Staphylococcus massiliensis isolated from human blood cultures, Germany, 2017–2020

Clinical and laboratory data on newly described staphylococcal species is rare, which hampers decision-making when such pathogens are detected in clinical specimens. Here, we describe Staphylococcus massiliensis detected in three patients at a university hospital in southwest Germany. We report the discrepancy of microbiological findings between matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, 16S-rRNA polymerase chain reaction, and whole-genome sequencing for all three isolates. Our findings highlight the diagnostic pitfalls pertinent to novel and non-model organisms in daily microbiological practice, in whom the correct identification is dependent on database accuracy.


Introduction
Coagulase-negative staphylococci (CoNS) are abundant inhabitants of the human skin and the mucosa with limited pathogenicity [1, 2], as opposed to Staphylococcus aureus [3] and other members of the S. aureus complex [4,5]. Device-related infections are one of the most common manifestations of CoNS infections, causing high morbidity and mortality [6,7]. With the advent of modern tools in routine microbiological diagnostics such as matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), pathogen detection has experienced a remarkable refinement within the past decade [8,9]. However, reliable species identification by MALDI-TOF MS relies greatly on the quality and accuracy of the employed database. Over the past years, the species identification and differentiation of clinical CoNS isolates have improved significantly, leading to the identification of novel or "rare" species in clinical specimens [10]. Indeed, the "true" clinical relevance of many CoNS species is frequently underappreciated. Staphylococcus massiliensis is a recently described CoNS first reported in 2010 from a cerebral abscess [11], and later from healthy human skin [12]. In 2012, Zong postulated to conceptualize S. massiliensis as part of the normal human skin microflora [13]. Until today, data pertaining to the pathogenic potential of this microorganism and possible treatment approaches is scarce. Here, we report the challenges and limitations in the identification of S. massiliensis in blood cultures from three patients at a university hospital in southwest Germany.

Microbiological characterization and MALDI-TOF MS species identification
We performed a retrospective search of our microbiological database to identify S. massiliensis isolates in blood cultures from January 2017 to December 2020. We screened the database for blood culture isolates deposited as "coagulase-negative staphylococci," suggesting that a definitive identification with MALDI-TOF MS failed in these isolates. All samples were processed in accordance with standard microbiological procedures, as described previously [4]. For isolate identification, a combination of MALDI-TOF MS (Biotyper™; Bruker Daltonics, Bremen, Germany, with the MBT Compass 4.1 database containing 7588 bacterial pathogens but not including S. massiliensis) according to the manufacturer's protocols and sequencing of the 16S-rRNA coding gene was utilized (see below). Antimicrobial susceptibility testing (AST) was carried out on a VITEK II (BioMérieux, Marcy l'Étoile, France) and interpreted in accordance with the guidelines of the European Committee on Antimicrobial Susceptibility Testing (EUCAST, version 10.0). Data on the number of positive blood culture bottles and the time to positivity was collected.

Sequencing of the 16S rRNA coding gene
A fragment of the 16S-rRNA coding gene was amplified, using the primer sequences (5′-3′) AGA GTT TGATC-MTGG CTC AG (forward) and CCG TCA ATTCMTTT GAG TTT (reverse), respectively. Subsequently, the PCR product was processed with the AMPure PCR purification kit (BeckmanCoulter, Krefeld, Germany) and dye-labeled using the DTCS Quick Start Kit (Sciex, Framingham, USA) with the primer sequences (5′-3′) AGA GTT TGATCMTGG CTC AG (forward) and GWA TTA CCG CGG CKGCTG (reverse), respectively. The dye-labeled products were purified with the CleanSEQ Dye-Terminator Removal Kit (BeckmanCoulter, Krefeld, Germany) and sequenced via capillary electrophoresis (GenomeLab GeXP; BeckmanCoulter, Krefeld, Germany). The forward and reverse strands were processed and analyzed individually. Sequencing and all kit reactions were performed according to the manufacturer's protocols. The PCR protocols are displayed in Table 1.

Whole-genome sequencing
DNA extraction and short-read Illumina sequencing were performed as previously described [14]. For correct species designation, core genome alignment was performed with publicly available Staphylococcus haemolyticus and Staphylococcus saprophyticus genomes, which showed the closest phylogenetic sequence homologies. The sequences were deposited at the NCBI GenBank under the Bioproject PRJNA750592.

Patient characteristics
Review of the patient charts was performed to extract clinical data, including basic demographics, comorbidities, concomitant infectious syndromes and pathogens detected, presence of foreign bodies, and the antibiotic treatment.

