Abstract
Escherichia coli is a major cause of bloodstream infections and death due to sepsis. It is the most frequent Gram-negative bacterial pathogen recovered from cultures of blood from both community-acquired and nosocomial cases. We set out to determine the relationships between E. coli virulence factors (VFs), phylogenetic groups, and antibiotic resistance and whether bacteremia cases had a community, health care-associated. or nosocomial origin. Isolates from consecutive episodes of E. coli bacteremia in 303 patients presenting to a university hospital were screened for their VFs, phylogenetic group, and antibiotic resistance. The majority of VFs present in the collection were equally distributed between antibiotic-susceptible and multiple-drug-resistant (MDR) isolates, but the overall VF score was higher for isolates of community and health care-associated origin than those of nosocomial origin (P = 0.0002 and P = 0.0172, respectively); the papA, papG allele II, hlyA, and hek VFs were more prevalent in this cohort. Most isolates belonged to phylogenetic group B2, which harbored a greater proportion of antibiotic-susceptible isolates than MDR isolates (P = 0.04). The community, health care-associated, or nosocomial origin of E. coli bacteremia determines the virulence capacity of an isolate better than the phylogenetic group does. This study provides new insights into the relationships between the pathogenesis and epidemiology of E. coli bacteremia.
Escherichia coli is a leading cause of bloodstream infections worldwide, and the associated rate of mortality is high (32). Extraintestinal pathogenic E. coli (ExPEC) is the predominant cause of these invasive infections, which often originate from the urinary tract. They possess a wide variety of specialized virulence factors (VFs) responsible for pathogenesis outside the gastrointestinal tract, including diverse adhesins, invasins, and protectins (48). ExPEC isolates are generally concentrated within phylogenetic group B2 and, to a lesser extent, group D, whereas less virulent and commensal E. coli isolates belong to groups A and B1 (25, 29, 37). However, the minimal requirements for bacterial invasion of the bloodstream have yet to be determined.
Bacteremic isolates harbor a significantly greater repertoire of VFs than gastrointestinal tract commensal E. coli isolates (46), and where their molecular and epidemiological environments have been further analyzed, it appears that antibiotic-susceptible and -resistant ExPEC isolates are fundamentally different bacterial populations (19, 21, 23, 25, 27, 37, 38, 42, 43, 50). Antibiotic-susceptible strains mostly derive from phylogenetic group B2 and are associated with higher VF prevalences than antibiotic-resistant strains, which are typically associated with shifts toward groups D and A. This decreased prevalence of VFs within resistant strains has been suggested to be a possible trade-off between resistance and virulence in ExPEC (8). An important epidemiological consideration is whether community-acquired (CA) or health care-associated (HCA) E. coli strains causing bacteremia are distinct from strains that cause nosocomially acquired (NA) bacteremia. The only study to address this to date found that CA bacteremic isolates were significantly associated with papC and papG fimbrial VFs, phylogenetic group B2, and antibiotic susceptibility compared to the associations for their NA counterparts (21).
In the study described here, we investigated a large collection of epidemiologically well documented E. coli bacteremia isolates to determine the relationships between VFs, phylogenetic group, and antibiotic resistance. We set out to draw a clearer distinction than has previously been achieved between fully antibiotic-susceptible isolates, those resistant to a limited set of antibiotics, and those resistant to multiple antibiotic classes on the basis of the fact that the last group would most closely represent a nosocomial population.
MATERIALS AND METHODS
Study setting, bacterial strains, and patient data.
This study was based at St. James's Hospital (SJH), which has 900 inpatient beds and which serves a large local inner-city community. It is also a tertiary referral center for hematopoietic stem cell transplantation, burns, and maxillofacial and plastic surgery and also provides supraregional services for HIV medicine, cardiothoracic surgery, and oncology. We studied a collection of consecutive E. coli bloodstream isolates recovered from 303 adult patients who presented with sepsis between January 2004 and December 2006; repeat isolates were not included. The identities of all isolates were confirmed in the diagnostic microbiology laboratory with an API 20 E kit (bioMerieux, Marcy l'Etoile, France), prior to inclusion in the study.
During this period, the antibiotics recommended in the hospital formulary for the treatment of patients with suspected/proven sepsis caused by a Gram-negative organism were either ciprofloxacin or piperacillin-tazobactam usually in combination with gentamicin; the recommended second-line antibiotics were meropenem and amikacin. Expanded-spectrum cephalosporins were not recommended for this indication. As part of the European Antibiotic Resistance Surveillance System (EARSS) (14), there was an Irish initiative in 2004 by the Irish EARSS Steering Group to gather clinical information generated at the time of each documented episode of bacteremia. The following were defined as part of this initiative: NA bacteremias were those in which cultures of blood from inpatients who became symptomatic >48 h after admission were positive. HCA bacteremias were those from (i) inpatients in SJH during the previous 90 days, (ii) outpatients in SJH during the previous 30 days, (iii) patients referred or transferred from another hospital, or (iv) nursing home residents. CA bacteremias were those not encompassed in the definitions of NA and HCA bacteremias. Prior approval for the retrieval and use of the clinical information presented above was obtained from the SJH Ethics Committee.
Virulence factor analysis.
