Mycoplasma hyopneumoniae surface-associated proteases cleave bradykinin, substance P, neurokinin A and neuropeptide Y

Mycoplasma hyopneumoniae is an economically-devastating and geographically-widespread pathogen that colonises ciliated epithelium, and destroys mucociliary function. M. hyopneumoniae devotes ~5% of its reduced genome to encode members of the P97 and P102 adhesin families that are critical for colonising epithelial cilia, but mechanisms to impair mucociliary clearance and manipulate host immune response to induce a chronic infectious state have remained elusive. Here we identified two surface exposed M. hyopneumoniae proteases, a putative Xaa-Pro aminopeptidase (MHJ_0659; PepP) and a putative oligoendopeptidase F (MHJ_0522; PepF), using immunofluorescence microscopy and two orthogonal proteomic methodologies. MHJ_0659 and MHJ_0522 were purified as polyhistidine fusion proteins and shown, using a novel MALDI-TOF MS assay, to degrade four pro-inflammatory peptides that regulate lung homeostasis; bradykinin (BK), substance P (SP), neurokinin A (NKA) and neuropeptide Y (NPY). These findings provide insight into the mechanisms used by M. hyopneumoniae to influence ciliary beat frequency, impair mucociliary clearance, and initiate a chronic infectious disease state in swine, features that are a hallmark of disease caused by this pathogen.

Mycoplasma hyopneumoniae is the etiological agent of porcine enzootic pneumonia, a highly infectious and globally distributed swine respiratory disease. Symptoms of enzootic pneumonia include growth rate retardation, reduced feed conversion, and higher susceptibility to secondary bacterial 1 and viral infections, including porcine reproductive and respiratory syndrome virus (PRRSV), swine influenza virus (SIV) and porcine circovirus type 2 (PCV2) 2,3 . Many broad-spectrum antibiotics are used to treat M. hyopneumoniae infections specifically, but a greater number of antibiotics are used to prevent polymicrobial respiratory infections 4 . Thus M. hyopneumoniae is a significant contributor to antibiotic use in swine production. In 1991, M. hyopneumoniae caused an estimated $1 billion economic loss in the USA alone 5 . This estimate did not account for the impact incurred by animal waste containing large quantities of multiple antibiotic resistant bacterial populations 6 . Animal waste including pig effluent is used as organic fertiliser on agricultural lands, particularly in China 7 , the world's largest producer of pork. While no current economic burden estimates have been published, a 2006 survey of nineteen large US pig production companies ranked the estimated losses associated with M. hyopneumoniae alone and M. hyopneumoniae in conjunction with PRRSV among the top challenges with the highest estimated average loss for all pigs 8 .
Bacterin vaccines are used to control M. hyopneumoniae in conjunction with antibiotics; however, their efficacy is limited due to a minimal reduction in pathogen transmission and high production cost 9 . There is a need to enhance our understanding of M. hyopneumoniae pathogenesis to develop more efficacious vaccines and therapeutics that seek to eradicate this pathogen by preventing colonisation of the respiratory tract and reducing reliance on antibiotics.
The mucociliary escalator is a major innate barrier to all infectious respiratory microorganisms. It lines the respiratory tract and is composed of mucus-secreting goblet cells and ciliated epithelium. Mucus traps inhaled particles that are then propelled to the pharynx by the synchronised beating of cilia to be either swallowed or

Results
Identification and localisation of MHJ_0659 and MHJ_0522 in M. hyopneumoniae. Despite a PSORT prediction for a cytoplasmic localisation for MHJ_0659 and MHJ_0522, mild enzymatic cell shaving experiments using trypsin identified tryptic peptides matching MHJ_0659 ( Fig. 1d; one peptide) and MHJ_0522 ( Fig. 2d; three peptides). In a second, independent set of experiments in which freshly cultured M. hyopneumoniae cells were labelled with biotin, and surface-exposed biotinylated proteins were recovered using avidin chromatography, LC-MS/MS identified peptides that also mapped to MHJ_0659 ( Fig. 1d; one peptide) and MHJ_0522 ( Fig. 2d; two peptides). Specificity of anti-sera generated against MHJ_0659 and MHJ_0522 was checked via Western Blotting (Figs 1b and 2b). To confirm the surface localisation of both proteases, we labelled the surface of freshly cultured M. hyopneumoniae with rabbit anti-MHJ_0659 and anti-MHJ_0522 antibodies and detected them with anti-rabbit antibodies conjugated with CF TM 568 (Figs 1c and 2c).

