点击显示 收起
Research Section, Infectious Diseases Section
Pathology Service, Louis Stokes Cleveland Veterans Affairs Medical Center, Case Western Reserve University
Division of Infectious Diseases, University Hospitals of Cleveland, Cleveland, Ohio
We used a mouse model to test the hypothesis that anaerobic microbiota in the colon inhibit the establishment of vancomycin-resistant enterococci (VRE) colonization by depleting nutrients within cecal contents and limiting the association of VRE with the mucus layer. Anaerobic growth of VRE was assessed in cecal contents and cecal mucus of mice that had received treatment with subcutaneous clindamycin or saline. VRE grew to high concentrations in cecal contents of clindamycin-treated mice and in cecal mucus of both groups but not in cecal contents of saline-treated mice, unless the cecal contents were autoclaved or converted into sterile filtrates. After orogastric inoculation of VRE, clindamycin-treated mice acquired high concentrations of VRE within the mucus layer, whereas saline-treated mice did not. These results suggest that colonic microbiota inhibit VRE by producing inhibitory substances or conditions rather than by depleting nutrients. The colonic mucus layer provides a potential niche for growth of VRE.
The indigenous microbiota of the colon inhibit the establishment of colonization by exogenously introduced microorganisms [1, 2]. Several mechanisms have been proposed to account for this important host defense, which has been termed "colonization resistance." First, depletion of nutrients by the indigenous microbiota contributed to the inhibition of exogenously introduced strains of Escherichia coli and Clostridium difficile in continuous-flow cultures that contained cecal microbiota from mice [2, 3]. Second, the prevention of access of invading microorganisms to attachment sites or niches associated with the lining of the colon may facilitate their removal [4]. Finally, the production of inhibitory substances or conditions by indigenous microbiota may inhibit the growth of invading microorganisms. For example, short-chain fatty acids (SCFAs) and branched-chain fatty acids (BCFAs) may inhibit colonization by prolonging the length of the lag phase of growth that bacteria commonly undergo when they enter a new environment [2, 5]. Hydrogen sulfide may reduce growth indirectly by restricting the range of substrates that an organism can efficiently utilize for anaerobic growth [2].
Vancomycin-resistant enterococci (VRE) are important nosocomial pathogens that colonize the intestinal tract [6]. We have provided evidence that the anaerobic component of the intestinal microbiota plays a crucial role in the inhibition of VRE [6, 7], but the mechanisms by which VRE are inhibited are not well defined. Because competition for nutrients plays an important role in the inhibition of other microorganisms, we examined the ability of VRE strains to utilize a variety of carbohydrates that have been shown to be major energy sources in the colon [8]. VRE strains were unable to utilize complex carbohydrates that are present in foods or mucin (a major component of the mucus layer of the gastrointestinal tract) as sole energy sources, but they were able to utilize a variety of monosaccharides, including some of those present in mucin (i.e., N-acetylgalactosamine and N-acetylglucosamine; authors' unpublished data). Because the mucus layer of the colon has been shown to provide an important source of nutrients and adhesion sites that facilitate the colonization of streptomycin-treated mice by E. coli and Salmonella species [912], we hypothesized that the mucus layer might provide a similar niche for VRE. We further hypothesized that anaerobic microbiota inhibit the establishment of VRE colonization by depleting nutrients within the lumen of the colon and by limiting the association of VRE with the mucus layer.
MATERIALS AND METHODS
The study protocol was approved by the Cleveland Veterans Affairs Medical Center's Animal Research Committee. Female CF-1 mice weighing 2530 g (Harlan Sprague-Dawley) were used in all experiments. Mice were housed in individual cages with plastic filter tops, to prevent cross-contamination among animals. The primary VRE isolate studied was C68, a vanB-type vancomycin-resistant Enterococcus faecium strain that we have used in previous mouse studies [7]. More limited experiments were conducted with 8 additional vancomycin-resistant E. faecium isolates representing different types that had been isolated from the Cleveland area by pulsed-field gel electrophoresis [13].
Growth of VRE in cecal mucus versus cecal contents.