Isolates were screened by PCR for 16 VFs with known or suspected clinical relevance to ExPEC pathogenesis (29). The reactions were analyzed in five pools, as follows: in pool 1, malX (pathogenicity-associated island [PAI]), papA (P fimbriae, structural subunit), fimH (type 1 fimbriae, adhesin molecule), and ibeA (invasin of brain endothelium); in pool 2, fyuA (yersiniabactin receptor), sfa/focDE (S/F1C fimbriae), iutA (aerobactin receptor), and papG allele III (P fimbriae, adhesin molecule); in pool 3, hlyA (α-hemolysin), papG allele I (P fimbriae, adhesin molecule), and kpsMT II (group 2 capsule lipopolysaccharide); in pool 4, traT (serum/complement resistance) and papG allele II (P fimbriae, adhesin molecule); and in pool 5, afa/draBC (Dr [blood group] fimbriae) and cnf1 (cytotoxic necrotizing factor type 1). A separate PCR was carried out for hek/hra (invasion of epithelial cells) with primer hek-F (CGAATCGTTGTCACGTTCAG) and primer hek-R (TATTTATCGCCCCACTCGTC). The VF score for each isolate was calculated as the sum of all VF genes for which the isolate tested positive (26).
Phylogenetic group.
Multiplex PCR amplifications employed three markers: (i) chuA, (ii) yjaA, and (iii) TSPE4.C2. Use of these markers allowed the classification of the E. coli isolates into one of four phylogenetic groups (group A, B1, B2, or D) by use of a dichotomous decision tree (7).
Antibiotic susceptibility testing.
The MICs of ampicillin, piperacillin, cefoxitin, ceftazidime, cefotaxime, gentamicin, amikacin, nalidixic acid, ciprofloxacin, and meropenem were determined by agar microdilution methods and those of amoxicillin-clavulanic acid, piperacillin-tazobactam, cefpodoxime, and trimethoprim-sulfamethoxazole were determined by disk diffusion methods, in accordance with Clinical and Laboratory Standards Institute guidelines (9-11). The double-disk diffusion test was employed to assess the isolates for the production of expanded-spectrum β-lactamases (ESBLs) (10). A susceptible isolate was defined as one that was fully susceptible to members of all antibiotic classes (penicillins, β-lactam-β-lactamase inhibitor combinations, cephalosporins, carbapenems, fluoroquinolones [FQs], aminoglycosides, folate pathway inhibitors); a resistant isolate was defined as one that was resistant to a member(s) of one to two antibiotic classes; and a multidrug-resistant (MDR) isolate was defined as one that was resistant to members of three or more antibiotic classes.
Molecular analysis of resistance genes.
Isolates resistant to one or more cephalosporins were screened for six plasmidic AmpC β-lactamase families (MOX, CIT, DHA, ACC, EBC, and FOX) by multiplex PCR (40). Full-length PCR amplicons used for sequencing were generated by using primers designed to flank entire blaCMY-2 (40) and blaACC alleles (primer ACC-F, CATCACGATCCCCATCTTCT; primer ACC-R, TACCTGTCTGGCAGGAACTG). The promoter and attenuator regions of their chromosomal ampC genes were analyzed by PCR (6).
Statistical analyses.
To compare the proportions of given characteristics between particular populations, Fisher's exact test with Yates' continuity correction was used. To analyze VF scores, the Mann-Whitney test was used. For both tests, P values of ≤0.05 were considered statistically significant. All tests were performed with InStat (version 3) software.
RESULTS
Clinical information.
The average age of the patients was 67 years: 2% were 19 to 24 years old, 37% were 25 to 64 years old, and 61% were 65 to 98 years old; 53% of the patients were female. The frequencies of CA, HCA, and NA bacteremias were 42%, 18%, and 40%, respectively, and the principal primary sources were the urinary tract (61%, 52%, and 36%, respectively), followed by the gastrointestinal tract (21%, 17%, and 21%, respectively). Other identified sources included central vascular catheters (7%), the respiratory tract (7%), surgical wounds (1%), and skin (1%).
Virulence genotypes.
Fifteen of the 16 VFs sought were detected, with the overall prevalences ranging from 4% (afa/draBC) to 98% (fimH) (Table 1). The exception was the papG allele I adhesin, which was not detected in any isolate, although its presence was confirmed in its respective control. In total, 63% of the isolates encoded two or more of the seven adhesins for which they were screened. Other key VFs detected were iron acquisition systems (fyuA and/or iutA siderophores), present in 93% of isolates, and protectins (kpsMT II and/or traT), present in 89%. The hlyA and/or cnf1 toxin was detected in a minority of isolates (33%), as was the ibeA and/or hek invasin, detected in 38% of the isolates. malX, a generic marker for uropathogenic PAIs, was detected in 58% of the isolates. E. coli isolates possessed an average of 6.7 VFs (range, 0 to 12 VFs).
TABLE 1.