Biochemical characterisation of functionally active rMHJ_0659. rMHJ_0659 was purified from
Escherichia coli as a polyhistidine fusion protein and resolved as a single band with a mass of approximately 40 kDa by SDS-PAGE (Fig. 1a). This protein was recovered from the gel, digested with trypsin and confirmed to be MHJ_0659 by LC-MS/MS (46% sequence coverage) (Fig. 1d). rMHJ_0659 was able to remove the N-terminal penultimate proline from BK (Fig. 3), SP (Fig. 4) and NPY (Fig. 5) in a manner that is typical of a PepP protease. rMHJ_0659 cleaves bradykinin (BK). The BK1-7 fragment [RPPGFSP] of bradykinin has a molecular mass of 757 Da. In the absence of rMHJ_0659, a single prominent peak at 757.40 Da representing the mass of BK1-7 was observed by MALDI-TOF/MS (Fig. 3a). Once BK1-7 was incubated with rMHJ_0659 two additional peaks were identified with masses of 601.30 Da and 504.27 Da (Fig. 3b) (Fig. 5a). The peak at 4091.83 Da was not evident when rMHJ_0659 was incubated with NPY at pH 5.0. Minor cleavage of NPY was evident at pH 6.3 in the presence of Mg 2+ , Ca 2+ and Mn 2+ . At pH 7.5, the metal ion cofactor Ca 2+ provided the most favourable conditions for cleaving NPY. rMHJ_0659 cleaved NPY with maximum efficiency at pH 8.8 in the presence of different metal ion cofactors. EDTA suppressed Mg 2+ mediated activity at pH 6.3, and Ca 2+ activity at pH 7.5 and 8.8 (Fig. 5b).

Biochemical Characterisation of rMHJ_0522
rMHJ_0522 (predicted MW of 71 kDa) resolves during SDS-PAGE as a single band with a mass of approximately 70 kDa (Fig. 2a). The protein band was digested with trypsin and confirmed to be rMHJ_0522 by LC-MS/MS (51% sequence coverage) (Fig. 2d). rMHJ_0522 cleaved full-length BK (Fig. 6), SP (Fig. 7) and NKA (Fig. 8). Unlike rMHJ_0659, rMHJ_0522 was not active at pH 5. rMHJ_0522 did not cleave BK1-7 nor did it cleave NPY, which is 36 amino acids in length. These data are consistent with rMHJ_0522 functioning as an oligoendopeptidase. rMHJ_0522 cleaves bradykinin (BK). Full-length BK [RPPGFSPFR] has a molecular mass of 1060.21 Da.
