Misidentification of meticillin-resistant Staphylococcus aureus by the Cepheid Xpert MRSA NxG assay, the Netherlands, February to March 2021

We describe two false-negative results in the detection of meticillin-resistant Staphylococcus aureus (MRSA) of sequence type 398 and spa type t011 using the Cepheid Xpert MRSA NxG assay. The isolates were recovered in late February and early March 2021 from two patients in different hospitals in the northern Netherlands. Variations between the two isolate genomes indicate that this MRSA strain might have been spreading for some time and could have disseminated to other regions of the Netherlands and other European countries.

We describe two false-negative results in the detection of meticillin-resistant Staphylococcus aureus (MRSA) of sequence type 398 and spa type t011 using the Cepheid Xpert MRSA NxG assay. The isolates were recovered in late February and early March 2021 from two patients in different hospitals in the northern Netherlands. Variations between the two isolate genomes indicate that this MRSA strain might have been spreading for some time and could have disseminated to other regions of the Netherlands and other European countries.
In this report, we describe two instances of falsenegative results in the detection of meticillin-resistant Staphylococcus aureus (MRSA) using the Cepheid Xpert MRSA NxG assay (Sunnyvale, California, United States (US)). The objective of this study was to elucidate a reason for the assay failure by applying whole genome sequencing. Moreover, the genome sequence data were used to assess the similarity between the study isolates.

Isolate identification
The first MRSA isolate (designated NL1) was obtained from a nasal sample of an asymptomatic veal farmer in his late 40s in February 2021, as a part of MRSA screening at a hospital in Zwolle, the Netherlands. The second MRSA isolate (designated UMCG578) was recovered from a sinus pus sample from a male patient in his early 70s with chronic purulent sinusitis at the beginning of March 2021 in a hospital in Groningen, the Netherlands. The two Dutch hospitals are located 104 km apart.
The isolates were identified as S. aureus by matrixassisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) (Bruker Daltonics, Billerica, Massachusetts, US). In both cases, the Xpert MRSA NxG assay was performed on pure colonies cultured and taken for testing from blood agar plates. The Xpert MRSA NxG assay detected the mecA target but not chromosome-SCCmec junction in both MRSA isolates.

Antibiotic resistance testing and results
Antibiotic resistance properties of isolates were further characterised. The minimum inhibitory concentration (MIC) values of 22 antibiotics were determined by Etest (bioMérieux, Marcy-l'Étoile, France) and the results were interpreted according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines [1]. Isolates NL1 and UMCG578 were phenotypically resistant to beta-lactams (benzylpenicillin, oxacillin and cefoxitin) but susceptible to ceftaroline ( Table 1). The isolates were also resistant to sulfamethoxazole/trimethoprim, gentamicin, kanamycin, tobramycin and tetracycline. Moreover, isolate NL1 was resistant to clindamycin and erythromycin.

Whole genome sequencing
The cells were lysed using the lysostaphin/lysozyme enzyme and total DNA was purified using the MagAttract HMW DNA Kit (Qiagen, Hilden, Germany). DNA was quantified using a Qubit 2.0 fluorometer (ThermoFisher Scientific, Waltham, Massachusetts, US) and the quality was assessed by the 2200 TapeStation software (Agilent Technologies, Santa Clara, California, US). A NanoDrop 2000c spectrophotometer (ThermoFisher Scientific) was used to measure the purity of extracted DNA.
To obtain the complete genome sequences for each isolate, Illumina genomic libraries were prepared using a Nextera XT kit (Illumina, San Diego, California, US) and sequenced on a MiSeq platform (Illumina) with a 2 × 300 bp paired-end protocol. Oxford Nanopore sequencing libraries were prepared using the Ligation Sequencing Kit (SQK-LSK109), and sequencing was carried out on a MinION device using flow cell type R9.4.1 (FLO-MIN106D, Nanopore, Oxford, United Kingdom). Nanopore reads were de novo assembled with SeqMan NGen assembler version 17.2.1.61 (DNASTAR, Madison, Wisconsin, US). Chromosomes were obtained as single contigs with a depth of coverage of 169.65 × for NL1 and 125.48 × for UMCG578. Chromosome consensus sequences were refined using the Illumina reads and the SeqMan NGen assembler with automated polishing workflow. Polished assemblies were manually corrected using SeqMan Pro (DNASTAR). The remaining Illumina reads, which were not mapped to the chromosomes during refinement step, were de novo assembled using SeqMan NGen. The resulting contigs were used in Basic Local Alignment Search Tool (BLASTn) against GenBank to identify plasmids. Automated genome annotation was performed using the NCBI Prokaryotic Genome Annotation Pipeline (https://www.ncbi.nlm. nih.gov/genome/annotation_prok).

Nucleotide sequence accession numbers
The sequences of the chromosome of S. aureus NL1 and its three plasmids (pNL1-01, pNL1-02 and pNL1-03) have been deposited in GenBank under accession numbers CP077741-CP077744. The sequences of the chromosome of S. aureus UMCG578 and its two plasmids (pUMCG578-01 and pUMCG578-02) have been deposited in GenBank under accession numbers CP077738-CP077740.

Molecular characterisation and single nucleotide polymorphism analysis
The complete chromosome sequence of isolate NL1 consisted of 2,911,340 bp and UMCG578 had 2,871,141 bp. Based on in silico analysis, the isolates were identified as multilocus sequence type (ST) 398 and spa type t011, which is the most frequent lineage of livestockassociated MRSA in the Netherlands [2,3]. The complete chromosome sequences of the two study isolates and 20 other ST398 isolates were uploaded to the CSI Phylogeny 1.4 server (https://cge.cbs.dtu.dk/services/CSIPhylogeny) in order to investigate their single nucleotide polymorphism (SNP)-based phylogeny. The selection of the ST398 isolates was obtained based  on the BLASTn analysis. Of all sequences deposited in the GenBank database, the selected 20 ST398 chromosomes were most genetically related to the chromosome of the NL1 isolate. Genome-wide SNP results revealed that the NL1 chromosome sequence was the most related to that of UMCG578 (Figure 1), although the two sequences differed by 59 SNPs.

