Comparison of Phenotypic MRSA Detection Methods with PCR for mecA Gene in the Background of Emergence of Oxacillin-Susceptible MRSA

Abstract

Background:

Phenotypic methods for detection of methicillin resistance in Staphylococcus aureus (MRSA) can be inaccurate due to heterogeneous expression of resistance and due to environmental factors that influence the expression of resistance. This study aims to compare various phenotypic methods of detection of methicillin resistance with polymerase chain reaction (PCR) for mecA gene and to detect the presence of oxacillin-susceptible MRSA (OS-MRSA).

Materials and Methods:

A total of 150 S. aureus isolates were tested using cefoxitin disk diffusion, oxacillin salt agar (OSA), latex agglutination test for penicillin binding protein 2a antigen, chromogenic MRSA ID agar, and mecA PCR.

Results:

Using PCR as the gold standard, 91 (60.66%) of 150 clinical S. aureus strains were identified as MRSA. Three oxacillin-susceptible (minimum inhibitory concentration ≤2 μg/mL) mecA-positive isolates were classified as OS-MRSA. Among the different phenotypic MRSA detection methods studied, latex agglutination had the highest sensitivity and specificity (98.9% and 98.3%), followed by cefoxitin disk diffusion (95.6% and 98.3%), MRSA ID (97.8% and 83.05%), and OSA (86.81% and 94.92%).

Conclusion:

The sensitivity of cefoxitin disk diffusion method may be reduced in areas with a high prevalence of OS-MRSA where a combination of cefoxitin disk diffusion test with MRSA ID agar or latex agglutination is recommended.

Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) is an important cause of infections globally and across the Indian subcontinent.^1,2 The treatment of MRSA infections widely differs from that of methicillin-sensitive S. aureus (MSSA) infections. Staphylococcus aureus strains that have oxacillin minimum inhibitory concentrations (MICs) >2 μg/mL or harbor the mecA gene are considered MRSA by Clinical and Laboratory Standards Institute (CLSI).³ The commonest resistance mechanism seen in MRSA is the synthesis of penicillin binding protein 2a (PBP2a), which is encoded by the mecA gene and has low affinity for β-lactam antibiotics. The gene is carried on the staphylococcal cassette chromosome mec (SCCmec).

Due to the heterogeneous nature of mecA-positive MRSA, the phenotypic expression of methicillin resistance is affected by temperature, incubation time, inoculum concentration, and salt content in the media.^4,5 Oxacillin-susceptible MRSA (OS-MRSA) is a subset of MRSA harboring the mecA gene but having oxacillin MICs of <2 μg/mL.⁶ They test susceptible by routine phenotypic tests such as oxacillin screen agar and cefoxitin disk diffusion, and could be misclassified as MSSA.

There are many options for the detection of methicillin resistance in S. aureus, including disk diffusion, MIC measurement (in broth or by Etest), chromogenic agar, latex agglutination, rapid screening methods, automated methods, and molecular methods.^2,6,7 Polymerase chain reaction (PCR) for mecA gene is considered the gold standard.⁷ It is rapid, with high specificity and sensitivity, but expensive and difficult to perform on a daily basis.

The clinical problem posed by isolates with heterogeneous expression of methicillin resistance is that during chemotherapy with beta-lactam antibiotics, production of PBP2a may be induced, rendering these strains oxacillin resistant.⁷ Hence, detection of mecA gene by PCR is a useful technique in clinical laboratories. This study involves a comparison of various phenotypic methods for detection of methicillin resistance using PCR for mecA gene as the gold standard.

Materials and Methods

The study was conducted in the Department of Microbiology at a 1,500-bedded tertiary care teaching hospital in South India. The study was approved by the institutional ethics committee of the hospital. Staphylococcus aureus was isolated from 150 samples, including blood, pus, tissue, pleural fluid, and nasal swabs. All isolates were identified by colony morphology and standard biochemical reactions such as DNAse, tube coagulase, and slide coagulase tests.⁸ Methicillin resistance was detected using several phenotypic methods.

Cefoxitin disk diffusion

Disk diffusion using cefoxitin (30 μg) disk (Becton Dickinson, Sparks, MD) on Mueller Hinton Agar (Himedia) was performed for all isolates and read at 16–18 hours of incubation (susceptible ≥22 mm, resistant ≤21 mm).

Oxacillin salt agar

Oxacillin salt agar (OSA) (prepared by incorporating Mueller Hinton agar with 4% NaCl and 6 μg/mL oxacillin) was inoculated using a 1 μL loop with a 0.5 McFarland suspension of the test colony over an area 10–15 mm in diameter and examined in transmitted light for colonies or light growth after 24 hours of incubation.

