Molecular basis of surface anchored protein A deficiency in the Staphylococcus aureus strain Wood 46

Protein A in Staphylococcus aureus is encoded by the spa (staphylococcal protein A) gene and binds to immunoglobulin (Ig). The S. aureus strain Wood 46 has been variously reported as protein A-deficient and/or spa negative and used as a control in animal models of staphylococcal infections. The results of this study indicate that Wood 46 has normal spa expression but transcribes very low levels of the srtA gene which encodes the sortase A (SrtA) enzyme. This is consistent with unique mutations in the srtA promoter. In this study, a low level of sortase A explains deficient anchoring of proteins with an LPXTG motif, such as protein A, fibrinogen-binding protein and fibronectin-binding proteins A and B on to the peptidoglycan cell wall. The activity of secreted protein A is an important consideration for use of Wood 46 in functional experiments and animal models.


Introduction
Protein A produced by S. aureus is a potent virulence factor and immune-modulator [1][2][3][4][5]. It is a 40-60 kDa cell-wall protein encoded by the spa gene [6,7]. It is synthesized by nearly all S. aureus isolates and binds to the Fc and F(ab) 2 regions of immunoglobulins [8,9], prevents opsonophagocytic killing [1,10,11] and serves as a B-cell superantigen [12][13][14]. Furthermore, marginal zone B cells and B-1 cells undergo, preferentially, induced cell death with supraclonal depletion and immune tolerance upon exposure to protein A [15]. Thus, protein A may interfere with vaccine efficacy that depends on opsonophagocytic antibody production by binding to B-cells and preventing their function. The S. aureus strain Wood 46 had been considered spa negative and protein A-deficient. For this reason, it was used as a negative control in studies involving S. aureus virulence [1,[16][17][18][19][20][21][22]. There is evidence that Wood 46 produces extracellular protein A during exponential growth phase and steady state [18]. This has been demonstrated indirectly by immunoglobulin binding experiments [18]. Extracellular protein A has also been isolated from Wood 46 using IgG Sepharose chromatography, although the amount of putative protein A recovered from Wood 46 is lower compared to S. aureus strain Cowan 1, a high producer of protein A [17,18]. S. aureus a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 produces a second immunoglobulin-binding protein (Sbi), which is also capable of binding to the Fc portion of IgG similar to protein A and protects the bacteria against innate immune responses [23].
Sortase A (SrtA) is a transpeptidase commonly produced by Gram-positive bacteria. It has specificity for proteins with an LPXTG motif and cleaves between the threonine (T) and glycine (G) residues to form an acyl enzyme intermediate that is relieved by the nucleophilic attack of the amino group in the pentaglycine cross bridge of the peptidoglycan precursor [24]. Modified proteins are subsequently incorporated into the peptidoglycan cell wall and displayed on the surface [24][25][26][27][28][29][30]. S. aureus isolates encode approximately 17-21 surface proteins with an LPXTG motif [4,31,32]. These include protein A, fibrinogen-binding proteinsclumping factors A and B (Clf A/B) and fibronectin binding proteins (FnBP A and FnBP B) that function as potent virulence factors [33]. S. aureus mutants lacking SrtA fail to covalently attach proteins with an LPXTG motif onto their cell wall, and are thus less virulent than their wild-type counterparts. They do not form abscess lesions in organs or cause bacteremia in mouse models [32,34]. In this study, the reason for protein A deficiency on the surface of Wood 46 was investigated and the defective expression of spa and srtA genes encoding functional proteins was examined as a possible cause.

Materials and methods
Bacterial strains and growth conditions S. aureus strains Wood 46 (ATCC 10832), Seattle 1945 (ATCC 25923) and Cowan 1 (ATCC 12598) were obtained from the American Type Culture Collection. The S. aureus Newman strain and its corresponding spa and sbi mutants (Newman WT, Nweman sbi::Em r , Newman spa::Ka r and Newman sbi::Em r spa::Ka r ) were a kind gift from Drs. Tim Foster and Joan Geoghegan (Department of Microbiology, Trinity College Dublin) [23]. Bacteria were grown overnight in Trypticase Soy Broth (TSB, Becton Dickinson, Franklin Lakes, New Jersey) at 37˚C in a shaker incubator and diluted 1:50 in fresh TSB to initiate log-phase cultures.

