Expression, immunogenicity and variation of iron-regulated surface protein A from bovine isolates of Staphylococcus aureus

Misra, Nb., Wines, T.Fa*, Knopp, C.La, McGuire, M.Ac. and J.K. Tinkerab#

Department of Biological Sciencesa and Biomolecular Ph.D. Programb, Boise State University, Boise, Idaho, USA; Department of Animal and Veterinary Science, University of Idaho, Moscow, Idaho, USAc

Running title: Staphylococcus aureus IsdA variation

# Address correspondence to Juliette K. Tinker, [email protected]

Abstract Staphylococcus aureus iron-regulated surface protein A (IsdA) is a fibrinogen and fibronectin adhesin that also contributes to iron sequestration and resistance to innate immunity. IsdA is conserved in human isolates and has been investigated as a human vaccine candidate. Here we report the expression of isdA, the efficacy of anti-IsdA responses, and the existence of IsdA sequence variants from bovine Staphylococcus. Clinical staphylococci were obtained from U.S. dairy farms and assayed by PCR for the presence and expression of isdA. isdA positive species from bovines included S. aureus, S. haemolyticus and S. chromogenes. Immunoassays on bovine milk and serum confirmed the induction and opsonophagocytic activity of anti-IsdA humoral responses. The variable region of isdA was sequenced and protein alignments predicted the presence of two main variants consistent with those from human S. aureus. Mouse antibodies against one IsdA variant reduced staphylococcal binding to fibronectin in vitro in an isotypedependent manner. Purified IsdA variants bound distinctly to fibronectin and fibrinogen. Our findings demonstrate that variability within the C-terminus of this adhesin affects immune

reactivity and binding specificity, but are consistent with the significance of IsdA in bovine disease and relevant for vaccine development.

Introduction Bovine mastitis is an inflammatory condition of the udder and regarded as one the most important diseases affecting dairy cattle worldwide. Multiple pathogens are capable of colonizing the udder; however, infection caused by Staphylococcus aureus is among the most relevant and economically impactful, resulting in clinical or subclinical disease that affects milk yield and somatic cell count (SCC) (Barkema et al., 2006, Bar et al., 2008, De Vliegher et al., 2012, Haran et al., 2012). Antibiotic therapy is often ineffective, and longitudinal transmission of resistant strains is an increasing concern with implications for veterinary and public health. While efforts have focused on management practices, an effective S. aureus vaccine for dairy heifers would improve animal health, increase milk production and reduce agricultural dependence on antibiotics. Vaccine strategies must include conserved virulence determinants and account for antigen sequence variation. The S. aureus iron-regulated surface determinant A (IsdA) is a fibrinogen (FG) and fibronectin (FN) adhesin that contributes to iron sequestration and the characteristic ability of S. aureus to disseminate (Clarke et al., 2004, Torres et al., 2006). The presence of IsdA is conserved among human isolates of S. aureus, and studies have described IsdA as a central human vaccine candidate (Clarke et al., 2006, Stranger-Jones et al., 2006, Kim et al., 2010, Arlian & Tinker, 2011, Stapleton et al., 2012). Full-length isdA is present in the bovine strain Newbould 305, as well as 100% of S. aureus isolates causing mastitis in ruminants from an international genotyping effort (Wolf et al., 2011, Bar-Gal et al., 2015). IsdA is also expressed at

2

higher levels in strains more likely to cause mastitis in ruminants (Le Marechal et al., 2011). S. aureus utilizes the Isd system, which is composed of IsdA and at least four additional cell wall anchored proteins (IsdB, C, H and I), to chelate iron from host heme. Each of these proteins contains conserved stretches of 125 amino acids called the near-iron transporter (NEAT) domain to bind heme for transport to the cytosol and iron-scavenging (Hammer & Skaar, 2011). IsdA is also a physiologically relevant extracellular matrix (ECM) adhesin to FG and FN that promotes binding to epithelial cells, and is required for colonization of rat nares (Clarke et al., 2004, Clarke et al., 2006, Clarke et al., 2009, Corrigan et al., 2009). The FG-binding site lies within the NEAT domain; however, the function of the IsdA C-terminus is less well understood. A cell-wall binding (LPxGT) motif is present, and the C-terminus is involved in surface hydrophobicity that confers resistance to fatty acids and antimicrobial peptides, such that IsdA is required for growth on human skin (Clarke et al., 2007). These reports indicate that IsdA is an important S. aureus virulence factor that may contribute to infection of the bovine udder. The aim of the current study was to confirm the presence and expression of isdA from bovine isolates of staphylococci as well as the efficacy of anti-IsdA responses. In addition, we identified isdA allelic variants that affect immune reactivity and binding specificity. These findings support the incorporation of IsdA into a multicomponent S. aureus bovine vaccine but acknowledge that sequence variation will be an important consideration.

