Proteomic profiling of Serratia marcescens by high-resolution mass spectrometry.

Introduction: Serratia marcescens, an opportunistic human pathogen, is reported as an important cause of nosocomial infection and outbreaks. Although the genome of S. marcescens (ATCC 13880) was completely sequenced by 2014, there are no studies on the proteomic profile of the organism. The objective of the present study is to analyze the protein profile of S. marcescens (ATCC 13880) using a high resolution mass spectrometry (MS).
Methods: Serratia marcescens ATCC 13880 strain was grown in Luria-Bertani broth and the protein extracted was subjected to trypsin digestion, followed by basic reverse phase liquid chromatography fractionation. The peptide fractions were then analysed using Orbitrap Fusion Mass Spectrometry and the raw MS data were processed in Proteome Discoverer software.
Results: The proteomic analysis identified 15 009 unique peptides mapping to 2541 unique protein groups, which corresponds to approximately 54% of the computationally predicted protein-coding genes. Bioinformatic analysis of these identified proteins showed their involvement in biological processes such as cell wall organization, chaperone-mediated protein folding and ATP binding. Pathway analysis revealed that some of these proteins are associated with bacterial chemotaxis and beta-lactam resistance pathway.
Conclusion: To the best of our knowledge, this is the first high-throughput proteomics study of S. marcescens (ATCC 13880). These novel observations provide a crucial baseline molecular profile of the S. marcescens proteome which will prove to be helpful for the future research in understanding the host-pathogen interactions during infection, elucidating the mechanism of multidrug resistance, and developing novel diagnostic markers or vaccine for the disease.

Introduction

Serratia marcescens is a gram-negative bacillus, belonging to the family Enterobacteriaceae. This opportunistic human pathogen may cause various infections ranging from urinary and respiratory tract infection, wound infection, endocarditis, meningitis, to conjunctivitis pneumonia, and may even lead to septicemia.[1,2] This pathogen is naturally present in soil and water, and accounts for about 1%-2% of the nosocomial infections and outbreaks occurring in intensive care units (ICUs), surgery units, medical wards, and neonatal ICUs.[3] A recent study reported that S.  marcescens  is the third most common pathogen involved in neonatal ICU outbreaks.[4] The overall rates of mortality due to S. marcescens vary from 25%-58%.[5,6] The drug resistance of S. marcescens to a broad spectrum of antibiotics complicates treatment and infection control in hospitals.[7] Recently, the World Health Organization (WHO) listed S. marcescens as one among the top ten global priority pathogens with multiple drug-resistance.[8,9] The complete genome sequence of S. marcescens strain ATCC 13880 was sequenced in 2014 and the total genome assembly constitutes 5 131 448 bp long including 4724 protein coding sequences.[10] Advances in high throughput technologies, such as proteomics, is emerging as an alternative strategy to address the multidrug resistance by understanding the mechanisms involved in drug resistance.[11,12] Proteomics can offer multiple advantages by the identification and characterization of novel proteins which are directly or indirectly involved in drug resistance that could be used as potential targets in future.[13-15]

In the last two decades, mass spectrometry (MS)-based proteomics has been effectively used to study proteome of microorganism, protein-protein interactions, differentially expressed proteins in order to better understand the molecular mechanism of microbial pathogenesis,[16] and antibiotic resistance mechanisms.[17] Compared to other important gram-negative bacteria, there are limited proteomic studies on S. marcescens.[18-22] Despite the previous efforts in S. marcescens proteome characterization, there are no studies involving a detailed proteomic profiling of S. marcescens ATCC 13880 strain.

In the present study, we have carried out a detailed proteome analysis of S. marcescens using high-resolution MS to provide the baseline data for future research. The study resulted in the identification of 2541 proteins, most of which were found to be associated with cell wall organization, chaperone-mediated protein folding, solute transport, signal transduction, ATP binding, metal ion binding, and protein transmembrane transporter activity. Proteins involved in important pathways including chemotaxis pathway, beta lactam resistance pathway, and bacterial secretion system pathway were also identified in the analysis. The information derived from this study could serve as a platform for future proteomic studies on the S. marcescens ATCC 13880.

