Insertion Mutagenesis of wca Reduces Acid and Heat Tolerance of Enterohemorrhagic Escherichia coliO157:H7

Construction of CA-deficient mutant.More than a hundredE. coli O157:H7 strains from our laboratory collection were screened on SMAC agar for EPS-forming colonies, a presumptive indication of CA production. CA-producing E. coli O157:H7 W6-13 was eventually selected as a parental strain because of its production of copious amounts of EPS. Two PCR fragments (upstream and downstream) were amplified from the W6-13 chromosome using the oligonucleotide primer pairs, P1 and P2 (5′-AAAGCTTAAACCGGACGTCACT-3′ and 5′-GAATTCTCCTCCACACCATGCCAAT-3′;EcoRI site underlined) and P3 and P4 (5′-AAAGAATTCCAAAGGCATTGC-3′ and 5′-TTCGCGTCAGCACACAATTC-3′; EcoRI site underlined), respectively. The PCR primers were derived based on GenBank sequences within wcaC, wcaD, and wcaE ofE. coli K-12 (GenBank accession no. U38473). Both primers 2 and 3 had an artificial EcoRI site near the 5′ end. Following PCR, the two amplified DNA fragments were digested withEcoRI and then ligated to each other to create anEcoRI restriction site. This fragment with theEcoRI site was subsequently cloned into plasmid vector pGEM-T (Promega Co., Madison, Wis.) to generate recombinant plasmid pYM37 (wcaΔDΔE). TheEcoRI restriction site on pYM37 was used to receive anEcoRI-EcoRI kanamycin resistance gene cassette which was PCR amplified from pNEO (Amersharm Pharmacia Biotech, Inc., Piscataway, N.J.) and restricted with EcoRI. The resulting plasmid was designated pYM3720 from which thewcaDE::kan fragment was amplified with primer pair P1 and P4. AmplifiedwcaDE::kan fragment was introduced intoE. coli O157:H7 W6-13 by electroporation. Recombinant colonies were selected on Luria-Bertani agar supplemented with kanamycin (100 μg/ml). One of the colonies, M4020 (W6-13wcaDE::kan), was isolated. The site of insertion was confirmed by PCR amplification and Southern hybridization with a kanamycin resistance gene probe. Insertion mutagenesis abolished the function of wca, whereby M4020 became defective in CA production.

wcaC, wcaD, and wcaE of E. coliO157:H7 were sequenced using an ABI Prism PCR Sequencer (Molecular Genetics Instrumentation Facility, University of Georgia). The sequences were deposited in GenBank (accession no. AF320069). Sequence analysis revealed that these genes share 98% homology with the same genes of E. coli K-12.

Quantification of CA production.CA produced by E. coli W6-13 and M4020 was quantified according to a procedure previously described by Junkins and Doyle (14). The UA/P ratio, defined as micrograms of uronic acid per milligram of protein, was used to estimate CA production of both strains on minimal glucose agar (MGA), MacConkey (MAC) agar, and SMAC agar. While colonies of W6-13 and M4020 were visible on agar plates after 16 h of incubation, development of a mucoid appearance by W6-13 required at least 24 h of incubation at 25°C. UA/P ratios of the same strain varied substantially among media used for cultivating the cells. However, CA production increased with incubation time, with the greatest amount being produced at day 2 on MGA and MAC agar and at day 4 on SMAC among the various days of incubation studied (Fig.1A). Colonies became extremely mucoid due to overproduction of CA (Fig. 1B). Greater amounts of CA were observed when W6-13 was grown on MGA than on complex media, such as MAC agar, SMAC agar (Fig. 1A), and MGA containing 0.5% Casamino Acids (wt/vol) (data not shown). This agrees with previous reports that CA production in E. coli K-12 is enhanced when cells are grown under less-than-optimal conditions for growth, such as in presence of low temperature, high salt concentrations, or high concentrations of utilizable carbon sources (18, 21, 27). Although many strains of E. coli O157:H7 produce CA, the amounts produced can vary greatly. Furthermore, CA production occurs when cultures grow at 25°C rather than at 37°C, which may be a feature to provide bacteria with a means to respond to stress and better survive conditions outside their human and animal hosts.

• Open in new tab
• Download powerpoint

Fig. 1.

UA/P ratio of E. coli O157:H7 W6-13 and its CA-deficient mutant M4020 when grown on different media, and CA production by colonies of W6-13 and M4020. (A) UA/P ratio of W6-13 and M4020 grown at 25°C for 1, 2, and 4 days on MGA, MAC agar, and SMAC agar. (B) CA production of E. coli O157:H7 W6-13 (left) and M4020 (right) when grown on MGA at 25°C for 4 days.

CA production by E. coli O157:H7 W6-13 was also compared to that of the nonpathogenic E. coli K-12 strain ZK2686 which is known to produce CA (6). The K-12 strain produced colonies with much less EPS than by E. coli O157:H7 W6-13 under the same conditions on MGA, supporting our earlier hypothesis that the overproduction of CA may play a more important role in greater resistance to environmental stresses by certain pathogenic than nonpathogenic microorganisms.

