Abstract
The prevalence and spread of mupirocin and antiseptic resistance among colonizing and infectious Staphylococcus aureus were determined. S. aureus isolated from anterior nares and infection sites of patients hospitalized in the largest tertiary care referral hospital in Malaysia was investigated for mupirocin and antiseptic susceptibility testing, and for PCR detection of mupA, qacA/B, and smr genes. Twelve isolates showed resistance to mupirocin by disk diffusion, of which 10 (3.8%) harbored the mupA gene. Minimum inhibitory concentrations (MICs) ranged from 64 to 768 μg/ml for mupA positive and below 46 μg/ml for negative isolates. The mupA was more common among ST239 isolates (70%). The qacA/B was carried in 67 out of 95 methicillin-resistant Staphylococcus aureus (MRSA) (70.5%) and 3 out of 164 methicillin-susceptible Staphylococcus aureus (MSSA) (1.8%), while smr was carried in 6 out of 95 MRSA (6.3%) strains. MICs ranged from 3.9 to 15.6 μg/ml for benzethonium chloride (BTC) and benzalkonium chloride (BKC), and from 10.3 to 20.7 μg/ml for chlorhexidine digluconate (CHG). Isolates with qacA/B and smr or qacA/B alone showed higher MIC (20.7 μg/ml for CHG and 15.6 μg/ml for BTC and BKC) than the isolates that lacked antiseptic resistance genes (10.3 μg/ml for CHG and 3.9 μg/ml for BTC and BKC). In 16 cases, ST239 was isolated from the infection site and the nares simultaneously, and shared identical resistance patterns (qacAB or qacAB+smr), suggesting possible endogenous infection. Spread of low-level mupirocin resistance expressing ST239 MRSA and high-level resistance expressing emerging ST1, co-existing with antiseptic-resistant genes showing elevated MICs, should be monitored for effective infection control.
Introduction
V
Mupirocin is a protein synthesis inhibitor that exerts antimicrobial activity by blocking the isoleucyl-tRNA synthetase (IRS). Resistance to mupirocin exists in two forms: low-level resistance (minimum inhibitory concentration [MIC] 8–256 μg/ml) due to mutations in the native IRS gene and high-level resistance (MIC≥256 μg/ml) mediated by the plasmid-associated mupA gene which encodes an extra IRS. 36
Antiseptics such as chlorhexidine and quaternary ammonium compounds exhibit antimicrobial action by cytoplasmic membrane disruption and congealing of the cytoplasm or damaging the phospholipid bilayers. 17 Resistance to antiseptics in staphylococci is mediated by multi-drug efflux pumps encoded by two families of PCR nondifferentiable genes (qacA, qacB and qacC, qacD, and smr) known as the qacA/B and smr families, respectively. 20 The qacA/B and smr genes may confer high and low level resistance to antiseptics and are transferable among different S. aureus strains, because they are located on conjugative plasmids.14,25
In Malaysia, where ST239 is the predominant nosocomial MRSA clone,9,19 although earlier studies have shown mupirocin 22 and antiseptic resistance 29 independently, characterization in contemporary S. aureus isolates, from both colonization and infection from the same group of patients is largely unknown.
To define the distribution of mupirocin and antiseptic resistance among infectious and colonizing MRSA, we investigated S. aureus strains isolated from hospitalized patients in Hospital Kuala Lumpur (HKL), the biggest tertiary referral center in Malaysia.
Materials and Methods
Study samples
This study was approved by the National Medical Research Register-Ministry of Health Research and Ethics Committee (NMRR ID, NMRR-10-488-5953 S1 R0). This study used S. aureus isolates collected from hospitalized patients (n=172) in HKL, Malaysia, during 2011. Patients with more than 1 week hospitalization period were investigated. A total of 259 S. aureus were collected from anterior nares and infection sites of 172 patients. Patients were subjected to four times nasal swab collection during the first week of admission. Among the 259, 149 (105 patients carried one S. aureus strain, 16 patients carried two different S. aureus strains, and 4 carried three different strains) were carrier isolates (isolated from the nares); while the remaining 110 were from infection sites (10 patients infected by two different S. aureus strains) such as tracheal aspirate (n=52), blood (n=40), pus (n=16), skin (n=1), and urine (n=1). All isolates were identified to the species level by the hospital laboratory and were reconfirmed in the microbiology laboratory at Universiti Putra Malaysia (UPM) by Gram staining, catalase, tube coagulation, mannitol testing, and MRSA screen latex agglutination testing (Denka Seiken Co., Ltd., Tokyo, Japan). All isolates were confirmed at the gene level as S. aureus by Sa442 PCR. 15 The MRSA isolates were confirmed by oxacillin and cefoxitin susceptibility testing according to Clinical and Laboratory Standard Institute (CLSI) guidelines 7 and mecA PCR. 24 All isolates were then stored at −80°C in Luria-Bertani broth that was supplemented with 20% glycerol awaiting further investigation. Chromosomal DNA was extracted using DNeasy blood and tissue kit (Qiagen Instruments, Hombrechtikon, Switzerland) according to the manufacturer's instructions and used for molecular investigation.