Materials and Methods

Materials

Sodium lauryl sulfate (SDS), dithiothreitol (DTT), iodoacetamide (IAA), formic acid, acetonitrile (ACN), triethylammonium bicarbonate (TEABC) were supplied. All chemicals used were of analytical grade and were obtained from Sigma-Aldrich (St. Louis, MO, USA). Luria-Bertani (LB) (HiMedia Laboratories Pvt. Ltd, Mumbai, India), Sequencing grade modified trypsin (Promega, Madison, WI, USA), Bicinchoninic acid assay (BCA) (Thermo Fisher Scientific, Rockford, IL, USA). J. T. Baker MS grade water (Fischer Scientific, Waltham, MA, USA), 30 kDa Molecular weight cutoff (MWCO) spin filters (EMD Millipore, Billerica, MA, USA), Sep-Pak C18 columns (Waters Corporation, Milford, MA, USA), XBridge C18, 5 µm 250 x 4.6 mm column (Waters, Milford, MA, USA), Trap column (75 µm x 2 cm) packed in-house with Magic C18 AQ (Michrom Bioresources, Inc., Auburn, CA, USA) were also purchased.

Equipment

New Brunswick Scientific Incubator Shaker (IL Exella E24) (Edison, NJ, USA), Branson 450 Digital Sonifier w/ Probe (Marshall Scientific, Hampton, NH, USA), Speed VacTM SPD1010 system (Thermo Fisher Scientific, Bremen, Germany), Agilent 1100 series HPLC system (Agilent, Santa Clara, CA, USA), Orbitrap Fusion Tribrid mass spectrometer (Thermo Fisher Scientific, Bremen, Germany), Eppendorf Centrifuge 5424 R (Eppendorf, Hamburg, Germany), Eppendorf ThermoMixer C (Eppendorf, Hamburg, Germany) were applied.

Preparation of Serratia marcescens culture

Serratia marcescens strain was purchased from American Type Culture Collection (ATCC number 13880) and was grown on LB agar plates (37°C overnight). The LB liquid medium was prepared in 500 mL Erlenmeyer flask. Then, colonies from the LB plate were picked and inoculated in 500 mL Erlenmeyer flask and incubated for 18-24 hours at 37°C with shaking (300 RPM). The final concentration of the bacterial suspension was adjusted to 0.5 McFarland standards (~1.5 x 108 CFU/mL at 600 nm wavelength) and then bacterial suspension was subjected to centrifugation for a period of 10 minutes at 1500 x g at 4°C. Further, these cell pellets were collected and washed four times with ice-cold PBS and stored at −80°C.

Sample preparation and fractionation

The cell pellets with buffer containing 4% SDS were subjected to five cycles of sonication for 30 seconds and 20% amplitude with an interval of 5 seconds each in Branson 450 Digital Sonifier (Marshall Scientific, Hampton, NH, USA). The lysates were centrifuged at 14 000 RPM for 10 minutes and the clear supernatant was collected and aliquoted. One aliquot was used for protein estimation using BCA assay and the other aliquot was stored at –80°C till further analysis. In-solution digestion and fractionation was performed as described previously.[23,24] In brief, 500 µg of protein was subjected to reduction using 100 mM DTT and alkylation using 10 mM IAA, followed by buffer exchange with 8 M urea to reduce the SDS concentration, using a 30 kDa MWCO spin filters (EMD Millipore, Billerica, MA). The buffer exchanged sample was diluted to bring down the concentration of urea to 2 M and proteins were then digested using sequencing grade modified trypsin (Promega, Madison, WI. Cat #: V5111) overnight at 37°C.

The peptide digest was cleaned using Sep-Pak C18 columns using two solvents; solvent A - 0.1% formic acid and solvent B - 40% ACN, 0.1% formic acid. Briefly, the Sep-Pack C18 column was activated by passing 5 mL of 100% ACN, and then equilibrated the column twice with 5 mL of 0.1% formic acid. The sample was reconstituted in 5 mL of 0.1% formic acid and loaded onto the Sep-Pack C18 column. The first flow through was collected and re-loaded onto the column and then, the column was washed twice with 5 mL of 0.1% formic acid. Peptides were eluted by passing 3 mL of 40% ACN, 0.1% formic acid and the eluate was collected in fresh Falcon 15 mL tube (This step was repeated). The eluate containing cleaned peptides collected from Sep-Pak C18 columns was vacuum-dried using SAVANT SPD 1010 Speed Vac Concentrator as reported previously.[25] The vacuum-dried samples were reconstituted in basic reverse phase liquid chromatography (bRPLC) solvent A containing 10 mM TEABC in water at pH 8.5 and loaded on XBridge C18, 5 µm 250 x 4.6 mm column connected to Agilent 1100 series HPLC system. The peptide digest was resolved using a gradient of 0 to 40% solvent B containing 10 mM TEABC in ACN at pH 8.5 in 50 minutes. The total fractionation time was 60 minutes. A total of 96 fractions were collected and concatenated to 6 fractions based on the hydrophobicity as reported earlier.[26] These fractions were then vacuum dried and stored at -20°C until further LC-MS/MS analysis was performed. All buffers and reagents used in the study are provided in Supplementary file 1.