Chromogenic agar MRSA ID

Chromogenic agar MRSA ID (bioMérieux, Inc., Durham, NC) was used for detection of MRSA. A 0.5 McFarland suspension of the test colony prepared from a 24 hour culture on 5% sheep blood agar was streaked on the MRSA ID agar. The presence of green colonies after 24 hours of incubation at 35°C was indicative of MRSA.

Latex agglutination test for PBP2a

Slidex MRSA (bioMérieux, Inc.) latex agglutination kit was used to detect PBP2a in accordance with the manufacturer's instructions. Induced growth from the cefoxitin zone margin was not used initially.

The test was repeated using induced growth (taken from a cefoxitin zone margin after 24 hour incubation on sheep blood agar in 5% CO₂) for one isolate that showed discordant results (mecA positive by PCR but latex agglutination for PBP2a negative).

PCR for mecA gene

PCR for mecA gene was performed on all isolates using a previously published protocol.⁹ Two standard strains, one MRSA (ATCC 43300) and one MSSA (ATCC 29213), were included in each batch of testing for all methods including PCR. Various validity parameters such as sensitivity, specificity, predictive value, and accuracy were compared with PCR as gold standard. Data were statistically analyzed using McNemar's chi-square test for association and kappa concordance measures for levels of concordance.

Oxacillin MICs

Oxacillin MICs were determined by Vitek-2 (bioMérieux, Inc.) for all isolates. For isolates that were cefoxitin susceptible, but mecA positive, oxacillin MICs were determined using Etest strips (bioMérieux, Inc.) placed on Mueller Hinton agar containing 2% NaCl and a 0.5 McFarland inoculum of the test strain. The plates were incubated at 35°C for 24 hours. Oxacillin MICs ≤2 μg/mL were considered susceptible.

Results

Of the 150 S. aureus isolates, 52 (34.6%) were cultured from pus, 50 (33.3%) from blood, 31 (20.7%) from tissue samples, 8 (5.3%) from sterile body fluids, 5 (3.3%) from respiratory samples, 3 (2%) from ear swabs, and 1 (0.6%) from urine.

PCR detected the mecA gene in 91 (60.7%) of the 150 S. aureus isolates. No phenotypic method showed 100% sensitivity when compared with PCR for mecA (Table 1).

Table 1.

Comparison of Phenotypic Methods for the Detection of Methicillin Resistance in Staphylococcus aureus with mecA Polymerase Chain Reaction as Gold Standard

Phenotypic test	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	Accuracy (%)
Latex agglutination test	98.9	98.31	98.9	98.3	98.6
Cefoxitin disk diffusion	95.6	98.3	98.86	93.5	97
MRSA ID chromogenic medium	97.8	83.05	89.9	96.07	92
Oxacillin salt agar	86.81	94.92	96.34	82.3	90

MRSA, methicillin resistance in Staphylococcus aureus; NPV, negative predictive value; PPV, positive predictive value.

Cefoxitin disk diffusion showed susceptible zones in three isolates (from different patients) that possessed the mecA gene. The oxacillin MICs of these strains (O-256, O-893, P-75) were <2 μg/mL as determined by oxacillin Etest (bioMérieux, Inc.) and by Vitek-2 (Table 2). Vitek cefoxitin screen, however, was positive.

Table 2.

Summary of Oxacillin-Susceptible Methicillin Resistance in Staphylococcus aureus Isolates in the Study

Test	O-256	O-893	P-75
Cefoxitin disk diffusion	Susceptible	Susceptible	Susceptible
Cefoxitin zone diameters—initial and repeat values	25 mm	24 mm	26 mm
Cefoxitin zone diameters—initial and repeat values	23 mm	25 mm	26 mm
Oxacillin salt agar	Negative	Negative	Negative
Latex agglutination for PBP2a	Positive	Positive	Positive
MRSA ID chromogenic agar	Positive	Positive	Positive
Oxacillin Etest MIC (μg/mL)—initial and repeat values	0.38	0.75	0.5
Oxacillin Etest MIC (μg/mL)—initial and repeat values	0.38	0.75	0.38
Vitek-2 oxacillin MIC (μg/mL)—initial and repeat values	0.5	1	2
Vitek-2 oxacillin MIC (μg/mL)—initial and repeat values	0.5	1	2
Cefoxitin screen (Vitek-2)	Positive	Positive	Positive
mecA PCR	Positive	Positive	Positive
Antimicrobial susceptibility to other classes	Susceptible to Va, Lz, Tig	Susceptible to Va, Lz, Tig, Cd	Susceptible to Va, Lz, Tig, Cd
Resistance	Resistant to Tmp-smx, Cd	Resistant to Tmp-Smx	Resistant to Tmp-Smx

Cd, clindamycin; Lz, linezolid; MIC, minimum inhibitory concentration; PBP2a, penicillin binding protein 2a; Tig, tigecycline; Tmp-Smx, Co-trimoxazole; Va, vancomycin.

