Extracellular Proteome Analysis Shows the Abundance of Histidine Kinase Sensor Protein, DNA Helicase, Putative Lipoprotein Containing Peptidase M75 Domain and Peptidase C39 Domain Protein in Leptospira interrogans Grown in EMJH Medium

Leptospirosis is a re-emerging form of zoonosis that is caused by the spirochete pathogen Leptospira. Extracellular proteins play critical roles in the pathogenicity and survival of this pathogen in the host and environment. Extraction and analysis of extracellular proteins is a difficult task due to the abundance of enrichments like serum and bovine serum albumin in the culture medium, as is distinguishing them from the cellular proteins that may reach the analyte during extraction. In this study, extracellular proteins were separated as secretory proteins from the culture supernatant and surface proteins were separated during the washing of the cell pellet. The proteins identified were sorted based on the proportion of the cellular fractions and the extracellular fractions. The results showed the identification of 56 extracellular proteins, out of which 19 were exclusively extracellular. For those proteins, the difference in quantity with respect to their presence within the cell was found to be up to 1770-fold. Further, bioinformatics analysis elucidated characteristics and functions of the identified proteins. Orthologs of extracellular proteins in various Leptospira species were found to be closely related among different pathogenic forms. In addition to the identification of extracellular proteins, this study put forward a method for the extraction and identification of extracellular proteins.


Introduction
Leptospirosis, the zoonotic disease once confined to posing a risk during agricultural activities, has been re-emerging due to increasing urbanization and slum areas that have increased the reservoir rodent population [1]. The increase in outbreaks during floods has been due to water getting contaminated with the urine from rats and several other domestic and wild animals that spread out during the floods. Humans exposed to such The abundance of secretory proteins across aqueous, detergent, pellet, wash, and supernatant fractions was calculated based on Intensity-Based Absolute Quantification (iBAQ) values (Table 1 and Supplementary Table S1). A protein was considered abundant in a fraction if the average iBAQ value of the protein from three replication was at least 1.5-fold higher than the average iBAQ value of the protein totaled out of all other fractions. This showed that 56 proteins were abundant out of 1176 proteins identified in the wash and supernatant together. Out of the 56 proteins, 19 were not found in cellular fractions (aqueous, detergent, and pellet proteins together). Out of the 19 proteins, six were found only in the wash fraction (surface), nine were found exclusively in the supernatant The abundance of secretory proteins across aqueous, detergent, pellet, wash, and supernatant fractions was calculated based on Intensity-Based Absolute Quantification (iBAQ) values (Table 1 and Supplementary Table S1). A protein was considered abundant in a fraction if the average iBAQ value of the protein from three replication was at least 1.5-fold higher than the average iBAQ value of the protein totaled out of all other fractions. This showed that 56 proteins were abundant out of 1176 proteins identified in the wash and supernatant together. Out of the 56 proteins, 19 were not found in cellular fractions (aqueous, detergent, and pellet proteins together). Out of the 19 proteins, six were found only in the wash fraction (surface), nine were found exclusively in the supernatant (secretory), and four were found in both with an abundance in the supernatant ( Figure 1B). Table 1. Abundance based sorting of extracellular proteins using iBAQ values. Selection of 56 extracellular out of 1195 proteins identified through LC-MS/MS in wash and supernatant of Leptospira grown in EMJH medium. The values shown are average of normalized iBAQ values from three replicates. The abundance is indicated as exclusively present in wash (Ex. W), exclusively present in supernatant (Ex. S), and present only in wash, while supernatant (Ex. EC) and those found in cellular fractions also are shown in values indicating a fold of abundance as being extracellular. The protein is not identified in the fraction (-). The proteins found abundant in the cellular fractions are submitted as Supplementary Table S1

