Logo of jcmPermissionsJournals.ASM.orgJournalJCM ArticleJournal InfoAuthorsReviewers
J Clin Microbiol. 2008 Jul; 46(7): 2231–2240.
Published online 2008 May 28. doi:  10.1128/JCM.01716-07
PMCID: PMC2446917

Capsular Polysaccharide Synthesis Regions in Klebsiella pneumoniae Serotype K57 and a New Capsular Serotype


Community-acquired pyogenic liver abscess caused by Klebsiella pneumoniae is an emerging infectious disease. We explored the capsular polysaccharide synthesis (cps) regions of three non-K1, non-K2 K. pneumoniae strains, A1142, A7754, and A1517, from Taiwanese patients experiencing pyogenic liver abscess. Two of the strains, A1142 and A7754, belonged to capsular serotype K57, while the third belonged to a new capsular serotype, different from the previously reported 77 serotypes. Deletion and complementation experiments suggested that a unique K57 gene, a homologue of wzy, was essential for K57 capsular synthesis and confirmed that this gene cluster was a genetic coding region for K57. Compared to K1 and K2 strains, the three strains were all serum sensitive, suggesting that host factors might also be involved in the three patients. PCR using primers from specific genes for K57 was more sensitive and specific than traditional serotyping. The remaining strain, A1517, did not react to the antisera from any of the 77 serotypes, and none of the 77 reference strains reacted to the serum against this strain. Moreover, PCR analyses using various primer pairs from the serotype-specific open reading frames did not reveal cross-reactivity to any of the 77 reference strains, suggesting that this strain likely represents a new capsular type. We conclude that sequences from these two cps regions are very useful in detecting K57 and the new cps genotype.

Klebsiella pneumoniae is an opportunistic hospital-acquired pathogen which usually causes urinary tract infections, pneumonia, and septicemia (1, 28). Over the past 2 decades, a new type of community-acquired K. pneumoniae associated with pyogenic liver abscess has been reported to occur in Taiwan, Japan, Europe, North America, and Korea (5, 7, 9, 16, 23, 24, 29, 34). This emerging disease is often complicated with septic meningitis and endophthalmitis. The tissue-invasive K. pneumoniae infections can attack healthy individuals, and only 50% of patients have a predisposing condition, such as diabetes mellitus (9).

The bacterial capsule is considered a major virulence factor of K. pneumoniae, with serotype-related variation in severity of infection being observed (8, 21). Previous studies have documented that the majority of K. pneumoniae strains causing liver abscess belong to serotypes K1 and K2 (12, 32), which are the most virulent of the known serotypes (21).

Previously, we reported on the isolation of 42 tissue-invasive strains from patients with pyogenic liver abscess (20). PCR-based detection of magA in capsular polysaccharide synthesis (cps) genotype K1 (6, 7, 11, 15, 22, 32) and cps PCR genotyping of other serotype-specific cps regions (10, 39) characterized these strains as belonging to cps genotypes K1 (n = 35), K2 (n = 2), K5 (n = 1), and K54 (n = 1). The remaining three strains were not characterized.

Presently, we investigated the cps genotypes of these three strains. The cps regions of two of them were fully sequenced. In addition to cps PCR genotyping, we also report on a new method of identifying the capsular serotypes more sensitively and inexpensively.


Bacterial strains and plasmids.

Forty-two clinical isolates of K. pneumoniae were obtained from patients admitted to the National Taiwan University Hospital (NTUH) with pyogenic liver abscess with or without septic complications, such as meningitis (20). The cps genotypes of the three previously uncharacterized strains (designated A1142, A7754, and A1517) were presently explored. Twenty-one nonblood isolates from nonseptic patients at the NTUH; 13 strains from patients at En Chu Kong Hospital (ECKH; Sansia, Taiwan); 34 strains obtained from patients at Far Eastern Memorial Hospital (FEMH; Banciao, Taiwan); 24 strains, including MGH78578, purchased from the American Type Culture Collection (ATCC) (9); 80 strains from Canada (Department of Medical Microbiology, University of Manitoba, Winnipeg, Manitoba, Canada); and 1 strain from Finland (Department of Clinical Microbiology, Kuopio University Hospital, Finland) were also used in this study. The bacterial strains and plasmids used in this paper are summarized in Table Table1.1. K. pneumoniae and Escherichia coli were cultured in Luria-Bertani (LB) medium supplemented with appropriate antibiotics, including ampicillin (100 μg/ml) or kanamycin (50 μg/ml).

Bacterial strains and plasmids used in this studya

Sequencing of cps.

According to the sequences of the cps genomic regions of a K2 serotype, K52 strain MGH78578 of the Genome Sequencing Center at Washington University Medical School (http://genome.wustl.edu/) (2, 30), and a K1 strain (6), we designed primers for conserved sequences flanking the cps region (Fig. (Fig.11 and Table Table2).2). PCR amplifications were performed with the Long and Accurate PCR system (Takara, Tokyo, Japan). We added 1 μg template genomic DNA to a solution containing 5 μl of 10× buffer, 5 μl of 25 mM MgCl2, 2.5 U of LA Taq polymerase, deoxynucleoside triphosphates at final concentrations of 0.5 mM each, and primers at final concentrations of 0.4 mM each in a final volume of 50 μl. The cycling program consisted of one denaturation step of 2 min at 94°C and 10 initial cycles of 10 s at 98°C, 30 s at 63°C, and 12 min at 68°C, followed by 20 iterative cycles of 10 s at 98°C, 30 s at 63°C, and 12 min plus 20 s for each new cycle at 72°C. A final elongation step of 10 min at 72°C was added.