3

Materials and Methods Bovine milk, serum and staphylococcal collection. Blood and quarter milk were taken from observationally healthy lactating Holstein cows from Dairy 1 (D1; n=53) and Dairy 2 (D2; n=50) in the Northwest U.S. All bovine sample collection protocols were pre-approved by the University of Idaho Institutional Animal Care and Use Committee (IACUC). Somatic cell counts (SCC) were determined using the Delaval method (Tumba, Sweden). Milk (100 l) was plated on mannitol salt agar (MSA) and MP2 Agar™ (Udder Health Systems Inc., Tacoma WA) and cultured at 37C for 24 h. Samples with yellow growth on MSA, plus growth in the absence of esculin hydrolysis on MP2 were presumptive staphylococci (Supplementary Table 1). Cultures were incubated in lysis buffer (20mM Tris Cl, 2 nM EDTA, 1.2% Triton X-100 + 20 mg/mL lysozyme, pH 8.0) for 30 min at 37°C prior to DNA extraction (DNeasy, Qiagen, Valencia, CA). Milk was centrifuged to remove fat layers. Blood was allowed to coagulate at room temperature. Milk and serum were diluted 1:10 in protease inhibitor (1:100 Thermo-Fisher HALT, 5% glycerol in 1X PBS) and frozen at -20 prior to ELISA. An additional 25 bovine S. aureus isolates from clinically mastitic cows on 15 farms in 6 U.S. states were obtained (Udder Health Systems, Inc.) These strains had been analyzed by mass spectrometry (MALDI). Species identification of all staphylococci was confirmed by 16S rRNA sequencing (GE Healthcare, SeqWright, Houston, TX). Gene presence/absence was determined by electrophoresis of PCR performed under optimized conditions, and amplicon sizes are as indicated (Table 1). RT-PCR, ELISA, western and opsonophagocytosis. Total RNA was isolated from milk as described (Mura et al., 2013). Briefly, milk was centrifuged at 3000 xg for 15 min at 4°C and pellets washed (1X PBS + 0.5 mM EDTA). RNA was extracted (RNeasy, Qiagen), with an additional DNase (Promega, Madison, WI) treatment (1U/100μl, 30 min). cDNA was transcribed as described by the manufacturer (High Capacity cDNA RT kit, Thermo Fisher) and included no RT controls. 16S rRNA and isdA PCR was conducted on 2 L of cDNA with primers (Table 1) and GAPDH control. PCR (15 L) was analyzed on 1% agarose with positive (Newbould 305) and negative (no template). For milk and serum ELISA, a HIS-IsdA fusion (pBA015, Table 1) was purified with cobalt chromatography as described (Arlian & Tinker, 2011). Plates (Nunc, Thermo-Fisher) were coated with 50 µL of 10 µg/mL IsdA, and incubated for 12 h at 4°C followed by washes (1x PBS + 0.05% Tween 20). Blocking buffer (1x PBS + 1% skim goat milk) was added for 2 h at 37°C. 2-fold dilutions, starting at 1:2 for milk, and 1:10 for serum, were added and incubated at 4°C for 12 h prior to washing and addition of HRP-anti-bovine IgG (1:10,000 Bethyl Laboratories, Montgomery, TX). Samples were developed with tetramethylbenzidine (TMB One, Thermo-Fisher). Endpoint titers were defined as the reciprocal of the dilution giving an O.D. of 0.2. Staphylococcus + and isdA +/- cows used in ELISA are shown (Supplementary Table 1). Ten cows from each farm, with both high and low SCC, and negative for Staphylococcus by culture, were also assayed (Staphylococcus-isdA-). ELISAs are by quarter for milk and by cow for blood. For westerns, IsdA (pBA015, 62 kD) was separated on 12% SDS-PAGE, transferred to nitrocellulose strips and incubated in blocking buffer (0.05% Tween-20 + 5% skim milk + 1X PBS) before probing with milk at 1:500. HRP-anti-bovine IgG (1:5000 Bethyl Labs, Montgomery, TX) was added and strips were developed with Immobilon Western HRP substrate (Millipore, Billerica, MA, USA). Opsonophagocytosis (OSP) was performed as described (Kurokawa et al., 2013) with the addition of purified IsdA. IsdA (pCK001) was purified using cobalt as above for pBA015. Equal volumes of 50g heat4