DNA extraction and conventional PCR
Bacterial DNA was extracted from overnight cultures using a commercial kit (MoBio Ultra Clean DNA Isolation Kit, MoBio, Carlsbad, CA) according to the manufacturer's protocol. PCR for spa, srtA and the -35 regulatory regions was performed using primers listed in Table 1. The following cycling conditions were performed: initial denaturation at 95˚C for 90s, annealing at 55˚C for 30s, extension at 72˚C for 1min (30 cycles) and a final extension at 72˚C for 5min. PCR products were sequenced at the Genomics Core Facility (University of Tennessee, Knoxville) and the sequences of spa, srtA and the -35 regulatory region were aligned and compared using Geneious (version 9.1.6) (Biomatters, Auckland, New Zealand).

RNA extraction and quantitative real-time PCR
Total RNA was extracted from log phase bacterial cultures using a commercial kit (MoBio Ultra Clean RNA Isolation kit, MoBio, Carlsbad, CA) according to the manufacturer's protocol. Quantitative Real-Time PCR (qPCR) for spa and srt A were performed using the TaqMan RNA-to-C T 1-Step kit on an ABI StepOne PCR System (Applied Biosystems, Foster City, CA). The primers and probe used for the qPCR are listed in Table 1. The following cycling conditions were performed: 48˚C for 30min, 95˚C for 10min, 95˚C for 15s, 60˚C for 1min (40 cycles). qPCR data were analyzed by the ΔΔC T method and normalized against 16S rRNA as the endogenous control as previously described [35].

Protein A ELISA
Secreted protein A in bacterial culture supernatants was measured using an antigen capture ELISA. Briefly, 96-well Costar plates (Corning Inc, Corning, NY) were coated with 1μg/ml of mouse monoclonal anti-protein A antibody clone SPA-27 (Sigma-Aldrich, St. Louis, MO) in PBS pH 7.2. Bacterial supernatant was added and incubated at 37˚C for 1 hour. Bound antigen was detected using a 1:500 dilution of HRP-conjugated chicken anti-protein A antibody (Gallus Immunotech, Cary, NC) followed by addition of 100μl/well of tetramethylbenzidine (TMB) substrate (Sigma-Aldrich, St. Louis, MO). Plates were washed in between each step with PBS containing 0.05% polysorbate-20 using an EL X 405 auto plate washer (BioTek Instruments Inc., Winooski, VT). The reaction was stopped by adding 50μl/well of 0.18M sulfuric acid. The optical density was measured at 450nm using an EL X 800 Universal Microplate Reader (BioTech Instruments Inc. Winooski, VT). Commercially available S.aureus Protein A (Thermo Scientific, Rockford, IL) was used to generate the standard curve.
Flow cytometry to measure cell wall-associated protein A, fibrinogenbinding protein and fibronectin-binding protein One milliliter of log phase bacterial cells were washed and incubated with either no conjugate (NC, negative control), 2μg/ml of chicken anti-protein A FITC conjugate (Gallus Immunotech, Cary, NC), 1μg/ml human fibrinogen FITC conjugate (Zedira GmbH, Darmstadt, Germany) or 1μg/ml bovine fibronectin HiLyte Fluor TM 488 conjugate (Cytoskeleton Inc., Denver, CO) for 30 min at room temperature in the dark. Excess conjugate was removed by washing and the amount of binding determined by measuring fluorescence using an Attune acoustic focusing cytometer (Applied Biosystems, Foster City, CA) at excitation/emission wavelengths of 488/519 nm.

Statistical analysis
A one-way ANOVA and multiple comparisons using Tukey's honest significant difference were performed to test if there was a significant difference in the expression of spa and srtA among bacterial isolates, and also to determine if there was a significant difference in the amount of surface-expressed protein A, fibrinogen-binding protein and fibronectin-binding protein. Each experiment was repeated at least three times and a p-value of <0.05 was Table 1. List of primers used in this study.

Gene
Primers Reference This study This study This study This study https://doi.org/10.1371/journal.pone.0183913.t001 considered significant. All analyses were conducted using the IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp, Armonk, NY).