FIG. 1.
Genetic alignment of the capsular polysaccharide synthesis (cps) region and the primers used in PCR amplification of cps regions. Arrowheads indicate primers for the first amplification. Black arrows and gray arrows indicate primers for extending the ...
Primers used in this study

These amplified products were cloned to a CopyControl pCC1 vector for long PCR product cloning (Epicentre, Madison, WI) in accordance with the manufacturer's instructions. We purified the plasmid and performed in vitro transposition with an EZ-Tn5 KAN-2 insertion kit (Epicentre), following the recommended procedures. Transposon insertion clones with randomly interspersed primer-binding sites provided a population of DNA-sequencing templates. Primers KAN-2 FP-1 and KAN-2 RP-1, with complementary sequences facing outward from the transposon, were used for sequencing. In addition, the gaps were filled by primer walking and multiple sequencing reactions to complete the entire sequences (3, 35).

To extend the 5′ and 3′ conserved regions (from galF to gnd), primers pre-galF-F, yegH, post gnd R, and ugd were designed according to the conserved regions upstream and downstream of the cps region; primers 9534 cps 5′R, 13-730F, 1517 conserve 5′R, and 1517 conserve 3′F were designed according to the A1142 or A1517 cps sequences that we read (Fig. (Fig.11 and Table Table2).2). The pre-galF-F and 9534 cps 5′R primers were used for the A1142 5′ cps region amplification, and the 13-730F and post gnd R primers were used for the A1142 3′ cps region. However, primers yegH and 1517 conserve 5′R and primers 1517 conserve 3′F and ugd were used for amplification of the A1517 5′ and 3′ cps regions, respectively, because we failed to obtain PCR products by using primers pre-galF-F and post gnd R. PCR amplifications were also performed with the Long and Accurate PCR system as described earlier in the text. The cycling program was 96°C for 3 min, followed by 30 temperature cycles of 96°C for 30 s, 52°C for 15 s, and 72°C for 2 to 5 min. A final elongation step of 10 min at 72°C was added. The products were sequenced to complete the sequences of the cps region (from galF to gnd). The full sequences of the cps regions in these strains were approximately 20 kb.

PCR-based genotyping of cps (cps PCR genotyping).

From the sequences of cps regions in A1142 and A1517, we designed primers for the cps variable region to clarify the cps genotypes of these strains. PCR was performed by using 77 serotype K reference strains (Statens Serum Institute, Copenhagen, Denmark) as a template and the designed primers for cps PCR genotyping. Primers 9471F and 9897R and primers 1142XF and 1142XR were used in A1142 cps PCR genotyping (see Fig. Fig.3A),3A), and six pairs of primers (1517XF and 1517XR, 1517YF and 1517YR, 12R STAR and 12 STAR, 1517YF and 15R-2, 7R STAR and 7 STAR, and 3-2 and 12R-2) were used in A1517 cps PCR genotyping (see Fig. Fig.3B).3B). In brief, 3 μl of an overnight bacterial culture was added to 10 μl of water and boiled for 15 min to release DNA. We added 2 μl of 10× buffer, 2.5 U Taq polymerase (Bioman, Taipei, Taiwan), deoxynucleoside triphosphates at final concentrations of 0.1 mM each, and primers at final concentrations of 0.4 mM each to the 13 μl of the boiled DNA to produce a final volume of 20 μl. The PCR conditions were 96°C for 3 min, followed by 30 temperature cycles of 96°C for 30 s, 53°C for 15 s, and 72°C for 30 s.

FIG. 3.
cps PCR genotyping of strains A1142, A7754, and A1517. (A) Genetic alignment of the A1142 cps region and the primers for cps PCR genotyping. Primer pair 1, 9471F and 9897R; primer pair 2, 1142XF and 1142XR. (B) Genetic alignment of the A1517 cps region ...

Construction of an unmarked K. pneumoniae deletion mutant.

To obtain cps mutant strains of A1142 (K57), the method of Link et al. (19) and a pKO3-Km plasmid with a temperature-sensitive origin were used. Briefly, a kanamycin resistance cassette from a pUC4K plasmid (33) was digested by AccI and ligated to an AccI-digested pKO3 plasmid to produce a plasmid designated pKO3-Km. Primers 9471F (952 bp upstream of wzy) and R5 (824 bp downstream of wzy) were designed to amplify the target gene wzy and its flanking region. The amplified products were cloned to a pGEM-T Easy vector, followed by inverse PCR and then self-ligation, resulting in the deletion of the entire wzy gene, and the flanking regions remained intact. The flanking regions of wzy were digested by NotI and ligated into a NotI-digested pKO3-Km vector. Plasmid pKO3-Km Δwzy was transformed into A1142 by electroporation. Integration (43°C) and excision (30°C) of this plasmid can result in wild-type or mutant strain formation, which is confirmed by PCR (19). The wzy mutant strain of A1142 was further confirmed using different pairs of primers.