In the absence of rMHJ_0522, a single prominent peak at 1060.21 Da was observed in MALDI-TOF MS spectra (Fig. 6a). When BK was incubated with rMHJ_0522, three additional peaks were observed, 652.74 Da, 572.66 Da, and 505.57 Da (Fig. 6b) 2+ and Zn 2+ and pH on the activity of rMHJ_0522 was determined. At pH 6.3, Mn 2+ and Mg 2+ produced the most intense peak for BK5-9, and www.nature.com/scientificreports www.nature.com/scientificreports/ both Ca 2+ and Zn 2+ produced the most intense peaks for BK1-5. A similar profile was seen at pH 7.3; only Mn 2+ joined Ca 2+ and Zn 2+ in producing the most intense peak for BK1-5. In basic conditions (pH ≥ 7.5), only Mg 2+ produced intense peaks for BK5-9, and Ca 2+ and Co 2+ produced the most intense peaks for BK1-5 ( Fig. 6c). EDTA supressed Ca 2+ mediated activity at pH 6.3 and pH 8.8, and Mn 2+ mediated activity at pH 7.5 ( Fig. 6c).  (Fig. 7b). The most prevalent cleavage fragment generated at all pHs (6.3, 7.3 & 8.8) and cofactors (Ca 2+ , Co 2+ , Mg 2+ , Mn 2+ , and Zn 2+ ) tested was SP1-8 ( Fig. 7c), however at pH 7.3 in the presence of Zn 2+ and Ca 2+ , SP1-9 and SP1-10 were also produced, albeit at lower intensities (Fig. 7c). This data indicates that rMHJ_0522 can produce cleavage events at RPKPQQFF↓G↓L↓M (first arrow indicates most prominent cleavage). EDTA suppressed Ca 2+ mediated activity at pH 6.3 and pH 8.8, and Zn 2+ mediated activity at pH 7.5 (Fig. 7c).  Fig. 8b). At pH 6.3 and only in the presence of Ca 2+ , Co 2+ and Mg 2+ did rMHJ_0522 produce non-negligible cleavage fragments of NKA. NKA1-6 and NKA1-7 were the most common cleavage fragment in the presence of Mg 2+ and NKA1-8 the most prominent in the presence of Co 2+ . At pH 7.3, Zn 2+ was the most effective cofactor, producing NKA1-8 as the most predominant fragment. Unlike BK and SP, no activity against NKA was observed at pH 8.8. EDTA suppressed Mg 2+ mediated activity at pH 6.3 and Zn 2+ mediated activity at pH 7.5 (Fig. 8c).   -7) represent the loss of an R residue, and loss of R and P residues respectively from the N-terminus of BK1-7. (c) At pH 5, the most prominent peak was the parent BK1-7. Smaller amounts of cleavage products BK2-7 and BK3-7 were detected at pH 5. At all pH levels tested, Ca 2+ followed by Mg 2+ produced the most intense peaks at a mass of 504.3 Da representing BK3-7. EDTA suppressed Ca 2+ mediated activity at all pH levels tested. Control refers to BK1-7 in the absence of rMHJ_0659. www.nature.com/scientificreports www.nature.com/scientificreports/ rMHJ_0659 and rMHJ_0522 share substrates and have the potential to cleave many biologically active proteins. rMHJ_0659 was shown to cleave at N-terminal penultimate prolines when the N-terminal amino acid was arginine (BK and SP), proline (BK second cleavage) and tyrosine (NPY). rMHJ_0522 cleaved BK at glycine and phenylalanine, SP at phenylalanine, glycine and leucine, and NKA at phenylalanine, valine, and glycine (Table 1). BK and SP were common substrates for both proteases. Interestingly, penultimate prolines are present at the N-terminal of many porcine innate immune system polypeptides, and the FXGLM-NH2 motif is common to all tachykinins, the largest family of neuropeptides (Table 2). We propose that these innate effector peptides and neuropeptides are potential substrates for PepP and PepF proteases on the surface of M. hyopneumoniae.