. Prediction of Subcellular Location of Extracellular Proteins Identified in Leptospira
Our results showed that CELLO v.2.5 predicted that 21.4% of the extracellular proteins would be same, while PSORTb predicted 12.5% and BUSCA predicted 8.9% to be extracellular ( Figure 2). Most of the predictions made by CELLO v.2.5 involved cytoplasmic (41%) and Outer Membrane Protein (OMP) proportions (26.8%) In the case of PSORTb, the predictions were that it was cytoplasmic (26.8%) and unknown (35.7%) while the BUSCA predicted that cytoplasmic (53.6%) and inner membrane proteins (35.7%) would be a major portion of Leptospira (Table 1).
Pathogens 2021, 10, x FOR PEER REVIEW 4 of 18 (secretory), and four were found in both with an abundance in the supernatant ( Figure  1B).

Prediction of Subcellular Location of Extracellular Proteins Identified in Leptospira
Our results showed that CELLO v.2.5 predicted that 21.4% of the extracellular proteins would be same, while PSORTb predicted 12.5% and BUSCA predicted 8.9% to be extracellular ( Figure 2). Most of the predictions made by CELLO v.2.5 involved cytoplasmic (41%) and Outer Membrane Protein (OMP) proportions (26.8%) In the case of PSORTb, the predictions were that it was cytoplasmic (26.8%) and unknown (35.7%) while the BUSCA predicted that cytoplasmic (53.6%) and inner membrane proteins (35.7%) would be a major portion of Leptospira (Table 1).  Table S1). * indicates proteins that were exclusively found on the surface and/or as secretory and were not found in any cellular fractions.    Table S1).

Prediction of Protein Function and Pathogenic Nature
Identification of the role of the proteins identified as extracellular is important to assess their usefulness in an anti-leptospirosis strategy. The KEGG database showed seven classes of functional groups in the 56 proteins ( Figure 3A). Four proteins were found under a metabolism that includes the Amino acid metabolism (1), Carbohydrate metabolism (1), Energy metabolism (1), and Metabolism of cofactors and vitamins (1). In the Cellular Processes under the Cellular community-prokaryotes, in the Genetic Information Processing under folding, sorting and degradation, and in Environmental Information Processing under Membrane transport one protein each category. The BRITE hierarchy showed 8 proteins which are distributed under the Protein families; genetic information processing 3, metabolism 3, signaling and cellular processes 2 proteins were present. Under Genetic Information Processing, two Replication and repair proteins and one Translation activity were present. There was one Unclassified signaling and cellular processes protein that is not included in any pathway or BRITE. hierarchy showed 8 proteins which are distributed under the Protein families; gene information processing 3, metabolism 3, signaling and cellular processes 2 proteins w present. Under Genetic Information Processing, two Replication and repair proteins a one Translation activity were present. There was one Unclassified signaling and cellu processes protein that is not included in any pathway or BRITE.  Table S2. Part (B) shows pathogenic/virulent proteins identified using MP3 online server represented as abundance-based groups where 'Abundant' proteins are those proteins found to be extracellular due to indicating a >1.5 times higher amount present in the wash (surface) and supernatant (secretory). * indicates proteins that were exclusively found on the surface and/or as secretory and were not found in any cellular fractions.
The severity of an infection caused by a pathogen is determined by the virulent and pathogenic proteins present in an organism. The MP3 online server predicted that out of 56 proteins, 32 were pathogenic/virulent and 24 were nonpathogenic ( Figure 3B). The abundant 37 proteins that were found to be cellular included 20 pathogenic proteins and 17 nonpathogenic proteins. Out of the 19 exclusive proteins found in wash and supernatant, 12 were pathogenic and seven were nonpathogenic. There were eight pathogenic proteins out of 10 exclusive proteins found in the wash.
These results showed that the proteins determined to be secretory may provide a balance between the pathogenic and non-pathogenic proteins which progressively determine the Leptospira based on the environmental conditions.

