concentration on the survival rate of enteric bacteria Strains
Survival rate (%) with glutamate (lM) 18
37
74
148
296
592
Escherichia coli EK227 Escherichia coli O157:H7 ATCC 43895 Enterobacter cloacae G07 Citrobacter brakii G15 Hafniae alvei G18 Proteus mirabilis G08 Morganella morganii G35 Morganella morganii G57 Morganella morganii G62 Providencia stuartii G22 Providencia rettgeri G37
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.002 <0.001 <0.001 <0.001
0.3 25.7 0.002 100.0 0.4 0.056 <0.001 17.8 1.00 <0.001 <0.001
17.8 25.7 0.018 25.7 1.8 0.1 0.1 38.9 4.6 0.4 0.6
10 25.7 0.1 45.8 0.4 0.6 1.0 100.0 2.6 56.2 21.9
25.7 0.1 45.8 1.8 1.0 0.4 38.9 2.6 56.2 38.9 25.7
308
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309
activity. The explanation for this unique trait could be due to unidentified potential defects in gene expression and/or protein synthesis. These GAD-positive strains were capable of excreting significant levels of GABA, the product of glutamate decarboxylation, and this production appeared to be linked to the disappearance of glutamate from the media. Among recent studies that have investigated the glutamate-mediated acid resistance, very few have reported the production of GABA and the consumption of glutamate as important factors. De Biase et al. [22] indicated that acid-resistant E. coli could produce as much as 1.3 mM GABA during an acid shock, but the glutamate consumption was not measured. The link of GABA production rate with an increased acid resistance remains to be determined, but in our study these parameters were useful to identify non-GAD glutamate-dependent acid-resistance systems. Among acid-resistance strains, gad genes and glutamic acid decarboxylation were not detected in strains belonging to the Proteeae Tribe of Enterobacteriaceae (Morganella, Providencia, and Proteus) and in an Enterobacter isolate. However, glutamate addition to the acid media significantly increased their survival after a pH-2.0 acid shock. The glutamate supplementation stimulated the survival rate of these bacteria to similar levels observed with GAD-positive E. coli and their response to glutamate concentrations was comparable to E. coli. Because the Morganella, Providencia and Proteus strains did not consume significant amounts of glutamate when subjected to acid shocks, it appeared that their acid-resistance system was metabolically different from that of E. coli. Waterman and Small [23] reported that the acid resistance of E. coli and Shigella could be completely inhibited if the glutamate-containing acid-shock media was supplemented with L -trans-pyrrolidine-2,4-dicarboxylic acid (L -PDC), an inhibitor of glutamate transporters. Similar to E. coli, the acid resistance of Morganella, Providencia, Proteus and Enterobacter was sensitive to the addition of L -PDC. The glutamate-dependent systems of these strains, however, did not appear to metabolize glutamate at the same rate as E. coli (90% vs. less than 25%) and they produced significantly less GABA. These comparisons suggest that Morganella, Providencia, and Proteus possess glutamate-dependent acid resistance, but markedly different from the gad-mediated acid resistance identified in E. coli. E. coli utilizes glutamate for osmotic as well as acid resistance. Uptake of potassium glutamate enable bacteria cells to resist high osmotic pressure by converting glutamate into proline, an osmoprotectant solute [24]. Ogahara et al. [25] suggested that the gad system is induced during an osmotic shock to maintain the charge balance during the rapid uptake of K+ by converting
glutamate into GABA. Based on these reports it appears that E. coli relies on glutamate for both hyperosmotic as well as acid tolerance, but the potential link between the glutamante-dependent acid resistance of the Proteeae group strains and their osmotolerance remains to be determined. Proteus, Morganella, and Providencia share some phenotypic properties rarely encountered in other enterobacteria. The most intriguing of these properties is the ability to oxidatively deaminate some amino acids to keto acids, which may function in iron transport [26]. Similar to other enterobacteria, these Proteeae group also have a urease that can hydrolyze urea and cause alkalinization of the media. We confirmed that all of the Proteeae strains that we used in this research produced urease (data not shown). These bacteria have been implicated in urinary tract infection because release of ammonia can cause crystal deposition, stone formation, and cellular damage to the renal epithelium [27]. Despite that all these strains have urease activity, only Morganella can survive extreme acid as low as pH 2.0 by generating ammonia from urea [10]. Because of the unique physiological characteristics in Proteeae group, commercial detection systems rarely have difficulty in providing a correct identification of these organisms [28]. However, these unusual biochemical characteristics would imply the special metabolism in these groups. These alternative metabolic pathways may compensate the relatively limited ability to ferment diverse carbon sources. A potentially new acid resistance relying on glutamate might be uniquely present in the Proteeae. Glutamate dependent acid resistance that have newly identified in Proteeae tribe will contribute to broadening the understanding of these bacteria itself. Further research will be undertaken to identify this mechanisms.