Discussion
M. hyopneumoniae enters the porcine respiratory tract on mucosal droplets expelled during bouts of coughing from infected animals. In naïve animals, these droplets are inhaled and are likely to be captured by the mucociliary escalator that forms an effective barrier to invading pathogens. The ability of this porcine-specific pathogen to survive as a species is contingent on being able to colonise cilia that beat with a frequency of ~13 hertz in normal uninfected tissue and overcome the mucociliary escalator 32 . A toxin that can disrupt mucociliary clearance by  www.nature.com/scientificreports www.nature.com/scientificreports/ slowing or blocking ciliary beat frequency has not been identified in M. hyopneumoniae, yet it is highly adept at colonising ciliary epithelium. M. hyopneumoniae devotes more than 5% of its reduced genome 23 to encoding multifunctional and highly-processed cilium adhesion proteins 16,[33][34][35][36] that can also bind glycosaminoglycans 35,37 , extracellular actin 38 , plasminogen 39,40 and fibronectin 39,41 , all important components of the extracellular matrix of host cells. The ability of M. hyopneumoniae to bind to porcine cilia is abrogated by the exogenous addition of the glycosaminoglycan heparin 42 , actin 38 and by treating cilia with heparinase 43 . Unsurprisingly, M. hyopneumoniae adhesins have been explored previously as vaccine targets, though they have proved less effective than the bacterin vaccines currently available 44,45 . This may be due to the growing discovery of M. hyopneumoniae surface proteins with adhesion moonlighting capacities 46,47 . Here we describe a novel enzymatic mechanism that potentially enables M. hyopneumoniae to inactivate four innate effector peptides that play a central role in controlling ciliary beat frequency, inflammation and other innate immune responses in the lungs. Specifically, we determined that MHJ_0569 and MHJ_0522 are retained on the cell surface of M. hyopneumoniae and that purified recombinant At pH 6, only the cleavage fragment SP1-8 was produced in the presence of each of the cofactors. At pH 7.3, the most common cleavage fragment was SP1-8. However, the presence of Zn 2+ produced additional cleavage fragments SP1-9 and SP1-10, as did Ca 2+ , albeit at lower intensities. At pH 8.8 only SP1-8 was produced. However, the intensities for Co 2+ , Mg 2+ and Mn 2+ were lower. EDTA suppressed cofactor mediated activity across all pH levels tested. Control refers to SP1-11 in the absence of rMHJ_0522. www.nature.com/scientificreports www.nature.com/scientificreports/ versions of these proteases can affect the function of NPY, BK, SP and NKA. rMHJ_0659 was shown to function as a PepP aminopeptidase that removes N-terminal amino acids that lie adjacent to a proline residue in BK, SP, and NPY. rMHJ_0522 (PepF) was shown to function as an oligopeptidase that can cleave BK and the tachykinins NKA and SP that carry a common tachykinin C-terminal motif. These observations provide insight into the mechanisms used by M. hyopneumoniae that alter the microenvironment in the upper respiratory tract of the host in a manner that favours colonisation of ciliated epithelium.
Despite lacking translocation signal peptides, MHJ_0659 and MHJ_0522 were identified on the cell surface of M. hyopneumoniae. Surface localisation was demonstrated by detecting tryptic peptides released by mild enzymatic cell shaving, digesting biotinylated surface proteins captured by avidin chromatography using LC-MS/ MS, and by immunofluorescence microscopy. MHJ_0659 and MHJ_0522 have unique access to host proteins including inflammatory peptides that regulate ciliary function on epithelial cells that line the upper respiratory tract, the preferred niche for this pathogen. Using a MALDI-TOF MS assay, we demonstrate that, collectively, rMHJ_0659 and rMHJ_0522 cleave four peptides that play a central role in modulating mucociliary clearance and inflammation.
BK is a nonapeptide involved in the coagulation cascade, which is now known to be an integral part of innate immunity 20 . Once BK is released from the D4 domain of high molecular weight kininogen via the serine protease kallikrein, it greatly enhances bronchoconstriction, accelerates tracheobronchial and mucociliary clearance, and induces cough 48 . BK binds to mast cells, releasing histamine, leukotriene, and prostaglandins, which lead to increased mucous secretions in the lungs 49 . BK also amplifies nitric oxide production leading to increased ciliary beat frequency 50 and can initiate a range of other immune responses 51 . rMHJ_0659 cleaved BK at the first two N-terminal amino acids R↓P↓P-G-F-S-P -F-R, and rMHJ0522 cleaved BK at R-P-P-G↓F↓S-P-F-R. The cleavage events demonstrated by both proteases mimic those known to inactivate BK 52 .