The three cefoxitin-sensitive mecA-positive isolates were classified as OS-MRSA. Their prevalence in this study was 2% among S. aureus isolates and 3% among MRSA. The three OS-MRSA isolates were cultured from three different patients. Two isolates (O-256 and O-893) were cultured from tissue samples; one from postoperative osteomyelitis and another from a nonhealing diabetic foot ulcer. The third isolate (P-75) was cultured from pus drained from a submandibular abscess in a pediatric patient. All the phenotypic tests (cefoxitin disk diffusion, susceptibility by Vitek-2, latex agglutination tests for PBP-2, and growth on MRSA screen ID agar) for these three isolates were repeated to confirm the results.

Oxacillin salt agar had a sensitivity and specificity of 86.8% and 94.92%, respectively, in comparison with PCR for mecA.

The sensitivity of chromogenic MRSA ID agar was 97.8%. It was more sensitive but less specific than cefoxitin disk diffusion. This test also identified the three OS-MRSA isolates as methicillin resistant.

The sensitivity of PBP2a latex agglutination test (without using induced growth) was 98.9% (90/91). One mecA-positive isolate (E14) that tested latex agglutination negative was tested again using induced growth from the zone margin of a cefoxitin disk on blood agar. It then agglutinated, but the agglutination was delayed, occurring after 3 minutes of observation recommended by the kit manufacturer. Hence, the strain was counted as latex agglutination negative. The three OS-MRSA strains were correctly identified as MRSA by latex agglutination test. The specificity of latex agglutination test was 98.3% with only one false positive. The isolate (B1786), cultured from blood, repeatedly tested positive with large clumps and clearing of the background. The control latex was negative.

The same isolate also tested resistant by MRSA ID agar and cefoxitin disk diffusion, though it did not have the mecA gene.

Latex agglutination was in almost perfect agreement with PCR in this study. Cefoxitin disk diffusion and MRSA ID also showed excellent results (Table 1).

Discussion

In our study, the sensitivity of cefoxitin disk diffusion was 95.6%. OSA had a lower sensitivity of 86.81%. Oxacillin MICs by Etest for three discordant isolates (cefoxitin susceptible by disk diffusion, OSA negative, mecA positive) were <2 μg/mL, and they were classified as OS-MRSA (Table 2). The CLSI 2020 M100 has described the detection of mec-A mediated resistance using oxacillin broth microdilution or oxacillin agar dilution with the addition of 2% NaCl to cation-adjusted Mueller Hinton broth or Mueller Hinton agar, respectively. Strains showing MICs of <2 μg/mL are to be considered mecA negative.¹⁰ It would be interesting to study these methods to see if they could correctly detect mecA-positive and negative isolates.

Previous studies have reported 100% concordance of cefoxitin disk diffusion with PCR.^11,12

Vitek-2 cefoxitin screen was positive for all the three OS-MRSA isolates identified in this study. Other studies have reported that some of their OS-MRSA isolates were misidentified as MSSA by Vitek-2.^13,14

Specificity of cefoxitin disk diffusion was 98.3% with one isolate from blood culture (B1786) testing resistant by all phenotypic tests (cefoxitin disk diffusion, chromogenic MRSA ID agar, and latex agglutination test) except OSA, in the absence of the mecA gene. The suspicion of MRSA was reported to the clinician. In view of the seriousness of the infection, the patient was treated with vancomycin.

The reason for discordance in this strain (B1786) could not be elucidated. A recently recognized form of MRSA, which encodes a divergent mec gene called mecC, can colonize and infect humans and a wide range of other host species.¹⁵ We did not look for mecC in our isolates.

MRSA ID had higher sensitivity (97.8%) than cefoxitin disk diffusion (95.6%). The three OS-MRSA isolates detected in this study were identified as methicillin resistant by this selective agar. Similar findings regarding the detection of OS-MRSA by MRSA ID have been reported previously.¹⁶

MRSA latex agglutination is rapid and easy to perform, though it is expensive. The sensitivity of latex agglutination for detection of MRSA was 98.9%, which was the highest among the phenotypic methods used in our study. Sakoulas et al. also reported that of the phenotypic methods for MRSA detection, latex agglutination test for PBP2a had the best sensitivity.¹⁷ We also found that latex agglutination test was positive in all our OS-MRSA isolates.