Distribution of virulence factors and phylogenetic groups among E. coli bloodstream isolates according to antibiotic susceptibility
| Characteristic | No. (%) of isolatesa |
Pb for comparison of: |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total (n = 303) | Susceptible (n = 52) | Resistant (n = 141) | MDR (n = 110) | FQs (n = 192) | FQr (n = 84) | Susceptible vs resistant | Susceptible vs MDR | Resistant vs MDR | FQs vs FQr | |
| Adhesins | ||||||||||
| papA | 161 (53) | 25 (48) | 89 (63) | 47 (43) | 107 (56) | 37 (44) | 0.0014 | |||
| papG | ||||||||||
| Allele I | 0 | 0 | 0 | 0 | 0 | 0 | ||||
| Allele II | 125 (41) | 20 (39) | 70 (50) | 35 (32) | 86 (45) | 27 (32) | 0.0047 | |||
| Allele III | 17 (6) | 2 (4) | 11 (9) | 4 (4) | 15 (8) | 0 | ||||
| sfa/focDE | 68 (22) | 11 (21) | 39 (28) | 18 (16) | 52 (27) | 5 (6) | 0.0478 | <0.0001 | ||
| afa/draBC | 11 (4) | 2 (4) | 7 (5) | 2 (2) | 11 (6) | 0 | ||||
| fimH | 297 (98) | 51 (98) | 139 (99) | 107 (97) | 189 (98) | 81 (96) | ||||
| Toxins | ||||||||||
| hlyA | 88 (29) | 16 (31) | 51 (36) | 21 (19) | 73 (38) | 7 (8) | 0.0032 | <0.0001 | ||
| cnf1 | 83 (27) | 12 (23) | 48 (34) | 23 (21) | 62 (32) | 10 (12) | 0.0242 | 0.0003 | ||
| Siderophores | ||||||||||
| fyuA | 264 (87) | 42 (81) | 131 (93) | 91 (83) | 172 (90) | 70 (83) | 0.0296 | 0.0163 | ||
| iutA | 208 (67) | 25 (48) | 98 (70) | 85 (77) | 123 (64) | 68 (81) | 0.0072 | 0.0003 | 0.0068 | |
| Protectins | ||||||||||
| kpsMT II | 188 (62) | 34 (65) | 91 (65) | 63 (57) | 119 (62) | 51 (61) | ||||
| traT | 224 (74) | 29 (56) | 108 (77) | 87 (79) | 141 (73) | 64 (76) | 0.007 | 0.0029 | ||
| Invasins | ||||||||||
| ibeA | 24 (8) | 10 (19) | 8 (6) | 6 (6) | 21 (11) | 1 (1) | 0.0093 | 0.0099 | 0.0035 | |
| hek | 103 (34) | 15 (29) | 55 (39) | 33 (30) | 71 (37) | 17 (20) | 0.0074 | |||
| PAI | ||||||||||
| malX | 177 (58) | 30 (58) | 93 (66) | 54 (49) | 127 (66) | 37 (44) | 0.0097 | 0.0008 | ||
| Avg VF score | 6.7 | 6.2 | 7.4 | 6.2 | 7.1 | 5.7 | 0.0336 | 0.0001 | <0.0001 | |
| Phylogenetic group | ||||||||||
| A | 33 (11) | 1 (2) | 17 (12) | 15 (14) | 16 (8) | 14 (17) | 0.0467 | 0.0221 | ||
| B1 | 1 (0.3) | 0 | 0 | 1 (1) | 0 | 1 (1) | ||||
| B2 | 182 (60) | 36 (69) | 90 (64) | 56 (51) | 132 (69) | 35 (42) | 0.0410 | <0.0001 | ||
| D | 65 (22) | 7 (14) | 26 (18) | 32 (29) | 30 (16) | 30 (36) | 0.0317 | 0.0004 | ||
| Nontypeable | 22 (7) | 8 (15) | 8 (7) | 6 (6) | 14 (7) | 4 (5) | 0.04 | <0.0001 | ||
The only isolate to be classified as intermediately resistant to ciprofloxacin was included in the FQr category. Data for isolates resistant to nalidixic acid only (n = 27) were excluded from this analysis.
Except for the VF scores, all P values were all analyzed by Fisher's exact test. VF scores were analyzed by the Mann-Whitney test. P values of ≤0.05 were considered significant.
Phylogenetic background.
Two hundred eighty-one isolates (93%) were successfully categorized (Table 1). The numbers of isolates belonging to phylogenetic groups A, B1, B2, and D were 33 (11%), 1 (0.3%), 182 (60%), and 65 (22%), respectively, while 22 (7%) isolates were nontypeable. The average VF scores were higher among the isolates in phylogenetic groups B2 and D (7.8 and 6.1, respectively) than among those in groups A and B1 and nontypeable isolates (4.9, 5.0, and 3.1, respectively).
Antibiotic resistance.
Fifty-two isolates (17%) were fully antibiotic susceptible, 141 (47%) were in the resistant category, and the remaining 110 isolates (36%) were MDR. Resistance to meropenem or amikacin was not detected. The highest resistance rates were detected in NA isolates, while the CA isolates were the most susceptible (Fig. 1).
FIG. 1.
Resistance rates for 303 E. coli isolates from CA, HCA, and NA bloodstream infections. AMP, ampicillin; PIP, piperacillin; AUG, amoxicillin-clavulanic acid; TAZ, piperacillin-tazobactam; FOX, cefoxitin; CAZ, ceftazidime; CTX, cefotaxime; CPD, cefpodoxime; GEN, gentamicin; NAL, nalidixic acid; CIP, ciprofloxacin; SXT, trimethoprim-sulfamethoxazole.
VFs are equally distributed between susceptible and MDR isolates and are concentrated in phylogenetic group B2.
VFs were detected in susceptible and MDR isolates at comparable frequencies, each having an average VF score of 6.2 (Table 1). The exceptions were ibeA, which was significantly more prevalent among susceptible isolates, and iutA and traT, which were significantly more prevalent among MDR isolates. The remaining 141 isolates, which belonged to the resistant category, had a significantly higher average VF score of 7.4 compared to the scores for the susceptible (P = 0.0336) and MDR (P = 0.0001) isolates. Irrespective of the deduced antibiotic susceptibility profile, the most prevalent phylogenetic group was B2, followed by group D. Together, these two phylogenetic groups accounted for 83%, 82%, and 80% of the susceptible, resistant, and MDR isolates, respectively.
Dominance of FQr isolates in phylogenetic group B2 with low levels of virulence.
The strong associations observed between MDR status, virulence genotypes, and phylogenetic group B2 prompted further analysis of fluoroquinolone (ciprofloxacin)-resistant (FQr) isolates, since high FQr rates (28%) were detected. Fluoroquinolone-susceptible (FQs) isolates had a significantly higher prevalence of VFs than their resistant counterparts, with average VF scores of 7.1 and 5.7, respectively (Table 1). Positive associations were observed between FQs isolates and adhesins, toxins, invasins, and malX. The main exception to this was the iutA siderophore, which was significantly more prevalent among FQr isolates. The majority of FQs and FQr isolates belonged to phylogenetic group B2, followed by group D. Together these groups accounted for 85% and 78% of the FQs and FQr isolates, respectively. Positive associations were observed between FQs isolates and phylogenetic group B2 and between FQr isolates and phylogenetic group D.