Identification of Orthologous Proteins
The NCBI protein-protein BLAST (blastp) retrieved orthologous proteins of 56 secretory proteins. The 'query cover' and 'identity' values were used to analyze protein similarity with the query sequence, which was the secretory protein of L. interrogans. The Leptospira species represented in the BLAST result were arranged according to the P1, P2, S1, and S2 groups, and the same sequence was shown of a phylogenetic tree constructed with the ppk gene sequences in Figure 4 as from an earlier study that describes 64 species of Leptospira [23]. The result showed the highest coverage and identity of 26 extracellular proteins with 17 species of pathogenic Leptospira (Figure 4). The intermediate, as well as saprophytic species, showed less than 50% identity though having good coverage against the query sequences.

Determination of Protein Interactions
As a representative study, four proteins: Histidine kinase sensor protein (LIC_11528), DNA helicase (LIC_11624), Putative lipoprotein containing peptidase M75 domain (LIC_10713), and Peptidase C39 domain protein (LIC_10511) showed an abundance of 1771-, 125-, 79-, and 76-times, respectively, in extracellular samples that were subjected to string analysis to find inter-correlation among the various proteins within the Leptospira group ( Figure 5). The C39 peptidase group was found to be associated with LIC 10510 and LIC 10512 where no functional analysis was interpreted. Also, putative lipoprotein (Imelysin domain) was predicted to associate with gene encoding proteins i.e., LIC 10712, LIC 11466, and LIC 10711, thereby resembling a neighborhood and co-occurrence among the species. Further, histidine kinase sensory proteins were found to be associated with other chemotaxis response regulator proteins. This association was also determined by neighborhood relationships within the species. Further, DNA helicase was found to be involved with uvrB, mutL, recQ, recA, pheT, polA, and lig proteins. These associations were determined experimentally with DNA helicase. As a representative study, four proteins: Histidine kinase sensor protein (LIC_11528), DNA helicase (LIC_11624), Putative lipoprotein containing peptidase M75 domain (LIC_10713), and Peptidase C39 domain protein (LIC_10511) showed an abundance of 1771-, 125-, 79-, and 76-times, respectively, in extracellular samples that were subjected to string analysis to find inter-correlation among the various proteins within the Leptospira group ( Figure 5). The C39 peptidase group was found to be associated with LIC 10510 and LIC 10512 where no functional analysis was interpreted. Also, putative lipoprotein (Imelysin domain) was predicted to associate with gene encoding proteins i.e., LIC 10712, LIC 11466, and LIC 10711, thereby resembling a neighborhood and co-occurrence among the species. Further, histidine kinase sensory proteins were found to be associated with other chemotaxis response regulator proteins. This association was also determined by neighborhood relationships within the species. Further, DNA helicase was found to be involved with uvrB, mutL, recQ, recA, pheT, polA, and lig proteins. These associations were determined experimentally with DNA helicase.

Discussion
Bacteria secrete a wide variety of proteins, enabling them to respond to their environment. These extracellular proteins have a diverse functional role such as degradation of substrates, response to environmental stimulus, migration, genetic exchange, feeding, ion-capturing, and sociobiological aspects like quorum sensing, biofilm formation, and host-pathogen interaction [24,25]. In this context, identification and characterization of extracellular proteins of Leptospira help in elucidating the functions of proteins that lead to pathogenesis and survival of the pathogen in the host. This study was conducted with the same sample that describes cellular proteins. Identification of proteins that are exclusively present in extracellular samples like wash and supernatant in spite of high-resolution analysis cellular fractions, which were separately analyzed in three (aqueous, detergent, and pellet) fractions, underlines the quality of extracellular protein preparations. Similarly, the exclusive proteins found

