Allocation of Klebsiella pneumoniae Bloodstream Isolates into Four Distinct Groups by ompK36 Typing in a Taiwanese University Hospital
ABSTRACT
The OmpK36 porin plays a role in carbapenem resistance and may contribute to bacterial virulence in Klebsiella pneumoniae. This study aimed to investigate the characteristics of different groups of K. pneumoniae separated by ompK36 typing. Among 226 nonduplicate K. pneumoniae bloodstream isolates collected at a Taiwanese hospital in 2011, four ompK36 types, designated types A, B, C, and D, were identified by PCR in 61, 28, 100, and 36 isolates, respectively; 1 isolate was untypeable. Statistical analysis showed significantly higher rates of antimicrobial resistance (all tested antibiotics except meropenem), extended-spectrum β-lactamases or DHA-1 (47.5% together), Qnr-type quinolone resistance determinants (50.8%), and IncFIIA-type plasmids (49.2%) in group A than in others. Seventeen isolates were identified as belonging to 3 international high-risk clones (4 sequence type 11 [ST11], 10 ST15, and 3 ST147 isolates); all isolates but 1 ST15 isolate were classified in group A. The significant characteristics of group C were hypermucoviscosity (62.0%) and a higher virulence gene content. This group included all serotype K1 (n = 30), K2 (n = 25), and K5 (n = 3) isolates, 6 of 7 K57 isolates, all isolates of major clones associated with pyogenic liver abscesses (29 ST23, 11 ST65, 5 ST86, 7 ST373, and 1 ST375 isolates), and 16 (94.1%) of 17 isolates causing bacteremic liver abscesses. Twelve (42.9%) of the group B isolates were responsible for bacteremic biliary tract infections. Group D was predominant (83.3%) among 12 K20 isolates. This study suggests that most clinical K. pneumoniae isolates can be allocated into four groups with distinct characteristics based on ompK36 types.
INTRODUCTION
Klebsiella pneumoniae is an important human opportunistic pathogen that causes a wide variety of community-acquired and nosocomial infections (1–4). The organism is still one of the leading causes of community-acquired pneumonia (3) and has been described as the major cause of pyogenic liver abscesses (PLA) in some Asia-Pacific countries (2, 4). Serotypes K1 and K2 in K. pneumoniae have been linked to invasive infections (2–9). Using multilocus sequence typing (MLST), Brisse et al. (10) identified two virulent K1 clones of K. pneumoniae, clonal complex (CC) 23 and CC82, which were strongly associated with PLA and respiratory infection, respectively.
K. pneumoniae is also an important reservoir of antibiotic resistance determinants (1). Several clones of K. pneumoniae with a high prevalence of carbapenem resistance or extended-spectrum β-lactamases (ESBLs) have been distributed worldwide (11–16). The epidemic drug-resistant clones, called “high-risk” clones (11), may play an important role in the dissemination of antimicrobial resistance. The major international high-risk clones identified by MLST belong to sequence type 11 (ST11), ST15, ST147, and ST258 (11–16). ST11 and ST258 were included in clonal group (CG) 258 (11, 14), while ST15 and ST147 were grouped in CG15 and CG147 (11, 13, 14), respectively.
High-level carbapenem resistance in K. pneumoniae may develop via the loss of the OmpK36 porin coupled with the expression of various β-lactamases (17–19). The OmpK36 porin in K. pneumoniae may also play a role in bacterial virulence (18). The correlation of OmpK36 porin variants with specific STs of K. pneumoniae was first described by Papagiannitsis et al. (20). We then found that K. pneumoniae isolates can be separated into four groups (designated groups A to D) by ompK36 genotyping (21). Interestingly, three major high-risk clones (ST11, ST15, and ST147) belonged to group A, while 6 PLA-associated STs (ST23, ST65, ST86, ST163, ST373, and ST375) were classified in group C (21). The present study aimed to investigate the characteristics of the four groups of K. pneumoniae isolates based on ompK36 genotyping.
MATERIALS AND METHODS
Bacterial isolates and data collection.
A total of 226 K. pneumoniae bloodstream isolates (1 isolate per patient) collected from the National Cheng Kung University Hospital in 2011 were analyzed. The hospital is a 1,300-bed tertiary teaching hospital in southern Taiwan and provides health care services, including all medical specialties, to 1.8 million people residing in Tainan City. Pertinent information regarding the infection foci of bacteremias caused by the study isolates was collected by chart review. A bloodstream infection was classified as community onset if the positive blood sample was collected within 48 h after admission to the hospital (22), while nosocomial bacteremia was defined as bacteremia occurring >48 h after admission. Health care-associated bloodstream infections were defined as described by Friedman et al. (23), and community-onset infections were classified as either health care associated or community acquired accordingly. The outcome measured was mortality within 30 days of the first blood culture positive for a study isolate. This retrospective study received institutional review board approval from the National Cheng Kung University Hospital (no. B-ER-103-024).
Microbiological tests.
Antimicrobial susceptibility testing was performed using the standard Kirby-Bauer disk diffusion method, and the results were interpreted based on the Clinical and Laboratory Standards Institute (CLSI) criteria (24). An isolate assigned to the intermediately susceptible, susceptible dose dependent, or resistant category was defined as nonsusceptible to the drug tested in this study. The production of ESBLs was detected by the CLSI-recommended disk diffusion method (24), and isolates negative in the ESBL confirmatory test were subjected to the double-disk synergy method to detect ESBLs coexisting with other classes of β-lactamases (25). Isolates were screened for AmpC and metallo-β-lactamases using the boronic acid disk method and the 2-mercaptopropionic acid double-disk method, respectively (25). The modified Hodge test was used for the detection of carbapenemase production (24). Isolates were classified phenotypically as mucoid or nonmucoid as described previously (6).
Detection of resistance genes.