CA isolates have higher VF scores and lower MDR frequencies than NA isolates.
The overall VF prevalence was significantly higher among both CA and HCA isolates than among NA isolates, with the average VF scores being 7.3, 7.1, and 6.0, respectively (Table 2). This was particularly evident for papA, papG allele II, hlyA, and hek. Phylogenetic group B2 predominated in the CA, HCA, and NA isolates, followed by group D. Antibiotic-resistant isolates were more frequently represented in CA and HCA bacteremias than in NA bacteremias (55% and 52% versus 35%, respectively), while MDR isolates were more frequent in NA and HCA bacteremias than in CA bacteremias (49% and 41% versus 22%, respectively), and FQr isolates prevailed in NA bacteremias compared to their prevalence in CA and HCA bacteremias (42% versus 15% and 26%, respectively).
TABLE 2.
Distribution of characteristics among E. coli bloodstream isolates according to community, health care-associated, or nosocomial origin
| Characteristic | No. (%) of isolates |
Pb for comparison of: |
||||
|---|---|---|---|---|---|---|
| CA (n = 127) | HCA (n = 54) | NA (n = 122) | ||||
| CA vs HCA | CA vs NA | HCA vs NA | ||||
| Virulence factora | ||||||
| papA | 84 (66) | 30 (56) | 47 (39) | <0.0001 | 0.0477 | |
| papG allele II | 70 (55) | 23 (43) | 32 (26) | <0.0001 | 0.0355 | |
| hlyA | 45 (35) | 19 (35) | 24 (20) | 0.0070 | 0.0361 | |
| hek | 51 (40) | 22 (41) | 30 (25) | 0.0102 | 0.0335 | |
| Avg VF score | 7.3 | 7.1 | 6.0 | 0.0002 | 0.0172 | |
| Phylogenetic group | ||||||
| A | 9 (7) | 4 (7) | 20 (16) | 0.0289 | ||
| B1 | 1 (1) | 0 | 0 | |||
| B2 | 82 (65) | 35 (65) | 65 (53) | |||
| D | 29 (23) | 9 (17) | 27 (22) | |||
| Nontypeable | 6 (5) | 6 (11) | 10 (8) | |||
| Antibiotic susceptibility category | ||||||
| Susceptible | 29 (23) | 4 (7) | 19 (16) | 0.0193 | ||
| Resistant | 70 (55) | 28 (52) | 43 (35) | 0.0022 | 0.0461 | |
| Multidrug resistant | 28 (22) | 22 (41) | 60 (49) | 0.0174 | <0.0001 | |
| Ciprofloxacin resistant | 19 (15) | 14 (26) | 51 (42) | <0.0001 | ||
Only data for those VFs that yielded P values of ≤0.05 are shown.
Except for the VF scores, all P values were all analyzed by Fisher's exact test. VF scores were analyzed by the Mann-Whitney test. P values of ≤0.05 were considered significant.
AmpC attenuator mutations predominate in phylogenetic group D isolates of community origin.
Twenty-four isolates (8%) were resistant to one or more cephalosporins; only two were phenotypically confirmed to be ESBL producers. Of the remainder, 21 overexpressed their AmpC β-lactamase (Table 3). Plasmidic AmpC β-lactamases were detected in four isolates: ACC-1 (n = 2) and CMY-2 (n = 2). These β-lactamases confer resistance to expanded-spectrum cephalosporins, while isolates harboring CMY-2 further conferred resistance to cefoxitin. These isolates belonged to phylogenetic groups D (n = 3) and B2 (n = 1). Sequence analysis of the chromosomal ampC gene revealed mutations in 18 isolates: promoter mutations at positions −32 (n = 4) and −42 (n = 1) and attenuator mutations (n = 13). Isolates encoding −42/−32 promoter mutations conferred resistance to cefoxitin and expanded-spectrum cephalosporins, while those with attenuator mutations conferred resistance to cefoxitin but remained susceptible to expanded-spectrum cephalosporins. Promoter mutations were more often found in NA isolates from phylogenetic group B2, whereas attenuator mutations were associated with CA isolates from group D.
TABLE 3.
Analysis of ampC-mediated resistance in cephalosporin-resistant E. coli bloodstream isolates
| Isolate no. | Phylogenetic group | Genotype |
Origin | |
|---|---|---|---|---|
| Plasmidic gene | AmpC mutation(s)a | |||
| 1 | A | −42 −18 −1 +21 +58 | NA | |
| 2 | B2 | −32 +58 + 63 | HCA | |
| 3 | B2 | CMY-2 | −32 +58 + 63 | HCA |
| 4 | B2 | −32 +58 + 63 | NA | |
| 5 | B2 | −13/−12* +6 | NA | |
| 6 | B2 | −32 +58 + 63 | CA | |
| 7 | B2 | +21 +58 + 63 | NA | |
| 8 | B2 | −28 +17 | NA | |
| 9 | B2 | −28 +17 | NA | |
| 10 | B2 | −28 | NA | |
| 11 | B2 | ESBL | −28 | CA |
| 12 | D | +22 + 26 + 27 + 32 | CA | |
| 13 | D | +22 + 26 + 27 + 32 | CA | |
| 14 | D | +22 + 26 + 27 + 32 | CA | |
| 15 | D | +22 + 26 + 27 + 32 | CA | |
| 16 | D | +22 + 26 + 27 + 32 | NA | |
| 17 | D | +22 + 26 + 27 + 32 | NA | |
| 18 | D | −22/−21* +22 + 26 + 27 + 32 | CA | |
| 19 | D | −28 +33 +58 | HCA | |
| 20 | D | +32 | CA | |
| 21 | D | ACC-1 | Wild-type AmpC | CA |
| 22 | D | ESBL | +72 | CA |
| 23 | D | CMY-2 | +65 | CA |
| 24 | D | ACC-1 | Wild-type AmpC | CA |
Compared with ampC promoter of E. coli K-12. The boldface font indicates promoter mutations causing AmpC overproduction; underlining indicates mutations in the attenuator causing AmpC overproduction. *, a 1-bp insertion in the ampC promoter.