Discussion
Bacteria secrete a wide variety of proteins, enabling them to respond to their environment. These extracellular proteins have a diverse functional role such as degradation of substrates, response to environmental stimulus, migration, genetic exchange, feeding, ion-capturing, and sociobiological aspects like quorum sensing, biofilm formation, and host-pathogen interaction [24,25]. In this context, identification and characterization of extracellular proteins of Leptospira help in elucidating the functions of proteins that lead to pathogenesis and survival of the pathogen in the host. This study was conducted with the same sample that describes cellular proteins. Identification of proteins that are exclusively present in extracellular samples like wash and supernatant in spite of high-resolution analysis cellular fractions, which were separately analyzed in three (aqueous, detergent, and pellet) fractions, underlines the quality of extracellular protein preparations. Similarly, the exclusive proteins found within wash and supernatant samples showed the discriminatory power of the preparations to distinguish between surface and secretory groups in extracellular proteins. Out of 1176 unique proteins identified from wash and supernatant samples, 56 proteins were found with >1.5-fold abundance that indicates that 4.76% were extracellular proteins. The 19 abundant proteins found exclusively in the extracellular group showed a 1.61% rate of proteins identified in the wash and supernatant. The surface proteins were extracted in PBS containing 5 mM MgCl2 that was easily detachable and also had the unique presence of six proteins without shedding out into the supernatant, which showed its sufficiently good binding on the surface of Leptospira. It is also worth noting that 57% of the extracellular proteins found were pathogenic in nature, as was predicted by MP3 tool with a 27.9% rate (826 pathogenic proteins out of the 2957 proteins identified) in the whole proteome of Leptospira [22]. Even though some of these proteins were found in lower quantities, the possibility of upregulation of the proteins under pathogenic conditions cannot be ignored. This underlines the significance of identification of these extracellular proteins.
The bioinformatics prediction of extracellular proteins using online tools did not achieve any appreciable level, as the true prediction with respect to 56 extracellular proteins made by CELLO v.2.5, PSORTb, and BUSCA were 21.4%, 12.5%, and 8.9%, respectively. Similarly, five proteins of the extra-cellular proteins were identified as extracellular proteins predicted in a previous report [25]. This shows the need for further improvements in the algorithms to predict extracellular proteins of Leptospira.
Though the species was arranged against a preexisting phylogenetic hierarchy, based on ppk gene sequences, the data matched with the clusters and sequence of species. This shows that the extracellular proteins can discriminate between Leptospira species and the unique extracellular proteins may have key functions in invasion, pathogenesis, or survival of the organism in the host. These can be used for diagnostic applications and identification of Leptospira species. The orthologous proteins in other species of Leptospira showed that all the 17 pathogenic proteins were closely related to the L. interrogans with respect to the 26 proteins with >90% coverage and >75% identity among the species. Three proteins: the WP_000141830.1 a Multispecies preprotein translocase subunit YajC, the WP_000587664.1 as a VOC family protein, and the WP_000658301.1 as a DNA starvation/stationary phase protection protein were found to be more closely related within the pathogenic forms. With respect to the preprotein translocase subunit YajC, it was reported that mice vaccinated with the yajC of Brucella abortus showed immune responses to YajC [26].