inactivated serum was mixed with 50 g purified IsdA and incubated for 1 hr at 37°C. Equal volumes of S. aureus Newbould 305 (2x10^6) and bovine PBMCs (2x10^6) were added to the reaction and incubated at 37°C. At 60 min, cells were plated on MSA in triplicate. Results are reported as bacterial survival in CFU/mL and compared to survival without PBMCs. Sequence alignments and predicted tertiary structure. The variable region of IsdA was predicted to comprise amino acids 222-308 (NCBI). Primers that encompass this region were used for isdA PCR and sequencing (GE Healthcare). Alignments and evolutionary analysis were completed in MEGA 7(Kumar et al., 2016) . A rooted dendrogram of exhaustive pairwise IsdA alignments of 53 sequences (Supplemental Table 2) was constructed using Maximum Likelihood based on the JTT matrix-based model (Jones et al., 1992). The I-TASSER server was used to visualize predicted IsdA tertiary structures (Yang et al., 2015). Model templates included the S. aureus hemoglobin-IsdH complex and the NEAT domain of IsdA from S. aureus (Grigg et al., 2007, Grigg et al., 2011, Honsa et al., 2013, Dickson et al., 2014). The Newbould 305 model has a C-score of -1.93 (estimated TM-score of 0.48±0.15), and the MRSA252 model has a C-score of -2.46 (estimated TM-score of 0.43 ± 0.14). In the I-TASSER server, the range of C-scores is 5 to 2, with >-1.5 indicating correct global topology. Anti-IsdA mouse serum. Mouse immunization protocols were pre-approved by the Boise State University IACUC. Immunizations were performed as described (Arlian & Tinker, 2011). Briefly, two groups of 6 female BALB/c mice, 7 to 9 weeks old (Taconic, Hudson, NY), were administered 17 g purified IsdA (MRSA252 variant III, pBA009) + 5 g native cholera toxin (CT, Sigma Aldrich, St. Louis, MO), or mock (PBS) in 10 L 1XPBS applied to each nare by pipette under light anesthesia on days 0 and 10. Blood samples were obtained by lateral tail vein on day 14. Blood was allowed to coagulate and serum diluted 10-fold in protease inhibitor (as above). IsdA-specific antibodies in serum were analyzed by ELISA as described and endpoint titers defined as the reciprocal of the dilution giving an O.D. of 0.2 (Arlian & Tinker, 2011). Adhesion blocking and ECM binding assays. For blocking assays, black walled (Nunc, Thermo-Fisher) 96-well plates were coated with 50 µL of 10 µg/mL bovine FN (R&D Systems, Minneapolis, MN ) for 12 h at 4°C. A 1:100 dilution of heat inactivated pooled mouse serum was mixed with 1x108 CFU of S. aureus MRSA252 or Newbould 305, grown in low iron media (LIM: 2 g NaCl, 1.2 g NaHCO3, 1.6 g yeast extract, 6 g proteose peptone/400 mL), for 30 mins at 37C. 100 L of the mixture was added to plates coated with FN and incubated at 37°C for 12 h prior to washing and addition of 150 μL Alamar Blue (0.1% reazzurin in 1X PBS). Plates were read at 530/590 nm after 24 h. For binding assays, IsdA from pBA009A and pCK001 was purified as above and separated on 12 % SDS-PAGE. Purified IsdA (100 g) was added in 2fold dilutions to 96-well plates coated with 10 g/mL bovine FN or bovine FG (Alfa Aerar, Reston, VA) and incubated at 37°C for 12 h prior to washing and addition of mouse anti-HIS (1:1000 Sigma) and anti-mouse HRP secondary (1:10,000 Promega). Plates were developed (TMB One) and read at 370 nm. Graphing and Statistical Analysis. Graphing and statistical analysis was performed with JMP SAS software (Cary, NC). PCR and RT-PCR is representative of three independent assays, ELISAs are reported as the means of three independent assays, and the OSP, adhesion and FN/FG binding assays are triplicate means of one assay that is representative of three 5

independent assays. Significance for ELISA was determined using the non-parametric Wilcoxon Rank Score (Mann-Whitney) test. Significance of OSP and adhesion assays was determined using the Student’s t-test. P-values are reported as p ≤ 0.05(*), p ≤ 0.01(**) or p ≤ 0.0001(****).