The intact wzy gene and its ribosomal binding site were cloned to a modified pGEM-T Easy vector, which contains an insertion of a kanamycin cassette from a pUC4K plasmid into the NdeI site. Additionally, a lac promoter was ligated to the SacII site upstream of the wzy gene. The plasmid was purified and transformed to the wzy mutant of A1142. trans-Complementation strains were selected by LB agar plates supplemented with 50 μg/ml kanamycin.

Serotyping with agglutination and double immunodiffusion.

Capsular serotypes were initially examined with Seiken antisera directed toward Klebsiella serotypes K1 to K6 (Denka Seiken, Tokyo, Japan) by agglutination, as previously described (6). Briefly, bacteria were cultured overnight in Worfel-Ferguson's medium (0.2% yeast extract, 0.2% sodium chloride, 0.1% potassium sulfate, 0.025% magnesium sulfate, and 2% sucrose). The bacterial cells were resuspended in saline. One loopful of a capsular suspension of the bacteria was mixed with serotype-specific antiserum (K1 to K6). Then, the agglutination results were read at 1 min.

Detailed serotyping was performed by using a double immunodiffusion assay (6). In brief, approximately 4 × 109 bacteria harvested to the extract capsule were exposed to serotype-specific antiserum. The 77 serotype K antisera and the 77 serotype K reference strains used in this study were purchased from Statens Serum Institute, Copenhagen, Denmark.

Serotyping with immunoblot analysis.

Ten microliters of each capsular extract was vacuum spotted onto a nitrocellulose membrane by means of a dot or slot blot device. The membrane was overlapped with a piece of filter, and both were rinsed with Western transfer buffer containing 47.8 mM Tris, 38.6 mM glycine, 20% methanol, and 0.037% sodium dodecyl sulfate. The membrane was dried, and nonspecific sites were blocked by soaking them in 1× phosphate-buffered saline with 0.5% Tween 20 (PBST) plus 5% milk for 1 h at room temperature. The membrane was then incubated with serotype-specific antiserum purchased from Statens Serum Institute or anti-A1517 antiserum (which was generated by immunizing a rabbit with formaldehyde-treated whole-bacterial cells as a primary antibody [1:5,000 dilution for antiserum]), dissolved in PBST plus milk at 4°C overnight, washed four times with PBST for 10 min each time, incubated with secondary antibody conjugated with horseradish peroxidase (HRP) (goat anti-rabbit immunoglobulin G [IgG]-HRP; 1:10,000) for 1 h at room temperature, and washed three times with PBST for 10 min each time. Enhanced chemiluminescence reagent was added for 3 min, and the membrane was exposed to X-ray film in the dark room.

cps PCR-RFLP analysis.

The cps PCR-restriction fragment length polymorphism (RFLP) method was performed as previously described (4). Briefly, the cps genomic regions of A1517, Canada 05-14, A1142, A7754, and the K57 reference strain were amplified with primers CPS-1 and rCPS or primers wzi2 and rCPS2 (Fig. (Fig.11 and Table Table2).2). The amplified products were digested by HincII, and the restriction DNA fragments were separated by electrophoresis in a 1.5% agarose gel.

Serum resistance assay.

The serum resistance assay of K. pneumoniae strains was performed as previously described (9). Briefly, an inoculum of 2.5 × 104 CFU bacteria was mixed with human sera from healthy volunteers at a 1:3 volume ratio. The mixture was incubated at 37°C for 3 h. After serial dilution and plating, the numbers of CFU were determined.

Nucleotide sequence accession numbers.

The nucleotide sequences of the cps regions in A1142 (K57) and A1517 were deposited in the GenBank database under accession numbers AB334776 and AB334777, respectively.


Capsular serotypes of A1142, A7754, and A1517.

Initially, the cps genotypes of the clinical isolates A1142, A7754, and A1517 were PCR screened with K1- or K2-specific primers (6, 7, 10, 11), and capsular serotyping was also performed using the Klebsiella antiserum Seiken, which has K1 to K6 serotype-specific antisera. However, none of these strains revealed positive results (data not shown).

cps sequencing.

Since these strains did not belong to capsular serotype K1 or K2, both of which are more prevalent among patients with liver abscess, we further compared the cps sequences which are conserved between different serotypes of K. pneumoniae and designed the conserved primers for PCR amplification of the cps regions in these non-K1, non-K2 strains.