With an abundance of 1700-times in wash with respect to cellular and supernatant fraction, histidine kinase sensor protein (LIC_11528) was identified as the most abundant and pathogenic protein identified on the surface of Leptospira in this study. The gene ontology was predicted to carry functions like signal transduction, and phosphorylation and molecular functions like phosphorylase sensor kinase activity and transferase activity by transferring phosphorus-containing groups. This shows that the protein is a two-component system with a histidine protein kinase (HPKs) and a response regulator protein [27]. The phosphorylation can induce conformational changes in the regulatory domain, resulting in the activation of the associated domain that affects the response. This shows that bacterial two-component pathways can control a dazzling array of functions like cell division, virulence, antibiotic resistance, response to environmental stress, sporulation, metabolite fixation and utilization, and taxis [28]. In addition, HPKs are unique signal transducers that are not common in animals, indicating this molecule can be a good target for an anti-leptospiral strategy [29]. Further, the STRING analysis identified the interaction of this domain with a cheY protein resembling the bacterial adaptions to the environment through the activation of specific sensory receptors along with signal processing proteins [30]. Apart from these properties, LIC_11528 is an ortholog of LA_2421 of L. interrogans serogroup Icterohaemorrhagiae serovar Lai and was found to be associated with the chemotaxis and signal transduction system [31]. Chemotaxis is one of the important mechanisms that drive a pathogen towards its target organs [32].
The second abundant protein was DNA helicase (LIC_11624,) which was found to be 38-fold greater on the surface and 87-fold greater in the supernatant than its entire quantity in the cell. This was found to be nonpathogenic in MP3 and the gene ontology showed molecular functions like DNA helicase activity, ATP binding, DNA binding, and hydrolase activity. The STRING analysis showed the association with uvr A and uvr B types of DNA repair gene homologs, which are involved in repairing of DNA that is damaged due to stress factors such as ultraviolet light [33,34]. DNA helicase is also associated with proteins like the DNA mismatch repair protein MutL, which determines the mismatched provoked excision step [35] and the RecQ helicase, which are widely conserved in bacteria [36] and helps in unwinding complementary strands of DNA required for the proper repair of DNA damage. However, the function of this DNA helicase as an extracellular entity is not yet known.
The putative lipoprotein containing peptidase M75 domain (LIC_10713) was found to be 78-times more abundant in the supernatant as a secretory molecule. The protein contains an Imelysin-like domain with a GxHxxE signature. This domain was distributed widely in bacteria and was found to be involved in iron transport [37]. This protein showed 100% Query Cover and 97.73% identity LruB, which was found to be playing a significant role in human and equine recurrent uveitis as well as antibodies against the protein are identified in patients diagnosed with Fuchs uveitis [38][39][40]. It is also found that the in vitro growth of Leptospira was significantly reduced when LruB is inactivated [41].
The gene LIC_10511 encoding the protein C39 peptidase, which has been found to be 75-fold abundant in the supernatant, was reported to be an endo-peptidase family that mostly serves as ABC transporters along with the translocation of the mature bacteriocin across the cytoplasmic membrane [42]. This protein was also abundantly present in the supernatant resembling type III secreted effectors (T3SEs), type IV secretion (T4SS), and a type VI secretion system [43][44][45]. The Type VI secretion system was widely reported in E. coli, which was found to be directly associated to pathogenesis leading to macrophage survival, which can further lead to events of lateral gene transfer [46,47]. This interconnected mechanism allows the bacteria to participate in metal uptake and provide an advantage during bacteria-bacteria competition. Further, it also allows the bacteria to widely deliver the effector toxic proteins directly into neighboring cells [48].
Apart from these four proteins, other proteins also showed a significant rise in quantity as secretory molecules as compared to their presence in aqueous, detergent, and pellet fractions. These proteins were found to carry functions like cell-cell signaling, determining nutritional requirement, stress response, external specific stimuli, and homeostasis, which were interpreted by InterProScan 5. Further, significant domain class proteins like chaperone proteins, VOC family, SGNH hydrolase, ComF, and ATP binding cassette proteins were also identified from our study. Previous studies reported the presence of these domains in Leptospira determining the functions like survival under stress conditions, cell-cell signaling, and binding to membrane receptors [49][50][51].