Acknowledgements Funding for this project was provided by the Minnesota Agricultural Experiment Station. The authors thank Dr. John W. Foster (University of South Alabama College of Medicine) for providing E. coli EK227 and EF522 strains, Dr. Patricia Ferrieri for providing enterobacteria, and Dr. Kun-Ho Seo for providing Enterobacter sakazakii.
References [1] Gorden, J. and Small, P.L.C. (1993) Acid resistance in enteric bacteria. Infect. Immun. 61, 364–367. [2] Small, P., Blankenhorn, D., Welty, D., Zinser, E. and Slonczewski, J.L. (1994) Acid and base resistance in Escherichia coli and Shigella flexneri: role of rpoS and growth pH. J. Bacteriol. 176, 1729–1737.
G.woo Park, F. Diez-Gonzalez / FEMS Microbiology Letters 237 (2004) 303–309 [3] Foster, J.W. (2000) Microbial responses to acid stress In: Bacterial Stress Responses (Storz, G., Hengge-Aronis, R., Eds.), pp. 99–115. ASM Press, Washington, DC. [4] Richard, H.T. and Foster, J.W. (2003) Acid resistance in Escherichia coli. Adv. Appl. Microbiol. 52, 167–186. [5] Lin, J., Lee, I.S., Frey, J., Slonczewski, J.L. and Foster, J.W. (1995) Comparative analysis of extreme acid survival in Salmonella typhimurium, Shigella flexneri, and Escherichia coli. J. Bacteriol. 177, 4097–4104. [6] Meng, S.Y. and Bennett, G.N. (1992) Nucleotide sequence of the Escherichia coli cad operon: a system for neutralization of low extracellular pH. J. Bacteriol. 174, 2659–2669. [7] Park, Y.K., Bearson, B., Bang, S.H., Bang, I.S. and Foster, J.W. (1996) Internal pH crisis, lysine decarboxylase and the acid tolerance response of Salmonella typhimurium. Mol. Microbiol. 20, 605–611. [8] Bearson, S., Bearson, B. and Foster, J.W. (1997) Acid stress responses in enterobacteria. FEMS Microbiol. Lett. 147, 173–180. [9] Matin, A., Zychlinsky, E., Keyhan, M. and Sachs, G. (1996) Capacity of Helicobacter pylori to generate ionic gradients at low pH is similar to that of bacteria which grow under strongly acidic conditions. Infect. Immun. 64, 1434–1436. [10] Young, G.M., Amid, D. and Miller, V.L. (1996) A bifunctional urease enhances survival of pathogenic Yersinia enterocolitica and Morganella morganii at low pH. J. Bacteriol. 178, 6487–6495. [11] Marquis, R.E. and Bender, G.R. (1987) Arginine deaminase system and bacterial adaptation to acid environments. Appl. Environ. Microbiol. 53, 198–200. [12] Waterman, S.R. and Small, P.L. (2003) Identification of the promoter regions and sigma (s)-dependent regulation of the gadA and gadBC genes associated with glutamate-dependent acid resistance in Shigella flexneri. FEMS Microbiol. Lett. 225, 155– 160. [13] Freir, P.A., Graves, M.H. and Kocka, F.E. (1976) A rapid glutamic acid decarboxylase test for identification of bacteria. Ann. Clin. Lab. Sci. 6, 537–539. [14] Cotter, P.D., OÕReilly, K. and Hill, C. (2001) Role of the glutamate decarboxylase acid resistance system in the survival of Listeria monocytogenes LO28 in low pH foods. J. Food Prot. 64, 1362–1368. [15] Sanders, J.W. and Leenhouts, K. (1998) A chloride-inducible acid resistance mechanism in Lactococcus lactis and its regulation. Mol. Microbiol. 27, 299–310.