LC-MS/MS analysis

MS analysis was performed using the Orbitrap Fusion Tribrid mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) coupled to an Easy-nLC-1200 nanoflow LC system (Thermo Scientific). The peptide digests from each fraction were reconstituted in 20 µL of solvent A (0.1% formic acid in water) and around 10 µL of this sample having 2 µg of peptide was loaded onto trap column (75 µm x 2 cm) packed in-house with Magic C18 AQ (Michrom Bioresources, Inc., Auburn, CA, USA, 5 µm particle size, pore size 100 Å) at a flow rate of 5 µL/min with solvent A (0.1% formic acid in water). The separation of peptides was performed on an analytical column (75 µm x 15 cm) using a gradient of solvent A (0.1% formic acid) and solvent B (90% ACN with 0.1% formic acid) over 90 minutes at a flow rate of 300 nL/min. Six LC-MS/MS runs were performed. Data-dependent MS acquisition in positive mode was achieved at a resolution of 120 000 for the top ten precursor ions with a scan range of 350–1600 m/z. For MS/MS, top intense precursor ions from a 3-second duty cycle were preferred and subjected to higher-energy collision dissociation with 32% normalized collision energy. The fragment ions were detected at a mass resolution of 15 000 at m/z of 200 as explained previously.[23,27]

Data analysis

The MS/MS searches were performed using SEQUEST (version 2.4.1) search algorithm through Proteome Discoverer (version 2.2) software (Thermo Scientific) against S. marcescens (ATCC 13880) NCBI protein list (https://www.ncbi.nlm.nih.gov/). The search database was created by downloading all the proteins using “Serratia marcescens subsp. marcescens ATCC 13880” as search key words from NCBI. The search workflow included SEQUEST (version 2.4.1), spectrum selector, search nodes, spectrum files, Sequest HT, and Percolator. Carbamidomethylation of cysteine was set as a fixed modification, acetylation of protein N-termini, and oxidation of methionine, and deamidation of asparagine and glutamine was set as a variable modification. MS and MS/MS mass tolerances were set to 10 ppm and 0.02 Da, respectively. The data was searched against the decoy database, and 1% false discovery rate (FDR) at both peptides and peptide spectral match levels was applied for the analysis.[28] The Bioinformatics analysis of the proteomic data obtained in this study was performed using PANTHER (http://www.pantherdb.org/),[29] DAVID (V. 6.8),[30,31] STRING (V. 11.0) (https://string-db.org/),[32] and KEGG Pathway (http://www. genome.ad.jp/kegg/).[33]

Data availability

Raw MS files obtained in this experiment have been uploaded to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org),[34] through the PRIDE partner repository with the dataset identifier PXD011776.

Results

Proteomic analysis of S. marcescens (ATCC 13880)

The objective of the present study was to perform proteomic profiling to make a comprehensive proteome map of S. marcescens (ATCC 13880). Overview of the proteomic workflow to study the proteome of S. marcescens is shown in Fig. 1. Organism grown in LB medium post sonication extraction yielded ~8 mg/mL protein. Around 500 µg of the protein was subjected to in-solution digestion. Digested peptides were fractionated using bRPLC and these fractionated peptides were concatenated to 6 fractions and analyzed on LC-MS/MS. The analysis of these fractions resulted in the acquisition of 130448 MS/MS spectra, and the database searches against S. marcescens (ATCC 13880) NCBI protein database using SEQUEST resulted in the identification of 16 306 peptide-spectrum matches. A total of 15 009 unique peptides were identified (1% FDR), which mapped to 2541 protein groups accounting for 54% of the S. marcescens (ATCC 13880) proteome. The identified proteins and peptides are given in Tables S1 and S2 (see Supplementary file 2 and 3). The majority (78%) of the reported proteins were identified based on two or more unique peptides. On average, 6 unique peptides were found for each of the identified proteins. Representative mass spectra are shown in Supplementary file 4.

Fig. 1

Overview of the proteomic workflow applied to study the proteome of S. marcescens: S. marcescens ATCC 13880 was cultured, protein was isolated from cell pellets and digested using trypsin enzyme. The digest was fractionated using bRPLC, acquired data on mass spectrometry and analyzed the data using Bioinformatics approach.

Gene Ontology analysis

The selected list of proteins was analysed for their gene ontology (GO) using PANTHER classification system. The identified proteins were classified as those involved in biological processes, molecular functions, and according to their cellular localization (Fig. 2 and Table 1). With reference to the biological process, majority of the proteins (31%) were involved in binding, 24% in catalytic activity, 16% molecular transduction activity, another 16% in transcription regulator activity, and 13% in transporter activity.