A multiplex PCR technique was used to detect plasmid-mediated ampC genes, including blaCMY, blaMOX, blaDHA, blaACC, blaACT, blaFOX, and blaCFE (26, 27). The PCR-NheI method was used to discriminate between blaSHV-ESBL and blaSHV-non-ESBL genes (28), and nucleotide sequencing of blaSHV was performed for isolates with inconsistent results of genotyping and phenotyping. The detection of blaCTX-M families was performed by multiplex PCR using primer pairs published previously (29), and blaCTX-M subtypes were determined by nucleotide sequencing (25). PCR and sequencing were used to identity blaTEM variants. The carbapenemase-encoding genes were detected by PCR, with searching for blaIMP, blaVIM, blaKPC, blaNDM, and blaOXA-48 (16, 25). PCR detection of plasmid-mediated quinolone resistance genes, including qnrA, qnrB, qnrC, qnrD, and aac (6′)-Ib-cr, was performed as described previously (25, 30).
Detection of virulence genes and capsular genotyping.
Ten virulence genes of K. pneumoniae, including wzy_K1 (previously designated magA) (4), allS, rmpA, mrkD, kfuBC, fimH, uge, wabG, and ureA, were amplified by PCRs using the primers described previously (10). The virulence score was calculated as the total number of positive virulence genes, ranging from 0 to 10 (31). The wzy_K1 gene was also used as the marker of the K1 serotype (4), while capsular serotypes K2, K20, K54, and K57 were determined by PCRs using primers described previously (5).
Bacterial typing.
Bacterial isolates were divided into groups A, B, C, and D by the PCR-based ompK36 typing method and subjected to nucleotide sequencing of ompK36 as described previously (21). MLST was performed as described by Diancourt et al. (32). Allelic profiles and STs were determined by querying the K. pneumoniae MLST database maintained by the Pasteur Institute (http://bigsdb.web.pasteur.fr/klebsiella/klebsiella.html). Novel alleles and STs identified in this study were submitted to the database for coding. CGs and CCs were determined by using the eBURST program (http://eburst.mlst.net) against the MLST database with 1,948 STs (as of 3 July 2015) as described previously (9, 13, 14). Pulsed-field gel electrophoresis (PFGE) of XbaI-digested genomic DNA was performed on a CHEF-DR III apparatus (Bio-Rad Laboratories, Hercules, CA) according to the instruction manual. Genetic relatedness based on PFGE typing was calculated based on the Dice coefficient.
Plasmid typing.
The PCR-based plasmid typing method based on replicons of the major plasmid incompatibility groups, including FIA, FIB, FIC, HI1, HI2, I1-Iγ, L/M, N, P, W, T, A/C, K, B/O, X, Y, F, and FIIA, was performed as described by Carattoli et al. (33).
Statistical analysis.
Statistical analyses were performed using the Statistical Package for Social Sciences for Windows, version 17.0 (SPSS, Chicago, IL). Categorical variables, expressed as numbers and percentages, were compared using the chi-square test. Continuous variables are expressed as means ± standard deviations and were compared with the Student t test. A P value of <0.05 was considered statistically significant, and all tests were two-tailed.
RESULTS
ompK36 typing.
Among the 226 study isolates, ompK36 types A, B, C, and D were identified by PCR in 61 (27.0%), 28 (12.4%), 100 (44.2%), and 36 (15.9%) isolates, respectively. The study isolates were separated into four groups based on the ompK36 types. Sequence analysis of ompK36 confirmed the result of PCR typing. PCR products were not observed with the 4 group-specific primers in 1 isolate. The comparative sequence analysis showed that the untypeable isolate possessed a rare ompK36 allele not belonging to the four major types (see Fig. S1 in the supplemental material).
MLST and PFGE analyses.
The MLST analysis showed 123 different STs among the 226 isolates (Table 1). Twenty-six new alleles and 44 novel STs were identified in this study. The three most common STs were ST23 (12.8%, 29 isolates), ST65 (4.9%, 11 isolates), and ST15 (4.4%, 10 isolates); the other STs represented 1 to 7 isolates. Five STs (ST15, ST34, ST107, ST776, and ST1544) comprised isolates with different ompK36 types. Further PFGE analysis showed that the ST15, ST34, ST107, and ST1544 isolates with different ompK36 types had <80% similarity (Fig. 1), suggesting the presence of subgroups in these STs.