Five primers in the 5′ cps conserved region and five primers in the 3′ cps conserved region were designed for PCR amplification (Fig. (Fig.1).1). Sharp banding of 10- to 20-kb DNA fragments in the A1142 and A7754 strains was obtained from primers CPS-F and CPS-R2, while that in A1517 was obtained from wzi-2 and gnd-R. After cloning and in vitro transposition, DNA sequencing was performed. Since A1142 and A7754 have similar sequences according to the results of DNA sequencing (2,451 bp upstream of gnd were read, and only 5 nucleotides were different), it was suggested that these strains belonged to the same cps genotype. For this reason, DNA sequencing was continued for A1142 and A1517. Finally, the 5′ and 3′ ends of the cps region were further amplified by PCR using the primers shown in Fig. Fig.11 and then sequenced. The open reading frames (ORFs) of the complete cps regions in A1142 and A1517 and their predicted functions, by comparison with the sequences of NTUH-K2044 (K1), MGH78578 (K52), and Chedid (K2) (Fig. (Fig.22 and Tables Tables33 and and4),4), suggested that this region was responsible for capsular polysaccharide synthesis.

FIG. 2.
Comparison of capsular polysaccharide synthesis (cps) regions between A1142, A1517, NTUH-K2044 (K1), MGH78578 (K52), and Chedid (K2). ORFs are shown by arrows. Black arrows indicate the ORFs conserved in these serotypes, and white arrows refer to the ...
Annotation of ORFs of the capsular polysaccharide synthesis (cps) region in A1142 (K57)a
Annotation of ORFs of the capsular polysaccharide synthesis (cps) region in A1517a

cps PCR genotyping.

Based on our sequencing results for serotype-specific regions, we designed primers for cps PCR genotyping. PCR was performed using 77 serotype K reference strains obtained from the Statens Serum Institute with A1142- or A1517-specific primers (Fig. 3A and B). Only reference strain K57 yielded a positive PCR result with primers 9471F and 9897R. The same result was obtained by another primer pair, 1142XF and 1142XR (Fig. (Fig.3A3A and data not shown). The data indicated that A1142 and A7754 belonged to cps genotype K57 and that the primers were specific for K57 cps PCR genotyping. Further PCR screening was performed with primers 9471F and 9897R among 21 NTUH nonblood isolates, 13 ECKH strains, 34 FEMH strains, 24 ATCC strains, 80 Canada strains, and 1 Finland strain. Only 2 of the 13 ECKH strains (designated E7 and E12) and the Finland strain have positive results (data not shown). Six pairs of specific primers were used in A1517 cps PCR genotyping (Fig. (Fig.3B).3B). Unexpectedly, none of these reference strains displayed positive results (data not shown), implicating A1517 as unique among the 77 serotypes (25). Further PCR screening was performed with primers 12R STAR and 12 STAR, among other isolates described above, and only one of the Canada strains (designated Canada 05-14) showed a positive result (data not shown), which suggested that they belonged to same cps genotype.

Serotyping of K57 cps genotype strains.

A double immunodiffusion assay was performed to confirm the cps PCR genotyping results for A1142. However, compared to what was found for the K57 reference strain, different dilutions of A1142 capsular extracts formed very weak precipitation lines with anti-K57 antiserum (Fig. (Fig.4),4), leading us to utilize an immunoblot strategy for serotyping. Immunoblot serotyping of the K57 reference strain, A1142, and A7754 revealed strong and clearly positive results (Fig. (Fig.5A,5A, ,1a1a to to3a),3a), compared with those for the negative controls NTUH-K2044 (K1) and MGH78578 (K52) (Fig. (Fig.5A,5A, ,4a4a and and5a).5a). Immunoblot serotyping was also performed by use of other PCR-positive strains, the E7, E12, and Finland strains, and four of the PCR-negative strains (randomly selected), ATCC 35597, 0708 (NTUH nonblood isolate), E13 (ECKH strain), and YD20 (FEMH strain). Similar results were obtained, as the E7, E12, and Finland strains showed strong positive signals compared to the PCR-negative strains (Fig. (Fig.5B5B).

FIG. 4.
Double immunodiffusion of Klebsiella pneumoniae serotype K57. Anti-K57 antiserum was added in the center well, while capsular extracts from overnight-cultured K. pneumoniae strains were loaded in the peripheral wells. Well 1, capsular extract of the K57 ...
FIG. 5.
Immunoblot serotyping of Klebsiella pneumoniae serotype K57. Anti-K57 antiserum was used as the first antibody and goat anti-rabbit IgG-HRP as the second antibody. (A) 1a, capsular extract of the K57 reference strain; 2a, A1142; 3a, A7754; 4a, NTUH-K2044 ...

Characterization of the K57 determinant gene of the cps region in A1142.

An unmarked deletion mutant of the wzy gene located in the cps variable region of A1142 (K57) was generated in this study. Presently, the wzy mutant of A1142 lost mucoviscosity with a string test (9), implying that capsule synthesis was attenuated. This mutant strain also showed negative results for anti-K57 antiserum by immunoblot serotyping; in complementation, experiments involving transformation by a wzy-carrying pGEM-T Easy-Km vector restored K57 positivity (Fig. (Fig.5A,5A, ,1b1b to to3b).3b). Therefore, wzy was essential for capsular serotype K57, and we also confirmed that this gene cluster was a K57 capsular synthesis region.

Serotyping of A1517.