DISCUSSION
We have investigated the relationships between virulence genotypes, phylogenetic background, and antibiotic resistance in 303 E. coli bloodstream isolates. Key among the VFs detected were fimH, fyuA, iutA, kpsMT allele II, and traT. These results confirm the roles of adhesins, iron acquisition systems, and protectins as the minimum prerequisites for bloodstream invasion by E. coli (3, 18, 28, 29, 42).
The majority of isolates belonged to phylogenetic groups B2 and D. While similar observations have been made with ExPEC strains originating from patients with bacteremia (26, 36, 42, 46), the urinary tract (24, 25, 35, 41, 43), both of these (4, 22, 37), or different sources (2, 20, 27), others have reported that ExPEC strains from the urinary tract and patients with bacteremia fall predominantly into phylogenetic groups A (15) and D (34), respectively. The study design and the number and source of the isolates, along with demographic and epidemiologic factors relating to the patient population, probably account for these discordant results. Although the findings of our study are representative for only one center, we believe that the strength of our study lies in the large number of isolates analyzed, the fact that the isolates were from consecutive episodes of bacteremia over a defined time period, and there was a clearer separation of the collection into three antibiotic susceptibility categories. While the majority of isolates from each of these categories belonged to phylogenetic group B2, significantly greater percentages of susceptible than MDR isolates were group B2 or nontypeable, while a relatively higher proportion of MDR isolates were in groups A and D.
The majority of VFs were equally distributed between susceptible and MDR isolates, although significant relationships between susceptible isolates and ibeA and between MDR isolates and iutA and traT were identified. Previous studies pointed to an association between MDR ExPEC and phylogenetic group A and/or D (5, 17, 23, 27, 38). Indeed, Jaureguy et al. (21) found non-B2 E. coli bacteremia isolates to be MDR significantly more often (P < 0.001). A dominance of MDR isolates in both phylogenetic groups A and B2 has been reported only once previously (43). Several studies also found that MDR isolates exhibit significantly reduced VF prevalences compared with those for non-MDR isolates (23, 38, 42).
As our MDR isolates also had a high frequency of ciprofloxacin resistance, which was predominantly due to chromosomally mediated mutations in gyrA and/or parC (data not shown), these isolates were investigated further to determine their role in the pathogenesis of ExPEC. Previous investigations showed that FQr isolates are associated with lower VF prevalences than FQs isolates. An exception was for iutA, which was universally detected at a higher frequency in FQr isolates (2, 27, 38, 41, 42, 47). Although we found that more FQr isolates belonged to phylogenetic group B2, they were proportionally less prevalent in this group than in group D. Previous studies of these particular relationships have produced conflicting results (2, 27, 38). The concentrated prescription of ciprofloxacin in our hospital may have progressively selected for ciprofloxacin-resistant E. coli mutants, which would explain the greater proportion of ciprofloxacin-resistant isolates detected in NA bacteremias than in those from the community.
The predominant phylogenetic group from CA, NA, and HCA isolates was group B2. The overall VF scores were significantly higher among both CA and HCA isolates, with an emphasis on papA, papG allele II, hlyA, and hek, all of which are PAI allele II-linked genes (16, 19). CA and HCA bacteremias also harbored more isolates in the resistant categories than NA bacteremias. In contrast, MDR isolates were more significantly associated with NA and HCA bacteremias than with CA bacteremias. This suggests that a population of MDR E. coli isolates with a lower virulence potential causes NA bacteremias in our institution, while CA infections appear to be due to highly virulent strains with reduced levels of antibiotic resistance, and those causing HCA infections to be both highly virulent and antibiotic resistant.
In comparison to reports from North American and European centers, our collection of isolates exhibited higher rates of resistance to the majority of antibiotics tested, with the exception of expanded-spectrum cephalosporins (1, 13, 14, 30, 31, 44, 49). We detected fewer ESBL producers (13, 14) but a higher proportion of MDR isolates (23, 39, 45). Not surprisingly, the resistance rates were the highest in NA isolates. Our finding that isolates overexpressing AmpC β-lactamases were mainly associated with virulent phylogenetic groups B2 and D is in contrast to the findings of previous studies, which associated AmpC overproducers with commensal phylogenetic groups A and B1 (12, 27, 33). With the emergence of this resistance mechanism and the concentration of AmpC overproducers predominantly being within virulent phylogenetic groups, antibiotic treatment will become increasingly challenging.
By analyzing our data on the basis of the community, health care-associated, or nosocomial acquisition of bacteremia, we have been able to show that NA E. coli isolates have a reduced VF content and a higher frequency of MDR. VFs have been suggested to be attractive targets for anti-infective therapies and vaccine development (6); however, the reduced prevalence of particular adhesins, toxins, and invasins in hospital isolates, which in our study were almost as common as CA isolates, could make this a more challenging component of future preventive strategies.
Acknowledgments
This study, including Niamh M. Cooke's Ph.D. studentship, was funded exclusively by the Department of Clinical Microbiology of Trinity College Dublin.