Fig. 2

Table 1

Functional analysis of the proteins identified in the study

Description	Molecular function	Biological processes	Cellular localization	Domain
Molecular chaperone proteins identified
Hsp33 family molecular chaperone HslO	Unfolded protein binding	Protein folding	Cytoplasm	Redox-active center
Chaperone protein papD	Chaperone	Cell wall organization, chaperone-mediated protein folding	Outer membrane-bounded periplasmic space	Signal
Periplasmic fimbrial chaperone focC	Chaperone	Cell wall organization, chaperone-mediated protein folding	Outer membrane-bounded periplasmic space	Signal
Chaperone protein FimB	Chaperone	Cell wall organization, chaperone-mediated protein folding	Outer membrane-bounded periplasmic space	Signal
Chaperone protein YcdY	Chaperone	Bacterial-type flagellum-dependent swarming motility	Cytosol	Coiled coil
Chaperone protein fimC	Chaperone	Cell wall organization, chaperone-mediated protein folding	Outer membrane-bounded periplasmic space	Signal
Molecular chaperone DnaJ	Unfolded protein binding	Protein folding, response to heat	Cytoplasm	Repeat, Zinc-finger
Molecular chaperone DnaK	Unfolded protein binding	Protein folding, response to heat	Cytoplasm	Repeat, Zinc-finger
Molecular chaperone HtpG	Unfolded protein binding	Protein folding	Cytoplasm	Unknown
Molecular chaperone OsmY	Unknown	Unknown	Unknown	Unknown
Molecular chaperone Skp	Unfolded protein binding	Signaling	Outer membrane-bounded periplasmic space	Coiled coil
Chaperonin GroEL	Unfolded protein binding	Protein refolding	Cytoplasm	Coiled coil
ATP-dependent chaperone ClpB	Unfolded protein binding	Protein folding	Cytoplasm	Coiled coil
Molecular chaperone HtpG	Unfolded protein binding	Protein folding	Cytoplasm	Coiled coil
Co-chaperone YbbN	Unknown	Cell redox homeostasis	Cell	Thioredoxin
Fe-S protein assembly chaperone HscA	Unknown	Unknown	Unknown	Unknown
Heat shock chaperone IbpB	Chaperone	Protein stabilization	Cytoplasm	Unknown
Co-chaperone GroES	Unknown	Unknown	Unknown	Unknown
Co-chaperone DjlA	Chaperone binding	Signaling	Integral component of membrane	Transmembrane
Co-chaperone HscB	ATPase activator activity	Protein complex oligomerization, protein folding	Unknown	Coiled coil
Chaperone-modulator protein CbpM	Unknown	Unknown	Unknown	Coiled coil
Outer membrane proteins
Outer membrane protein A	Porin activity	Conjugation, ion transport	Cell outer membrane	Signal
Outer membrane assembly protein AsmA	Unknown	Regulation of protein targeting to membrane	Periplasmic space	Signal
Outer membrane autotransporter barrel domain-containing protein	Serine-type endopeptidase activity	Unknown	Outer membrane	Coiled coil
Outer membrane channel protein TolC	Efflux transmembrane transporter activity	Unknown	Outer membrane	Signal
Outer membrane lipid asymmetry maintenance protein MlaD	Unknown	Phospholipid transport	Integral component of membrane	Transmembrane,
Outer membrane lipoprotein chaperone LolA	Lipoprotein transporter activity	Protein transport	Periplasmic space	Signal
Outer membrane protein assembly factor	Unknown	Gram-negative-bacterium-type cell outer membrane assembly	Cell outer membrane	Signal
Outer membrane protein assembly factor BamA	Unknown	Gram-negative-bacterium-type cell outer membrane assembly	Cell outer membrane	Repeat
Outer membrane protein assembly factor BamB	Unknown	Gram-negative-bacterium-type cell outer membrane assembly	Cell outer membrane	Signal
Outer membrane protein assembly factor BamC	Unknown	Gram-negative-bacterium-type cell outer membrane assembly	Cell outer membrane	Signal
Outer membrane protein assembly factor BamD	Unknown	Gram-negative-bacterium-type cell outer membrane assembly	Cell outer membrane	Signal
Outer membrane protein assembly factor BamE	Unknown	Gram-negative-bacterium-type cell outer membrane assembly	Cell outer membrane	Signal
Outer membrane protein OmpW	Unknown	Transmembrane transport	Cell outer membrane	Signal
Outer membrane protein OmpX	Identical protein binding	Protein insertion into membrane from inner side	Outer membrane-bounded periplasmic space	Signal
Outer membrane usher protein	Fimbrial usher porin activity	Pilus assembly	Cell outer membrane	Transmembrane
Two component systems proteins
Two-component sensor histidine kinase BarA	Phosphorelay sensor kinase activity	Unknown	Integral component of membrane	Transmembrane
Two-component system QseEF-associated lipoprotein QseG	Unknown	Unknown	Unknown	Coiled coil
Two-component system response regulator ArcA	DNA binding	Phosphorelay signal transduction system	Unknown	Unknown
Two-component system response regulator BaeR	DNA-binding transcription factor activity	Regulation of response to drug	Cytosol	Coiled coil
Two-component system response regulator BasR	DNA binding	Phosphorelay signal transduction system	Unknown	Unknown
Two-component system response regulator CreB	DNA binding	Phosphorelay signal transduction system	Unknown	Unknown
Two-component system response regulator DcuR	ATP binding	Regulation of response to drug	Integral component of membrane	Transmembrane
Two-component system response regulator GlrR	DNA binding	Phosphorelay signal transduction system	Unknown	Unknown
Two-component system response regulator NarL	DNA binding	Phosphorelay signal transduction system	Unknown	Unknown
Two-component system response regulator OmpR	DNA-binding transcription factor activity	Phosphorelay signal transduction system	Cytosol	Unknown
Two-component system response regulator PhoP	DNA-binding transcription factor activity	Phosphorelay signal transduction system	Cytosol	Unknown
Two-component system response regulator RstA	DNA-binding transcription factor activity	Phosphorelay signal transduction system	Cytosol	Unknown
Two-component system response regulator UvrY	DNA binding	Phosphorelay signal transduction system	Cytoplasm	Unknown
Two-component system response regulator YehT	DNA binding	Phosphorelay signal transduction system	Cell inner membrane	Unknown
Two-component system sensor histidine kinase CpxA	ATP binding	Signal transduction by protein phosphorylation	Integral component of membrane	Unknown
Two-component system sensor histidine kinase DcuS	ATP binding	Phosphorelay signal transduction system	Integral component of membrane	Transmembrane
Two-component system sensor histidine kinase EnvZ	ATP binding	Phosphorelay signal transduction system	Integral component of membrane	Transmembrane
Two-component system sensor histidine kinase PhoQ	ATP binding	Phosphorelay signal transduction system	Integral component of membrane	Transmembrane
Two-component system sensor histidine kinase PhoR	ATP binding	Cellular response to phosphate starvation	Integral component of membrane	Transmembrane
Two-component system sensor histidine kinase QseC	ATP binding	Cellular response to phosphate starvation	Integral component of membrane	Transmembrane
Two-component system sensor histidine kinase RcsC	ATP binding	Cellular response to phosphate starvation	Integral component of membrane	Transmembrane

