Previous Article | Next Article ![]()
Journal of Clinical Microbiology, September 2006, p. 3225-3230, Vol. 44, No. 9
0095-1137/06/$08.00+0 doi:10.1128/JCM.00524-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Division of Infectious Disease Control, Norwegian Institute of Public Health, P.O. Box 4404 Nydalen, NO-0403 Oslo,1 Department of Oral Biology, University of Oslo, Oslo, Norway2
Received 10 March 2006/ Returned for modification 12 May 2006/ Accepted 5 July 2006
|
|
|---|
|
|
|---|
The major antibacterial action of penicillin is derived from its ability to bind to and inactivate penicillin-binding proteins (PBPs). PBPs are membrane-associated serine peptidases that catalyze the polymerization and transpeptidation of glycan strands during the final steps of peptidoglycan biosynthesis. Penicillin resistance in S. pneumoniae is caused by alterations in the PBPs that lead to organisms with reduced affinities for penicillin and other ß-lactam antibiotics (9). S. pneumoniae contains six different PBPs; and changes in types 1A, 2B, and 2X have been reported to be the most important in the development of penicillin resistance (9). The pbp genes have a mosaic structure, and alterations are caused by the uptake and incorporation of DNA fragments from closely related species by genetic transformation (9, 19).
Norway is still in a favorable position regarding resistance to antimicrobial drugs, with less than 3% of pneumococcal strains causing systemic infections showing resistance to penicillin (24). Consequently, penicillin still remains the drug of choice for the treatment of pneumococcal infections.
In this study, we analyzed non-penicillin-susceptible strains isolated in Norway in the period from 1995 to 2001 to determine the nature of penicillin resistance in our pneumococcal population. The strains were serotyped and genotyped by multilocus sequence typing (MLST); and the genetic heterogeneity of their PBPs was analyzed by restriction fragment length polymorphism analysis of their pbp1a, pbp2b, and pbp2x genes.
|
|
|---|
Antimicrobial susceptibility. The drug susceptibility patterns of the strains were initially tested by the disc diffusion method for penicillin G (by applying 1-µg oxacillin depots), lincomycin, erythromycin, tetracycline, and trimethoprim-sulfamethoxazole (Neo-Sensitabs; Rosco Diagnostics, Taastrup, Denmark). Isolates showing reduced zone sizes for any one of these antibiotics were further tested by the Etest method (AB Biodisk, Solna, Sweden) to determine the MICs of penicillin, erythromycin, chloramphenicol, doxycycline, cefotaxime, and clindamycin. For both methods, the Biodisk PDM medium with 5% horse blood was used and the plates were incubated in 5% CO2. For control purposes we used S. pneumoniae TIGR4 (NIPH 7/87), ATCC 49619, and NIPH 76/00; these strains are wild-type, intermediately penicillin-sensitive, and penicillin-resistant isolates, respectively (24). Strains were defined as antibiotic susceptible, intermediately resistant, or resistant, in accordance with the Clinical and Laboratory Standards Institute guidelines (6).
When several isolates were obtained from a patient during the same disease episode, only one of them was included in the analysis. A total of 125 isolates showed reduced susceptibility to penicillin (MICs > 0.064 µg/ml); 50 of these isolates were from patients with systemic disease and 75 were from patients with localized infections. All these strains were included in the study.
Serotyping. Serotyping was performed by the Quellung reaction with antisera produced by the Statens Seruminstitut, Copenhagen, Denmark. All strains were serogrouped and serotyped by using factor-specific sera to identify types within groups.
MLST. The strains were analyzed by MLST. The sequences of the seven housekeeping loci included in the pneumococcal MLST scheme were determined for each isolate, as described elsewhere (12). The sequence types (STs) were obtained by using the MLST website (http://www.mlst.net). New alleles and new STs were submitted to the curator of the database and assigned designations.
To view the genetic relationships among the different strains, a dendrogram was generated from the distance matrix between STs by using the unweighted pair group method with arithmetic averages. The relationship of our strains to the 26 international clones was analyzed with the clustering algorithm BURST by using the programs on the website http://pubmlst.org/.