Subcutaneous clindamycin (1.4 mg/kg) or normal saline was administered daily for 3 days. Four mice were included in each experiment group. Food was removed from the cages 2 h before each mouse was killed by CO2 asphyxiation; preliminary experiments demonstrated that results were similar when food was removed 2 or 16 h before they were killed (data not shown). Clindamycin was used because it inhibits anaerobes in the colon without reducing levels of facultative gram-negative bacilli or vancomycin-susceptible enterococci [7], and we have shown that this agent consistently promotes the overgrowth of VRE in mice and colonized patients [6, 7]. The cecum was removed and opened longitudinally. Cecal contents were collected for analysis of pH, concentrations of SCFAs and BCFAs, and bacterial cultures. SCFAs (acetic acid, butyric acid, and propionic acid) and BCFAs (isobutyric acid, isovaleric acid, 2-methylbutyric acid, and valeric acid) were measured by gas chromatography, as described elsewhere [14]; the BCFAs isovaleric acid and 2-methylbutyric acid were not separated by use of these methods. Total numbers of anaerobes, facultative gram-negative bacilli, and enterococci were quantified by plating of serially diluted specimens onto Brucella, MacConkey, and Enterococcosel agar (Becton Dickinson), respectively [14]. Cultures of total anaerobes were performed inside an anaerobic chamber (Coy Laboratories). The cecal mucus layer was removed by scraping with a spatula, as described by Cohen and Laux [15]. The mucus was diluted 1 : 4 (vol : vol) in PBS; preliminary experiments demonstrated that growth was similar in undiluted and diluted mucus.
To examine the ability of VRE to grow in cecal contents or mucus under anaerobic conditions, growth curves were performed inside the anaerobic chamber. Cecal contents and mucus samples were transferred to the anaerobic chamber within 5 min of mouse death. Then, 104 cfu/mL VRE were inoculated into the samples, and the samples were incubated at 37°C. Quantitative cultures were performed at specified intervals by plating of serially diluted samples onto Enterococcosel agar that contained 20 g/mL vancomycin [7].
Mechanisms of the inhibition of VRE by cecal contents.
To further evaluate the mechanisms by which VRE growth is inhibited by cecal contents, additional anaerobic growth curves were performed after modification of the cecal contents of 4 saline-treated mice. Modifications included sterilization by autoclave for 15 min, the addition of brain-heart infusion broth (10%), and the preparation of sterile filtrates. Sterile filtrates were produced by centrifugation at 5000 g for 10 min, followed by filtration of the supernatant through a 0.22-m filter.
Because SCFAs and BCFAs have been shown to inhibit the growth of some pathogens [2, 5], we evaluated their possible role in the inhibition of VRE. Growth curves were performed in brain-heart infusion broth with or without supplementation by SCFAs and BCFAs, to reproduce the levels present in the cecal contents of saline-treated mice; the pH was adjusted to approximate the range of pH values obtained from the cecal contents of saline-treated mice.
In vivo association of VRE C68 with the cecal mucus layer.
To evaluate whether clindamycin treatment facilitated the association of VRE with the mucus layer, mice that received clindamycin or saline treatment as described above were killed 4, 6, 8, 24, or 48 h after inoculation with 106 cfu of VRE C68 diluted in 0.5 mL of PBS. Three mice were included in each treatment group at each time point. The cecum was opened longitudinally, and cecal contents were removed. The cecal lining was rinsed gently 3 times with sterile PBS, and mucus was scraped into HEPES-Hanks buffer (as described above). Quantitative cultures for VRE were performed by plating of serially diluted samples of cecal contents or mucus onto Enterococcosel agar that contained 20 g/mL vancomycin.
Additional mice (3 mice/group) were killed 24 h after inoculation with VRE, to evaluate pathologically whether VRE C68 became associated with the mucus layer. The cecum was prepared as described above, except that the mucus layer was not removed. Each cecum was placed into 3 mL of fixing solution that contained methanol, chloroform, and acetic acid (6 : 3 : 1) for 4050 min and then transferred to 3 mL of fresh fixing solution for an additional 45 min. The tissue was then removed from the fixing solution, placed into 70% ethanol, and stored at 4°C until it was ready for sectioning. Once the tissue was ready for sectioning, it was embedded in paraffin, and 5-m-thick cross sections were prepared and were placed onto multiple glass slides. To remove the paraffin, the slides were placed into fresh xylene for 1015 min, followed by 2 washes in 95% ethanol for 3 min each. After air drying, separate slides were stained with mucicarmine or tissue Gram stain.
To confirm that some of the gram-positive cocci visualized by light microscopy were E. faecium, a commercially available fluorescence in situ hybridization kit specific for E. faecium (Microscreen) was used. The slides were processed as described above. The cecal sections were circumscribed with an ImmEdge pen (Vector Laboratories) and hybridized with the CY3-labeled probe from the commercial kit by use of the conditions described by Miranda et al. [16], except that incubation was performed overnight at 50°C. After hybridization, the sections were viewed by fluorescent and confocal microscopy.