Wood 46 expresses lower amounts of cell wall-associated protein A
The S. aureus isolate Wood 46 had been reported as spa-negative and protein A-deficient although there have been other studies that demonstrated that it binds small quantities IgG suggesting that it may have a low density of protein A on its surface [16][17][18][19][20][21][22]. Flow cytometry using chicken anti-protein A antibody was performed to determine the amount of cell wallassociated protein A.  (Fig 2). These data indicate that Wood 46 produces and secretes protein A, although the amount is lower compared to Seattle 1945 or wild-type Newman isolates. The lower limits of detection and quantification of the assay were 0.125ng/ml and 0.3ng/ml respectively.
Gene sequence analysis of spa, srtA and regulatory region of srtA It was hypothesized that if the spa gene in Wood 46 contains a mutation, it may lead to a truncated or non-functional protein. In order to test this hypothesis, the spa gene in Wood 46 and control S. aureus isolates was amplified using PCR and the products were sequenced and analyzed. Results from the PCR and DNA sequence analysis of the spa gene showed that Wood 46 not only possesses the full-length spa gene, but also the sequence of the spa gene is in agreement with those of Cowan 1, Seattle 1945 and Newman WT. Therefore, mutation in the spa gene was ruled out as the possible cause for lower expression of Protein A in Wood 46. Sequence analysis of the srtA gene in Wood 46, however, showed that it has a 12bp deletion at the 3' end of the gene that results in a four amino acid difference in the protein compared to S. aureus Cowan 1, Seattle 1945 and wild-type Newman (Fig 3A). Also, analysis of the upstream promoter and regulatory region showed that Wood 46 contains mutations in both the -10 and -35 regions (Fig 3B). These mutations likely adversely affect RNA polymerase binding and transcription as mutations in the upstream promoter and regulatory regions have been associated with reduced gene expression [38][39][40].

mRNA levels of spa and srtA in Wood 46
If the transcription of spa is adversely affected, it could lead to lower expression of protein A on the bacterial surface. Reverse transcriptase qPCR was used to analyze the expression of spa in S. aureus isolates. The results showed that Wood 46 transcribes spa mRNA although the levels were lower compared to Cowan 1 which was assigned as the calibrator strain for this experiment (Fig 4A). The expression levels of spa in Seattle 1945, Newman WT and Δsbi were 0.74 fold, 3.9 fold and 4.18 fold respectively. As expected, the Δspa and Δsbi-Δspa mutants did not produce detectable spa mRNA. This suggests that inefficient spa transcription was not the reason for low surface expression of Protein A in Wood 46.
Conversely, the expression level of srtA mRNA in Wood 46 was below the limit of detection of the assay. Cowan 1 was assigned as the calibrator strain for this experiment. The expression levels of Seattle 1945 and Newman WT were 0.26 fold and 0.28 fold relative to Cowan 1, while that of the deletion mutants Δsbi, Δspa and Δsbi,Δspa were 0.14 fold, 1.03 fold and 0.51 fold respectively (Fig 4B). Taken together, the results suggest the lack of sortase A could be responsible for the deficient anchoring of protein A in Wood 46. The spa and srtA expression levels were normalized against 16S rRNA as the endogenous control and the results were expressed as fold change in mRNA levels using the ΔΔC T method. To ensure equal amplification efficiencies of the target and control genes, four 10-fold serial dilutions of the target RNA were tested. The efficiencies of both the target gene and the endogenous control were comparable. Wood 46 expresses lower amounts of the LPXTG motif-containing fibrinogen-binding protein and fibronectin-binding protein Sortase A is responsible for anchoring proteins with an LPXTG motif on to the peptidoglycan cell well [24,27,28,30,41]. If lack of this enzyme is indeed the reason for low surface density of protein A in Wood 46, there should be a global decrease in the expression of all surface proteins that bear the LPXTG motif. In order to test this hypothesis, two well-studied proteins with LPXTG motifs were chosen. Fibrinogen-binding protein (Clf A and B) and fibronectinbinding protein (FnBP A and B) are anchored on the bacterial surface and known immune evasion factors in S. aureus. A binding assay was performed with FITC-labeled human fibrinogen to detect surface-expressed fibrinogen-binding protein, and HiLyte Fluor TM 488 conjugated bovine fibronectin to detect surface-expressed fibronectin-binding protein. Flow cytometry results showed that Wood 46 bound only 0.49% of labeled fibrinogen on its surface  compared to the Δspa isolate which bound the highest amount of fibrinogen in this study ( Fig  5A). Also, Wood 46 bound only 11.4% of the labeled fibronectin relative to Cowan 1 (considered as 100%) and Seattle 1945 (31.2%) (Fig 5B). The Newman isolate and its corresponding deletion mutants were not tested in the fibronectin-binding assay because they have truncations in both FnBP A and FnBP B [42]. These data suggest that lack of sortase A in Wood 46 likely results in lower amounts of not just protein A, but other proteins that contain the LPXTG motif.