Results Presence and expression of isdA in bovine isolates of staphylococci. Blood and milk were obtained from 103 cows at two U.S. dairies that differed in size and operation. Presumptive staphylococci from milk were assayed by PCR for the presence of isdA and 16S rRNA species identification. Four isolates of 34 (12%) were S. aureus and 19 were isdA positive (56%). Twenty-five additional clinical bovine S. aureus isolates were obtained. The presence of isdA, and 6 additional virulence factors, was assessed by PCR in all staphylococci (Table 2). isdA was conserved in S. aureus (total of 28/29, 96.6%), and also present in S. haemolyticus (11/17, 64.7%), S. chromogenes (2/5, 40%) and S. devreise (1/1, 100%). isdB was conserved in all S. aureus (29/29, 100%), and other species harboring isdA. The adhesin clumping factor A (clfA) was also highly conserved (28/29, 96.6%). To determine the expression of isdA during mammary infection , we identified the presence of isdA mRNA from total milk RNA. As shown in Figure 1(A), the majority of isdA positive staphylococci infected milk samples (Staphylococcus+ isdA+) contained isdA mRNA. The staphylococcal culture from 14FL (D1) was negative for isdA by PCR, and the milk sample was negative by RT-PCR. Two Staphylococcus+isdA+ samples had low CFUs, and were inconclusive by RT-PCR (D1-26FR and D2-43FL). The expression of IsdA in vivo was assessed by anti-IsdA ELISA of milk and serum. As shown in Figure 1(B and C), Staphylococcus+ isdA+ milk and serum samples had heightened anti-IsdA IgG responses over samples negative for 6

Staphylococcus (Staphylococcus- isdA-). A western blot of purified IsdA probed with Staphylococcus+ isdA+ milk (cows D2 43FR and D2 37FR) compared to Staphylococcus – isdA – milk (cows D2 1FL and D2 6FL) supported the immunogenicity of IsdA in vivo (Figure 1D). To assess the efficacy of anti-IsdA serum antibodies we determined opsonophagocytosis (OSP) of S. aureus Newbould 305 (Figure 1E). Bacterial cells were mixed with bovine PBMCs and Staphylococcus-isdA- serum (low anti-IsdA titer, light grey bars) or Staphylococcus+isdA+ serum (high anti-IsdA titer, dark grey bars). Serum was preincubated in the presence or absence of purified IsdA to block antibody activity. High anti-IsdA serum stimulated more bacterial killing than low anti-IsdA serum (p≤0.05), and purified IsdA significantly blocked bacterial killing in the presence of high anti-IsdA serum (p ≤ 000.1). IsdA sequence analysis. Comparison of published S. aureus IsdA indicates that 86 amino acids within the C-terminus are variable. This region is distinct from conserved iron-binding, hemebinding and NEAT domains of IsdA, as well as the C-terminal cell wall sortase signal (Figure 2A). The 380 bp variable region from all isdA+ staphylococcal isolates was amplified for sequencing and alignment. Figure 2 shows IsdA alignments from staphylococci of the two dairies (B) and 23 S. aureus clinical isolates (C). One S. aureus was negative for isdA by PCR and one was isdA+ but sequence could not be obtained. Nucleotide differences within this region resulted in two main allelic variants that align with bovine (Newbould 305, variant I) and human (MRSA252, variant III) IsdA. Notably, variant III contains a characteristic four amino acid insertion near the C-terminus. Predicted tertiary structure models were constructed using IsdA from Newbould 305 and MRSA252 (Figure 2D) (Yang et al., 2015). Models predict that the variable region of IsdA has a coiled secondary structure with a solvent accessibility of 3-7 (0=buried residue, 9=highly exposed).