Gene Ontology analysis: A histogram depicting the distribution of Serratia marcescens proteins among various (A) biological processes and (B) molecular functions and (C) subcellular localizations.

Functional enrichment analysis of the identified proteins

The identified proteins were further categorized into different classes by sorting them based on their names manually. These classes include molecular chaperone proteins, outer membrane proteins, two-component systems proteins, and type VI secretion system proteins.

Protein network and pathway analysis

To understand the biological network complex involved, STRING analysis was performed.[32] The analysis showed protein–protein interaction networks that are associated with protein export, cellular localization, bacterial secretion, protein insertion into membrane, membrane organization, and signal transductions. The detailed interactions of these networks are given in Fig. 3. Pathway analysis of the S. marcescens proteomic data resulted in the identification of two component systems pathway, chemotaxis pathway (Fig. 4), beta-lactam resistance pathway (Fig. 5), and bacterial secretion system pathways (Table 2).

Fig. 3

Fig. 4

Fig. 5

Table 2

List of Type VI secretary proteins identified

Description	Gene symbol	T6SS subunit	Domain
Hcp1 family Type VI secretion system effector	hcp1	Hcp1	Unknown
Type VI secretion system ATPase TssH	clpV	TssH	Coiled coil
Type VI secretion system baseplate subunit TssF	vasA	TssF	Unknown
Type VI secretion system baseplate subunit TssG	tssG	TssG	Unknown
Type VI secretion system baseplate subunit TssK	tssK	TssK	Unknown
Type VI secretion system contractile sheath large subunit	tssC	tssC	Unknown
Type VI secretion system contractile sheath small subunit	tssB	tssB	Unknown
Type VI secretion system lipoprotein TssJ	tssJ	TssJ	Unknown
Type VI secretion system membrane subunit TssM	icmF	TssM	Transmembrane
Type VI secretion system protein TssA	tssA	TssA	Unknown
Type VI secretion system protein TssL	tssL	TssL	Transmembrane
Type VI secretion system tip protein VgrG	tssI	VgrG	Unknown
Type VI secretion system-associated FHA domain protein TagH	tagH	TagH	Unknown
Type VI secretion system-associated protein TagF	tagF	TagF	Unknown

Interaction network analysis: STRING interaction network depicts the association among the proteins. Color balls are representing individual protein. Red: involved in beta-lactam resistance. Green: Bacterial secretion system, Blue: Bacterial chemotaxis and Yellow: Two-component systems.