TABLE 1
| ompK36 type | Capsular type (no. of isolates) | CGa,b | ST (no. of isolates)a,c |
|---|---|---|---|
| A | K20 (1) | Miscellaneous | ST776 (1) |
| Untyped (60) | CG15 | ST15 (9), ST709 (4) | |
| CG22 | ST152 (1) | ||
| CG23 | ST1905 (2) | ||
| CG34 | ST34 (1) | ||
| CG37 | ST37 (6), ST726 (1), ST884 (1), ST1198 (1), ST1906 (1), ST1931 (1) | ||
| CG45 | ST45 (1) | ||
| CG65 | ST685 (1) | ||
| CG147 | ST147 (3), ST1922 (2) | ||
| CG292 | ST200 (1), ST1925 (1) | ||
| CG347 | ST919 (1) | ||
| CG1074 | ST1914 (1) | ||
| Miscellaneous | ST6 (1), ST11 (4), ST20 (1), ST322 (1), ST367 (1), ST500 (1), ST534 (1), ST1584 (1), ST1770 (1), ST1897 (1), ST1898 (1), ST1900 (1), ST1910 (1), ST1913 (1), ST1917 (1), ST1919 (1), ST1924 (2) | ||
| B | K54 (1) | Miscellaneous | ST29 (1) |
| K57 (1) | Miscellaneous | ST1899 (1) | |
| Untyped (26) | CG292 | ST292 (2), ST1180 (1) | |
| CG347 | ST355 (2), ST1903 (1) | ||
| CG1612 | ST1612 (1), ST1915 (1), ST1947 (1) | ||
| CG1928 | ST1918 (1), ST1928 (1) | ||
| Miscellaneous | ST29 (1), ST605 (1), ST906 (1), ST1304 (1), ST1308 (1), ST1891 (1), ST1892 (1), ST1894 (1), ST1895 (1), ST1901 (1), ST1904 (2), ST1909 (1), ST1912 (1), ST1945 (1) | ||
| C | K1 (30) | CG23 | ST23 (27), ST1893 (1) |
| Miscellaneous | ST1526 (1), ST1932 (1) | ||
| K2 (25) | CG65 | ST25 (1), ST65 (11), ST375 (1) | |
| CG86 | ST86 (5), ST373 (7) | ||
| Miscellaneous | |||
| K5 (3) | CG76 | ST76 (1) | |
| Miscellaneous | ST1049 (2) | ||
| K20 (1) | CG420 | ST1544 (1) | |
| K57 (6) | CG23 | ST23 (2) | |
| CG34 | ST34 (1), ST592 (3) | ||
| Untyped (35) | CG22 | ST1911 (2) | |
| CG34 | ST34 (1) | ||
| CG35 | ST35 (1), ST1948 (1) | ||
| CG65 | ST1927 (1) | ||
| CG76 | ST490 (1) | ||
| CG107 | ST107 (1), ST219 (1) | ||
| CG292 | ST412 (1) | ||
| CG1074 | ST660 (2) | ||
| Miscellaneous | ST1 (1), ST12 (1), ST14 (3), ST39 (1), ST42 (1), ST48 (1), ST111 (1), ST184 (1), ST294 (1), ST715 (1), ST964 (1), ST1040 (1), ST1526 (1), ST1589 (1), ST1764 (1), ST1886 (1), ST1896 (1), ST1902 (1), ST1923 (1), ST1926 (1), ST1929 (1) | ||
| D | K20 (10) | CG268 | ST268 (3), ST1590 (1) |
| CG420 | ST420 (1), ST1544 (5) | ||
| Untyped (26) | CG15 | ST15 (1), ST277 (1) | |
| CG22 | ST1808 (1) | ||
| CG45 | ST1540 (1) | ||
| CG107 | ST107 (1) | ||
| Miscellaneous | ST36 (7), ST133 (1), ST182 (1), ST252 (1), ST397 (1), ST587 (1), ST776 (2), ST873 (1), ST1409 (1), ST1516 (1), ST1916 (1), ST1920 (1), ST1930 (1), ST1946 (1) | ||
| Untypeable | Untyped (1) | Miscellaneous | ST1921 (1) |
a
CGs or STs that included isolates with different ompK36 types are in bold.
b
Only CGs containing two or more different STs are listed.
c
Novel STs obtained in this study are underlined.
FIG 1
The eBURST analysis with the 123 STs identified in this study alone showed 8 single-locus variant (SLV) pairs (ST35-ST1948, ST76-ST490, ST86-ST373, ST107-ST219, ST147-ST1922, ST355-ST1899, ST420-ST1544, and ST1918-ST1928), 7 ST clusters (ST14-ST15-ST709, ST23-ST1905-ST1893, ST37-ST726-ST1906-ST1198-ST884, ST65-ST1927-ST25-ST375, ST268-ST36-ST1914-ST1590, ST292-ST1925-ST1180, and ST1612-ST1915-ST1947), and 82 singletons. ST14 was an SLV of ST15, but they were assigned to different ompK36 groups. ST14 and ST15 were subgroup founders in the eBURST analysis using the whole MLST database and have been allocated to CG14 and CG15 (34). Moreover, 3 ST clusters (ST23-ST1905-ST1893, ST268-ST36-ST1914-ST1590, and ST292-ST1925-ST1180) included members with different ompK36 types. The remaining three ST clusters and all SLV pairs contained STs with the same ompK36 types (see Fig. S2 in the supplemental material).
Nineteen CGs were found to contain two or more STs detected in this study by the eBURST analysis, including CG15, CG22, CG23, CG34, CG35, CG37, CG45, CG65, CG76, CG86, CG107, CG147, CG268, CG292, CG347, CG420, CG1074, CG1612, and CG1928 (Table 1). In general, most isolates within a single CG had the same ompK36 type. All CG37 and CG147 isolates and all CG15 isolates except 1 ST15 and 1 ST277 isolates belonged to ompK36 group A. All CG1612 isolates and both CG1928 isolates were allocated to group B. All CG23 isolates except 2 ST1905 isolates, all CG34 isolates except 1 ST34 isolate, both CG35 and CG76 isolates, all CG65 isolates except 1 ST685 isolate, and all CG86 isolates were assigned to group C. All CG268 isolates and all CG420 isolate except 1 ST1544 isolate belonged to group D.
Hypermucoviscous phenotype, capsular typing, and virulence gene content.
Among the 226 isolates, hypermucoviscosity was observed in 78 (34.5%) isolates, of which 62 (79.5%) isolates were classified in group C. Three (3.8%), 1 (1.3%), and 12 (15.4%) of the 78 mucoid isolates were in groups A, B, and D, respectively. The hypermucoviscous phenotype was predominant (62.0%) among the group C isolates, and the prevalence of the phenotype was significantly different among the four ompK36 groups (Table 2).
