Prevalence and molecular characterization of extended-spectrum β-lactamase-producing Klebsiella pneumoniae in Riyadh, Saudi Arabia

BACKGROUND AND OBJECTIVES: Reports on extended-spectrum β-lactamases (ESBL) production by Enterobacteriaceae, and especially in Klebsiella pneumoniae, are few in Saudi Arabia. Therefore, we determined the prevalence of ESBL in K pneumoniae from Riyadh and characterized the predominant β-lactamase gene in these isolates. METHODS: A total of 400 K pneumoniae samples were isolated from two hospitals in Riyadh during 2007 and screened for production of ESBL using ESBL-E-strips and combined disk methods. PCR assay was used to detect blaTEM, blaSHV, and blaCTX-M genes. RESULTS: Phenotypic characterization identified a high ESBL rate of 55% of K pneumoniae isolates. ESBL-producing K pneumoniae were PCR positive for SHV, TEM and CTX-M β-lactamase genes with prevalences 97.3%, 84.1% and 34.1%, respectively. Within the CTX-M family, two groups of enzymes, CTX-M-1 and CTX-M-9-like genes were found with prevalences of 60% and 40%, respectively. CONCLUSIONS: This study confirms the high rate of ESBL in K pneumoniae clinical isolates in hospitals in Riyadh. This study demonstrates the worldwide spread of blaCTX-M genes. This first report of the presence of the blaCTX-M gene in clincial isolates in Saudi Arabia is evidence of the continuing worldwide spread of this gene.

ESBL found in many regions of the world, including Asia, South America, Europe and Africa. [7][8][9][10][11] The CTX-M enzymes form a rapidly growing family of currently over 69 enzymes (http://www.lahey.org/studies) catego-rized in five different groups called CTX-M-1, CTX-M-2, CTX-M-8, CTX-M-9 and CTX-M-25. 7 In Saudi Arabia, few studies have been undertaken to determine the prevalence of ESBL-producing K pneumoniae [12][13][14] However, b-lactamases were not mo-lecularly characterized in these studies. Therefore, this study was devoted to determining the prevalence of ESBL in K pneumoniae and the molecular characteriza-tion of ESBLs in K pneumoniae isolates from Riyadh.

Gene Primers Nucleotide sequence Position
C-1 5'-AACACGGATTGACCGTCTTG-3' F C-2 5'-TTACAGCCCTTCGGCGAT-3' B F= forward, B=backward inhibitory concentration (MIC) was determined by an E-strip test (AB BIODISK, Solana, Sweden) as de-scribed by the manufacturer. A laboratory control strain, Escherichia coli ATCC 25922, was used in the sensitiv-ity test and in the MIC determination. Phenotypic detection of ESBL was carried out by two different methods: 1) the combined disc method using discs con-taining cefotaxime and ceftazidime with and without clavulanate (Becton Dickinson, USA) with the ESBL phenotype defined as an increase of ≥5 mm in the zone around the disc containing clavulanate compared to the zone of corresponding discs without clavulanate; and 2) using the E-test ESBL-strip (AB BIODISK, Solana, Sweden) with a ceftazidime gradient at one end and a ceftazidime plus clavulanic acid gradient at the other end. These strips were applied according to the proce-dure described by the manufacturer. ESBL was detect-ed if the ratio of the MIC of ceftazidime to the MIC of ceftazidime plus clavulanic acid was ≥8. PCR methods were used to detect bla TEM , bla SHV , and bla CTX-M using the primers and methods previously described. 9,16,17 A further group-specific CTX-M PCR was performed to differentiate between the CTX-M-1, -2, -8 and -9 groups of enzymes using primers and methods previ-ously described. [18][19][20] The primers used in this study are listed in Table 1. All PCRs were conducted under stan-dard conditions using plasmid DNA as templates.