SP and NKA are neuropeptides belonging to the tachykinin superfamily, characterised by a common C-terminal sequence FXGLM- NH2 53 . SP is the most extensively researched tachykinin and is best known for its ability to function as a neurotransmitter. However, it can also induce a repertoire of innate immune effector cells and inflammatory mediators 54,55 , and concentrations of SP are increased within inflamed tissue 56 . Invading bacteria stimulate primary sensory neurons and activate SP NK1 receptors in tissue cells promoting a local influx of inflammatory and immune cells 57 . Additionally, SP has been shown to; (i) increase intracellular calcium which in turn increases ciliary beat frequency 58 ; (ii) regulate proinflammatory cytokines 59 ; and (iii) stimulate airway submucosal gland secretion 60 . Interestingly, the ability of SP to boost airway submucosal gland secretion is ten-fold greater in pigs than in humans 60 . SP-induced bronchoconstriction and cough have both been reported 61 . However, the ability to induce cough has since been contested 62 . Like SP, NKA induces bronchoconstriction and mucus production in the lungs as well as neurogenic inflammation 63 . Additionally, NKA is a potent attractor of T-cells 57 .  Table 2. Porcine innate immune system polypeptides that exhibit penultimate proline (left) residues and tachykinin C-terminal sequences (right).

Substrate Cleavage site
Bradykinin R↓P↓P-G↓F↓S-P-F-R Substance P R↓P-K-P-Q-Q-F-F↓G↓L↓M-NH2

Neurokinin A H-K-T-D-S-F↓V↓G↓LM-NH2
Neuropeptide Y Y↓P-S-K-P-D-N-P-G… www.nature.com/scientificreports www.nature.com/scientificreports/ To exert their inflammatory responses, SP and NKA need to bind NK1 and NK2 receptors present in airway smooth muscle, epithelium and on macrophages 64 . rMHJ_0659 cleaved SP at the anticipated N-terminal arginine residue due to the presence of a penultimate proline. While this cleavage does not inactivate SP 65 , it does remove the protective conformation that penultimate proline provides, leaving the peptide vulnerable to potential degradation by other M. hyopneumoniae surface exposed aminopeptidases, such as leucine aminopeptidase 47 and glutamyl aminopeptidase 46 . On the other hand, rMHJ_0522 efficiently cleaved SP at R-P-K-P-Q-Q-F-F↓G-L-M-NH2. This cleavage event removes the C-terminal G-L-M-NH2 sequence essential for SP to bind tachykinin receptors NK1 and NK2. At pH 7.3 in the presence of Zn 2+ , rMHJ_0522 produced two additional cleavages at R-P-K-P-Q-Q-F-F↓G↓L↓M-NH2 thus wholly degrading the G-L-M-NH2 sequence. SP fragments are known to be biologically active 66 , and previous reports have demonstrated that F-G-L-M-NH2 and G-L-M-NH2 can bind NK1 and NK2 receptors, albeit their binding affinity is 1000 times less than that of an intact SP molecule 67 . rMHJ_0522 had a similar activity to SP against NKA producing cleavages at H-K-T-D-S-F↓V↓G↓L-M-NH2 again cleaving at the biologically significant C-terminal sequence. Our results indicate that MHJ_0522 can potentially eradicate inflammation mediated by SP and NKA.
NPY is a neuropeptide that has well-recognised roles in the brain; however, NPY has also been identified as a potent mediator of inflammation 68 . Similarly to both BK and SP, NPY increases the release of proinflammatory cytokines and chemokines 69 , but is also a key activator of antigen presenting cell (APC) function 68 . Additionally, allergic airway inflammation in NPY deficient mice is significantly reduced 70 . rMHJ_0659 removed N-terminal tyrosine off NPY. All cofactors tested greatly enhanced rMHJ_04659 activity against NPY at pH 8.8. At a more neutral pH, Ca 2+ was the most efficient cofactor. The most prominent NPY receptors in the lungs are Y1 receptors, which potentate local inflammatory responses and are also widely expressed on T cells, B cells and APCs 68 . Importantly, studies have demonstrated that removing Y from the N-terminus of NPY diminishes Y1 receptor-binding efficiency and the resultant NPY 2-36 fragment is a selective Y2 receptor agonist 71 . Y2 receptor agonists have been linked to a reduction in food intake 72 . Our data suggest that cleavage events caused by the action of MHJ_0659 have the potential to minimise Y1 receptor effects and increase Y2 receptor effects.