Study Design
The Leptospira culture in complete EMJH medium at standard culture condition was used for a complete subcellular proteomic analysis using Triton X-114, as shown in Figure 6. The study comprised of two parts: (1) subcellular proteome which includes all the fractions of Triton X-114 fractionation considered as cellular proteins; and (2) the extracellular proteins obtained from a wash of the Leptospira pellet (surface) and proteins enriched from the culture supernatant (secreted). The total amount of each protein from Triton X-114 fractions identified in part one was added together and considered as a cellular portion of the protein in contrast to the number of the same extracellular proteins found, which is part two. Pathogens 2021, 10, x FOR PEER REVIEW 12 of 18

Leptospira Strain and Culture
L. interrogans Copenhageni stain Fiocruz L1-130 was obtained from the repository of the ICMR-Regional Medical Research Centre, Port Blair, India. This is a WHO Collaborating Center for diagnosis, reference, research, and training in leptospirosis. Leptospira were cultured in EMJH medium supplemented with 1% bovine serum albumin at 30 °C with intermittent checking for contamination and growth. Afterwards, they were harvested at the mid-log phase for further protein extraction.

Enrichment of Surface and Extracellular/Secreted Protein
The mid-log phase culture of Leptospira was centrifuged at 2500× g for 30 min at 4 °C to obtain the culture supernatant for the separation of secretory protein. This supernatant was further centrifuged at 6000× g for 30 min to remove any Leptospira left in the medium and the supernatant was again centrifuged at 12,000× g for 30 min. The three-step centrifugation was to avoid tight packing and rupture of Leptospira while pelleting. This supernatant was used to separate secretory/extracellular proteins.

Extraction of Extracellular/Secreted Proteins
The secretory proteins present in low abundance in comparison with the BSA or serum proteins. Enrichment of secretory proteins was carried out using ProteoMiner™ Bio-Rad protein enrichment technology based on binding of proteins to a library of combinatorial peptide ligands that act as unique binders for proteins [52,53].
The supernatant was dialyzed against PBS (containing 150 mM NaCl, pH 7.4) to facilitate optimum binding condition to ProteoMiner™. Slurry from ProteoMiner™ Large-Capacity column (100 μL settled beads) was washed two times with 1 mL of PBS and added to 100 mL of the supernatant and allowed to bind overnight (>8 h) under shaking at 4 °C. After binding, the beads were allowed to settle and we removed the clear volume of supernatant, repacked in the ProteoMiner™ column, and carried out 2 × 100 μL washes with PBS. The elution 2 × 20 μL was made using Elution Reagent (8 M urea, 2% CHAPS) supplied by the manufacturer. The eluted secretory protein was subjected to

Leptospira Strain and Culture
L. interrogans Copenhageni stain Fiocruz L1-130 was obtained from the repository of the ICMR-Regional Medical Research Centre, Port Blair, India. This is a WHO Collaborating Center for diagnosis, reference, research, and training in leptospirosis. Leptospira were cultured in EMJH medium supplemented with 1% bovine serum albumin at 30 • C with intermittent checking for contamination and growth. Afterwards, they were harvested at the mid-log phase for further protein extraction.

Enrichment of Surface and Extracellular/Secreted Protein
The mid-log phase culture of Leptospira was centrifuged at 2500× g for 30 min at 4 • C to obtain the culture supernatant for the separation of secretory protein. This supernatant was further centrifuged at 6000× g for 30 min to remove any Leptospira left in the medium and the supernatant was again centrifuged at 12,000× g for 30 min. The three-step centrifugation was to avoid tight packing and rupture of Leptospira while pelleting. This supernatant was used to separate secretory/extracellular proteins.

Extraction of Extracellular/Secreted Proteins
The secretory proteins present in low abundance in comparison with the BSA or serum proteins. Enrichment of secretory proteins was carried out using ProteoMiner™ Bio-Rad (Hercules, CA, USA) protein enrichment technology based on binding of proteins to a library of combinatorial peptide ligands that act as unique binders for proteins [52,53].
The supernatant was dialyzed against PBS (containing 150 mM NaCl, pH 7.4) to facilitate optimum binding condition to ProteoMiner™. Slurry from ProteoMiner™ Large-Capacity column (100 µL settled beads) was washed two times with 1 mL of PBS and added to 100 mL of the supernatant and allowed to bind overnight (>8 h) under shaking at 4 • C. After binding, the beads were allowed to settle and we removed the clear volume of supernatant, repacked in the ProteoMiner™ column, and carried out 2 × 100 µL washes with PBS. The elution 2 × 20 µL was made using Elution Reagent (8 M urea, 2% CHAPS) supplied by the manufacturer. The eluted secretory protein was subjected to quantification, electrophoretic characterization, and trypsin digestion to obtain peptides and further high-resolution LC-MS/MS based proteomics.