7

To determine if observed variation is consistent with human S. aureus IsdA, phylogenetic analysis was completed with 53 published sequences (Figure 3). Results included four cow, four pig and one chicken isolate. Sequences were largely conserved, however differences within the variable region resulted in clustering within two groups represented by MRSA252 (III, blue) and Newbould 305 (I, red). A smaller subgroup of consisted of those that contained only the characteristic 4 amino acid C-terminal insertion (variant II, green). Two strains (CA-347 and SK1585) are closely related to variant II, but do not contain this C-terminal insertion. Characterization of IsdA variants. To assess if genetic variation of isdA could contribute to differences in protein immune reactivity we produced antibodies to variant III (MRSA252) in mice (Figure 4A) and used pooled serum to block S. aureus adhesion to FN. Antibodies produced against variant III reduced FN binding of S. aureus MRSA252, but did not reduce binding of Newbould 305 (Figure 4B). Variants I and III were purified from E.coli and found to run at distinct sizes on SDS-PAGE (Figure 4C). ECM binding assays revealed that variant I (Newbould 305) and variant III (MRSA252) had consistently distinct interactions with FN, while variants bound similarly to FG (Figure 4, D and E).

8

Discussion Despite much research on the pathogenesis of S. aureus, this bacterium remains an elusive target for prevention. The current study was undertaken to evaluate IsdA expression and sequence variation to inform bovine vaccine development. The presence of isdA was found to be conserved in bovine S. aureus. We also observed isdA in bovine coagulase negative staphylococci (CNS). While these species tend to produce less severe disease than S. aureus, they are often associated with mastitis and clinically relevant (Waller et al., 2011). The presence of isdB and clfA was also conserved . Previous reports suggest that IsdB is not a viable vaccine antigen but ClfA could be further evaluated for a role in bovine disease (Fowler et al., 2013). Our findings that isdA mRNA is expressed, and anti-IsdA is induced in milk and serum, indicate that this protein is surface exposed during infection. Additionally, OSP assays provide evidence that IsdA antibodies are able to trigger antigen uptake and effective S. aureus killing. The predicted sequence of IsdA revealed polymorphisms within the C-terminus, and two main allelic variants. While variation may be more extensive than our limited sample size and geographic area reveals, the identified sequences correspond to published bovine (I, Newbould 305) and human (III, MRSA252) strains, with variant I predominating from both hosts. Most substitutions within the variable region of IsdA are conserved, however the four amino acid insertion and a proline to histidine substitution at amino acid 303 are not. These changes affect IsdA immune reactivity as well as binding specificity. The function of the C-terminal region of IsdA is reported to decrease bacterial hydrophobicity to support human skin colonization (Clarke et al., 2007). Our results are consistent with the additional involvement of this region in FN binding, with FG binding occurring within the NEAT domain as reported (Clarke et al., 2004, Clarke et al., 2007). While our sampling does not support the conclusion that IsdA allelic

9

variants are specific to host, or to species, they indicate a significant role for this adhesin in host interaction and bovine disease. Sequence variation will require further consideration; however these results support the continued exploration of IsdA as an antigenic target for bovine vaccines.

Acknowledgements We would like to thank Janet E. Williams from the University of Idaho, and Liwen Yang, Shandra Jeffries and Emily Price from Boise State University for technical support. We also thank Dr. Allan Britten (Udder Health Systems, Inc., Meridian, ID) for strains and continued support. This work was supported by a 2012 Idaho State Board of Education HERC Incubation Fund (#IF13-006, PI-Tinker, Co-PI McGuire), a 2013 USDA AFRI standard grant (#201301189, PI-Tinker, Co-PI McGuire), and an pilot grant to J.K.T. from an Institutional Development Awards (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health (#P20GM103408 and P20GM109095). Conflicts of interest The corresponding author owns an unlicensed patent for the use of cholera toxin chimera as a staphylococcal vaccine (Tinker, US 13/328,686). Although this patent is related to the subject matter of this article, the author will not realize any financial gain from this association.

References Arlian BM & Tinker JK (2011) Mucosal immunization with a Staphylococcus aureus IsdA-cholera toxin A2/B chimera induces antigen-specific Th2-type responses in mice. Clin Vaccine Immunol 18: 1543-1551. Bar-Gal GK, Blum SE, Hadas L, Ehricht R, Monecke S & Leitner G (2015) Host-specificity of Staphylococcus aureus causing intramammary infections in dairy animals assessed by genotyping and virulence genes. Vet Microbiol 176: 143-154.