We thank the staff of the blood culture section, St. James's Hospital Department of Microbiology, for providing the isolates used in the study.
Footnotes
Published ahead of print on 27 January 2010.
REFERENCES
- 1.Al-Hasan, M. N., B. D. Lahr, J. E. Eckel-Passow, and L. M. Baddour. 2009. Antimicrobial resistance trends of Escherichia coli bloodstream isolates: a population-based study, 1998-2007. J. Antimicrob. Chemother. 64:169-174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bert, F., X. Panhard, J. Johnson, H. Lecuyer, R. Moreau, J. Le Grand, B. Johnston, M. Sinegre, D. Valla, and M. H. Nicolas-Chanoine. 2008. Genetic background of Escherichia coli isolates from patients with spontaneous bacterial peritonitis: relationship with host factors and prognosis. Clin. Microbiol. Infect. 14:1034-1040. [DOI] [PubMed] [Google Scholar]
- 3.Bingen-Bidois, M., O. Clermont, S. Bonacorsi, M. Terki, N. Brahimi, C. Loukil, D. Barraud, and E. Bingen. 2002. Phylogenetic analysis and prevalence of urosepsis strains of Escherichia coli bearing pathogenicity island-like domains. Infect. Immun. 70:3216-3226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bonacorsi, S., V. Houdouin, P. Mariani-Kurkdjian, F. Mahjoub-Messai, and E. Bingen. 2006. Comparative prevalence of virulence factors in Escherichia coli causing urinary tract infection in male infants with and without bacteremia. J. Clin. Microbiol. 44:1156-1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bukh, A. S., H. C. Schonheyder, J. M. Emmersen, M. Sogaard, S. Bastholm, and P. Roslev. 2009. Escherichia coli phylogenetic groups are associated with site of infection and level of antibiotic resistance in community-acquired bacteraemia: a 10 year population-based study in Denmark. J. Antimicrob. Chemother. 64:163-168. [DOI] [PubMed] [Google Scholar]
- 6.Caroff, N., E. Espaze, D. Gautreau, H. Richet, and A. Reynaud. 2000. Analysis of the effects of −42 and −32 ampC promoter mutations in clinical isolates of Escherichia coli hyperproducing AmpC. J. Antimicrob. Chemother. 45:783-788. [DOI] [PubMed] [Google Scholar]
- 7.Clermont, O., S. Bonacorsi, and E. Bingen. 2000. Rapid and simple determination of the Escherichia coli phylogenetic group. Appl. Environ. Microbiol. 66:4555-4558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Clermont, O., M. Lavollay, S. Vimont, C. Deschamps, C. Forestier, C. Branger, E. Denamur, and G. Arlet. 2008. The CTX-M-15-producing Escherichia coli diffusing clone belongs to a highly virulent B2 phylogenetic subgroup. J. Antimicrob. Chemother. 61:1024-1028. [DOI] [PubMed] [Google Scholar]
- 9.Clinical and Laboratory Standards Institute. 2009. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard, 8th ed. Document M07-A8. Clinical and Laboratory Standards Institute, Wayne, PA.
- 10.Clinical and Laboratory Standards Institute. 2009. Performance standards for antimicrobial disk susceptibility tests; approved standard, 10th ed. CLSI document M02-A10. Clinical and Laboratory Standards Institute, Wayne, PA.
- 11.Clinical and Laboratory Standards Institute. 2009. Performance standards for antimicrobial susceptibility testing; 19th informational supplement. CLSI document M100-S19. Clinical and Laboratory Standards Institute, Wayne, PA.
- 12.Corvec, S., A. Prodhomme, C. Giraudeau, S. Dauvergne, A. Reynaud, and N. Caroff. 2007. Most Escherichia coli strains overproducing chromosomal AmpC beta-lactamase belong to phylogenetic group A. J. Antimicrob. Chemother. 60:872-876. [DOI] [PubMed] [Google Scholar]
- 13.Deshpande, L. M., R. N. Jones, T. R. Fritsche, and H. S. Sader. 2006. Occurrence of plasmidic AmpC type beta-lactamase-mediated resistance in Escherichia coli: report from the SENTRY Antimicrobial Surveillance Program (North America, 2004). Int. J. Antimicrob. Agents 28:578-581. [DOI] [PubMed] [Google Scholar]
- 14.European Antimicrobial Resistance Surveillance System. 2009. EARSS interactive database access. European Antimicrobial Resistance Surveillance System, Bilthoven, Netherlands. http://www.rivm.nl/earss/database.