Discussion
S. aureus produces a number of virulence factors including secreted toxins, enzymes and cell wall-associated proteins that bind to and interact with host extracellular matrix [43]. In an in vitro environment, the production of virulence factors is growth-phase dependent [44]. The secreted proteins are produced during post-exponential (stationary) phase wheras the cell wall-associated proteins are predominantly produced during the logarithmic growth phase [45,46]. Many of the known virulence factors such as protein A, fibrinogen-binding proteins like clumping Factors A and B, fibronectin-binding proteins A and B are synthesized and anchored onto the peptidoglycan cell wall by the transpeptidase enzyme, sortase A [27,30,41,47].
The S. aureus isolate Wood 46 has played an important role in studies designed to examine the biological function of protein A and its role in pathogenesis and virulence. There have been contradictory reports on the production, display and secretion of protein A [18,48]. The results have either been ambiguous or lacked definitive experimental data. Wood 46 has been described as a spa-negative and protein A-deficient strain although there is evidence that Wood 46 expresses low amounts of protein A on its surface as evidenced by its ability to bind low amounts of IgG [18,49]. It is also reported that Wood 46 secretes small amounts protein A into the culture supernatant during exponential growth phase [17,18]. Recently, several mutant strains have replaced Wood 46 as the preferred choice for negative control in both in vitro and in vivo studies. However, the mechanism for low expression of protein A on the surface of Wood 46 had not been previously investigated.
Cell wall-associated protein A binds to IgG via its Fc region, thereby preventing it from opsonization and phagocytosis [50]. In this study, it was observed that Wood 46 had a lower density of protein A on its surface, and secreted less protein A compared to the Cowan 1, Seattle 1945 or Newman isolates. It was observed that even the Δspa isolate bound some anti-protein A antibody. This could be due to the ability of the antibody to react with other immunoglobulin-binding proteins, such as Sbi, on the surface of the bacterium. Secreted protein A functions as a superantigen (SAg) [13]. It binds to VH 3 -clan B-cell receptors and limits their abilities to produce antibody. Also, marginal zone B-cells and B1-cells undergo apoptosis upon exposure to protein A [13-15, 29, 51, 52]. There have been reports of secreted protein A in Wood 46 activating human B-cells in a mechanism that is different from that seen in Cowan 1 [53]. Therefore, it is important to consider the toxic effects of secreted protein A, if Wood 46 is used as a control in animal models to test vaccine efficacy.
Conventional PCR and sequencing showed that Wood 46 possesses the spa gene with a complete open reading frame. The gene sequence is in agreement with that of Seattle 1945, Cowan 1 and Newman. qPCR for spa expression showed that Wood 46 does transcribe spa mRNA but at lower amounts compared to that of Cowan 1, Seattle 1945 and Newman isolates. Because this did not explain the low surface levels of protein A, the transpeptidase enzyme sortase A responsible for anchoring surface proteins on the peptidoglycan cell wall, was examined. Conventional PCR with srtA-specific primers showed that Wood 46 contains the srtA gene with a complete open reading frame. However, it harbors a 12bp deletion at the 3'-end of the gene resulting in a corresponding loss of four amino acids at the N-terminus of the protein sequence. It is not known if the loss of the amino acids affects the function of the protein in any way. Further analysis of the upstream region of the srtA gene shows mutations in the -10 and -35 regions of the promoter and regulatory that would be expected to adversely affect binding of RNA polymerase and transcription. The functional effects of these mutations are yet to be evaluated and may be pursued at a later stage. Promoter site mutations have been associated with impaired transcription and in turn affect protein expression [38][39][40][54][55][56][57]. These mutations could affect the expression of srtA which in turn may be responsible for the low surface density of protein A. If this is true, there should be a decrease in the expression of other surface proteins that harbor the LPXTG motif. This was confirmed by examining the surface