TABLE 2
| Characteristic | ompK36 typea | P valueb | |||
|---|---|---|---|---|---|
| A (n = 61) | B (n = 28) | C (n = 100) | D (n = 36) | ||
| Hypermucoviscosity | 3 (4.9) | 1 (3.6) | 62 (62.0) | 12 (33.3) | <0.001 |
| Capsular type | |||||
| K1 (wzy_K1) | 0 (0) | 0 (0) | 30 (30.0) | 0 (0) | <0.001 |
| K2 | 0 (0) | 0 (0) | 25 (25.0) | 0 (0) | <0.001 |
| K5 | 0 (0) | 0 (0) | 3 (3.0) | 0 (0) | 0.284 |
| K20 | 1 (1.6) | 0 (0) | 1 (1.0) | 10 (27.8) | <0.001 |
| K54 | 0 (0) | 1 (3.6) | 0 (0) | 0 (0) | 0.070 |
| K57 | 0 (0) | 1 (3.6) | 6 (6.0) | 0 (0) | 0.116 |
| Virulence gene | |||||
| allS | 3 (4.9) | 1 (3.6) | 16 (16.0) | 0 (0) | 0.008 |
| rmpA | 5 (8.2) | 2 (7.1) | 64 (64.0) | 13 (36.1) | <0.001 |
| mrkD | 61 (100) | 28 (100) | 97 (97.0) | 36 (100) | 0.284 |
| kfuBC | 18 (29.5) | 11 (39.3) | 46 (46.0) | 10 (27.8) | 0.102 |
| cf29a | 61 (100) | 28 (100) | 100 (100) | 36 (100) | |
| fimH | 61 (100) | 28 (100) | 97 (97.0) | 36 (100) | 0.284 |
| uge | 49 (80.3) | 25 (89.3) | 93 (93.0) | 32 (88.9) | 0.112 |
| wabG | 61 (100) | 28 (100) | 99 (99.0) | 36 (100) | 0.740 |
| ureA | 61 (100) | 28 (100) | 99 (99.0) | 36 (100) | 0.740 |
| Virulence scorec | 5.2 ± 0.8 | 5.4 ± 0.6 | 6.4 ± 1.6 | 5.5 ± 0.9 | <0.001d |
a
Data are expressed as no. (%) of isolates in each ompK36 group except for the virulence score.
b
P values refer to differences between the four groups and were obtained by chi-square tests except for the virulence score.
c
Total number of positive virulence genes. Data are given as means ± standard deviations.
d
Student's t test.
Thirty (13.3%) of the 226 isolates were identified as the K1 serotype by multiplex PCRs, 25 (11.1%) as K2, 3 (1.3%) as K5, 12 (5.3%) as K20, 1 (0.4%) as K54, and 7 (3.1%) as K57 (Table 1). All K1, K2, and K5 isolates, 6 (85.7%) of the 7 K57 isolates, and 1 (8.3%) of the 12 K20 isolates belonged to group C (Table 2). Group D was predominant among the 12 K20 isolates (83.3%, 10 isolates). The K54 isolate and 1 K57 isolate belonged to group B, and 1 K20 isolate was in group A.
Among the 10 tested virulence genes, wzy_K1 (magA), allS, and rmpA were significantly different among the four ompK36 groups, and they were most frequent among group C isolates (Table 2). Group C isolates had higher mean virulence scores than isolates of the other groups.
Antimicrobial resistance and resistance genes.
ESBLs were detected by the phenotypic and genotypic tests in 33 (14.6%) of the 226 isolates. Group A isolates were predominant among the 33 ESBL producers (66.7%, 22 isolates), and group C and D isolates accounted for 15.1% (n = 5) and 18.2% (n = 6) of all ESBL producers. The SHV-type ESBLs were identified in 26 (11.5%) of the 226 isolates; 17 (65.4%), 4 (15.4%), and 5 (19.2%) of the 26 isolates belonged to groups A, C, and D, respectively. Eight (3.5%) of the 226 isolates had the CTX-M-type ESBLs, including 2 CTX-M-3, 3 CTX-M-14, and 3 CTX-M-15; 6 (75.0%), 1 (12.5%), and 1 (12.5%) of the 8 CTX-M producers belonged to groups A, C, and D, respectively. The DHA-1 AmpC enzyme was detected in 20 (8.9%) of the 226 isolates. Group A was also predominant among the 20 DHA-1 producers (70.0%, 14 isolates); 2 (10.0%) and 4 (20.0%) isolates were in groups B and C, respectively. The most common ST among 44 isolates producing ESBLs and/or AmpC was ST15 (20.5%, 9 isolates), followed by ST11 (9.1%, 4 isolates) and ST709 (6.8%, 3 isolates). The other STs represented 1 to 2 isolates (see Table S1 in the supplemental material). Carbapenemase-producing isolates were not detected among the study isolates.
Overall, qnr genes were detected in 54 (23.9%) of the 226 isolates, including 1 qnrA1-like-positive, 6 qnrB1-like-positive, 18 qnrB4-like-positive, and 37 qnrS1-like-positive isolates. Also, group A isolates were the most common and accounted for 31 (57.4%) of the 54 isolates. Three (5.6%), 11 (20.4%), and 9 (16.7%) isolates were in groups B, C, and D, respectively. Group A was predominant among the qnrB-positive isolates (75.0%, 18 isolates) and the most common among the qnrS-positive isolates (45.9%, 17 isolates). Two isolates had the aac (6′)-Ib-cr gene, including 1 group A and 1 group B isolate.
Statistical analysis showed significant differences in antimicrobial resistance and resistance gene content among the four ompK36 groups (Table 3). Group A isolates had the highest nonsusceptibility rates against all tested antibiotics except meropenem and had the highest prevalence rates of SHV-type ESBLs, CTX-M ESBLs, the DHA-1-type AmpC enzyme, qnrB, and qnrS.