Statistical analysis.
Analysis of variance was performed to compare the densities of VRE among the growth curve groups. Student's t test was used to compare the pH values, volatile fatty-acid concentrations, and bacterial densities among saline-treated versus clindamycin-treated mice. Computations were performed with Stata software (version 5.0; StataCorp). P < .05 was considered to be significant.
RESULTS
Table 1 shows a comparison of pH, SCFA and BCFA concentrations, and microbiota of cecal contents obtained from clindamycin- and saline-treated mice. Clindamycin treatment resulted in significant reductions in the densities of total anaerobes as well as in the concentrations of acetic, butyric, and propionic acids.
(1 of 2 images)
DISCUSSION
Our findings suggest that the anaerobic microbiota of the colon inhibit the growth of VRE by producing inhibitory substances or conditions rather than by depleting nutrients. VRE did not replicate in cecal contents of saline-treated mice under anaerobic conditions, but they grew rapidly to high concentrations when the contents were autoclaved or converted into sterile filtrates (figure 2), which demonstrates that nutrients were available to support enterococcal growth. The addition of nutrient media to the cecal contents did not facilitate the growth of VRE, which further demonstrates that replication was inhibited even in the presence of a large surplus of nutrients. In contrast, VRE grew rapidly in the cecal contents of clindamycin-treated mice after an initial lag phase of 6 h (figure 1). Clindamycin treatment resulted in marked suppression of cecal anaerobes but did not reduce the levels of facultative gram-negative bacilli or cocci (table 1).
Further work is needed to identify the specific inhibitory substances and/or conditions that prevent the growth of VRE in cecal contents of saline-treated mice and to determine the mechanisms by which growth is inhibited. Our data suggest that pH and SCFAs or BCFAs do not play a significant role in the inhibition of VRE. Supplementation of brain-heart infusion broth with SCFAs and BCFAs to approximate the levels found in the cecal contents of saline-treated mice did not result in the inhibition of VRE at pH values present in the cecum (figure 3). The loss of inhibition during the preparation of sterile filtrates suggests the possibility that a volatile inhibitor could be present in the cecal contents [4]. Freter et al. [4] similarly found that the growth of E. coli was inhibited in a continuous-flow culture that contained mouse cecal microbiota but not in a sterile filtrate prepared from the culture. It was shown that hydrogen sulfide in the culture was lost during the preparation of sterile filtrates and that restoring the levels of this gas in the filtrates to equal the levels in the culture resulted in the inhibition of the growth of E. coli [4]. Freter et al. [4] found that hydrogen sulfide inhibited growth indirectly by restricting the range of substrates that an organism could efficiently utilize for anaerobic growth. An alternative explanation for our findings could be that physical contact between living obligate anaerobes and VRE is necessary to obtain the inhibitory effect.
We found that VRE isolates grew to high concentrations in vitro in the cecal mucus of saline-treated mice (figure 1); however, after oral inoculation, minimal in vivo association of VRE C68 with the cecal mucus layer was observed on the basis of quantitative cultures and histological findings (figure 4). In contrast, relatively high concentrations of VRE became associated with the cecal mucus layer of clindamycin-treated mice. These data provide support for the hypothesis that the anaerobic microbiota may reduce the ability of ingested VRE to associate with the cecal mucus layer. Such reductions in association with the mucus layer could occur because growth is inhibited within the lumen of the cecum or because the anaerobic microbiota block access of VRE to adherence sites or niches within the mucus layer. Jin et al. [17] demonstrated that E. faecium strains may adhere to specific receptors in the ileal mucus of piglets.
Although our data suggest that the cecal mucus layer could provide a potential niche for the growth of VRE, growth in mucus was not essential for the establishment of VRE colonization in clindamycin-treated mice. In these mice, VRE replicated in vitro in cecal contents, and much higher in vivo concentrations of VRE were present in cecal contents, compared with those in mucus. On the basis of our previous studies, however, we hypothesize that growth in mucus could possibly play a role in the persistence of colonization. We have shown that the anaerobic microbiota of mice recover within 10 days after discontinuation of clindamycin treatment, and mice are not susceptible to establishment of VRE colonization at that time [18]. However, mice that receive clindamycin treatment in conjunction with oral VRE develop high-density colonization that persists for 35 weeks after the clindamycin is discontinued [7]. Similarly, VRE-colonized patients may have detectable levels of colonization for weeks or months after antibiotic therapy has been discontinued [6].