Pathway map representation of identified proteins involved in bacterial chemotaxis pathway.

Pathway map representation of identified proteins involved in beta-lactam resistance pathway.

Discussion

The sequencing of bacterial genome has contributed a lot to elucidating various aspects of bacterial infection. However, a major obstruct in utilizing the bacterial genomic sequence for the study of pathogenesis is the fact that there are many parts of the genes with unknown function in each organism sequenced to date.[35] Hence, a detailed proteomic profiling of such organisms are important to draw conclusions on the host-pathogen interaction during an infection. Recent advances in MS instrumentation and the application of these sensitive high-throughput analytical instruments to study the proteins is opening new perspectives in microbiology.[36] In the last two decades much attention has been paid to study the proteome of different human pathogenic bacteria.[37,38] S. marcescens is one of the pathogenic bacteria and many strains of this bacterium have been sequenced at genome and proteome levels.[10,18,20,2239-41] To the best of our knowledge, the genome of S. marcescens (ATCC 13880) sequence is available and the proteome of the same is not available till date.[10] Hence, this study was performed to understand the proteome of S. marcescens (ATCC 13880). This strain was initially isolated from pond water and later known to be pathogenic to humans especially children and the immunocompromised patients in the ICU units.[42,43] This is the first detailed proteomic analysis for S. marcescens (ATCC 13880) in which we were able to identify 2541 protein groups accounting for 54% of the S. marcescens (ATCC 13880) proteome. In this study, we reported the identification of important class of proteins which are essential for the growth and function of S. marcescens including molecular chaperone proteins, outer membrane proteins, two-component system proteins and type VI secretion system proteins. Different studies have shown that these classes of proteins play an important role in the physiology and virulence of S. marcescens.[44-49]

Analysis of molecular chaperone proteins

In this study, we identified 21 molecular chaperone proteins and analysis of these proteins showed that they are mainly involved in cell wall organization, chaperone-mediated protein folding and response to heat. Further, we found that most of them are located in cytoplasm and outer membrane-bounded periplasmic space having coiled coil or signal domain. These classes of proteins include 17 KD protein Skp and molecular chaperone protein dnaK (DnaK). Molecular chaperone 17 KD protein Skp is a trimeric periplasmic 17 KD protein that helps outer membrane proteins in their folding and incorporation into membranes.[50] Skp is a significantly important protein in the chaperone system which has been distinguished for its function in outer membrane protein (OMP) biogenesis,[51] where it interacts with unfolded proteins.[52] Skp assists in releasing the translocated proteins from the plasma membrane and corrects folding of numerous OMPs.[53,54] However, the molecular mechanisms of how Skp performs its chaperone role have remained unclear,[55,56] and it has been also reported that it does not show sequence similarity to any other chaperone.[50]

The protein DnaK, another important protein in the category, is a heat shock protein 70 (hsp70) which is a member of the molecular chaperones family. Studies show that hsp70s are essential part of the cell’s protein folding, degradation and repair machinery.[57] DnaK proteins coat polypeptide chains as they come out from membranes or ribosomes, averting accumulation reactions and premature folding.[58] DnaK helps cell to prevent denaturation of proteins and support the deprivation of the damaged proteins during stress conditions and play a vital role in chromosomal DNA replication (Table 1).[18]

Analysis of outer membrane proteins

The outer membrane of gram-negative bacteria plays a role in membrane permeability and avoid the entry of toxic compounds and allow the influx of nutrient molecules.[59] OMPs perform multiple functions including solute transport, membrane structure, signal transduction, stability and catalysis.[60] In this study, we found 15 such outer membrane proteins. Functional analysis of the important proteins of this class includes outer membrane protein A (OmpA) and Outer membrane protein X (OmpX).