TABLE 3
| Characteristic | ompK36 typea | P valueb | |||
|---|---|---|---|---|---|
| A (n = 61) | B (n = 28) | C (n = 100) | D (n = 36) | ||
| Nonsusceptible to: | |||||
| Cefazolin | 31 (50.8) | 3 (10.7) | 10 (10.0) | 6 (16.7) | <0.001 |
| Cefoxitin | 16 (26.2) | 3 (10.7) | 7 (7.0) | 1 (2.8) | 0.014 |
| Cefotaxime or ceftazidime | 30 (49.2) | 2 (7.1) | 8 (8.0) | 6 (16.7) | <0.001 |
| Cefepime | 25 (41.0) | 0 (0) | 5 (5.0) | 6 (16.7) | <0.001 |
| Ertapenem | 14 (23.0) | 1 (3.6) | 5 (5.0) | 0 (0) | <0.001 |
| Imipenem | 12 (19.7) | 0 (0) | 4 (4.0) | 1 (2.8) | 0.001 |
| Meropenem | 1 (1.6) | 0 (0) | 0 (0) | 0 (0) | 0.440 |
| Gentamicin | 25 (41.0) | 1 (3.6) | 5 (5.0) | 6 (16.7) | <0.001 |
| Amikacin | 8 (13.1) | 0 (0) | 1 (1.0) | 0 (0) | 0.001 |
| Nalidixic acid | 40 (65.6) | 5 (17.9) | 30 (30.0) | 18 (50.0) | <0.001 |
| Ciprofloxacin | 32 (52.5) | 0 (0) | 7 (7.0) | 7 (19.4) | <0.001 |
| Resistance gene | |||||
| blaSHV-ESBL | 17 (27.9) | 0 (0) | 4 (4.0) | 5 (13.9) | <0.001 |
| blaCTX-M | 6 (9.8) | 0 (0) | 1 (1.0) | 1 (2.8) | 0.018 |
| blaDHA | 14 (23.0) | 2 (7.1) | 4 (4.0) | 0 (0) | <0.001 |
| ESBL and/or AmpC gene(s) | 29 (47.5) | 2 (7.1) | 7 (7.0) | 6 (16.7) | <0.001 |
| qnr | 31 (50.8) | 3 (10.7) | 11 (11.0) | 9 (25.0) | <0.001 |
| qnrA | 1 (1.6) | 0 (0) | 0 (0) | 0 (0) | 0.440 |
| qnrS | 17 (27.9) | 3 (10.7) | 9 (9.0) | 8 (22.2) | 0.010 |
| qnrB | 18 (29.5) | 1 (3.6) | 4 (4.0) | 1 (2.8) | <0.001 |
| aac(6′)-Ib-cr | 1 (1.6) | 0 (0) | 1 (1.0) | 0 (0) | 0.911 |
| Plasmid Inc group | |||||
| FIA | 0 (0) | 0 (0) | 0 (0) | 1 (2.8) | 0.153 |
| I1-1γ | 1 (1.6) | 0 (0) | 3 (3.0) | 0 (0) | 0.568 |
| L/M | 0 (0) | 0 (0) | 1 (1.0) | 1 (2.8) | 0.518 |
| N | 4 (6.6) | 1 (3.6) | 2 (2.0) | 1 (2.8) | 0.499 |
| P | 0 (0) | 0 (0) | 1 (1.0) | 1 (2.8) | 0.518 |
| A/C | 6 (9.8) | 0 (0) | 3 (3.0) | 1 (2.8) | 0.102 |
| F | 1 (1.6) | 0 (0) | 0 (0) | 1 (2.8) | 0.392 |
| FIIA | 30 (49.2) | 9 (32.1) | 11 (11.0) | 10 (27.8) | <0.001 |
a
Data are expressed as no. (%) of isolates in each ompK36 group.
b
P values were obtained by chi-square tests and refer to differences between the four groups.
Plasmid typing.
Eight incompatibility groups of plasmids, IncFIA, I1-Iγ, L/M, N, P, A/C, F, and FIIA, were detected by the PCR-based replicon typing method in 1 (0.4%), 4 (1.8%), 2 (0.9%), 8 (3.5%), 2 (0.9%), 10 (4.4%), 2 (0.9%), and 60 (26.5%) of the 226 isolates, respectively. The four groups of K. pneumoniae differed only in the carriage of IncFIIA (Table 3), and the prevalence rate of IncFIIA-type plasmids was highest in group A. It was noted that 22 (84.6%) of the 26 blaSHV-ESBL-positive isolates, 4 (50.0%) of the 8 CTX-M-producers, and 15 (75.0%) of the 20 DHA-1 producers had the IncFIIA-type plasmids, suggesting that the IncFIIA-type plasmids might play an important role in the spread of blaESBL and ampC in K. pneumoniae in this institution.
Classification of bloodstream infections.
Bloodstream infections caused by the isolates of the four ompK36 groups were analyzed, and statistical analysis showed significant differences in the possible site of infection acquisition between the four groups (Table 4). Group C isolates were more likely to be community acquired (40.0%) than the isolates of the other groups, while group A isolates tended to be nosocomially acquired (50.8%). Twelve (42.9%) of the 28 group B isolates were responsible for bacteremic biliary tract infections, and the rate was significantly higher than the rates in other groups (8.2 to 14.0%) (Table 4). Twenty-one currently available STs were identified among the 44 isolates causing bacteremic biliary tract infections; each of the STs represented 1 to 2 isolates (see Table S2 in the supplemental material). Seventeen isolates were responsible for PLA-associated bacteremias, and 16 (94.1%) of them were in group C. ST23 was predominant among the 17 PLA-associated isolates (76.5%, 13 isolates) (see Table S2 in the supplemental material). There were no significant differences in bacteremias related to pneumonia, urinary tract infections, peritonitis, and soft tissue infections among the four groups. The majority of bacteremic PLAs, biliary tract infections, soft tissue infections, peritonitis, and urinary tract infections were community onset, and the vast majority (82.4%) of PLAs were community acquired (see Table S3 in the supplemental material). Most of the bacteremic pneumonias were either nosocomial acquired (50.0%) or health care-associated community onset (38.9%). Fifty-four patients who died within 30 days of the first positive blood culture with K. pneumoniae and 158 patients known to have survived for more than 30 days were analyzed. The 30-day mortality rates of groups A, B, C, and D were 28.1% (16 of 57 patients), 22.2% (6 of 27 patients), 28.7% (27 of 94 patients), and 14.7% (5 of 34 patients), respectively, and the difference was not statistically significant between the four groups (P = 0.399).