Molecular epidemiological analysis by spa typing and MLST
All S. aureus isolates, regardless of whether they were MRSA or methicillin-susceptible Staphylococcus aureus (MSSA) were subjected to spa typing according to the method and using the primers described earlier. 31 The spa type for each isolate was assigned by using the spa-server data (Ridom GmbH, Würzburg, Germany), as described by Harmsen et al. 10 A representative strain from each spa type was subjected to multilocus sequence typing (MLST) to determine sequence types (STs) of MSSA and MRSA isolates. 8 STs and clonal complexes were assigned using the S. aureus MLST database (www.mlst.net) hosted by Imperial College in London.
Mupirocin susceptibility testing and detection of the mupA gene
Susceptibility to mupirocin was determined by disc diffusion method 7 using a 200-μg mupirocin disk (Oxoid, Cambridge, United Kingdom). To classify the resistant strains as “low level” and “high level,” MICs were determined using an Etest (bioMérieux, Marcy l'Etoile, France). Strains were classified as “low level” when the MIC was between 8 and 256 μg/ml and as “high level” when the MIC was ≥256 μg/ml. 36 All strains were further tested for the presence of the mupA gene by PCR according to primers described by Yun et al. 36
Antiseptic susceptibility testing and detection of qacA/B and smr genes
All isolates were screened for the presence of qacA/B and smr genes as described by Noguchi et al. 20 Susceptibility to quaternary ammonium compounds such as benzethonium chloride (BTC), benzalkonium chloride (BKC), and biguanide compounds such as chlorhexidine digluconate (CHG; Sigma-Aldrich, Steinheim, Germany) was determined using the Müeller–Hinton broth micro dilution method. 20
Statistics
Differences per groups were tested for significance by the Fischer exact test. p-Values<0.05 were considered significant.
Results
The 259 S. aureus isolates contained 164 MSSA and 95 MRSA. A total of 73 different spa types were detected among all 259 isolates; 83 t037/ST239 (25 from nose and 58 from infection), 21 t127/ST1 (16 from nose and 5 from infection), 15 t189/ST188 (10 from nose and 5 from infection), 10 t5500/ST96 (7 from nose and 3 from infection), 8 t084/ST15 (7 from nose and 1 from infection), and 6 t701/ST152 (5 from nose and 1 from infection) were the major spa types/MLSTs (Table 1).
Resistance gene: antiseptic or mupirocin resistance gene.
MLST, multilocus sequence typing; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-susceptible Staphylococcus aureus; NT, nontypeable.
Most MRSA strains were typed as t037/ST239/n=79 (23 from nose and 56 from infection), 3 t421/ST239 (2 from nose and 1 from infection), t4184/ST22 (3 from infection), and t657/ST772 (1 from nose and 1 from infection).
Mupirocin resistance
Twelve (4.6%) out of 259 S. aureus isolates showed resistance to mupirocin by disk diffusion test; of that, 10 (83%) harbored the mupA gene (8 MRSA and 2 MSSA). The prevalence of the mupA gene was higher (p=0.005) in MRSA (8.4%; n=8/95) than in MSSA (1.2%; n=2/164). MICs determined by the mupirocin Etest ranged from 64 to 768 μg/ml for mupA-positive isolates and was below 46 μg/ml for mupA-negative strains. The highest MIC (768 μg/ml) was observed for t127/MSSA and t6811/MRSA (Table 2). For t037 strains in general that carried the mupA gene, the MIC ranged from 64 to 128 μg/ml. Although mupA was more common among ST239 isolates (n=7; 70%), acquisition of mupA was not clonal as it was also detected among other clones (Table 2). Six mupA-positive MRSA strains were isolated from infection sites (five from blood and one from pus), and two were isolated from nares. Of the eight MRSA isolates, four blood and two nare isolates belonged to t037/ST239, and one from pus was typed as t6811/NT. One blood isolate was nontypeable. The two MSSA strains were isolated from the nose and were typed as t127/ST1 and t315/ST361 (Table 2).
Controls, MIC of strains negative for the resistant genes. Numbers are given between brackets.
—, absence of gene; BKC, benzalkonium chloride; BL, blood; BTC, benzethonium chloride; CHG, chlorhexidine digluconate; MUP, mupirocin; TA, tracheal aspirate.