The data presented here and elsewhere provide a body of evidence to suggest that proteases play an essential role in the pathogenesis of M. hyopneumoniae by processing adhesins, lipoproteins and other putative pathogenicity factors 15 , destroying host innate immunity effector peptides and neuropeptides, recycling amino acids for nutrients 42,43 and by inducing apoptosis 14 . Here we demonstrate novel functions for M. hyopneumoniae surface proteases. Our work provides insight into how genome-reduced pathogens colonise and persist in their chosen host.

Material and Methods
Materials. Ethylenediaminetetraacetic acid (EDTA), tributylphosphine (TBP), bovine serum albumin (BSA), SP, BK, NKA and NPY were purchased from Sigma (Australia). MS grade trypsin was purchased from Promega (USA). Acrylamide was purchased from Bio-Rad (USA). Immunofluorescence dyes, pre-cast gels, buffers, molecular weight markers and all standard molecular biology reagents were purchased from Life Technologies (Australia) unless otherwise noted.
Mycoplasma hyopneumoniae culture conditions. M. hyopneumoniae strain J was cultured in modified Friss 73 medium for 48 h at 37 °C and harvested by centrifugation at 12 000 × g for 15 min. Pellets were stored at −80 °C until use. cell surface analyses of Mycoplasma hyopneumoniae proteins. All surface exposed M. hyopneumoniae proteins were identified using both cell surface biotinylation and trypsin cell surface shaving experiments as previously described 36  Western blot. Once SDS-PAGE gel containing a freshly purified recombinant protease sample was electrophoretically transferred, the membrane was washed in PBS with 0.1% Tween [Bio-Rad, USA] for 20 min and then blocked with PBS, 0.1% Tween and 5% skim milk powder for 30 min to prevent non-specific antibody binding. To check specific binding to generated anti-recombinant protease sera, the membrane was then placed in a solution containing primary antibody at a 1:1000 dilution in PBS for 90 min before being washed three times with PBS and probed with peroxidase conjugated anti-rabbit secondary antibody [Sigma Aldrich, USA] diluted to 1:10,000 in blocking solution for 30 min at room temperature. The membrane was again washed briefly three times in PBS before being developed with 3,3′-Diaminobenzidine (DAP) [Sigma Aldrich, USA].
Immunofluorescence microscopy. Microscopy was performed as previously described 46 with minimal modifications. A 35 mm WPI Fluorodish was coated with 0.01% poly-l-lysine for 30 min and allowed to air dry for 30 min. An overnight M. hyopneumoniae culture was centrifuged (11 000 × g for 10 min) and washed twice with sterile PBS. Washed cells were added to the Fluorodish and allowed to attach for 30 min. Excess cells (2019) 9:14585 | https://doi.org/10.1038/s41598-019-51116-w www.nature.com/scientificreports www.nature.com/scientificreports/ were removed by washing once with PBS followed by fixation in 2% paraformaldehyde at 4 °C overnight. Excess aldehydes were quenched using 100 mM glycine in PBS for 5 min, followed by washing three times with PBS. Non-specific binding sites were blocked with 2% BSA in PBS for 1 h at room temperature, followed by washing three times with PBS. Polyclonal rabbit rMHJ_0522 and rMHJ_0659 antisera were diluted 1:500 with PBS containing 2% BSA, incubated for 1 hr at room temperature and washed three times in PBS. A 1:1000 dilution of goat anti-rabbit antibodies conjugated to CF ™ 568 (Sigma-Aldrich) was prepared in PBS containing 2% BSA and incubated for one hr at room temperature and washed three times in PBS. The liquid was removed from the Fluorodish which was allowed to air dry for 15 min. Two drops of VECTASHIELD were added to the Fluorodish followed by imaging on a Nikon Ti inverted epifluorescence microscope, capturing images using a DS-Qi2 microscope camera (Nikon Instruments). These same samples were imaged on a V3 DeltaVision OMX 3D-Structured Illumination Microscopy Imaging System (Applied Precision, GE Healthcare) as previously described 76 . Images were processed using Imaris Scientific 3D/4D image processing software (Bitplane AG).