expression of two known LPXTG proteins, fibrinogen-binding protein (Clf A and Clf B) and fibronectin-binding protein (FnBP A and FnBP B). Both these proteins in S. aureus are virulence factors similar to protein A [24,27,[29][30][31]58]. Wood 46 expressed very low amounts of both fibrinogen-binding protein and fibronectin-binding protein on its surface suggesting that it may express lower amounts of all surface proteins that have an LPXTG motif. Previous studies have shown that a portion of SpA is released into the supernatant with an intact sorting signal, and release of SpA was reduced when the native sorting signal of SpA was replaced with the corresponding region of another sortase-anchored protein, namely, SdrE [59]. Since replacing the native sorting signal and determining its effects on secreted protein A were beyond the scope of this study, we instead addressed this concern by examining the sequence of secreted protein A in Wood 46 and one of the control strains, Cowan 1 by mass spectroscopy. In both Cowan 1 and Wood 46, we found peptide hits that matched with the LPETG motif suggesting that it is indeed the low level of sortase A in Wood 46 that is responsible for the low surface level of LPXTG proteins in this strain (S1 Fig). The protein abundance from mass spectrometry in each strain showed that Cowan 1 had a higher amount of protein A in the supernatant compared to Wood 46 (S1 Table) which could be correlated to the ELISA data. LytM triggers polypeptide release from the bacterial envelope. These released proteins have a modified anchor structure [36]. If LytM is inhibited, then there should be lower amounts of released proteins with the modified anchor structure. We confirmed this by mass spectrometry. We obtained peptide hits to protein A with an LPETG motif both in Cowan 1 and Wood 46 suggesting that it was indeed low amounts of sortase A in Wood 46 that is responsible for defective anchoring of protein A on the bacterial cell wall (S2 and S3 Figs). Nacetylglucosamine is a known inhibitor of LytM [60]. If LytM was inhibited, there would be decrease in the amount of secreted protein A in bacterial culture supernatants. In order to test this, bacterial strains Cowan 1 and Wood 46 were grown to mid log-phase in the presence and absence of 10mM N-acetylglucosamine. The amount of secreted protein A was measured by capture ELISA. It was found that although there was a slight reduction in the amount of protein A produced in each strain, the difference was not significant. If murein wall hydrolases were responsible for defective anchoring/secretion of protein A, there would have been a significant difference in the amount of cell wall-associated and secreted protein A in both Cowan 1 and Wood 46. Since this was not the case, it was concluded that the low level of sortase A in Wood 46 is responsible for defective anchoring of protein A on the surface of Wood 46. Taken together, the results strongly suggest that low amounts of sortase A are responsible for impaired surface expression of proteins with an LPXTG motif.
The complete coding sequence of the Wood 46 srtA gene along with the upstream promoter and regulatory region has been deposited in GenBank (accession number: KY514391). The whole genome sequence of Wood 46 has been determined and deposited in GenBank (accession number: MTFQ00000000, version MTFQ00000000.1). This could provide a wealth of information about other immunoglobulin-binding proteins with or without an LPXTG motif. It will also help identify novel genes whose products may be involved in immune evasion. Sequence information about promoters and regulatory elements of virulence genes could help explain why some strains of S. aureus are innately less virulent compared to other highly virulent strains.
In conclusion, the results from this study suggest that defective expression of srtA in Wood 46 accounts for the failure of protein A anchoring to its surface with the majority of the protein A being secreted. This may be accompanied by deficient anchoring of all LPXTG-bearing surface proteins. The Wood 46 promoter at both the -10 and -35 regions contained mutations compared to the control strains used in this study. Promotors that vary from consensus sequences generally are associated with decreased gene expression [54,55,57]. Understanding the nature of surface protein expression and its role in virulence is crucial towards successful vaccine development against S. aureus.