Description of the staphylococcal cassette chromosome mec composite island
Genome-wide analysis revealed a novel organisation of the SCCmec composite island (SCCmec-CI) in these two isolates ( Figure 2 Additionally, the SCC-like region displayed a high degree of mosaicism. It was composed of several sub-regions, each of which showed the highest similarity to different staphylococcal species (Figure 2). The SCCmec IVa region in both NL1 and UMCG578 was the same size (24,164 bp) and the sequences were almost identical, differing by only a single nucleotide. Moreover, BLAST searches in the GenBank database showed that the SCCmec IVa sequences from both isolates were almost identical to that of S. aureus isolate CA05 (JCSC1968; identity: 99.98%; GenBank accession number: AB063172) ( Figure 2).

Genetic diversity of the chromosomes
The overall gene content showed some differences in these two isolates (

Plasmid content
The NL1 and UMCG578 isolates had different plasmid profiles. Both isolates harboured a small cryptic plasmid of 1,373 bp in size (designated pNL1-01 or pUMCG578-01) with an identical nucleotide sequence.
This plasmid was highly divergent from any nucleotide sequence deposited in the GenBank database. In isolate NL1, two other plasmids, pNL1-02 and pNL1-03, were also found. Plasmid pNL1-02 had a size of 11,899 bp and carried the tet(L) gene (conferring resistance to tetracycline) and erm(T) genes (conferring resistance to clindamycin and erythromycin). This plasmid also carried the cadD gene encoding cadmium resistance determinant. Plasmid pNL1-02 showed the highest similarity (query cover: 83%; identity: 99.98%) to plasmid pUR2941 from the human isolate of MRSA ST398 [4]. The biggest plasmid of NL1 was designated pNL1-03 and had a size of 17,455 bp. This plasmid shared the highest identity (query cover: 87%; identity: 99.98%) with plasmid 2 from S. aureus strain NZ15MR0322. The pNL1-03 plasmid possessed the gene encoding CadD family cadmium resistance transporter and the gene encoding tetronasin resistance protein. The Cad proteins of plasmids pNL1-02 and pNL1-03 were not identical and shared 84% amino acid similarity. The antibiotic compound tetronasin is used as a growth-promotant in animal farms with activity against Gram-positive bacteria. Moreover, plasmid pNL1-03 carried the multi-copper oxidase gene mco and the copper-translocating P-type ATPase gene copA, which are involved in copper resistance. The second of the two UMCG578 plasmids, designated pUMCG578-02 with a size of 3,048 bp, was essentially identical to plasmid pRIVM1295-2 (query cover: 100%;   identity: 99.88%) from the S. aureus strain RIVM1295 [5].

Ethical statement
As the data presented in this article are results of the routine diagnostic investigations and patient details are not described, ethical approval was not needed.

Discussion
The SCCmec elements precisely integrate into the S. aureus chromosome at a locus designated attB, located within the 3' end of the orfX gene [6,7]. The Xpert MRSA NxG assay is a real-time PCR-based method designed to detect MRSA by targeting extremities of the chromosome-SCCmec junction (i.e. the orfX gene in S. aureus and the J3 region in SCCmec) [8]. This approach ensures a discrimination of MRSA from meticillinsusceptible S. aureus (MSSA) and meticillin-resistant coagulase-negative staphylococci if present together in a specimen. The Xpert MRSA NxG assay contains additional primers and probes that target a sequence in the mecA/mecC genes, reducing the possibility of a false-positive result [9]. The PCR primers designed for the Xpert MRSA NxG assay should efficiently hybridise to the targets in the NL1 and UMCG578 isolates. However, our investigation revealed that an integration of the SCC-like element at the 3' end of the orfX gene resulted in the separation of orfX from SCCmec IVa, preventing amplification of the orfX-SCCmec target region.  [11]. Both groups came to the same conclusion that the false-negative results produced by the assays likely resulted from the large insertions in the orfX/SCCmec integration site. Therefore, we can assume that the two MRSA ST398 isolates with the new SCCmec-CI characterised in our study may be misidentified in the orfX/SCCmec junction assays, which do not utilise a polymerase optimised for long-range PCR and increased extension time during target amplification.
Our analysis revealed several differences between the genomes of the NL1 and UMCG578 isolates. First, we only found the Staphylococcus prophage 96 in the NL1 isolate, while the StauST398-2 phage differed between isolates by several partial or entire gene deletions. Also, the isolates had different plasmid profiles, which underlie different antibiotic susceptibility patterns. Finally, the gene encoding a putative Ig domaincontaining protein differed between isolates by a number of 267 nt repeats, and at least four transpositions of IS256 had occurred. These alterations suggest that the ST398 strain bearing the SCC-like element between orfX and SCCmec IVa has circulated for a longer time in the northern Netherlands, as it had already undergone microevolution.

Conclusion
We alert that this 'false-negative' MRSA strain could have already spread to other regions of the Netherlands and to other neighbouring countries. This could create a public health risk as this MRSA clone has been shown to have zoonotic potential, causing infections in people who come into contact with the carrier animals.
Our study demonstrates that some MRSA ST398 strains may be missed using real-time PCR detection methods and highlights that molecular screening approaches should be performed in combination with culturebased identification and conventional antibiotic susceptibility testing.