The initial serological assay of A1517 with 77 antisera from Statens Serum Institute showed only a preliminary reaction to capsular serotype 79. However, confirmation by double immunodiffusion and immunoblot analysis was negative with anti-K79 antiserum (Statens Serum Institute) (data not shown). Moreover, antiserum against A1517 failed to react with 77 reference strains in immunoblot analysis (only K25, K53, K55, K56, and K58 had weak reactions with anti-A1517 antiserum) (Fig. (Fig.6A,6A, ,4d;4d; B, 5d; and C, 1a, 2a, and 4a, respectively), and Canada 05-14, which was considered to be of the same cps genotype as A1517 by cps PCR genotyping, showed a strong positive reaction with anti-A1517 antiserum (Fig. (Fig.6C,6C, 8d). Therefore, both the cps genotype and the capsular serotype observation suggested that A1517 was probably a new serotype, not belonging to any of the 77 documented serotypes.

FIG. 6.
Immunoblot serotyping of Klebsiella pneumoniae A1517. Anti-A1517 antiserum was used as the first antibody and goat anti-rabbit IgG-HRP as the second antibody. (A) 1a to 7a, K1 to K7; 1b to 7b, K8 to K14; 1c to 7c, K15 to K21; 1d to 6d, K22 to K27; 7d, ...

cps PCR-RFLP analysis.

The A1517 and Canada 05-14 strains were amplified with primers CPS-1 and rCPS. The patterns of HincII-digested fragments of these strains were undistinguishable (data not shown). Moreover, the cps PCR-RFLP pattern of A1517 was very distinct from those of the reference strains of K25, K53, K55, K56, and K58, which had weak reactions with anti-A1517 antiserum (4). Also, sequences of the cps variable region of A1517 were very similar to those of Canada 05-14 (only 1-nucleotide difference in 2,479 bp upstream of gnd) but very different from those of K25, K53, K55, K56, K58, and K79 (<10% nucleotide similarity in 1.8 kb upstream of the gnd sequences), implying that A1517 belonged to a new cps genotype, not among the 77 reference strains.

Because we failed to amplify the cps region of the reference K57 strain with primers CPS-1 and rCPS, amplified products from the reference K57 strain with primers wzi2 and rCPS2 were used. Together with those of A1142 and A7754, their cps PCR-RFLP patterns were compared. The pattern of HincII-digested fragments of K57 was quite different from that of A1142 or A7754, while those of A1142 and A7754 were similar to each other (Fig. (Fig.7).7). Nevertheless, sequencing results showed that the cps variable region of the reference K57 strain was very similar to that of A1142 (4,323 bp from wbaP to ORF10′ were read, and only 44 nucleotides were different) and represented the same gene product.

FIG. 7.
cps PCR-RFLP analysis of the A1142, A7754, and K57 reference strains. cps regions were amplified with primers wzi2 and rCPS2. Strain names are indicated above the lanes. Lane M, marker (in base pairs).

Serum sensitivity of A1142, A7754, and A1517.

After incubation with serum for 3 h, the CFU counts in A1142, A7754, and A1517 decreased to about 1%, 20%, and 4%, respectively, of those in the initial inoculum. Therefore, the three strains were all serum sensitive.


A previous survey from Australia on the serotypes of 293 K. pneumoniae isolates (13) reported that 88 isolates (30%) were nontypeable by countercurrent immunoelectrophoresis (27), while 54 had a positive reaction for more than one serotype. Likewise, we had similar difficulties in serotyping using double immunodiffusion (26), as this study revealed. We initially confirmed the cps PCR genotyping results of the A1142 (K57) strain by double immunodiffusion. However, an ambiguous precipitation line was formed by our clinical isolate in contrast to that for the K57 reference strain. This may have been due to some differences in the presented antigens recognized by the specific antiserum between the reference and test strains. Compared with double immunodiffusion and countercurrent immunoelectrophoresis, which are usually used for identifying the serotypes of K. pneumoniae, the modified immunoblot serotyping method presently employed showed an increased sensitivity and reduced the consumption of antiserum. Nevertheless, this method did not increase specificity. The result of immunoblot analysis with anti-A1517 antiserum also showed several cross-reactions. However, DNA sequences were different among A1517 and these known serotypes. These observations were consistent with our cps PCR genotyping result. Therefore, cps PCR genotyping seems to be a more sensitive and specific way for detecting this serotype.

A recent study has reported a molecular serotyping method, cps PCR-RFLP analysis, which has a higher discriminatory power than classical serotyping (4). Notably, cps PCR-RFLP pattern variations were found among strains with the same K serotype (4). This phenomenon was also observed in our study. The cps PCR-RFLP pattern of reference strain K57 was quite different from those of our clinical isolates A1142 and A7754. Nevertheless, sequencing results revealed that their sequences in the cps variable region were very similar and carried the same gene products. Therefore, strains belonging to the same capsular serotype but having different cps PCR-RFLP patterns may be differentiated by this strategy; however, this may also make the interpretation of cps genotyping more complicated. Thus, PCR-based cps genotyping may provide a simple way to detect capsular type, at least for this serotype, in clinical use.