TABLE 4
| Characteristic | ompK36 typea | P valueb | |||
|---|---|---|---|---|---|
| A (n = 61) | B (n = 28) | C (n = 100) | D (n = 36) | ||
| Onset of infection | |||||
| Nosocomial | 31 (50.8) | 11 (39.3) | 22 (22.0) | 12 (33.3) | 0.002 |
| Community onset | 30 (49.2) | 17 (60.7) | 78 (78.0) | 24 (66.7) | 0.002 |
| Community-acquired | 5 (8.2) | 6 (21.4) | 40 (40.0) | 6 (16.7) | <0.001 |
| Health care associated | 25 (41.0) | 11 (39.3) | 38 (38.0) | 18 (50.0) | 0.657 |
| Primary site of infection | |||||
| Biliary tract | 5 (8.2) | 12 (42.9) | 14 (14.0) | 3 (8.3) | <0.001 |
| Pyogenic liver abscess | 0 (0) | 0 (0) | 16 (16.0) | 1 (2.8) | <0.001 |
| Pneumonia | 13 (21.3) | 3 (10.7) | 14 (14.0) | 6 (16.7) | 0.537 |
| Urinary tract | 14 (23.0) | 3 (10.7) | 14 (14.0) | 10 (27.8) | 0.146 |
| Peritonitis | 1 (1.6) | 1 (3.6) | 5 (5.0) | 2 (5.6) | 0.709 |
| Skin and soft tissue | 1 (1.6) | 1 (3.6) | 4 (4.0) | 1 (2.8) | 0.865 |
a
Data are expressed as no. (%) of isolates in each ompK36 group.
b
P values were obtained by chi-square tests and refer to differences between the four groups.
DISCUSSION
Genetic polymorphism of ompK36 has been described in K. pneumoniae (19, 20), and the ompK36 variants can be divided into four major groups (21). The present study demonstrated that the vast majority of clinical K. pneumoniae isolates can be allocated into four distinct groups based on ompK36 types; only one isolate with a rare ompK36 type was not assigned to the four groups by the PCR typing method. Moreover, this study only enrolled isolates from a single institution in Taiwan. It is thus very likely that there are still other rare ompK36 alleles or groups that cannot be defined by the PCR typing method.
Group A included all isolates of three genetically unrelated epidemic resistant clones (ST11, ST15, and ST147) except 1 ST15 isolate. Among the 17 isolates of the three high-risk clones, 13 (76.5%) isolates were ESBL and/or AmpC producers. The inclusion of major high-risk clones may explain partially the distinct characteristics of group A: high prevalences of antimicrobial resistance, ESBLs, AmpC, Qnr-type quinolone resistance determinants, and the IncFIIA-type plasmids. Moreover, among the four groups, group A was most likely associated with nosocomial infections. The characteristics of this group suggest that group A isolates have been more successfully adapted to health care environments than others. The important epidemic clone ST258 (an SLV of ST11) is not common in Taiwan (16) and was not detected in our present and previous studies (21). The ompK36 type of ST258 and the possibility that clones other than ST11, ST15, and ST147 in this group become high-risk clones deserve further investigation.
The capsular serotypes K1 and K2 and the hypermucoviscous phenotype are important virulence determinants of K. pneumoniae (4, 6, 7), and community-onset invasive infections are often caused by isolates with those characteristics (4, 5, 8, 9). Previous studies have demonstrated that among non-K1/K2 K. pneumoniae isolates, serotype K5, K20, K54, and K57 isolates are common causes of community-onset pneumonia and PLA, and the majority of them had the hypermucoviscous phenotype (5, 9). It is interesting to find that the majority of mucoid isolates, all K1, K2, and K5 isolates, and most K57 isolates were included in group C. ST23, ST65, ST86, ST373, and ST375 have been reported to be associated with PLA (4, 7), and all isolates of these virulent clones (29 ST23, 11 ST65, 5 ST86, 7 ST373, and 1 ST375 isolates) were also assigned to this group (see Table S2 in the supplemental data). The inclusion of hypervirulent clones can explain the association of this group with community-onset infection and PLA.
It is also interesting to find that as many as 42.9% of group B isolates were responsible for bacteremic biliary tract infections, and no endemic clone was identified in this group. K20 was the third most common serotype in this study, and 11 K20 isolates were detected in this study. Ten of the 11 K20 isolates, which belonged to four STs, were classified in group D. The rate of bacteremic urinary tract infection-associated isolates was the highest in group D; however, the difference among the four ompK36 groups was not statistically significant. A better understanding the characteristics shared by different clones of K. pneumoniae linked to a specific infection might be helpful to identify microbiological factors contributing to the specific type of infection. One major limitation of this study was that only bloodstream isolates, which might represent more virulent strains, were analyzed. Another limitation was the lack of an adequate number of patients for subset studies to investigate the usefulness of our typing method in clinical practice. More studies using isolates from various sample types are therefore needed to confirm the link between specific types of infection and ompK36 allele groups in K. pneumoniae and to investigate if our typing method can be utilized to predict the progression of primary infection or the prognosis of bacteremia in various clinical subsets.
The OmpK36 porin plays an important role in noncarbapenemase-mediated carbapenem resistance (17–19). Moreover, previous studies revealed the association of OmpK36 with bacterial virulence (18, 35). The ompK36-deficient mutant was shown to confer a lower virulence in a mouse peritonitis model (18), and the loss of OmpK36 may confer decreased resistance to neutrophil phagocytosis and increased resistance to serum killing (35). It is therefore likely that the separation of high-risk resistant clones and virulent clones into two distinct groups and similar ompK36 types among genetically unrelated epidemic clones were the consequence of convergent evolution driven under the selective pressures created by antimicrobial use and host factors.
A prevalent multidrug-resistant clone may either derive from a preexisting successful, but antibiotic-susceptible, strain with subsequently acquired resistance determinants or come from an initially minor strain with resistance as the key driver for its prevalence (11). ST11, ST15, ST37, and ST147 are important STs among carbapenem-resistant K. pneumoniae isolates in Taiwan, and they were reported to account for >70% of carbapenem-resistant isolates by Chiu et al. (16). In this study, the four STs only accounted for 24 (10.6%) of all study isolates (Table 1), suggesting that these high-risk resistant clones had been minor strains rather than evolving from preexisting successful strains. Active continuous surveillance followed by appropriate infection control measures is needed to prevent the further spread of these clones and dissemination of resistance determinants, and the ompK36 typing methods may be useful for screening of the high-risk clones.