Analysis of PBPs. PCR fragments of the pbp1a, pbp2b, and pbp2x genes were amplified by using the following primers: Pbp 1A-F (dCGG CAT TCG ATT TGA TTC GCT TCT), Pbp 1A-R (dTCG TAC TAT TAT TTG TGC TTG GAG T) (11), Pbp 2B-F (dGAT CCT CTA AAT GAT TCT CAG GTG G), Pbp 2B-R (dCAA TTA GCT TAG CAA TAG GTG TTG G) (10), Pbp 2x-F (dTGC CAA TTC ACA CGA TTT GC), and Pbp 2X-R (dTCA CAA TTC CAG CAC TGA TG) (13). The PCRs were run for 30 cycles with an annealing temperature of 54°C for 2 min. The PCR products were cleaned by using QIAquick purification columns (QIAGEN, Oslo, Norway), and the products were eluted in 30 µl distilled water.
The purified PCR products were digested separately with 10 U HinfI (Invitrogen, Carlsbad, CA), according to the manufacturer's instructions, as described by Coffey et al. (7). The digested products were separated on a 1.8% (wt/vol) agarose gel at 100 V for 4 h. DNA was visualized following ethidium bromide staining. The restriction fragment length patterns for each gene were grouped by visual judgment. Gels were also scanned, and the different patterns were compared by using BioNumerics software (Applied Maths BVBA, Sint-Martens-Latem, Belgium) by using the Dice coefficient of similarity and cluster analysis with the unweighted pair group method with arithmetic averages. The position tolerance and the optimization were set equal to 0.6 and 1.0, respectively.
|
|
|---|
Of the 125 strains, 77 were intermediately resistant (MICs
0.064 to 1 µg/ml) and 48 were fully resistant to penicillin (MICs
2 µg/ml). Resistant strains were recovered from patients in all age groups (Table 1). Isolates from children between 0 and 4 years old represented nearly 25% of the resistant isolates, and this age group was especially overrepresented among the patients with nonsystemic cases of infection due to resistant isolates (35.7%). Most of the strains resistant to penicillin showed additional resistance to one or several other drugs, especially erythromycin, doxycycline, clindamycin, and, to a lesser extent, chloramphenicol (Table 2).
|
View this table: [in a new window] |
TABLE 1. Non-penicillin-susceptible S. pneumoniae isolates in relation to age of the patients, Norway, 1995 to 2001
|
|
View this table: [in a new window] |
TABLE 2. Numbers of S. pneumoniae isolates resistant or intermediately resistant to penicillin and simultaneously resistant to other antibiotics, Norway, 1995 to 2001
|
All but one patient had only one disease episode. The exceptional patient had five blood cultures positive for non-penicillin-susceptible S. pneumoniae in the period from April 1999 to November 2000. The five isolates were included in the analysis and were serotypes 23F (two isolates), 19F, 21, and 9V and ST-277 (two isolates), ST-336, ST-375, and ST-156, respectively. This patient, a refugee from Kosovo, was a 30-year-old man with myelomatosis, a type of cancer that begins in plasma cells.
Serotypes. Seven serotypes (serotypes 14, 9V, 23F, 19F, 6B, 19A, and 3) were present among the fully resistant strains, while one strain was nontypeable. The intermediately resistant strains belonged to 20 different serotypes, and four strains were nontypeable. The most frequent serotypes among the intermediately resistant strains were 19F, 9V, 14, 23F, and 6B (Table 3).
|
View this table: [in a new window] |
TABLE 3. Serotypes among penicillin-resistant and intermediately penicillin resistant isolates of S. pneumoniae, Norway, 1995 to 2001
|
![]() View larger version (33K): [in a new window] |
FIG. 1. Genetic relationships among the 74 STs of 119 non-penicillin-susceptible S. pneumoniae isolates from Norway, 1995 to 2001. The dendrogram was generated from the distance matrix between STs by using the unweighted pair group method with arithmetic averages. The serotypes of the strains, the number of isolates, and the STs isolated from patients with systemic disease are indicated. S, isolated from patients with systemic disease; S/N, isolated from both patients with systemic and patients with nonsystemic disease.