Determination of the genetic composition of the K57 cps region has identified the specific genes for capsular serotype K57 by comparison with the sequences of NTUH-K2044 (K1), MGH78578 (K52), and Chedid (K2) (2, 6, 30). galF, ORF2, wzi (orfX), wza, wzb, wzc, and gnd were conserved genes among the different serotypes, even in different species, while wbaP and wbaZ were relatively specific in some serotypes (10). The ORFs from wzx to ORF14′, except ORF13′ (transposase) in A1142 (K57), were variable regions specific for serotype K57.

In the conserved region, galF, ORF2, and gnd are considered to be involved in carbohydrate metabolism (36); wzi (orfX), wza, wzb, and wzc are considered to be responsible for the translocation and surface assembly of the capsule (30, 36).

The structure of the capsular polysaccharide from Klebsiella serotype K57 has been resolved as

equation M1

This (14) is agreement with the predicted functions of some genes in the cps region of K57. For example, one of the genes within the cluster shares high homology (78% identity) with wbaP, one of a family of undecaprenolphosphate hexose-1-P transferase enzymes that initiate the formation of a variety of cell surface polysaccharides in both gram-negative and gram-positive bacteria. These polymers all contain glucose or galactose in their repeat unit structures (36). Also evident is a homologue (71% identity) of wbaZ encoding mannosyltransferase that may be involved in a mannose-containing linkage region in the repeat unit of Klebsiella K57. The wzy gene family encodes an O-polysaccharide polymerase that recognizes and extends the O-antigen polysaccharide-repeating units (31). This protein was also postulated as a polymerase responsible for lipid-linked repeat unit polymerization in the capsular synthesis process (36). In this study, the deletion mutant of the wzy gene in A1142 decreased mucoviscosity and turned negative for anti-K57 antiserum, suggesting that this gene was essential for capsular synthesis of serotype K57. Therefore, the predicted functions of the cps region and the genetic result imply that this gene cluster was a K57 capsular synthesis region.

In addition to the conserved genes, galF, ORF2, wzi (orfX), wza, wzb, wzc, and gnd, wbaP was also present in the A1517 cps region. The rest of the genes, ORFs 8″ to 15″, seemed to be unique for this cps genotype, encoding three hypothetical proteins with unknown functions, one putative polysaccharide export protein, and four putative glycosyltransferases (one was predicted as rhamnosyltransferase, implying that there exists rhamnose in the structure) that are probably responsible for the unique repeat unit synthesis.

Patients with pyogenic liver abscess caused by non-K1 K. pneumoniae are considered to be associated with diabetes mellitus, compared to those with infections caused by serotype K1 (10). Presently, we observed that strains A1142, A7754, and A1517 display serum sensitivity by a serum resistance assay (9). For this reason, they were considered less virulent than serotype K1 and K2, which are resistant to serum killing. Moreover, strains A7754 and A1517 were obtained from patients whose diabetes mellitus was poorly controlled (HbA1C > 9%), implicating host factors as more predominant than bacterial virulence in invasive infection by these K. pneumoniae strains.

The current study provided a rapid and accurate PCR-based cps genotyping method for detecting cps genotype K57 and a new cps genotype, A1517. The variable regions seem to be unique for each serotype, including K1, K2, K5, K20, and K54 (10, 32, 37, 39), and PCR-based cps genotyping has been widely used since the DNA sequences of the cps region have been available (7, 11, 15, 22, 32). Therefore, if the cps sequences of the remaining reported serotypes can also be resolved, primers specific for their variable regions can be used in cps PCR genotyping. This may provide a new way to “genotype” an unknown capsular type strain instead of serotyping but will require validation with several strains of each serotype to determine specificity and sensitivity.


We thank Po-Ren Hsueh (NTUH, Taipei, Taiwan), Shih-Si Wang (ECKH, Sansia, Taiwan), Jann-Tay Wang (FEMH, Banciao, Taiwan), and Ann-Mari Rissanen (Department of Clinical Microbiology, Kuopio University Hospital, Finland) for providing K. pneumoniae strains and Yi-Li Liu and I-Ching Huang for technical support on DNA sequencing (supported in part by the Department of Medical Research in NTUH).

This study was supported by grants from the National Science Council and the Liver Disease Prevention and Treatment Research Foundation in Taiwan.


Published ahead of print on 28 May 2008.