ACKNOWLEDGMENTS
This work was partially supported by grant NCKUH-10301001 from the National Cheng Kung University Hospital, Taiwan, and grant NSC 103-2320-B-006-018-MY3 from the Ministry of Science and Technology, Taiwan.
We thank the team of curators of the Institut Pasteur MLST and whole-genome MLST databases for curating the data and making them publicly available at http://bigsdb.web.pasteur.fr/.
We declare no conflicts of interest.
Supplemental Material
File (zjm999094513so1.pdf)
- Download
- 512.36 KB
REFERENCES
1.
Podschun R, Ullmann U. 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin Microbiol Rev 11:589–603.
2.
Keynan Y, Rubinstein E. 2007. The changing face of Klebsiella pneumoniae infections in the community. Int J Antimicrob Agents 30:385–389.
3.
Song JH, Thamlikitkul V, Hsueh PR. 2011. Clinical and economic burden of community-acquired pneumonia amongst adults in the Asia-Pacific region. Int J Antimicrob Agents 38:108–117.
4.
Siu LK, Yeh KM, Lin JC, Fung CP, Chang FY. 2012. Klebsiella pneumoniae liver abscess: a new invasive syndrome. Lancet Infect Dis 12:881–887.
5.
Fang CT, Lai SY, Yi WC, Hsueh PR, Liu KL, Chang SC. 2007. Klebsiella pneumoniae genotype K1: an emerging pathogen that causes septic ocular or central nervous system complications from pyogenic liver abscess. Clin Infect Dis 45:284–293.
6.
Yu VL, Hansen DS, Ko WC, Sagnimeni A, Klugman KP, von Gottberg A, Goossens H, Wagener MM, Benedi VJ, International Klebsiella Study Group. 2007. Virulence characteristics of Klebsiella and clinical manifestations of K. pneumoniae bloodstream infections. Emerg Infect Dis 13:986–993.
7.
Lin JC, Koh TH, Lee N, Fung CP, Chang FY, Tsai YK, Ip M, Siu LK. 2014. Genotypes and virulence in serotype K2 Klebsiella pneumoniae from liver abscess and non-infectious carriers in Hong Kong, Singapore and Taiwan. Gut Pathog 6:21.
8.
Luo Y, Wang Y, Ye L, Yang J. 2014. Molecular epidemiology and virulence factors of pyogenic liver abscess causing Klebsiella pneumoniae in China. Clin Microbiol Infect 20:O818−O824.
9.
Liu YM, Li BB, Zhang YY, Zhang W, Shen H, Li H, Cao B. 2014. Clinical and molecular characteristics of emerging hypervirulent Klebsiella pneumoniae bloodstream infections in mainland China. Antimicrob Agents Chemother 58:5379–5385.
10.
Brisse S, Fevre C, Passet V, Issenhuth-Jeanjean S, Tournebize R, Diancourt L, Grimont P. 2009. Virulent clones of Klebsiella pneumoniae: identification and evolutionary scenario based on genomic and phenotypic characterization. PLoS One 4:e4982.
11.
Woodford N, Turton JF, Livermore DM. 2011. Multiresistant Gram-negative bacteria: the role of high-risk clones in the dissemination of antibiotic resistance. FEMS Microbiol Rev 35:736–755.
12.
Damjanova I, Tóth A, Pászti J, Hajbel-Vékony G, Jakab M, Berta J, Milch H, Füzi M. 2008. Expansion and countrywide dissemination of ST11, ST15 and ST147 ciprofloxacin-resistant CTX-M-15-type β-lactamase-producing Klebsiella pneumoniae epidemic clones in Hungary in 2005—the new ‘MRSAs’? J Antimicrob Chemother 62:978–985.
13.
Baraniak A, Izdebski R, Fiett J, Sadowy E, Adler A, Kazma M, Salomon J, Lawrence C, Rossini A, Salvia A, Vidal Samso J, Fierro J, Paul M, Lerman Y, Malhotra-Kumar S, Lammens C, Goossens H, Hryniewicz W, Brun-Buisson C, Carmeli Y, Gniadkowski M, MOSAR WP2 and WP5 Study Groups. 2013. Comparative population analysis of Klebsiella pneumoniae strains with extended-spectrum β-lactamases colonizing patients in rehabilitation centers in four countries. Antimicrob Agents Chemother 57:1992–1997.
14.
Breurec S, Guessennd N, Timinouni M, Le TA, Cao V, Ngandjio A, Randrianirina F, Thiberge JM, Kinana A, Dufougeray A, Perrier-Gros-Claude JD, Boisier P, Garin B, Brisse S. 2013. Klebsiella pneumoniae resistant to third-generation cephalosporins in five African and two Vietnamese major towns: multiclonal population structure with two major international clonal groups, CG15 and CG258. Clin Microbiol Infect 19:349–355.
15.
Lee MY, Ko KS, Kang CI, Chung DR, Peck KR, Song JH. 2011. High prevalence of CTX-M-15-producing Klebsiella pneumoniae isolates in Asian countries: diverse clones and clonal dissemination. Int J Antimicrob Agents 38:160–163.
16.
Chiu SK, Wu TL, Chuang YC, Lin JC, Fung CP, Lu PL, Wang JT, Wang LS, Siu LK, Yeh KM. 2013. National surveillance study on carbapenem non-susceptible Klebsiella pneumoniae in Taiwan: the emergence and rapid dissemination of KPC-2 carbapenemase. PLoS One 8:e69428.
17.
Tsai YK, Liou CH, Fung CP, Lin JC, Siu LK. 2013. Single or in combination antimicrobial resistance mechanisms of Klebsiella pneumoniae contribute to varied susceptibility to different carbapenems. PLoS One 8:e79640.
18.