Protective effect of CA on acid and thermal tolerance.In this study, W6-13 and M4020 were subjected to acid treatment with pH levels ranging from 4.5 to 6.5 and to heat treatment at 55 and 60°C to evaluate the effect of CA expression on acid and thermal tolerance. For the acid treatment, W6-13 and M4020 were inoculated onto MGA and incubated at 25°C for 24 h. Colonies were removed with minimal glucose broth (MGB), and cell suspensions of W6-13 and M4020 were inoculated (1%) (vol/vol) into fresh MGB containing 5% Casamino Acids adjusted to pH 4.5, 5.5, and 6.5 with 0.1 M HCl. The cultures were incubated at 37°C, and viable counts of E. coli O157:H7 were determined at selected intervals. Approximately 10⁸CFU/ml of cell suspension of each strain was heated at 55 and 60°C for the thermal inactivation studies. Viable counts of E. coli O157:H7 were determined by sampling at selected time intervals, plating bacteria on MGA containing 0.5% Casamino Acids and incubating the plates at 37°C for 24 h. The D-value, which is defined as the time at a specific temperature that is required to inactivate 1 log of the bacterial population (12), was calculated by linear regression analysis.

Although insertional mutagenesis abolished the function ofwca, mutant M4020 at 37°C grew at a similar rate and survived as well as its isogenic parental strain W6-13. During acid treatment with pH levels ranging from 4.5 to 6.5, both W6-13 and M4020 grew in MGB at a similar rate at pH 6.5 (Fig.2A); however, the growth of M4020 lagged slightly at pH 5.5 (Fig. 2B). The growth of W6-13 at pH 4.5 was not affected substantially compared to growth at pH 5.5 or 6.5; however, the growth of M4020 was completely inhibited at pH 4.5 (Fig. 2C). Similarly, M4020 was more susceptible than W6-13 to sublethal temperatures as reflected by significantly lower D-values at 55 and 60°C (Table 1).

• Open in new tab
• Download powerpoint

Fig. 2.

Viability of E. coli O157:H7 W6-13 (●) and M4020 (○) at 37°C in MGB containing 0.5% Casamino Acids and adjusted to pH 4.5, 5.5, or 6.5 with 0.1 M HCl.

It has been suggested based on the results of previous studies that CA expressed on cell surfaces may simply act as a physical barrier to protect cells from hostile environmental conditions, such as enabling cells to retain water in a dry environment (1, 18, 21,27). It was determined in another study that E. coliO157:H7 cells are rapidly inactivated by a rapid accumulation of protons (13). In an acidic environment, the lower the pH, the greater the concentration of protons. CA confers a strong negative charge to the cell surface which, we suggest may serve as a buffer by neutralizing protons at the cell surface, whereby preventing positively charged chemical groups from accumulating on cell envelopes and from penetrating into cells. We further hypothesize that the amount of CA on cell surfaces determines the buffering capacity of cells. When cells lose their ability to produce CA, cell surfaces become less negatively charged and thereby have reduced buffering capacity. When negatively charged cell surfaces are neutralized, protons will accumulate and enter cells freely. Such a change in intracellular pH will impair cell metabolism, causing cell death.

Studies revealed that the expression of CA is regulated by several direct and indirect mechanisms (9, 16, 19). The transcription of the wca operon is positively regulated by RcsA, RcsB, and RcsF and negatively regulated by ATP-dependent protease Lon (3, 11, 16, 29). RcsC can be activated by environmental stimuli and both positively and negatively regulates CA biosynthesis (29, 31). RcsC and RcsB act as a sensor and an effector, respectively, in a two-component regulatory system (9, 29, 31). Several heat shock proteins, including Dnak (Hsp70), GrpE, and DnaJ, also influence CA biosynthesis (16). A recent study revealed that DjlA, a member of the DnaJ family, is a DnaK cochaperone, and the induction of the colanic acid requires this DnaK-DjlA interaction to stimulate the RcsB-RcsC two-component signaling system in E. coli (7). Although CA on cell surfaces may act as a physical barrier and affect heat conductivity, thereby preventing heat from reaching cells readily, the influence of several heat shock proteins on the regulation of CA induction could further suggest a role for CA in the heat response ofE. coli O157:H7.

Results of this study demonstrated that colanic acid is inducible inE. coli O157:H7, mostly by certain environmental stimuli, and that the induction of colanic acid has a substantial protective effect on the pathogen's acid and thermal tolerance. However, there are likely other factors of E. coli O157:H7 that confer to the pathogen additional protection against stressful environmental conditions. The mechanism of stress response in E. coliO157:H7 is complex, and other factors in addition to the induction of colanic acid are likely involved in conferring tolerance to the pathogen under stress-imposed conditions. Although there is limited information addressing the association of CA production with pathogenicity (1), a recent report suggests a role for colanic acid in the architecture of biofilms in the K-12 strain (6). Additional studies are under way to further elucidate the role of CA in the pathogen's exceptional tolerance to other forms of environmental stress, as well as in bacterial attachment and biofilm formation.