1. Abbot, S. L. 2003. Klebsiella, Enterobacter, Citrobacter, Serratia, Plesiomonas, and other Enterobacteriaceae, p. 684-700. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. A. Pfaller, and R. H. Yolken (ed.), Manual of clinical microbiology, 8th ed. ASM Press, Washington, DC.
2. Arakawa, Y., R. Wacharotayankun, T. Nagatsuka, H. Ito, N. Kato, and M. Ohta. 1995. Genomic organization of the Klebsiella pneumoniae cps region responsible for serotype K2 capsular polysaccharide synthesis in the virulent strain Chedid. J. Bacteriol. 1771788-1796. [PMC free article] [PubMed]
3. Berleman, J. E., B. M. Hasselbring, and C. E. Bauer. 2004. Hypercyst mutants in Rhodospirillum centenum identify regulatory loci involved in cyst cell differentiation. J. Bacteriol. 1865834-5841. [PMC free article] [PubMed]
4. Brisse, S., S. Issenhuth-Jeanjean, and P. A. D. Grimont. 2004. Molecular serotyping of Klebsiella species isolates by restriction of the amplified capsular antigen gene cluster. J. Clin. Microbiol. 423388-3398. [PMC free article] [PubMed]
5. Cheng, D. L., Y. C. Liu, M. Y. Yen, C. Y. Liu, and R. S. Wang. 1991. Septic metastatic lesions of pyogenic liver abscess. Their association with Klebsiella pneumoniae bacteremia in diabetic patients. Arch. Intern. Med. 1511557-1559. [PubMed]
6. Chuang, Y. P., C. T. Fang, S. Y. Lai, S. C. Chang, and J. T. Wang. 2006. Genetic determinants of capsular serotype K1 of Klebsiella pneumoniae causing primary pyogenic liver abscess. J. Infect. Dis. 193645-654. [PubMed]
7. Chung, D. R., S. S. Lee, H. R. Lee, H. B. Kim, H. J. Choi, J. S. Eom, J. S. Kim, Y. H. Choi, J. S. Lee, M. H. Chung, Y. S. Kim, H. Lee, M. S. Lee, and C. K. Park. 2007. Emerging invasive liver abscess caused by K1 serotype Klebsiella pneumoniae in Korea. J. Infect. 54578-583. [PubMed]
8. Cortés, G., N. Borrell, B. de Astorza, C. Gómez, J. Sauleda, and S. Albertí. 2002. Molecular analysis of the contribution of the capsular polysaccharide and the lipopolysaccharide O side chain to the virulence of Klebsiella pneumoniae in a murine model of pneumonia. Infect. Immun. 702583-2590. [PMC free article] [PubMed]
9. Fang, C. T., Y. P. Chuang, C. T. Shun, S. C. Chang, and J. T. Wang. 2004. A novel virulence gene in Klebsiella pneumoniae strains causing primary liver abscess and septic metastatic complications. J. Exp. Med. 199697-705. [PMC free article] [PubMed]
10. Fang, C. T., S. Y. Lai, W. C. Yi, P. R. Hsueh, K. L. Liu, and S. C. Chang. 2007. Klebsiella pneumoniae genotype K1: an emerging pathogen that causes septic ocular or central nervous system complications from pyogenic liver abscess. Clin. Infect. Dis. 45284-293. [PubMed]
11. Fang, F. C., N. Sandler, and S. J. Libby. 2005. Liver abscess caused by magA+ Klebsiella pneumoniae in North America. J. Clin. Microbiol. 43991-992. [PMC free article] [PubMed]
12. Fung, C. P., F. Y. Chang, S. C. Lee, B. S. Hu, B. I. Kuo, C. Y. Liu, M. Ho, and L. K. Siu. 2002. A global emerging disease of Klebsiella pneumoniae liver abscess: is serotype K1 an important factor for complicated endophthalmitis? Gut 50420-424. [PMC free article] [PubMed]
13. Jenney, A. W., A. Clements, J. L. Farn, O. L. Wijburg, A. McGlinchey, D. W. Spelman, T. L. Pitt, M. E. Kaufmann, L. Liolios, M. B. Moloney, S. L. Wesselingh, and R. A. Strugnell. 2006. Seroepidemiology of Klebsiella pneumoniae in an Australian tertiary hospital and its implications for vaccine development. J. Clin. Microbiol. 44102-107. [PMC free article] [PubMed]
14. Kamerling, J. P., B. Lindberg, J. Lonngren, and W. Nimmich. 1975. Structural studies of the Klebsiella type 57 capsular polysaccharide. Acta Chem. Scand. B 29593-598. [PubMed]
15. Keynan, Y., J. A. Karlowsky, T., Walus, and E. Rubinsten. 2007. Pyogenic liver abscess caused by hypermucoviscous Klebsiella pneumoniae. Scand. J. Infect. Dis. 39828-830. [PubMed]
16. Ko, W. C., D. L. Paterson, A. J. Sagnimeni, D. S. Hansen, A. Von Gottberg, S. Mohapatra, J. M. Casellas, H. Goossens, L. Mulazimoglu, G. Trenholme, K. P. Klugman, J. G. McCormack, and V. L. Yu. 2002. Community-acquired Klebsiella pneumoniae bacteremia: global differences in clinical patterns. Emerg. Infect. Dis. 8160-166. [PMC free article] [PubMed]
17. Reference deleted.
18. Reference deleted.
19. Link, A. J., D. Phillips, and G. M. Church. 1997. Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: application to open reading frame characterization. J. Bacteriol. 1796228-6237. [PMC free article] [PubMed]
20. Ma, L. C., C. T. Fang, C. Z. Lee, C. T. Shun, and J. T. Wang. 2005. Genomic heterogeneity in Klebsiella pneumoniae strains is associated with primary pyogenic liver abscess and metastatic infection. J. Infect. Dis. 192117-128. [PubMed]