Expression and purification of rMHJ_0659 and rMHJ_0522. The mhj_0659 and mhj_0522 genes encoding PepP and PepF respectively were ligated into expression vector PS100030, conveying both a hexahistidine tag and ampicillin resistance, by Blue Heron Biotech (USA). All in-frame TGA codons were substituted to TGG. The recombinant construct was transformed into E. coli BL21 (Invitrogen) as per the manufacturer instructions and grown overnight in LB supplemented with 100 mg/mL ampicillin. LB supplemented with 100 mg/mL ampicillin was inoculated with overnight transformation culture and grown until mid-log growth phase. Cultures were then induced with IPTG and grown to optimal protein expression levels (4 hrs for rMHJ_0522, 6 hrs for rMHJ_0659). Cells were harvested by centrifugation at 3000 × g for 10 min at 4 °C. Pellets were resuspended in Lysis Buffer [50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8) at 2 mL per gram of wet weight and lysozyme (Sigma Aldrich) was added at 1 mg/mL to break down peptidoglycan. Resuspended pellets were incubated on ice for 30 min and then probe sonicated with 6 × 10-sec bursts at 400 W and 10 sec cooling period on ice between bursts. Cell debris was removed by centrifugation at 10,000 × g for 30 min at 4 °C and the cleared cell lysate was added to 50% slurry of Profinity immobilized metal affinity chromatography (IMAC) Ni 2+ charged resin (Bio-Rad) at a volume of 1 mL slurry per 1 L starting culture, and left overnight on a rotary shaker at 4 °C. Both recombinant proteins were then purified under native conditions using imidazole, dialysed against PBS in 10 K MWCO dialysis tubing, and stored at 4 °C. 1D SDS-PAGE was used to separate purified protein samples.

proteomics.
To analyse purified protein samples separated by 1D SDS-PAGE, peptide preparation by trypsin in-gel digestion, and LC-MS/MS analysis parameters were followed as described previously 46 . A novel assay was used to determine substrate cleavage. From a stock solution (1 mg/ml) of BK, SP, NKA or NPY (Sigma Aldrich) 1 µl was diluted in 8.5 µL 50 mM Tris-HCl buffer (pH 6.3-8.8) or 50 mM sodium acetate (pH 5) and 0.5 µL 100 mM cofactor (Ca 2+ , Co 2+ , Mg 2+ , Mn 2+ , or Zn 2+ ). Purified rMHJ_0659 or rMHJ_0522 was added in a 1:20 protease to substrate concentration and incubated for 1 hr at 37 °C. The peptides were then desalted and captured using C18 ZipTips (Millipore). 1 µl of peptide sample was later spotted onto a clean 384-well OptiTOF target plate (AB Sciex) followed by 1 µl of 5 mg/ml α-Cyano-4-hydroxycinnamic acid (CHCA) dissolved in 50% ACN, 0.1% TFA, 10 mM NH 4 H 2 PO 4 and allowed to dry. Spotted samples were then analysed using a 5800 MALDI-TOF/TOF MS in positive reflector mode for BK, NKA and SP, and linear mode for NPY due to substrate size. Laser intensity was set to 2600 for MS parent ion scans. 400 laser shots were averaged for MS scans. MS parent ion scans were calibrated using a 6-peptide mixture (Sciex). The resulting MS spectral data were then manually inspected to explain the ions present concerning their amino acid sequence and the cleavage events caused by PepP and PepF proteolysis. For inhibition studies, rMHJ_0659 or rMHJ_0522 was pre-incubated with 10 mM metal chelating agent EDTA for 20 min at 37 °C, before repeating substrate cleavage assays using the cofactor that had previously demonstrated highest activity at each pH level tested.