10

Bar D, Tauer LW, Bennett G, González RN, Hertl JA, Schukken YH, Schulte HF, Welcome FL & Gröhn YT (2008) The cost of generic clinical mastitis in dairy cows as estimated by using dynamic programming. J Dairy Sci 91: 2205-2214. Barkema HW, Schukken YH & Zadoks RN (2006) Invited Review: The role of cow, pathogen, and treatment regimen in the therapeutic success of bovine Staphylococcus aureus mastitis. J Dairy Sci 89: 1877-1895. Clarke SR, Wiltshire MD & Foster SJ (2004) IsdA of Staphylococcus aureus is a broad spectrum, ironregulated adhesin. Mol Microbiol 51: 1509-1519. Clarke SR, Andre G, Walsh EJ, Dufrene YF, Foster TJ & Foster SJ (2009) Iron-regulated surface determinant protein A mediates adhesion of Staphylococcus aureus to human corneocyte envelope proteins. Infect Immun 77: 2408-2416. Clarke SR, Mohamed R, Bian L, Routh AF, Kokai-Kun JF, Mond JJ, Tarkowski A & Foster SJ (2007) The Staphylococcus aureus surface protein IsdA mediates resistance to innate defenses of human skin. Cell Host Microbe 1: 199-212. Clarke SR, Brummell KJ, Horsburgh MJ, et al. (2006) Identification of in vivo-expressed antigens of Staphylococcus aureus and their use in vaccinations for protection against nasal carriage. J Infect Dis 193: 1098-1108. Corrigan RM, Miajlovic H & Foster TJ (2009) Surface proteins that promote adherence of Staphylococcus aureus to human desquamated nasal epithelial cells. BMC Microbiol 9: 22. De Vliegher S, Fox LK, Piepers S, McDougall S & Barkema HW (2012) Invited review: Mastitis in dairy heifers: nature of the disease, potential impact, prevention, and control. J Dairy Sci 95: 1025-1040. Dickson CF, Kumar KK, Jacques DA, Malmirchegini GR, Spirig T, Mackay JP, Clubb RT, Guss JM & Gell DA (2014) Structure of the hemoglobin-IsdH complex reveals the molecular basis of iron capture by Staphylococcus aureus. J Biol Chem 289: 6728-6738. Fowler VG, Allen KB, Moreira ED, et al. (2013) Effect of an investigational vaccine for preventing Staphylococcus aureus infections after cardiothoracic surgery: a randomized trial. JAMA 309: 1368-1378. Grigg JC, Mao CX & Murphy ME (2011) Iron-coordinating tyrosine is a key determinant of NEAT domain heme transfer. J Mol Biol 413: 684-698. Grigg JC, Vermeiren CL, Heinrichs DE & Murphy ME (2007) Haem recognition by a Staphylococcus aureus NEAT domain. Mol Microbiol 63: 139-149. Hammer ND & Skaar EP (2011) Molecular mechanisms of Staphylococcus aureus iron acquisition. Annu Rev Microbiol 65: 129-147. Haran KP, Godden SM, Boxrud D, Jawahir S, Bender JB & Sreevatsan S (2012) Prevalence and characterization of Staphylococcus aureus, including methicillin-resistant Staphylococcus aureus, isolated from bulk tank milk from Minnesota dairy farms. J Clin Microbiol 50: 688-695. Honsa ES, Owens CP, Goulding CW & Maresso AW (2013) The near-iron transporter (NEAT) domains of the anthrax hemophore IsdX2 require a critical glutamine to extract heme from methemoglobin. J Biol Chem 288: 8479-8490. Jones DT, Taylor WR & Thornton JM (1992) The rapid generation of mutation data matrices from protein sequences. Comput Appl Biosci 8: 275-282. Kim HK, DeDent A, Cheng AG, McAdow M, Bagnoli F, Missiakas DM & Schneewind O (2010) IsdA and IsdB antibodies protect mice against Staphylococcus aureus abscess formation and lethal challenge. Vaccine 28: 6382-6392. Kumar S, Stecher G & Tamura K (2016) MEGA7: Molecular Evolutionary Genetics Analysis Version 7.0 for Bigger Datasets. Mol Biol Evol 33: 1870-1874. Kurokawa K, Jung DJ, An JH, et al. (2013) Glycoepitopes of staphylococcal wall teichoic acid govern complement-mediated opsonophagocytosis via human serum antibody and mannose-binding lectin. J Biol Chem 288: 30956-30968. 11