The OmpA belongs to the family of OMPs and it is conserved among gram-negative bacteria.[61] OmpA proteins have been reported to play important pathogenic roles such as adhesion to mucosal surfaces, intracellular survival or invasion, stimulators of proinflammatory cytokine production, and antimicrobial peptide resistance.[62] In S. marcescens, OmpA is an outer membrane protein with porin activity and conjugation or ion transport processes. The OmpA has been reported to play structural and ion-permeable porin roles which help in bacterial survival during osmotic stress conditions.[63]

OmpX proteins are small molecular weight (15 to 18 KD) proteins and belong to a family of conserved bacterial proteins where they support adhesion to and entry into mammalian cells.[64,65] These proteins fold in an eight-beta-barrel structure through membrane-spanning domains that project from the cell surface.[66] OmpX are involved in physiological processes such as antibiotic resistance, invasion, signal transduction, internalization in epithelial cells, binding external proteins, contributing to channeling and virulence.[64,67-69]

Analysis of two-component system proteins

The ability to sense and respond to its surroundings is fundamental to the survival of a bacterium. Two-component regulatory systems are reported as key mechanism through which bacteria perceive and respond to their environment.[70,71] Two component system proteins have been studied to be associated with physiological functions including competence, antibiotic resistance, cell-cell communication, adaptation to environments, and sporulation.[72] In this study we reported 22 such two-component system proteins. PhoP/PhoQ system is one among the 22 component system proteins reported in this study. It is a two-component system that controls the expression of genes associated with virulence, the adaptation to Mg 2+ restrictive environments, and controls various cellular activities in numerous gram-negative species.[73] Studies reported that in non-pathogenic bacteria PhoP/PhoQ system plays a role in the control of the magnesium homeostasis, whereas in pathogenic bacteria it plays an essential role in the expression of vital virulence phenotypes.[74] It is also reported that activity of the PhoP/PhoQ system permits Serratia to survive scarce environmental Mg 2+ availability, acidic pH, and the action of the cationic antimicrobial compound polymyxin B.[75]

Analysis of type VI secretion system proteins

Gram-negative bacteria have evolved several secretion systems to deliver toxins and proteins into target cell, which are essential for virulence and survival within hosts. Type VI secretion system is widely used by gram-negative bacteria and it functions as nano-machine which expands from cytoplasm to the outer membrane to inject effector proteins into target cells.[76,77] T6SS is composed of at least 13 core-component proteins and is associated with auxiliary component proteins.[77] These core-component proteins are categorized into different subunits: the membrane complex (TssJLM), the baseplate complex (TssAEFGK), needle sheath complex (TssBC), sheath recycling ATPase (TssH), and the injection apparatus (Hcp, VgrG).[78] In the present study we found 12 such core-component proteins (TssABC, Hcp1, TssFGH, VgrG, TssJKLM) and two additional auxiliary component proteins (TagH and TagF) (Table 2). Interestingly, most of these proteins found to be in the cytoplasmic region which suggests that these cytoplasm proteins can be further investigated for enhanced understanding of secreted protein transport during the pathogenesis.[19] These proteins have been reported to play a role in various diseases.[2,79]

Proteins associated with bacterial chemotaxis pathway and beta lactam resistance pathway were identified by the proteomic profiling of S. marcescens chemotaxis, the capacity of gram-negative bacteria to move in response to external stimuli has long been concerned in the bacterial colonization and pathogenicity.[80] In this study we identified 7 proteins which belong to chemotaxis pathway (Fig. 4). The prokaryotic chemotaxis pathway senses the external chemoeffector concentrations by the transmembrane receptor methyl-accepting chemotaxis protein (MCP).[81] The binding of chemoeffector ligand to a chemoreceptor is signaled across the membrane to regulate the autophosphorylation activity of the CheA kinase with the help of an adaptor protein CheW. Once CheA is phosphorylated, the phosphoryl group is transferred to response regulator CheY. Phosphorylated CheY diffuses away from the signaling cluster and binds with the flagella motor switch proteins FliG, FliM and FliN to cause a change in the rotation direction of the motor. Interruption of any component of the chemotaxis signal transduction pathway (MCP receptor-CheA/CheW-CheY-flagella cascade) reduces the ability of pathogens to colonize and cause disease.[82] Drug resistant bacterial infections have significant impact on the morbidity and mortality, as a consequence of this WHO has raised serious concern to address the growing multidrug resistance in gram-negative bacteria.[9] Bacteria have developed several strategies of defense against classes of antibiotics, including elimination of antibiotics through efflux pumps, enzymatic inactivation of antibiotics (production of beta-lactamase enzymes) and expression of low affinity targets (penicillin-binding proteins).[83] In this study we reported ten key proteins which belong to beta-lactam resistance pathway that may play an important role under these three strategies. TolC-outer membrane channel protein and AcrAB-multidrug efflux pump transcriptional activator found to play a role in efflux pumps (RND family),[84,85] AmpG- muropeptide transporter AmpG, nagZ-beta N-acetylhexosaminidase, AmpR- LysR family transcriptional regulator, SRT/SST family class C beta-lactamase, and class B metallo-beta-lactamases associated with enzymatic inactivation of antibiotics,[86] whereas, carboxypeptidase/penicillin-binding protein 1a, penicillin-binding protein 1b and penicillin-binding protein 2 play a role in the expression of low affinity targets.[87,88] Further studies are essential to understand the detailed role of these proteins in drug resistance mechanism.