Extraction of Surface Proteins
The pellet obtained after separation of supernatant was washed 3 × with PBS containing 5 mM MgCl2 and we collected the wash supernatant by centrifuging the leptospires at 2500× g for 5 min at room temperature. The wash was again centrifuged at 12,000× g for 30 min to remove any trapped leptospires and the supernatant was designated as a 'Wash fraction' that contains washable surface proteins of Leptospira and used for further processing to carry out LC-MS/MS.

Triton X-114 Extraction
The Leptospira pellet obtained after wash was used for Triton X-114 fractionation as described earlier [22]. The extraction buffer containing 10 mM Tris (pH 8) carrying 1% Triton X-114 and 150 mM NaCl at 4 • C at the rate of 1 mL of extraction buffer per amount of pellet derived from a 25 mL mid-log phase culture was used for extraction. The extract was centrifuged at 12,000× g for 30 min at 4 • C and the pellet was saved as a 'pellet fraction' and the supernatant was used for phase separation. The Triton X-114 concentration of the supernatant was increased to 2% by the addition of an adequate amount of Triton x-114, depending on the volume, mixed well, and incubated at 37 • C for 1 h for phase separation, then it was subsequently centrifuged at 1500× g for 5 min to separate the upper aqueous phase from the lower detergent phase. The undissolved proteins from the pellet from the TritonX-114 extraction step, which contained a cytoplasmic cylinder, were further extracted using a buffer containing 10 mM Tris-Cl (pH8), 8 M urea, 4 mM dithiothreitol, and 1% sodium dodecyl sulfate. Following centrifugation at 12,000× g for 30 min at 4 • C, the supernatant was used as the pellet fraction. Similar fractions of all four replicated were polled and the protein concentrations were estimated using the BCA method in the aqueous, detergent, and pellet fractions, which were then stored at −20 • C. This protein was used for mass spectrometry. Data from three replications of the same kind was used for further analysis.

Basic pH RPLC Based Fractionation
Lyophilized peptides were subjected to basic pH reverse phase chromatography (bR-PLC) fractionation. The samples were reconstituted in 1 mL of 10 mM Triethylammonium bicarbonate (TEABC) and separated on an XBridge C18 column (Waters Corporation, Milford, MA, USA; 130, 5 m, 250 × 4.6 mm) attached to a Hitachi LaChrom Elite HPLC system over 120 min using a linear gradient increase from 5% to 100% of 10 mM TEABC with 90% acetonitrile. Initially, 96 fractions were collected, which were then concatenated to 6 fractions and dried before desalting with C18 cartridges. Desalted peptides were vacuum dried and stored in a deep freezer at −80 • C prior to LC-MS/MS analysis.

LC-MS/MS Analysis
The tryptic peptides from bRPLC fractionation were analyzed on a Thermo Fischer Scientific Orbitrap Fusion Tribrid mass spectrometer (Thermo Fischer Scientific, Bremen, Germany) connected with an Easy-nLC-1200 nanoflow liquid chromatography system (Thermo Fischer Scientific). The lyophilized peptides were reconstituted in 0.1% formic acid and loaded onto a 2 cm trap column (nanoViper, 3 µm C18 Aq) (Thermo Fisher Scientific). Peptides were separated using a 15 cm analytical column (nanoViper, 75 µm silica capillary, 2 µm C18 Aq) at a flow rate of 300 nl/min. For data-dependent acquisition, solvent gradients were set as the linear gradient of 5-35% solvent B (80% acetonitrile in 0.1% formic acid) over 90 min through 120-min run time. MS analysis was carried out at a scan range of 400-1600 m/z mass range (120,000 mass resolutions at 200 m/z). The maximum injection time was 10 ms. For MS/MS analysis, data were acquired at top speed mode with 3 s cycles and subjected to a higher collision energy dissociation with 32% normalized collision energy. MS/MS scans were carried out at a range of 100-1600 m/z using Orbitrap mass analyzer at a resolution of 30,000 at 200 m/z. The maximum injection time was 200 ms.