Antiseptic resistance
Seventy-two (27.0%) of the 259 strains carried at least one antiseptic resistance gene. The qacA/B and smr were detected in 70 out of 259 (27%) and 6 out of 259 (2.3%) isolates, of which 67 out of 95 (70.5%) were MRSA and 3 out of 164 (1.8%) were MSSA (p=0.0001). Smr was detected only in MRSA strains. Five strains carried both qacA/B and smr, of which four were t037/ST239 and one was t6811/NT (Table 2). Apart from ST239, qacA/B was also observed among other MSSA and MRSA sequence types (ST573, ST96, and ST1) (Table 2).
MICs ranged from 3.9 to 15.6 μg/ml for BTC and BKC and from 10.3 to 20.7 μg/ml for CHG (Table 2). In strains with qacA/B and smr or qacA/B alone, regardless of MSSA or MRSA, all strains showed MIC as 20.7 μg/ml for CHG and as 15.6 μg/ml for BTC and BKC. One strain that carried only the smr gene had an MIC of 10.3 μg/ml for CHG, while MIC for BTC and BKC were 7.8 μg/ml. In ten isolates that did not carry any of the antiseptic resistance genes, the MIC was 10.3 μg/ml for CHG and 3.9 μg/ml for both BTC and BKC (Table 2).
Antiseptic and mupirocin resistance genes in colonization versus infection
Eighteen out of 26 MRSA strains isolated from the nose were t037/ST239. Among them, one carried qacA/B, smr, and mup; another isolate carried smr and mupA; while 16 strains possessed only qacA/B (Table 1).
Among 19 study cases where MRSA was isolated from infection and nares simultaneously, in 16 cases, both strains were ST239 and shared similar resistance patterns (qacAB or qacAB+smr). In one case, the strain isolated from the nares carried smr and mupA, while the strain from infection was qacA/B positive. In another case, the patient who carried qacA/B, smr, and mupA-positive t6811/NT in the nares (colonization) was infected with a qacA/B-positive t037/ST239. Nine out of the 10 mupirocin-resistant MRSA isolates shared the combination of mupA and qacA/B or smr genes in both carrier and infection strains (Table 2).
Discussion
S. aureus, especially MRSA, continues to be a global cause of hospital-acquired infections. This limits treatment options and results in substantial morbidity and mortality. In humans, S. aureus has been shown to preferentially colonize the nasal cavity, and infections are often endogenous. Decolonization and decontamination by using topical mupirocin ointment and antiseptic body wash prevents S. aureus infections in hospitals to a certain degree.3,11 We investigated the phenotypic and genotypic prevalence of mupirocin and cationic antiseptic resistance in S. aureus strains isolated from the anterior nares and infection sites of hospitalized patients. Several studies have reported the prevalence of mupirocin and antiseptic resistance in MRSA, but this is the first study that compared the contemporary carriage and infection isolates from the same patients.
The vast majority of our MRSA isolates belonged to ST239 (t37/n=80; t421/n=3; t966/n=1), which again emphasizes that ST239 is the predominant clone in Malaysian hospitals. 9 This is true for most other Asian countries except Korea and Japan. 12 For the first time, we also found the colonization of ST239 in the nares of patients (n=22).