Ethics
This study was approved by the local ethics committee (Number 147/21). Due to the retrospective nature of the study and the anonymization of patient data, the need for informed consents was waived.

Microbiological characterization
Between January 2017 and December 2020, a total of 203 isolates were found to be deposited as "coagulase-negative staphylococci" in our microbiological database. Of these, 181 were in a second step identified per MALDI-TOF MS as Staphylococcus epidermidis or other, common species, and in 13 cases, a polymicrobial culture was found and/or the initial suspicion of CoNS was refuted. In the remaining nine cases, MALDI-TOF MS yielded either Staphylococcus sp. or no reliable identification. Out of this cohort of nine isolates, four blood culture bottles from three different patients were positive for S. massiliensis. The clinical characteristics are provided in Table 2. MALDI-TOF MS yielded no reliable identification of all three isolates (all scores ≤ 1.47, for detailed information refer to Table 3), which were subsequently diagnosed as S. massiliensis via sequencing of the 16S-rRNA coding gene and further investigated by whole-genome sequencing (WGS). All isolates were methicillin-susceptible (Table 4).

WGS
The draft genome was analyzed using KmerFinder and two different species were suggested as best hit: S. haemolyticus and S. saprophyticus. S. massiliensis is not included in the database of KmerFinder; therefore, we compared our sequences to the 3 publicly available genomes of S. massiliensis as well as the complete genome publicly available in the refseq database. The core genome used for the comparison was relatively small (27 genes, 1661 polymorphic sites) but revealed a close phylogentic relationship between our isolates and the 3 genomes belonging to S. massiliensis ( Fig. 1A; Table 5). To validate these results, we refined the core genome (1987 genes, 3093 polymorphic sites) using only the common genes among those 6 isolates. This analysis validated the genetic identity of our isolate as S. massiliensis with a higher confidence. These results are also in line with the average nucleotide identity (ANI) of 99.29% (99.19-99.38%) between our 3 isolates and the 3 available genomes of S. massiliensis, with an ANI value > 95% being a reliable indicator for species identity [15]. Although our analyzed isolates all belonged to S. massiliensis, they showed a different resistome pattern (Fig. 1B). We found a high level of resistance with 6 resistance genes in isolate P3, while the reference genomes were overall more susceptible to antibiotics.

Discussion
We describe here the incongruency of bacterial species identification for S. massiliensis using three different methods, namely sequencing of the 16S rRNA coding gene, MALDI-TOF MS, and WGS. While Sanger sequencing of the 16S rRNA coding gene correctly identified S. massiliensis, both mass spectrometry and next-generation sequencing were unable to clearly distinguish S. massiliensis from closely related staphylococcal species. Indeed, Zong previously reported the misidentification as Staphylococcus simulans by the MicroScan Walkaway 96 system [13]. Our results demonstrate that even modern and high-resolution identification techniques, such as MALDI-TOF MS and WGS, are dependent on the quality and currentness of the database.
Although WGS is frequently regarded as a new gold standard for the molecular characterization of bacteria, the used databases are not necessarily reliable for newly described  [16]. The first description of S. massiliensis dates back to 2010, when Al Malsalma and colleagues reported the detection of this staphylococcal species from clinical specimen of a 52-year-old male patient (from 2005) with a brain tumor [11]. He developed a cerebral abscess after neurosurgery, which was drained and sent for microbiological diagnostics. Indeed, CoNS can cause post-surgical intracranial infections, rendering the findings of Al Malsalma et al. plausible. However, S. massiliensis has also been suggested as commensal of the human skin as it has been found in the skin of a fresh wound and healthy human skin [12,13]. On  [17], we detected the presence of multiple resistance genes in one of our isolates. The presence of clinically relevant antimicrobial resistance genes in skin commensals may pose therapeutic challenges and may be relevant for the acquisition of antibiotic resistances in facultative pathogens [18], such as members of the S. aureus complex, which warrants further investigations. In conclusion, our data add to the scarce body of evidence that S. massiliensis may be detected in clinical samples for routine microbiological diagnostics. Thus far, the clinical relevance is still unclear and further clinical and experimental studies are needed to elucidate the pathogenic properties and virulence of S. massiliensis. The technical advancement in microbiological diagnostics will introduce new depths in species identification and specification of novel organisms within the next years. Code availability Not applicable.

Declarations
Ethics approval This study was approved by the local ethics committee (Number 147/21).

Consent to participate
Due to the retrospective nature of the study and the anonymization of patient data, the need for informed consents was waived.

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.