MS/MS Data Analysis
Mass spectrometry-derived data were searched against the L. interrogans serogroup Icterohaemorrhagiae serovar Copenhageni (strain Fiocruz L1-130) reference protein database obtained from NCBI (3667 protein entries), with common contaminants added to the protein database (115 contaminants entries). The mass spectrometry data was analyzed with Mascot (versions 2.5.1; Matrix Science, London, UK) and SEQUEST-HT search algorithms in the Proteome Discoverer software suite, version 2.2 (PD 2.2) (Thermo Fischer Scientific, Bremen, Germany). The search parameters used were: (a) trypsin as the proteolytic enzyme (with up to one missed cleavage); (b) fragment mass error tolerance of 0.05 Da; (c) peptide mass error tolerance of 10 ppm; (d) oxidation of methionine as a variable modification; (e) carbamidomethylation of cysteine as a fixed modification. A false discovery rate (FDR) was set to 1% at PSM and peptide levels. The iBAQ (Intensity Based Absolute Quantification) value was generated using the iBAQ algorithm that estimates the relative abundance of the proteins within each sample [54].

Data Availability
The proteomics data of these mass spectrometry analyses have been deposited to the ProteomeXchange Consortium via the PRIDE [55], the partner repository with the dataset identifier.

Prediction of Protein Function and Pathogenic Nature
Functional annotation of the secretory protein is important to know the role of these proteins. Identification of functions and metabolic pathways of these proteins were made using the KEGG database (https://www.genome.jp/kegg-bin/show_brite?lic, accessed on 28 February 2021) [60]. Similarly, virulence and pathogenicity are key determinants of the severity of infection caused by a pathogen. To identify virulent proteins, we used MP3 (http://metagenomics.iiserb.ac.in/mp3/application.php, accessed on 12 April 2020) [61]. This tool is an SVM-based method to characterize the pathogenic proteins from the nonpathogenic ones.

Identification of Orthologous Proteins
A search for orthologous proteins of extracellular within 64 species of Leptospira was carried out using algorithm blastp (protein-protein BLAST) under online NCBI BLAST search at default parameters to retrieve top 1000 hits. Orthologous proteins of the highest score from each Leptospira species, irrespective of strains, were selected along with their 'query cover' and 'identity' values with respect to the query sequence.

Analysis of Interacting Proteins
To predict the protein-protein interaction among the species of Leptospira, we used String database (http://version10.string-db.org/, accessed on 21 May 2021) version 10. Under the search option, we entered the unique protein uniport ID and selected the autodetect option [62].

Conclusions
This study aimed to identify extracellular proteins of L. interrogans from protein-rich EMJH medium. It shows that the surface and secretory proteins can be easily identified with reference to the cellular proteins in quantitative terms. The extraction method was found to be easy, rational, and justified with the identification of exclusive molecules and significant times of abundance with respect to the cellular fraction of the proteins, though it was analyzed at higher resolutions due to three Triton X-114 fractions. Identification of pathogenic proteins and the correlation with pathogenic species shows the significance of the identified proteins. Similarly, a huge number (57%) of pathogenic proteins present as secretory molecules also highlight the significance of extracellular proteins. These key molecules identified can implement various functions like nutrient acquisition, cell-cell communication, detoxification of environment, and attaching to potential inhibitors. In this regard, the article presents an efficient method for extraction and analysis of extracellular proteins for other organisms too, as well as identification of extracellular proteome of L. interrogans.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/pathogens10070852/s1, Table S1: List of proteins identified in LC-MS/MS, Table S2: KEGG based search for function of extracellular proteins.  Institutional Review Board Statement: Ethical review and approval was not applicable for this study as not involving humans or animals.
Informed Consent Statement: Informed Consent was not applicable for this study as not involving humans or animals.

Data Availability Statement:
The LC-MS/MS proteomics dataset is available through ProteomeXchange dataset identifiers PXD009050, PXD016204 and PXD026044.