RESULTS
ESBL phenotype was detected in 220 (55%) of 400 isolates. The antibiotic resistance rate and MIC distri-bution of ESBL-producing K pneumoniae are listed in Table 2. The resistance rate to cefotaxime, ceftazidime and amoxicillin/clavulanate were 97% (n=215/220), 95% (n=210/220) and 86% (n=190/220), respectively. A fourth-generation cephalosporin, cefepime, showed moderate activity (47%), but 4.5% (n=10/220) were resistant to cefoxitin and all ESBL-producing isolates were susceptible to imipenem. The resistance to cefoxi-tin in these isolates may be due to alteration in ompK35 or ompK36 and may not be due to AmpC enzymes be-cause the MIC of b-lactam/b-lactamse inhibitors are markedly reduced in our isolates, but AmpC was not.

DISCUSSION
K pneumoniae is the most frequent ESBL-producing species worldwide. Production of ESBLs was detected in 220 (55%) of 400 clinical isolates of K pneumoniae iso-lated from hospital-acquired infections in Riyadh. There was no difference between the combined disk method and the E-test strip method in the phenotypic detec-tion of ESBL for K pneumoniae isolates from Riyadh. Reports on ESBL production by Enterobacteriaceae, and especially in K pneumoniae, are few in Saudi Arabia. The overall prevalence of ESBL producers was found to vary greatly in different geographical areas and in differ-ent institutes in Saudi Arabia. In our study, the rate of ESBLs-producing isolates collected from two hospitals in Riyadh was relatively high (55%) and was similar to other rates reported from previous studies in Riyadh. 12 However, lower rates for the prevalence of ESBL-pro--ducing K pneumoniae were detected in Abha and Al-Khobar (27.5% and 12.2%, respectively). 13,21 A higher rate of ESBL production was observed in Riyadh (40% to 55%) followed by Abha (27.5%) and a lower rate was reported in Al-Khobar (12.2%). The variation is difficult to explain, but may be due to differences in the type and volume of consumption of antibiotics and differences in the time of collection of isolates. In Europe, a survey of 1610 Escherichia coli and 785 K pneumoniae isolates from 31 centers in 10 European countries found that the prevalence of ESBL in these organisms ranged from as low as 1.5% in Germany to as high as 39% to 47% in Russia, Poland, and Turkey. 22 Higher figures of 30% to 60% have been reported for the South American coun-tries of Brazil, Venezuela and Colombia. [23][24][25] In Asia, ESBL production in Klebsiella has also been reported to be as low as 5% in Japan and ESBL rates in India for E coli, K pneumoniae and K oxytoca were 79.0%, 69.4% and 100%, respectively. ESBL-positive E coli rates were also relatively high in China (55.0%) and Thailand (50.8%) and Australia with higher rates of 20% to 50% in other parts of the continent. 3,26,27 In Pakistan, the prevalence of ESBL-producing K pneumoniae was very high (70%). 28 In the Arabian Peninsula, 23.5% of ESBL-producing K pneumoniae were identified as having the ESBL phenotype in Kuwait. 29 However, in the United Arab Emirates, 36% of K pneumoniae was found to harbor the ESBL phenotype. 30 In Jordan, 70% of K pneumoniae isolates express ESBL phenotypes and this prevalence is alarming. 31 Resistance to b-lactams, especially third-genera-tion cephalosporins and non-b-lactams, among clinical isolates of gram-negative bacteria is increasing world-wide. 32, 33 The overall resistance rate of ESBL-produc-ing K pneumoniae isolates studied was alarmingly high to most antibiotics tested including gentamicin, ami-kacin, cefepime, and trimethoprim/sulfamethoxazole. Imipenem and cefoxitin followed by ciprofloxacin were the antibiotics most active against ESBL-producers. K pneumoniae isolates harboring ESBLs are signifi-cantly more often resistant to flouroquinolones. 34,35 In the present study only 11.3% of ESBL-producing K pneumoniae were resistant to ciprofloxacin. In Iran, re-sistance to ciprofloxacin was found among 32% of the ESBL-producing K pneumoniae strains. 