|
BURST analysis of the 74 STs, together with the 26 international clones, identified 11 groups of STs and 33 singletons. Nine of the groups were related to 1 of the 26 international clones: (i) 1 serotype 14 isolate was an SLV of the England14-9 clone; (ii) 30 serotype 14 or 9V isolates were related to the Spain9V-3 clone; (iii) 5 isolates represented the DCC-Pn1476 clone, an SLV (ST-374), or a double-locus variant (DLV) of ST-374 (ST-1545); (iv) ST-66 was an SLV of the Tenessee14-18 clone; (v) ST-340 was an SLV of the Hungary19A-6 clone (vi) in addition to the 7 strains of the Spain23F-1 clone, a novel ST, ST-326, an SLV of this clone that also expressed serotype 23, was represented by 1 strain; (vii) 11 serotype 6B or serotype 3 isolates were related to the Spain6B-2 clone (ST-90); (viii) the Taiwan19F-14 clone (ST-236) was represented by 1 strain, and an SLV of it, ST-271, a known multiresistant clone from Korea, was represented by 2 strains; (ix) finally, the Colombia23F-26 clone (ST-338) was represented by 2 strains, and 6 additional isolates represented SLVs or DLVs of this clone. Thus, in our strain collection, 37 isolates belonged to 1 of 6 of the international clones, and a total of 74 isolates (62%) were SLVs or DLVs of the 26 international clones.
PBPs. To study the diversity of PBPs 1A, 2B, and 2X of the Norwegian non-penicillin-susceptible strains of S. pneumoniae, the PCR products of their respective genes from 79 of the isolates were digested with the enzyme HinfI. PCR fragments showed 12, 12, and 19 different patterns for pbp1a, pbp2b, and pbp2x, respectively. A total of 43 different pbp profiles were distinguished among the 79 isolates by the combination of the restriction patterns of the three pbp genes. Seven profiles were identified more than once, and these were designated by roman numerals I to VII (Table 4). The largest group (profile I) comprised 27 strains of nine different STs belonging to serotypes 9V, 23F, and 14; and the penicillin MICs ranged from 1 to 4 µg/ml. Profile II included six serotype 6B strains and three strains of three genetically closely related STs. Profile III contained one serotype 14 strain and one nontypeable strain; these two strains had different STs. Profile IV included two strains of ST-230 and of serotypes 17F and 19A. Profile V included two strains of ST-277, serotype 23F; profile VI included two serotype 6B strains of ST-90, and profile VII contained two strains with two different STs (ST-668 and ST-1545) and were serotypes 19A and 3, respectively.
|
View this table: [in a new window] |
TABLE 4. pbp profiles represented by multiple isolates of non-penicillin-susceptible S. pneumoniae, Norway, 1995 to 2001
|
|
View this table: [in a new window] |
TABLE 5. pbp profiles of STs of non-penicillin-susceptible S. pneumoniae represented by multiple isolates, Norway, 1995 to 2001
|
|
|
|---|
To get a wider perspective on the non-penicillin-susceptible pneumococci present in Norway, a further 75 nonsusceptible isolates from patients with localized infections were included in the analysis. The fully resistant strains were recovered from patients of all age groups but were more often found in children 0 to 4 years of age. This is in accordance with the findings of studies that show that children, especially those attending day care centers, are at increased risk of carrying resistant strains (15, 25, 30). There was no difference in the proportion of cases of pneumococcal disease caused by resistant strains by gender. This is in accordance with the data on the occurrence of pneumococcal disease in general in Norway (24).
One patient presented with five blood cultures positive for non-penicillin-susceptible S. pneumoniae within a 20-month period. These five isolates showed four different STs and four different serotypes. This indicates that, despite the general level of low resistance, there are niches in our society as well where the prevalence of resistant strains is high. This particular patient was a refugee. In the beginning of their stay in our country, refugees live in centers where resistant strains are overrepresented due to continual import from various countries, and crowded conditions facilitate spread between individuals.