The reasons for the dissonance between the methicillin-susceptible phenotype of OS-MRSA and their mecA-positive genotype have been researched extensively. A recent study that subjected seven OS-MRSA isolates to whole-genome sequencing found that mutations within the mecA gene caused them to resemble MSSA phenotypically, and that exposure to subinhibitory concentrations of β-lactams induced secondary mutations that caused reversion to resistance.¹⁸ Another reason for the discordance is the difference in expression of oxacillin resistance between S. aureus strains of classes I, II, III, and IV described by Tomasz et al. Only class IV is homogenously resistant and it is difficult to detect class I heteroresistance by phenotypic methods.^19,20 Most clinical isolates of S. aureus are thought to show some heterogeneity in expression of oxacillin resistance. In class I heteroresistant strains, the frequency of highly resistant subpopulations is extremely low (in the order of 10^–7 to 10^–8), but these resistant cells can thrive under antibiotic pressure.²⁰

OS-MRSA strains have been shown to be functionally oxacillin resistant by oxacillin bactericidal assays. Studies have reported the selection of resistant subpopulations on exposure of these strains to β-lactams.¹⁷ Hence, there is a possibility of failure of β-lactam therapy in patients infected with OS-MRSA.^17,21

The rates of OS-MRSA reported across the world vary widely.¹⁹ Within Asia, the prevalence of these strains among S. aureus causing bovine mastitis is high, and has been reported to be 48.7% in India and 35.9% in China.^22,23 OS-MRSA constituted 3.8% of the S. aureus isolates obtained from hospitalized patients from 10 cities across China.²⁴ Brahma et al. reported that 14.86% of the S. aureus isolated from ocular infections in their hospital in South India were OS-MRSA.²⁵

The prevalence of OS-MRSA in our study was 2% among S. aureus isolates. This finding suggests that cefoxitin disk diffusion used in isolation would miss several OS-MRSA out of the hundreds of S. aureus cultured from clinical samples in our center every year, possibly leading to failure of antibiotic therapy. The combination of cefoxitin disk diffusion with either latex agglutination test or chromogenic MRSA ID agar would however detect all OS-MRSA among our isolates (100% sensitivity).

As latex agglutination test was positive, the three OS-MRSA isolates in our study were reported as methicillin resistant and the infections treated with anti-MRSA drugs such as linezolid. Further characterization of the isolates by multilocus sequence typing to determine clonality could not be performed.

Conclusion

Our study showed that Oxacillin salt agar has lower sensitivity than other phenotypic methods for MRSA detection. Latex agglutination had the best sensitivity, specificity, and concordance with mecA PCR. The sensitivity of cefoxitin disk diffusion was reduced due to the presence of a few unusual MRSA phenotypes, which were cefoxitin sensitive and mecA positive. These OS-MRSA isolates tested methicillin resistant by latex agglutination and chromogenic MRSA ID agar. Based on our findings, we recommend a combination of cefoxitin disk diffusion with either MRSA ID chromogenic medium or latex agglutination for PBP-2a, to achieve sensitivity and specificity levels similar to mecA PCR in laboratories that do not perform mecA PCR as a routine.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

This study received no specific grant from any funding agencies.

References

Woodford

, and Livermore

D.M.

. 2009. Infections caused by Gram-positive bacteria: a review of the global challenge. J. Infect. 59:S4–S16.

Mohanasoundaram

K.M.

, and Lalitha

M.K.

. 2008. Comparison of phenotypic versus genotypic methods in the detection of Methicillin resistance in Staphylococcus aureus. Indian J. Med. Res. 127:78–84.

CLSI. 2018. Performance Standards for Antimicrobial Susceptibility Testing; Twenty-Eighth Informational Supplement. CLSI Document M100-S28. Clinical and Laboratory Standards Institute, Wayne, PA.

De Lencastre

, and Tomasz

. 1994. Reassessment of the number of auxiliary genes essential for expression of high-level methicillin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 38:2590–2598.

Montanari

, Tonin

, Biavasco

, and Varaldo

. 1991 Further characterization of borderline methicillin-resistant Staphylococcus aureus and analysis of penicillin-binding proteins. Antimicob. Agents Chemother. 35:632–639.

Petinaki

, Kontos

, and Maniatis

A.N.

. 2002. Emergence of two oxacillin- susceptible mecA positive S. aureus clones in a Greek Hospital. J. Antimicrob. Chemother. 50:1090–1091.

Chambers, H.F. 1997. Methicillin resistance in Staphylococcus: molecular and biochemical basis and clinical implications. Clin. Microbiol. Rev. 10:781–791.