- 15.Grude, N., N. I. Potaturkina-Nesterova, A. Jenkins, L. Strand, F. L. Nowrouzian, J. Nyhus, and B. E. Kristiansen. 2007. A comparison of phylogenetic group, virulence factors and antibiotic resistance in Russian and Norwegian isolates of Escherichia coli from urinary tract infection. Clin. Microbiol. Infect. 13:208-211. [DOI] [PubMed] [Google Scholar]
- 16.Hejnova, J., U. Dobrindt, R. Nemcova, C. Rusniok, A. Bomba, L. Frangeul, J. Hacker, P. Glaser, P. Sebo, and C. Buchrieser. 2005. Characterization of the flexible genome complement of the commensal Escherichia coli strain A0 34/86 (O83:K24:H31). Microbiology 151:385-398. [DOI] [PubMed] [Google Scholar]
- 17.Ho, P. L., W. W. Poon, S. L. Loke, M. S. Leung, K. H. Chow, R. C. Wong, K. S. Yip, E. L. Lai, and K. W. Tsang. 2007. Community emergence of CTX-M type extended-spectrum beta-lactamases among urinary Escherichia coli from women. J. Antimicrob. Chemother. 60:140-144. [DOI] [PubMed] [Google Scholar]
- 18.Horcajada, J. P., S. Soto, A. Gajewski, A. Smithson, M. T. Jimenez de Anta, J. Mensa, J. Vila, and J. R. Johnson. 2005. Quinolone-resistant uropathogenic Escherichia coli strains from phylogenetic group B2 have fewer virulence factors than their susceptible counterparts. J. Clin. Microbiol. 43:2962-2964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Houdouin, V., S. Bonacorsi, P. Bidet, M. Bingen-Bidois, D. Barraud, and E. Bingen. 2006. Phylogenetic background and carriage of pathogenicity island-like domains in relation to antibiotic resistance profiles among Escherichia coli urosepsis isolates. J. Antimicrob. Chemother. 58:748-751. [DOI] [PubMed] [Google Scholar]
- 20.Houdouin, V., S. Bonacorsi, P. Bidet, and E. Bingen. 2007. Antimicrobial susceptibility of 136 Escherichia coli isolates from cases of neonatal meningitis and relationship with virulence. Clin. Microbiol. Infect. 13:1207-1210. [DOI] [PubMed] [Google Scholar]
- 21.Jaureguy, F., E. Carbonnelle, S. Bonacorsi, C. Clec'h, P. Casassus, E. Bingen, B. Picard, X. Nassif, and O. Lortholary. 2007. Host and bacterial determinants of initial severity and outcome of Escherichia coli sepsis. Clin. Microbiol. Infect. 13:854-862. [DOI] [PubMed] [Google Scholar]
- 22.Johnson, J. R., O. Clermont, M. Menard, M. A. Kuskowski, B. Picard, and E. Denamur. 2006. Experimental mouse lethality of Escherichia coli isolates, in relation to accessory traits, phylogenetic group, and ecological source. J. Infect. Dis. 194:1141-1150. [DOI] [PubMed] [Google Scholar]
- 23.Johnson, J. R., M. A. Kuskowski, A. Gajewski, D. F. Sahm, and J. A. Karlowsky. 2004. Virulence characteristics and phylogenetic background of multidrug-resistant and antimicrobial-susceptible clinical isolates of Escherichia coli from across the United States, 2000-2001. J. Infect. Dis. 190:1739-1744. [DOI] [PubMed] [Google Scholar]
- 24.Johnson, J. R., M. A. Kuskowski, A. Gajewski, S. Soto, J. P. Horcajada, M. T. Jimenez de Anta, and J. Vila. 2005. Extended virulence genotypes and phylogenetic background of Escherichia coli isolates from patients with cystitis, pyelonephritis, or prostatitis. J. Infect. Dis. 191:46-50. [DOI] [PubMed] [Google Scholar]
- 25.Johnson, J. R., M. A. Kuskowski, T. T. O'Bryan, R. Colodner, and R. Raz. 2005. Virulence genotype and phylogenetic origin in relation to antibiotic resistance profile among Escherichia coli urine sample isolates from Israeli women with acute uncomplicated cystitis. Antimicrob. Agents Chemother. 49:26-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Johnson, J. R., M. A. Kuskowski, T. T. O'Bryan, and J. N. Maslow. 2002. Epidemiological correlates of virulence genotype and phylogenetic background among Escherichia coli blood isolates from adults with diverse-source bacteremia. J. Infect. Dis. 185:1439-1447. [DOI] [PubMed] [Google Scholar]
- 27.Johnson, J. R., M. A. Kuskowski, K. Owens, A. Gajewski, and P. L. Winokur. 2003. Phylogenetic origin and virulence genotype in relation to resistance to fluoroquinolones and/or extended-spectrum cephalosporins and cephamycins among Escherichia coli isolates from animals and humans. J. Infect. Dis. 188:759-768. [DOI] [PubMed] [Google Scholar]
- 28.Johnson, J. R., S. L. Moseley, P. L. Roberts, and W. E. Stamm. 1988. Aerobactin and other virulence factor genes among strains of Escherichia coli causing urosepsis: association with patient characteristics. Infect. Immun. 56:405-412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Johnson, J. R., and A. L. Stell. 2000. Extended virulence genotypes of Escherichia coli strains from patients with urosepsis in relation to phylogeny and host compromise. J. Infect. Dis. 181:261-272. [DOI] [PubMed] [Google Scholar]
- 30.Jones, R. N. 2003. Global epidemiology of antimicrobial resistance among community-acquired and nosocomial pathogens: a five-year summary from the SENTRY Antimicrobial Surveillance Program (1997-2001). Semin. Respir. Crit. Care Med. 24:121-134. [DOI] [PubMed] [Google Scholar]
- 31.Karlowsky, J. A., L. J. Kelly, C. Thornsberry, M. E. Jones, and D. F. Sahm. 2002. Trends in antimicrobial resistance among urinary tract infection isolates of Escherichia coli from female outpatients in the United States. Antimicrob. Agents Chemother. 46:2540-2545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Laupland, K. B., D. B. Gregson, D. L. Church, T. Ross, and J. D. Pitout. 2008. Incidence, risk factors and outcomes of Escherichia coli bloodstream infections in a large Canadian region. Clin. Microbiol. Infect. 14:1041-1047. [DOI] [PubMed] [Google Scholar]