36 TEM-and SHV-ESBL are derived from parental TEM-1 and SHV-1 by point mutations. TEM-1 and SHV-1 are non-ESBL; however CTX-M enzymes are not derived from non ESBL and consequently all CTX-M enzymes are ESBL. In the present study, the domi-nant b-lactamase was SHV with a prevalence of 97.3% followed by TEM and CTX-M with prevalences of Table 2. Minimum inhibitory concentration (MIC) of 220 extended-spectrum b-lactamase-producing K pneumoniae isolates. 84.1%, and 34.1%, respectively. In the present study the prevalence of CTX-M enzymes was 34%, but the preva-lence of CTX-M enzymes in a study under preparation was 85%. This may be because the present study was done on the isolates were collected during 2007 while in the other study the isolates collected during 2008. In addition to an insertion sequence (ISEcp1), which en-hances the mobilization of the CTX-M gene, was de-tected in 60% of CTX-M-producing Klebsiella (study under preparation), while the prevalence of ISEcp1 was relatively low (10%) in isolates included in the pres-ent study (Data not included). However, 32 (74%) of the 43 ESBL-producing isolates from Kuwait carried bla CTX-M . 37 In Iran, the prevalence of bla SHV and bla TEM among ESBL-producing K pneumoniae was 69.6% and 32.1%, respectively. 36 The prevalence of CTX-M in the present study was lower than the prevalence in Kuwait. However, the prevalence of TEM and SHV was higher than reported in Iran. The higher prevalence of CTX-M in Kuwait is due to most of the CTX-M being de-tected in non-Kuwaiti immigrants, mainly from South Asia, where CTX-M is endemic. 37  The bla SHV gene was only found alone in 6.8% (n=15/220) of ESBL-producing isolates with elevated MIC for both cefotaxime and ceftazidime (≥256 mg/ L). The presence of SHV b-lactamase alone suggests that these SHV genes are responsible for resistance to extended-spectrum cephalosporins in 6.8% of ESBLproducing K pneumoniae isolates. However, the bla SHV gene was found to be associated with the bla TEM gene in 56.8% (n=125/220) with three different patterns of resistance. Seventy-five of 125 (44%) isolates showed a higher MIC for both (≥256 mg/L), while 35 (20%) of 125 isolates exhibited an increase in MIC with ce-fotaxime and a decrease in MIC with ceftazidime, and 15 of 125 (7.2%) isolates exhibited a decrease in MIC with cefotaxime and an increase in MIC with ceftazi-dime. From this result either SHV or TEM or both are the ESBLs in 125 of 220 isolates producing ESBL. On the other hand, SHV b-lactamase was present with the CTX-M enzyme in 9.1% (n=20/220) of ESBL-producers. However, the SHV b-lactamase gene was present with both TEM and CTX-M in 25% (55/220). Thirty-five isolates belonged to the CTX-M-1 group with a higher MIC for cefotaxime and cef-tazidime, while 20 isolates belonged to the CTX-M-9 group with a decrease in MIC with cefotaxime and an increase in MIC with ceftazidime.

Antibiotics
CTX-M enzymes (CTX-M-1 like genes) were copresent with TEM in 5 isolates (2.27%) with an in-crease in MICs with cefotaxime and ceftazidime (≥256 mg/L). However, the CTX-M family of ESBLs has been increasingly detected worldwide. 5 In the present study, the occurrence of CTX-M-producing K pneum m moniae was relatively high (34.1%). In India, higher percentages (72%) of ESBL-producers harbor bla CTX-M genes. 39 In this study, we did not determine which blactamase was responsible for resistance to extendedspectrum cephalosporins because these genes were not sequenced. The DNA sequence of these genes must be done to know the type of b-lactamase gene and the prevalence of ESBL genes in ESBL-producing iso-lates in Saudi Arabia. To our knowledge, there is no published report of the discovery of bla CTX-M genes in Saudi Arabia.
In conclusion, this study confirms a high rate of ESBLs in K pneumoniae in Riyadh, Saudi Arabia, and further demonstrates the worldwide spread of genes coding for CTX-M enzymes in clinical isolates. Most ESBL producers were resistant to oxyimino-cephalo-sporins and other non-b-lactam agents at high levels.