21. Mizuta, K., M. Ohta, M. Mori, T. Hasegawa, I. Nakashima, and N. Kato. 1983. Virulence for mice of Klebsiella strains belonging to the O1 group: relationship to their capsular (K) types. Infect. Immun. 4056-61. [PMC free article] [PubMed]
22. Nadasy, K. A., R. Domiati-Saad, and M. A. Tribble. 2007. Invasive Klebsiella pneumoniae syndrome in North America. Clin. Infect. Dis. 45e25-e28. [PubMed]
23. Ohmori, S., K. Shiraki, K. Ito, H. Inoue, T. Ito, T. Sakai, K. Takase, and T. Nakano. 2002. Septic endophthalmitis and meningitis associated with Klebsiella pneumoniae liver abscess. Hepatol. Res. 22307-312. [PubMed]
24. Okano, H., K. Shiraki, H. Inoue, T. Kawakita, N. Yamamoto, M. Deguchi, K. Sugimoto, T. Sakai, S. Ohmori, K. Murata, and T. Nakano. 2002. Clinicopathological analysis of liver abscess in Japan. Int. J. Mol. Med. 10627-630. [PubMed]
25. Ørskov, I., and M. A. Fife-Asbury. 1977. New Klebsiella capsular antigen K82 and the deletion of five of those previously assigned. Int. J. Syst. Bacteriol. 27386-387.
26. Ouchterlony, O. 1958. Diffusion-in-gel methods for immunological analysis. Prog. Allergy 51-78. [PubMed]
27. Palfreyman, J. M. 1978. Klebsiella serotyping by counter-current immunoelectrophoresis. J. Hyg. 81219-225. [PMC free article] [PubMed]
28. Podschun, R., and U. Ullmann. 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin. Microbiol. Rev. 11589-603. [PMC free article] [PubMed]
29. Rahimian, J., T. Wilson, V. Oram, and R. S. Holzman. 2004. Pyogenic liver abscess: recent trends in etiology and mortality. Clin. Infect. Dis. 391654-1659. [PubMed]
30. Rahn, A., J. Drummelsmith, and C. Whitfield. 1999. Conserved organization in the cps gene clusters for expression of Escherichia coli group 1 K antigens: relationship to the colanic acid biosynthesis locus and the cps genes from Klebsiella pneumoniae. J. Bacteriol. 1812307-2313. [PMC free article] [PubMed]
31. Schild, S., A. K. Lamprecht, and J. Reidl. 2005. Molecular and functional characterization of O antigen transfer in Vibrio cholerae. J. Biol. Chem. 28025936-25947. [PubMed]
32. Struve, C., M. Bojer, E. M. Nielsen, D. S. Hansen, and K. A. Krogfelt. 2005. Investigation of the putative virulence gene magA in a worldwide collection of 495 Klebsiella isolates: magA is restricted to the gene cluster of Klebsiella pneumoniae capsule serotype K1. J. Med. Microbiol. 541111-1113. [PubMed]
33. Taylor, L. A., and R. E. Rose. 1988. A correction in the nucleotide sequence of the Tn903 kanamycin resistance determinant in pUC4K. Nucleic Acids Res. 16358. [PMC free article] [PubMed]
34. Wang, J. H., Y. C. Liu, S. S. Lee, M. Y. Yen, Y. S. Chen, J. H. Wang, S. R. Wann, and H. H. Lin. 1998. Primary liver abscess due to Klebsiella pneumoniae in Taiwan. Clin. Infect. Dis. 261434-1438. [PubMed]
35. Ware, J., L. Moran, J. Foster, J. Posfai, T. Vincze, D. Guiliano, M. Blaxter, J. Eisen, and B. Slatko. 2002. Sequencing and analysis of a 63 kb bacterial artificial chromosome insert from the Wolbachia endosymbiont of the human filarial parasite Brugia malayi. Int. J. Parasitol. 32159-166. [PubMed]
36. Whitfield, C., and I. S. Roberts. 1999. Structure, assembly and regulation of expression of capsules in Escherichia coli. Mol. Microbiol. 311307-1319. [PubMed]
37. Yeh, K. M., F. Y. Chang, C. P. Fung, J. C. Lin, and L. K. Siu. 2006. magA is not a specific virulence gene for Klebsiella pneumoniae strains causing liver abscess but is part of the capsular polysaccharide gene cluster of K. pneumoniae serotype K1. J. Med. Microbiol. 55803-804. [PubMed]
38. Reference deleted.
39. Yu, W. L., C. P. Fung, W. C. Ko, K. C. Cheng, C. C. Lee, and Y. C. Chuang. 2007. Polymerase chain reaction analysis for detecting capsule serotypes K1 and K2 of Klebsiella pneumoniae causing abscesses of the liver and other sites. J. Infect. Dis. 1951235-1236. [PubMed]

Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • Nucleotide
    Primary database (GenBank) nucleotide records reported in the current articles as well as Reference Sequences (RefSeqs) that include the articles as references.
  • Protein
    Protein translation features of primary database (GenBank) nucleotide records reported in the current articles as well as Reference Sequences (RefSeqs) that include the articles as references.
  • PubMed
    PubMed citations for these articles
  • Taxonomy
    Taxonomy records associated with the current articles through taxonomic information on related molecular database records (Nucleotide, Protein, Gene, SNP, Structure).
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...