Few studies have reported the proteomic analysis of this species, which helped in elucidating the molecular mechanisms underlying the resistance to, and removal of, some toxic elements such as manganese from the environment. Zakeri et al identified 130 proteins from the strain ZF03 of S. marcescens under radium stress condition using MS/MS analysis.[18] In another study Fritsch et al used LC-MS/MS to identify 842 proteins from the type VI secretion system (T6SS) dependent secretome of S. marcescens Db11 strain.[19] Yi et al used 2D MALDI-TOF MS analysis and reported 23 expressed proteins in S. marcescens ATCC 14041.[20] Cianfanelli et al identified 1025 proteins using LC-MS/MS based proteomic approach to measure the secretion of T6SS and VgrG dependent effector proteins in S. marcescens Db10 strain.[21] Recently, Tuveng et al reported the identification of 232 secreted proteins from S. marcescens BJL200 strain.[22] In one of the recent studies, Queiroz et al reported the alterations in the proteomic composition of S. marcescens in response to manganese (II).[89] They performed the comparative study with the presence and absence of Mn (II). They reported the identification of structural proteins and enzymes related to major cellular and metabolic processes required for growth and survival of gram-negative bacteria. Additionally, they have shown the differential expression of these proteins in the absence and presence of Mn II. Through this study, they report the application of S. marcescens LG1 strain in bioremediation applications, in particular removal of Mn (II) from the drinking water and industrial water.

The highlight of the present study is the use of high-sensitive mass-spectrometry to report a large number of proteins in S. marcescens (ATCC 13880) which was not available earlier. We believe that the proteins reported in this study may be used to compare the proteomic changes in the stress-induced/disease conditions and other situations, which may help to draw the inferences on the role of this organism in antibiotic resistance.

Conclusion

Serratia marcescens, one of the important human pathogens is ubiquitously distributed which causes nosocomial infections and outbreaks especially in children and immunocompromised patients. The situation is further worsened by the development of resistance to beta-lactam antibiotics of S. marcescens. Recent advances and successful microbial proteomic experiments with MS-based proteomics proves that MS can be used as a tool to understand the interaction between pathogens and their host such as that in infections and drug resistance. It involves the study of protein–protein interactions on a small and proteome-wide scale, defining proteome of several organelles, analysis of post-translational modifications, and quantitative estimation of change in the abundance of proteins. MS can be used to identify and quantify thousands of proteins in samples which have lots of applications in the field of clinical microbiology. LC-MS/MS-based proteomic studies will set the stage for the identification of proteins that are involved in virulence and pathogenesis of this organism and the host-pathogen interaction during infection. In the present study, we performed a detailed proteomic analysis of S. marcescens ATCC 13880, using MS-based advanced proteomic and bioinformatic approach to elucidate the proteome profile of S. marcescens ATCC 13880 which resulted in the identification of 2541 proteins. To the best of our knowledge, this study provides an in-depth coverage of the proteome and functional analysis of identified proteins. The study supplements the microbial proteome project and presents an essential knowledge of the S. marcescens ATCC 13880 proteome.

Acknowledgments

Authors thank the Management and Principal of MS Ramaiah Institute of Technology, Bangalore for their support and facility.

Funding sources

Nil.

Ethical statement

Not applicable.

Competing interests

The authors declare there is no conflict of interests.

Authors’ contribution

SR, GS, and BSG conceived and planned the project, the main conceptual ideas, and design of the experiments. BSG implemented the experiments. SR, GS, and BSG contributed to data analysis and interpretation of the results. BSG wrote the manuscript. SR and GS reviewed final version of the manuscript. All authors provided helpful feedback.

Supplementary file 1, 2, 3 and 4 contain buffers and reagents used in the study, Table S1, Table S2 and representative mass spectra, respectively.

Click here for additional data file.

Click here for additional data file.

Click here for additional data file.

Click here for additional data file.

√ The complete genome sequence of S. marcescens strain ATCC13880 is available and computationally predicted to have 4724 protein coding sequences.

√ Proteome of other strains of S. marcescens is available where as the whole proteome of S. marcescens ATCC 13880 remains largely unexplored.

√ The study provided experimental evidence at the level of protein expression for the computationally predicted gene annotations for the genome of S. marcescens strain ATCC 13880.

√ The study provided the list of 2541 unique protein groups, which corresponds to ~54 % of the computationally predicted protein-coding genes with peptide level evidence and their associated Bioinformatics analysis.

√ The data and outcome of the study is valuable in addition to the microbial proteome project.