- 33.Mammeri, H., F. Eb, A. Berkani, and P. Nordmann. 2008. Molecular characterization of AmpC-producing Escherichia coli clinical isolates recovered in a French hospital. J. Antimicrob. Chemother. 61:498-503. [DOI] [PubMed] [Google Scholar]
- 34.Martinez, J. A., S. Soto, A. Fabrega, M. Almela, J. Mensa, A. Soriano, F. Marco, M. T. Jimenez de Anta, and J. Vila. 2006. Relationship of phylogenetic background, biofilm production, and time to detection of growth in blood culture vials with clinical variables and prognosis associated with Escherichia coli bacteremia. J. Clin. Microbiol. 44:1468-1474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Moreno, E., A. Andreu, C. Pigrau, M. A. Kuskowski, J. R. Johnson, and G. Prats. 2008. Relationship between Escherichia coli strains causing acute cystitis in women and the fecal E. coli population of the host. J. Clin. Microbiol. 46:2529-2534. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Moreno, E., I. Planells, G. Prats, A. M. Planes, G. Moreno, and A. Andreu. 2005. Comparative study of Escherichia coli virulence determinants in strains causing urinary tract bacteremia versus strains causing pyelonephritis and other sources of bacteremia. Diagn. Microbiol. Infect. Dis. 53:93-99. [DOI] [PubMed] [Google Scholar]
- 37.Moreno, E., G. Prats, I. Planells, A. M. Planes, T. Perez, and A. Andreu. 2006. Characterization of Escherichia coli isolates derived from phylogenetic groups A and B1 causing extraintestinal infection. Enferm. Infecc. Microbiol. Clin. 24:483-489. [DOI] [PubMed] [Google Scholar]
- 38.Moreno, E., G. Prats, M. Sabate, T. Perez, J. R. Johnson, and A. Andreu. 2006. Quinolone, fluoroquinolone and trimethoprim/sulfamethoxazole resistance in relation to virulence determinants and phylogenetic background among uropathogenic Escherichia coli. J. Antimicrob. Chemother. 57:204-211. [DOI] [PubMed] [Google Scholar]
- 39.Peralta, G., M. B. Sanchez, J. C. Garrido, I. De Benito, M. E. Cano, L. Martinez-Martinez, and M. P. Roiz. 2007. Impact of antibiotic resistance and of adequate empirical antibiotic treatment in the prognosis of patients with Escherichia coli bacteraemia. J. Antimicrob. Chemother. 60:855-863. [DOI] [PubMed] [Google Scholar]
- 40.Perez-Perez, F. J., and N. D. Hanson. 2002. Detection of plasmid-mediated AmpC beta-lactamase genes in clinical isolates by using multiplex PCR. J. Clin. Microbiol. 40:2153-2162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Piatti, G., A. Mannini, M. Balistreri, and A. M. Schito. 2008. Virulence factors in urinary Escherichia coli strains: phylogenetic background and quinolone and fluoroquinolone resistance. J. Clin. Microbiol. 46:480-487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Rijavec, M., M. Muller-Premru, B. Zakotnik, and D. Zgur-Bertok. 2008. Virulence factors and biofilm production among Escherichia coli strains causing bacteraemia of urinary tract origin. J. Med. Microbiol. 57:1329-1334. [DOI] [PubMed] [Google Scholar]
- 43.Rijavec, M., M. Starcic Erjavec, J. Ambrozic Avgustin, R. Reissbrodt, A. Fruth, V. Krizan-Hergouth, and D. Zgur-Bertok. 2006. High prevalence of multidrug resistance and random distribution of mobile genetic elements among uropathogenic Escherichia coli (UPEC) of the four major phylogenetic groups. Curr. Microbiol. 53:158-162. [DOI] [PubMed] [Google Scholar]
- 44.Sahm, D. F., N. P. Brown, Y. C. Yee, and A. T. Evangelista. 2008. Stratified analysis of multidrug-resistant Escherichia coli in US health care institutions. Postgrad. Med. 120:53-59. [DOI] [PubMed] [Google Scholar]
- 45.Sahm, D. F., C. Thornsberry, D. C. Mayfield, M. E. Jones, and J. A. Karlowsky. 2001. Multidrug-resistant urinary tract isolates of Escherichia coli: prevalence and patient demographics in the United States in 2000. Antimicrob. Agents Chemother. 45:1402-1406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Sannes, M. R., M. A. Kuskowski, K. Owens, A. Gajewski, and J. R. Johnson. 2004. Virulence factor profiles and phylogenetic background of Escherichia coli isolates from veterans with bacteremia and uninfected control subjects. J. Infect. Dis. 190:2121-2128. [DOI] [PubMed] [Google Scholar]
- 47.Takahashi, A., T. Muratani, M. Yasuda, S. Takahashi, K. Monden, K. Ishikawa, H. Kiyota, S. Arakawa, T. Matsumoto, H. Shima, H. Kurazono, and S. Yamamoto. 2009. Genetic profiles of fluoroquinolone-resistant Escherichia coli isolates obtained from patients with cystitis: phylogeny, virulence factors, PAIusp subtypes, and mutation patterns. J. Clin. Microbiol. 47:791-795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Wiles, T. J., R. R. Kulesus, and M. A. Mulvey. 2008. Origins and virulence mechanisms of uropathogenic Escherichia coli. Exp. Mol. Pathol. 85:11-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Zhanel, G. G., M. DeCorby, K. A. Nichol, A. Wierzbowski, P. J. Baudry, J. A. Karlowsky, P. Lagace-Wiens, A. Walkty, M. R. Mulvey, and D. J. Hoban. 2008. Antimicrobial susceptibility of 3931 organisms isolated from intensive care units in Canada: Canadian National Intensive Care Unit Study, 2005/2006. Diagn. Microbiol. Infect. Dis. 62:67-80. [DOI] [PubMed] [Google Scholar]
- 50.Zhao, L., X. Chen, X. Zhu, W. Yang, L. Dong, X. Xu, S. Gao, and X. Liu. 2009. Prevalence of virulence factors and antimicrobial resistance of uropathogenic Escherichia coli in Jiangsu province (China). Urology 74:702-707. [DOI] [PubMed] [Google Scholar]