Le Marechal C, Hernandez D, Schrenzel J, Even S, Berkova N, Thiery R, Vautor E, Fitzgerald JR, Francois P & Le Loir Y (2011) Genome sequences of two Staphylococcus aureus ovine strains that induce severe (strain O11) and mild (strain O46) mastitis. J Bacteriol 193: 2353-2354. Mura MC, Daga C, Bodano S, Paludo M, Luridiana S, Pazzola M, Dettori ML, Vacca GM & Carcangiu V (2013) Development of a RNA extraction method from milk for gene expression study in the mammary gland of sheep. Mol Biol Rep 40: 2169-2173. Stapleton M, Wright L, Clarke S, Moseby H, Tarkowski A, Vendrengh M & Foster S (2012) Identification of Conserved Antigens from Staphylococcal and Streptococcal Pathogens. J Med Microbiol. Stranger-Jones YK, Bae T & Schneewind O (2006) Vaccine assembly from surface proteins of Staphylococcus aureus. Proc Natl Acad Sci U S A 103: 16942-16947. Torres VJ, Pishchany G, Humayun M, Schneewind O & Skaar EP (2006) Staphylococcus aureus IsdB is a hemoglobin receptor required for heme iron utilization. J Bacteriol 188: 8421-8429. Waller KP, Aspán A, Nyman A, Persson Y & Andersson UG (2011) CNS species and antimicrobial resistance in clinical and subclinical bovine mastitis. Vet Microbiol 152: 112-116. Wolf C, Kusch H, Monecke S, Albrecht D, Holtfreter S, von Eiff C, Petzl W, Rainard P, Broker BM & Engelmann S (2011) Genomic and proteomic characterization of Staphylococcus aureus mastitis isolates of bovine origin. Proteomics 11: 2491-2502. Yang J, Yan R, Roy A, Xu D, Poisson J & Zhang Y (2015) The I-TASSER Suite: protein structure and function prediction. Nat Methods 12: 7-8.

12

Figure 1. In vivo expression and immunogenicity of Staphylococcus IsdA from bovine milk and serum. A) Expression of isdA from total RNA in bovine milk: RT-PCR of (a) isdA (385 bps) and (b) 16S rRNA (530 bps). B) IsdA-specific IgG responses in milk and C) serum as determined by ELISA on samples obtained from Dairies 1 and 2. Samples are grouped as those from cows that were negative or positive by culture of Staphylococcus in milk and those that are negative or positive for isdA by PCR. Significance using the nonparametric Wilcoxon Rank Score test (Mann-Whitney), p ≤ 0.05(*) or p ≤ 0.01(**), between cow groups is shown. D) Western blot (top) and corresponding SDS-PAGE (bottom) of anti-IsdA responses in milk. Purified IsdA (62 13

kD) was probed with Staphylococcus-isdA- (D2-1FL, D2-6Fl) and Staphylococcus+isdA+ (D243FR, D2-37FR) milk samples. E) Opsonophagocytosis of S. aureus Newbould 305 with PBMCs and serum from Staphylococcus –isdA- ( light grey bars) and Staphylococccus +isdA+ (dark grey bars) cows. Significance using the pairwise Student’s t-test; p ≤ 0.05(*) between serum types or p ≤ 0.0001 (****) between treatment groups, is indicated.

Figure 2. IsdA alignment and structure from bovine isolates. A) Schematic representation of IsdA structure including the conserved NEAT domain (red), variable C-terminal regions (blue) and cell-wall associated sortase signal (LPxTG, green). Predicted amino acid alignment of the IsdA variable region from; B) 19 isdA+ Staphylococcus from Dairies 1 and 2, and C) 23 isdA + 14

S. aureus from mastitic bovine milk across the U.S., showing percent positive identity to S. aureus Newbould 305 and MRSA252. D) I-TASSER predicted tertiary structures of IsdA from Newbould 305 (variant I) and MRSA252 (variant III) S. aureus. Colors are representative of the same regions as those in A).