309
[16] Swanson, K.M.J., Petran, R.L. and Hanlin, J.H. (2001) Culture Methods for Enumeration of Microorganisms. Microbiological Examination of Foods. American Public Health Association, Washington, DC. [17] McDaniels, A.E., Rice, E.W., Reyes, A.L., Johnson, C.H., Haugland, R.A. and Stelma Jr., G.N. (1996) Confirmational identification of Escherichia coli, a comparison of genotypic and phenotypic assays for glutamate decarboxylase and b-D -glucuronidase. Appl. Environ. Microbiol. 62, 3350–3354. [18] Rice, E.W., Johnson, C.H., Dunnigan, M.E. and Reasoner, D.J. (1993) Rapid glutamate decorboxylase assay for detection of Escherichia coli. Appl. Environ. Microbiol. 59, 4347–4349. [19] Higuchi, T., Hayashi, H. and Abe, K. (1997) Exchange of glutamate and c-aminobutyrate in a Lactobacillus strain. J. Bacteriol. 179, 3362–3364. [20] Beutler, H.-O. (1983) L -glutamate, colorimeteric method with glutamate dehydrogenase and diaphorase In: Metabolites 3: Lipids, Amino Acids and Related Compounds (Grabl, M., Ed.), pp. 369–376. VCH Verlagsgesellschaft mbH, Weinheim. [21] Moore, D.S. and McCabe, G.P. (1999) Introduction to the Practice of Statistics. W.H. Freeman and Co., New York, USA. [22] De Biase, D., Tramonti, A., Bossa, F. and Visca, P. (1999) The response to stationary-phase stress conditions in Escherichia coli: role and regulation of the glutamic acid decarboxylase system. Mol. Microbiol. 32, 1198–1211. [23] Waterman, S.R. and Small, P.L. (2003) The glutamate-dependent acid resistance of Escherichia coli and Shigella flexneri is inhibited in vitro by L -trans-pyrrolidine-2,4-dicarboxylic acid. FEMS Microbiol Lett. 224, 119–125. [24] Booth, I.R. and Higgins, C.F. (1990) Enteric bacteria and osmotic stress: intracellular potassium glutamate as a secondary signal of osmotic stress. FEMS Microbiol. Rev. 75, 239–246. [25] Ogahara, T., Ohno, T., Takayama, M., Igarashi, K. and Kobayashi, H. (1995) Accumulation of glutamate by osmotically stressed Escherichia coli is dependent on pH. J. Bacteriol. 177, 5987–5990. [26] Ewing, W.H. (1986) Edwards and EwingÕs Identification of Enterobacteriaceae, 4th Ed.. Elsevier Inc., New York. [27] Mobley, H.L.T. and Warren, J.W. (1987) Urease-positive bacteriuria and obstruction of long-term urinary catheters. J. Clin. Microbiol. 25, 560–564. [28] Janda, J.M. and Abbott, S.L. (1998) The Enterobacteria. Lippincott-Raven Publishers, Philadelphia, PA.