Mupirocin is a topical antimicrobial agent introduced in 1987 to treat staphylococcal skin infections, and it is widely used for eliminating MRSA from the nose. However, decolonization failure due to the presence of mupirocin-resistant strains has been reported.26,32,34 In the current study, mupirocin resistance was observed in 8.4% MRSA (8 of 95) and 1.2% MSSA (2 of 164) isolates. This resistance rate is higher than mentioned in previous reports from Malaysia, where 2.8%, 1.3%, 3.8%, and 5% were reported in 2000, 2001, 2003, and 2010, respectively.13,22,23,27 This shows that mupirocin resistance is on the rise. 27 However, the resistance rates reported here are lower than in other Asian countries, including Korea (17%) and Singapore (11%) 6 but higher than in China (6.6%). 12 Despite the fact that the majority of ST239 strains remained mupA negative, among the few mupirocin-resistant MRSA isolates, most were ST239 and showed low-level resistance. A recent study 1 in India identified ST239 as the predominant clone and carried high-level mupirocin resistance (MIC>512 ng/ml; mupA positive). This is in contrast with our results where all ST239 isolates were low-level mupirocin resistant. The clinical significance of low-level mupirocin resistance in decolonization failure is not well understood, 28 but caution is necessary as ST239 is the predominant clone in Malaysia. In HKL, patients are not routinely screened for MRSA carriage on admission. Only when there is an outbreak, health care workers (HCWs) and the patients in that particular ward and associated wards are screened for MRSA carriage. Decolonization is performed for MRSA nasal carriers (patients) only if they develop infection. Similarly, HCWs that are positive for MRSA carriage are treated only when they are identified as the source for outbreak or transmission. Hence, the use of mupirocin is limited in the hospital. The low prevalence of MRSA in the community and the limited use of mupirocin in the hospital explain the low prevalence of mupirocin-resistant MRSA in Malaysia.5,18
The incidence of qacA/B among MRSA isolates in our study was 69.4% (66/95 MRSA). In a previous study carried out on MRSA strains from Malaysia, 83.3% was reported. 29 This is higher than the rates detected in the current study. However, both these carriage rates were higher than the average rate of qacA/B detected among MRSA isolated from 11 different Asian countries (38.5%) 21 and 12 European countries (62.6%). 16 An increasing trend of qacA/B resistance observed in a recent study from Korea (59%) implies that detailed monitoring is mandatory. 13
In comparison to qacA/B, the smr gene was less common among our isolates; the rates have increased from 1.6% in Malaysia in the previous report to 6.3% in the present study. 29 In other countries (Asia and Europe), the incidences of smr among MRSA isolates ranged from 0% to 7%.16,21,30 Surprisingly, a recent investigation from Korea showed an unusually high prevalence of smr (71%) in MRSA. 13 Among the 71% smr-positive MRSA isolated, although the vast majority (85.5%) was ST5, 14.5% were ST239 and showed low-level mupirocin resistance. In our study, the majority of the qacA/B- (89%) and smr (83%)-positive strains were ST239, and all these strains showed low-level mupirocin resistance. This clearly shows the spread of antiseptic-resistant ST239 MRSA strains with low-level mupirocin resistance in Malaysia.
Since differences in MICs ranges against antiseptic reagents were narrow, defining a resistance criteria was not possible but strains with qacA/B and smr or qacA/B alone showed an increase in MICs against quaternary ammonium compounds (BTC, BKC) and biguanide compound (CHG) in comparison with strains that did not carry resistance genes. Strains with smr alone showed less increase in MIC level to BTC and BKC when compared with qacA/B-positive strains. High prevalence of resistance genes (qacA/B) and elevated MICs among MRSA strains harboring these genes emphasizes the fact that qacA/B is the major gene responsible for antiseptic resistance among MRSA strains in Malaysia. The reported qacA/B prevalence among strains in different Asian and European countries shows that the genes are widely spread around the world and are not limited to any geographic area. 16 When qacA/B was detected in 70.5% of MRSA, a few MSSA strains (1.8%) carried qacA/B concurrently. This is in concordance with the results of Mayer et al., where higher prevalence of qacA/B among MRSA (63%) compared with MSSA (12%) was reported. 16
In 21 patients, antiseptic-resistant genes were detected from strains isolated from both nose and infection sites and the majority of strains (n=17; 81%) were ST239 MRSA (t037 and t421). Despite the fact that the MICs for antiseptic reported here are still lower than the concentration which is used in the hospital, the high prevalence of MRSA among the hospitalized patients clearly shows that disinfectants containing chlorhexidine or quaternary ammonium compounds can lose their effectiveness in patients colonized with MRSA containing qacA/B and smr genes. 2 Although nonclonal distribution of antiseptic- and mupirocin-resistant genes among S. aureus strains has been observed elsewhere13,16,21 due to the mobility of plasmids, 25 the higher prevalence of low-level mupirocin and antiseptic-resistant ST239 in Malaysia need to be considered with care.
In conclusion, ST239 is still the predominant MRSA clone in Malaysia. The high prevalence of antiseptic-resistant genes among the MRSA isolates underscores the need for prudent use of antiseptics in the hospitals. Sharing a similar combination of resistant genes (antiseptic/mupirocin) in both carrier and infection strains from the same patients suggests possible endogenous infection and strongly emphasizes that MRSA decolonization should not be taken lightly. Spread of low-level mupirocin resistance expressing ST239 MRSA strains and high-level resistant ST1 MRSA and MSSA should be continuously monitored.
Footnotes
Acknowledgments
This work was funded by the Ministry of Higher Education (Malaysia) through the Fundamental Research Grant scheme (1068 FR). The authors thank all staff and patients of Hospital Kuala Lumpur who contributed to this study. The authors also thank Mateq Alreshidi (Faculty of Medicine, Alrajhi colleges, Saudi Arabia) for kindly providing resistance gene-positive strains.
Disclosure Statement
None of the authors have any conflicts of interest and financial interest relevant to this article.