15

Figure 3. Dendrogram of pairwise IsdA alignments of 53 published sequences. Evolutionary analysis was performed with the Maximum Likelihood method based on the JTT matrix-based model (MEGA, Version 7). Variants align into three main groups, based upon sequence differences within the variable C-terminus, representative of Newbould 305 (I, red), MRSA252 (III, blue) and LGA251 (II, green). Isolates from cow (*), pig (**) and chicken (***) are noted.

Figure 4. Characterization of IsdA variants. Immune reactivity of variants assessed by blocking S. aureus adhesion to FN in vitro with mouse antiserum: A) IgG anti-IsdA titers from mice vaccinated with IsdA (variant III, MRSA252) or unvaccinated, and B) adhesion of S. aureus Newbould 305 or S. aureus MRSA252 to FN after incubation with pooled mouse serum from mice vaccinated with IsdA MRSA252 (dark grey bars) or unvaccinated serum (light grey bars). Significance using the Student’s t-test, p ≤ 0.01(**), between serum types is indicated. Binding specificity of IsdA variants assessed by ECM binding assay: C) SDS-PAGE of purified IsdA from Newbould 305 (lane 1) and MRSA252 (lane 2), and binding of purified IsdA from MRSA252 and Newbould 305 to D) FN and E) FG.

16

Table 1. Bacterial strains, plasmids and primers used in this study Bacterial strains E.coli TE1 S. aureus Newbould 305 S. aureus MRSA USA300 S. aureus MRSA252 #75

Genotype or characteristics

Reference or source

∆endA derivative of TX1 Bovine clinical isolate Human clinical isolate Human clinical isolate

(Tinker et al., 2003) (Tinker et al., 2003) (Prasad LB, 1968) (Highlander et al., 2007) (Arlian & Tinker, 2011)

Plasmids pBA015 pBA009A pCK001

Gene

Vector

Reference or source

isdA (MRSA252) isdA (MRSA252) isdA (Newbould)

pET40(b) pTRCHisA pTRCHisA

(Arlian & Tinker, 2011) (Arlian & Tinker, 2011) This study

Primers

FW CGGTTCAACCAAAACCTGCT RV GCGAAGGCAACTGTGCTAAT FW CAAATGGCGAAGCACAAGCAG RV CAAATGGCGAAGCACAAGCAG FW AGCAGCACTGCAACAAATCC RV CCATGGACCATTGGATCT FW CCTATGCCAGTAGCCAATGTC RV GCACCAAGCAGGTTATGTC FW GTGAAAAACAATCTTAGGTAC RVTATCAATAGCTGATGAATCCG FW GCGATTGATGGTGATACGGTT RV AGCCAAGCCTTGACGAACTAAAGC FW GGGATCATAGCGTCATTATTC RV AACGATTGTGACACGATAGCC FW GGCCGTGTTGAACGTGGTCAAATCA RV TACCATTTCAGTACCTTCTGGTAA FW 8F AGAGTTTGATCCTGGCTCAG RV 534R ATTACCGCGGCTGCTGGC

Gene

Amplicon (bp)

Reference or Source

isdA

380

This study

isdB

138

This study

isdH

557

This study

clfA

318

This study

fnbpA

1750

(Loughman et al., 2008)

nuc

270

(Brakstad et al., 1992)

mecA

527

tstaG

370

(Technical University of Denmark, 2008) (Morot-Bizot et al., 2004)

16s rRNA

530

(Nossa et al., 2010)

Table 2. PCR analysis of adhesins from bovine staphylococcal isolates. # positive isolates (percent positive) Sequence ID # of S. aureus S. haemolyticus S. chromogenes S. agnetis S. xylosus S. devriesei S. saprophyticus S. spp. Newbould 305 USA300

total isolates 29 17 5 2 2 1 1 3

isdA

isdB

isdH

clfA

fnbpA

nuc

mecA

28(96.6) 11(64.7) 2(40) 0 0 1(100) 0 1 (33.3) + +

29(100) 11(64.7) 2(40) 0 0 1(100) 0 1(33.3) + +

25(86.2) 0 0 0 0 0 0 0 + +

28(96.6) 0 2(40) 0 0 1(100) 0 0 + +

25(86.2) 4(23.5) 2(40) 0 0 0 0 0 + +

26(89.7) 0 0 0 0 0 0 0 + +

0 0 3(60) 0 0 0 1(100) 1 (33.3)

17

+