Many of the strains resistant to penicillin showed additional resistance to erythromycin, clindamycin, doxycycline, and, sometimes, chloramphenicol. The reasons why pneumococci develop simultaneous resistance to several antimicrobial classes are not quite clear, but some resistance determinants have been shown to be transferred together on the same transposon (1). Thus, the emergence of penicillin resistance has been associated with the emergence of multidrug resistance. Most of the known internationally resistant clones are multidrug resistant (21).
Serotypes 6B, 9V,14, 19A, 19F, and 23F have been associated with a high burden of resistance (26). Of the Norwegian isolates with reduced susceptibilities, serotypes 9V, 14, 19F, 23F, and 6B accounted for 73.6% of the cases. These serotypes are known to have the ability to colonize the upper respiratory tracts of healthy carriers and persist there for a long time (18, 28, 29), making them prone to the development of antibiotic resistance selection and gene transfer (22). The seven-valent pneumococcal vaccine could cover 64% of the invasive cases caused by nonsusceptible strains in Norway.
MLST analyses showed that the non-penicillin-susceptible strains in Norway were a very heterogeneous group with a high degree of genetic diversity; 42% of the strains had genotypes not previously identified elsewhere. However, several of the internationally known resistant clones were also identified, and 62% of the isolates were related to 1 of the 26 international clones. International resistant clones are an important cause of the increase of antibiotic resistance among pneumococci. In several countries most of the resistance observed is caused by a relatively few clones circulating in the community (3, 4, 16, 20, 27). The international Spain9V-3 clone, ST-156, was the most frequent resistant clone present in Norway, but it appeared only in solitary cases; only one to four cases occurred each year, and these were in different geographic regions. The Spain23F-1 clone, the Taiwan19F-14 clone, a cluster containing the Spain6B-2 clone, a variant of the Sweden15A-25 clone, and the Colombia23F-26 clone were also all present; but fortunately, none of them have yet had any success in spreading in the Norwegian population. The reason why these resistant strains do not establish themselves in Norway might be the relatively low selective pressure for resistant strains (2). Norway has a relatively strict routine for the prescription of antibiotics, and the consumption of penicillin was reported to be 7 defined daily doses/1,000 inhabitants/day in the period from 1995 to 2001 (23); in comparison, for example, the consumption of penicillin in France during that period was 16 defined daily doses/1,000 inhabitants/day (14). In addition, broad-spectrum penicillins are little used in Norway (14).
The majority of the STs identified by MLST were represented by only one strain. The high degree of diversity shown among the non-penicillin-susceptible strains suggests that many of the resistant cases in Norway result from imported strains rather than from the spread of resistant clones within the Norwegian community. This hypothesis is further supported by the analysis of the PBP patterns, which also revealed a high degree of genetic diversity in the pbp genes among the non-penicillin-susceptible strains. This high degree of diversity indicated that resistance has developed by genetic transformation of the pbp genes on many occasions rather than from the interclone transfer of modified genes. The fact that strains of the same serotype and strains of the same ST showed several different PBP patterns strengthens this point of view.
The PBP analyses revealed seven different profiles represented by more than one isolate. Isolates with profile I showed a MIC range of 1 to 4 mg/liter. This difference might indicate that mechanisms other than the PBPs are influencing penicillin resistance. Alterations in PBP 2X have been recognized as the most important factor in high-level penicillin resistance. Twenty of the strains showing the type 1 pattern of digested pbp2x were fully resistant to penicillin, with the MIC range being from 2 to 4 mg/liter, while seven strains had an MIC of 1 mg/liter.
So far, the spread of non-penicillin-susceptible pneumococcal strains seems to be limited in Norway. The level of resistance shown in our material seems to be a result of multiple, independent imports of resistant strains. As long as the occurrence of resistant pneumococci in the world around us is increasing, it is probable that we will have to live with the continual import of resistant strains at a level revealed in this study. Continuous monitoring for resistance and evaluation of the characteristics of the resistant strains in Norway are important so that we may be able to detect an eventual change in this favorable situation and rapidly try to control a possible clonal spread within the country.
Funding was provided by the Norwegian Research Council (grant 143253/310 to D.A.C.).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»