Acknowledgments
We thank Rolf Freter and Paul Cohen, for helpful discussions.
References
1. Bohnhoff M, Drake BL, Miller CP. Effect of streptomycin on susceptibility of intestinal tract to experimental Salmonella infection. Proc Soc Exp Biol Med 1954; 86:1327. First citation in article
2. Freter R, Brickner H, Botney M, Cleven D, Aranki A. Mechanisms that control bacterial populations in continuous-flow culture models of mouse large intestinal flora. Infect Immun 1983; 39:67685. First citation in article
3. Wilson KH. The microecology of Clostridium difficile. Clin Infect Dis 1993; 16(Suppl 4):S2148. First citation in article
4. Freter R, Brickner H, Fekete J, Vickerman MM, Carey KE. Survival and implantation of Escherichia coli in the intestinal tract. Infect Immun 1983; 39:686703. First citation in article
5. Que JU, Casey SW, Hentges DJ. Factors responsible for increased susceptibility of mice to intestinal colonization after treatment with streptomycin. Infect Immun 1986; 53:11623. First citation in article
6. Donskey CJ, Chowdhry TK, Hecker MT, et al. Effect of antibiotic therapy on the density of vancomycin-resistant enterococci in the stool of colonized patients. N Engl J Med 2000; 343:192532. First citation in article
7. Donskey CJ, Hanrahan JA, Hutton RA, Rice LB. Effect of parenteral antibiotic administration on persistence of vancomycin-resistant Enterococcus faecium in the mouse gastrointestinal tract. J Infect Dis 1999; 180:38490. First citation in article
8. Cummings JH, Englyst HN. Fermentation in the human large intestine and the available substrates. Am J Clin Nutr 1987; 45:124355. First citation in article
9. Wadolkowski EA, Laux DC, Cohen PS. Colonization of the streptomycin-treated mouse large intestine by a human fecal Escherichia coli strain: role of growth in mucus. Infect Immun 1988; 56:10305. First citation in article
10. Wadolkowski EA, Laux DC, Cohen PS. Colonization of the streptomycin-treated mouse large intestine by a human fecal Escherichia coli strain: role of adhesion to mucosal receptors. Infect Immun 1988; 56:103643. First citation in article
11. McCormick BA, Stocker BAD, Laux DC, Cohen PS. Roles of motility, chemotaxis, and penetration through and growth in intestinal mucus in the ability of an avirulent strain of Salmonella typhimurium to colonize the large intestine of streptomycin-treated mice. Infect Immun 1988; 56:220917. First citation in article
12. Moller AK, Leatham MP, Conway T, et al. An Escherichia coli MG1655 liposaccharide deep-rough core mutant grows and survives in mouse cecal mucus but fails to colonize the mouse large intestine. Infect Immun 2003; 71:214252. First citation in article
13. Donskey CJ, Schreiber JR, Jacobs MR, et al. A polyclonal outbreak of predominantly vanB vancomycin-resistant enterococci in Northeast Ohio. Clin Infect Dis 1999; 29:5739. First citation in article
14. Donskey CJ, Hume ME, Callaway TR, Das SM, Hoyen CK, Rice LB. Inhibition of vancomycin-resistant enterococci by an in vitro continuous-flow competitive exclusion culture containing human stool flora. J Infect Dis 2001; 184:16247. First citation in article
15. Cohen PS, Laux DC. Bacterial adhesion to and penetration of intestinal mucus in vitro. Methods Enzymol 1995; 253:30915. First citation in article
16. Miranda RL, Conway T, Leatham MP, et al. Glycolytic and gluconeogenic growth of Escherichia coli 0157:H7 (EDL933) and E. coli K-12 (MG1655) in the mouse intestine. Infect Immun 2004; 72:166676. First citation in article
17. Jin LZ, Marquardt RR, Zhao X. A strain of Enterococcus faecium (18C23) inhibits adhesion of enterotoxigenic Escherichia coli K88 to porcine small intestine mucus. Appl Environ Microbiol 2000; 66:42004. First citation in article
18. Stiefel U, Pultz NJ, Helfand MS, Donskey CJ. Susceptibility to vancomycin-resistant Enterococcus intestinal colonization persists after completion of anti-anaerobic antibiotic treatment in mice. Infect Control Hosp Epidemiol 2004; 25:3739. First citation in article