Baird

2006. Staphylococci: Cluster forming Gram positive cocci, Chapter 11. In: J.G. Collee, A.G. Fraser, B.P. Marmion, and A. Simmons (eds.), Mackie and McCartney. Practical medical microbiology. 14th ed. Churchill Livingstone, Edinburgh,, UK, pp. 245–261.

Zhang

, Sparling

, Chow

B.L

, et al. 2004 New quadriplex PCR assay for detection of methicillin and mupirocin resistance and simultaneous discrimination of Staphylococcus aureus from coagulase-negative staphylococci. J. Clin. Microbiol. 42:4947–4955.

10.

CLSI. 2020. Performance Standards for Antimicrobial Susceptibility Testing; CLSI Document M100. 30th ed. Clinical and Laboratory Standards Institute, Wayne, PA.

11.

Anand

K.B.

, Agarwal

, Kumar

, and Kapila

. 2009.Comparison of cefoxitin disc diffusion test, oxacillin screen agar, and PCR for mecA gene for detection of MRSA. Indian J. Med. Microbiol. 27:27–29.

12.

Velasco

, del Mar Tomas

, Cartelle

, et al. 2005.Evaluation of different methods for detecting methicillin (oxacillin) resistance in Staphylococcus aureus. J. Antimicrob. Chemother. 55:379–382.

13.

Ikonomidis

, Michail

, Vasdeki

, et al. 2008. In vitro and in vivo evaluations of oxacillin efficiency against mecA-positive oxacillin-susceptible Staphylococcus aureus. Antimicrob. Agents Chemother. 52:3905–3908.

14.

Kong

L.Y.

, Jean

, Wong

, et al. 2015. Bacteremia caused by a mecA-positive oxacillin-susceptible Staphylococcus aureus strain with inducible resistance. Diagn. Microbiol. Infect. Dis. 83:377–378.

15.

Paterson

G.K.

, Harrison

E.M

, and Holmes

M.A.

. 2014. The emergence of mecC methicillin-resistant Staphylococcus aureus. Trends Microbiol. 22:42–47.

16.

Kumar

V.A.

, Steffy

, Chatterjee

, et al. 2013. Detection of oxacillin-susceptible mecA-positive Staphylococcus aureus isolates by use of chromogenic medium MRSA ID. J. Clin. Microbiol. 51:318–319.

17.

Sakoulas

, Gold

H.S

, Venkataraman

, DeGirolami

P.C

, Eliopoulos

G.M

, and Qian

. 2001. Methicillin-resistant Staphylococcus aureus: comparison of susceptibility testing methods and analysis of mecA-positive susceptible strains. J. Clin. Microbiol. 39:3946–3951.

18.

Goering

R.V.

, Swartzendruber

E.A

, Obradovich

A.E

, Tickler

I.A

, and Tenover

F.C.

. 2019. Emergence of oxacillin resistance in stealth methicillin-resistant Staphylococcus aureus due to mecA sequence instability. Antimicrob. Agents Chemother. 63:e00558-19.

19.

Tenover

F.C.

, and Tickler

I.A.

. 2015. Is that Staphylococcus aureus isolate really methicillin susceptible? Clin. Microbiol. News, 37:79–84.

20.

Tomasz

, Nachman

, and Leaf

. 1991. Stable classes of phenotypic expression in methicillin-resistant clinical isolates of staphylococci. Antimicrob. Agents Chemother. 35:124–129.

21.

Saeed

, Dryden

, and Parnaby

. 2010. Oxacillin-susceptible MRSA, the emerging MRSA clone in the UK? J. Hosp. Infect. 76:267–268.

22.

Mistry

, Sharma

, Mahato

, Saravanan

, Kumar

P.A

, and Bhandari

. 2016. Prevalence and characterization of oxacillin susceptible mecA-positive clinical isolates of Staphylococcus aureus. causing bovine mastitis in India. PLoS One 11: e016, 2256.

23.

, Su

, Li

, Yang

, Deng

, and Ni

. 2014. High incidence of oxacillin-susceptible mecA-positive Staphylococcus aureus (OS-MRSA) associated with bovine mastitis in China. PLoS One, 9:e88134.

24.

, Chen

, Zhao

, Zhang

, and Wang

. 2013. Prevalence and molecular typing of oxacillin-susceptible mecA-positive Staphylococcus aureus from multiple hospitals in China. Diagn. Microbiol. Infect. Dis. 77:267–269.

25.

Brahma

, Sharma

, Murthy

, et al. 2019. Decreased expression of femXAB genes and fnbp mediated biofilm pathways in OS-MRSA clinical isolates. Sci. Rep. 9:16028.