Tsai YK, Fung CP, Lin JC, Chen JH, Chang FY, Chen TL, Siu LK. 2011. Klebsiella pneumoniae outer membrane porins OmpK35 and OmpK36 play roles in both antimicrobial resistance and virulence. Antimicrob Agents Chemother 55:1485–1493.
19.
Clancy CJ, Hao B, Shields RK, Chen L, Perlin DS, Kreiswirth BN, Nguyen MH. 2014. Doripenem, gentamicin, and colistin, alone and in combinations, against gentamicin-susceptible, KPC-producing Klebsiella pneumoniae strains with various ompK36 genotypes. Antimicrob Agents Chemother 58:3521–3525.
20.
Papagiannitsis CC, Giakkoupi P, Kotsakis SD, Tzelepi E, Tzouvelekis LS, Vatopoulos AC, Miriagou V. 2013. OmpK35 and OmpK36 porin variants associated with specific sequence types of Klebsiella pneumoniae. J Chemother 25:250–254.
21.
Yan JJ, Wang MC, Zheng PX, Tsai LH, Wu JJ. 2015. Associations of the major international high-risk resistant clones and virulent clones with specific ompK36 allele groups in Klebsiella pneumoniae in Taiwan. New Microbes New Infect 5:1–4.
22.
Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. 1988. CDC definitions for nosocomial infections, 1988. Am J Infect Control 16:128–140.
23.
Friedman ND, Kaye KS, Stout JE, McGarry SA, Trivette SL, Briggs JP, Lamm W, Clark C, MacFarquhar J, Walton AL, Reller LB, Sexton DJ. 2002. Health care-associated bloodstream infections in adults: a reason to change the accepted definition of community-acquired infections. Ann Intern Med 137:791–797.
24.
Clinical and Laboratory Standards Institute. 2014. Performance standards for antimicrobial susceptibility testing; 24th informational supplement. Document M100-S24. Clinical and Laboratory Standards Institute, Wayne, PA.
25.
Wu JJ, Ko WC, Wu HM, Yan JJ. 2008. Prevalence of Qnr determinants among bloodstream isolates of Escherichia coli and Klebsiella pneumoniae in a Taiwanese hospital, 1999–2005. J Antimicrob Chemother 61:1234–1239.
26.
Pérez-Pérez FJ, Hanson ND. 2002. Detection of plasmid-mediated AmpC β-lactamase genes in clinical isolates by using multiplex PCR. J Clin Microbiol 40:2153–2162.
27.
Nakano R, Okamoto R, Nakano Y, Kaneko K, Okitsu N, Hosaka Y, Inoue M. 2004. CFE-1, a novel plasmid-encoded AmpC β-lactamase with an ampR gene originating from Citrobacter freundii. Antimicrob Agents Chemother 48:1151–1158.
28.
Nüesch-Inderbinen MT, Hächler H, Kayser FH. 1996. Detection of genes coding for extended-spectrum SHV beta-lactamases in clinical isolates by a molecular genetic method, and comparison with the E test. Eur J Clin Microbiol Infect Dis 15:398–402.
29.
Edelstein M, Pimkin M, Palagin I, Edelstein I, Stratchounski L. 2003. Prevalence and molecular epidemiology of CTX-M extended-spectrum β-lactamase-producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. Antimicrob Agents Chemother 47:3724–3732.
30.
Zhou TL, Chen XJ, Zhou MM, Zhao YJ, Luo XH, Bao QY. 2011. Prevalence of plasmid-mediated quinolone resistance in Escherichia coli isolates in Wenzhou, Southern China, 2002-2008. Jpn J Infect Dis 64:55–57.
31.
Jung SW, Chae HJ, Park YJ, Yu JK, Kim SY, Lee HK, Lee JH, Kahng JM, Lee SO, Lee MK, Lim JH, Lee CH, Chang SJ, Ahn JY, Lee JW, Park YG. 2013. Microbiological and clinical characteristics of bacteraemia caused by the hypermucoviscosity phenotype of Klebsiella pneumoniae in Korea. Epidemiol Infect 141:334–340.
32.
Diancourt L, Passet V, Verhoef J, Grimont PA, Brisse S. 2005. Multilocus sequence typing of Klebsiella pneumoniae nosocomial isolates. J Clin Microbiol 43:4178–4182.
33.
Carattoli A, Bertini A, Villa L, Falbo V, Hopkins KL, Threlfall EJ. 2005. Identification of plasmids by PCR-based replicon typing. J Microbiol Methods 63:219–228.
34.
Bialek-Davenet S, Criscuolo A, Ailloud F, Passet V, Jones L, Delannoy-Vieillard AS, Garin B, Le Hello S, Arlet G, Nicolas-Chanoine MH, Decré D, Brisse S. 2014. Genomic definition of hypervirulent and multidrug-resistant Klebsiella pneumoniae clonal groups. Emerg Infect Dis 20:1812–1820.
35.
Chen JH, Siu LK, Fung CP, Lin JC, Yeh KM, Chen TL, Tsai YK, Chang FY. 2010. Contribution of outer membrane protein K36 to antimicrobial resistance and virulence in Klebsiella pneumoniae. J Antimicrob Chemother 65:986–990.
Information & Contributors
Information
Published In
Journal of Clinical Microbiology
Volume 53 • Number 10 • October 2015
Pages: 3256 - 3263
Editor: P. Bourbeau
PubMed: 26224840
Copyright
© 2015 American Society for Microbiology.
History
Received: 28 April 2015
Returned for modification: 25 May 2015
Accepted: 22 July 2015
Published online: 16 September 2015
Contributors
Editor
P. Bourbeau
Editor
Metrics & Citations
Metrics
Note:
- For recently published articles, the TOTAL download count will appear as zero until a new month starts.
- There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
- Citation counts come from the Crossref Cited by service.
Citations
Citation
2015.
Allocation of Klebsiella pneumoniae Bloodstream Isolates into Four Distinct Groups by ompK36 Typing in a Taiwanese University Hospital. J Clin Microbiol
53:.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.