K. pneumoniae was obtained from American Type Culture Condition (ATCC). Escherichia coli OP50 was provided by Caenorhabditis Genetic Center (CGC), MN, USA. Both the cultures were inoculated in Luria Bertani medium and incubated at 37°C.

The C. elegans wild type and mutant strains were originally obtained from the CGC, MN, USA and grown on a solid nematode growth medium (NGM) plates containing E. coli OP50 as a standard food source. The C. elegans strains used in the present study were Bristol N2 (wild-type), VC2312 (let-363) mTOR mutant, BQ1 (akt-1), and VC858 (pdi-2). The age-synchronized L4 C. elegans were used in all bioassays. To obtain L4 staged C. elegans, the gravid hermaphrodites containing eggs inside the uterus were treated with commercial bleach solution containing 5 M potassium hydroxide in a ratio of 1:1 and the resulting solution containing eggs were transferred to the fresh NGM plates seeded with E. coli OP50. The plates containing eggs were incubated at 20°C for favoring the hatching. Thus, obtained stage-synchronized C. elegans were allowed to reach the L4 stage and used in all bioassays.

The age-synchronized L4 stage C. elegans were infected with K. pneumoniae for different time points (12, 24, and 36 h). Here, the nematodes fed on E. coli OP50 at respective time points were served as controls. After infection, the nematodes were washed thoroughly with sterile M9 medium to remove the bacterial contaminants. The samples were then flash frozen in lysis buffer [7 M urea, 2 M thiourea, 4% CHAPS, and 30 mM Tris-HCl, pH 8.5, and protease inhibitor cocktail (Sigma)] and stored at −86°C. When required, the samples were homogenized on ice for 1 min with 5 s pulses for every 5 s by employing sonicator. The insoluble cellular debris was separated by centrifugation at 12,000 rpm for 15 min at 4°C. The protein concentrations were determined with a Bradford assay Kit (Bio-Rad) using bovine serum albumin as a standard.

The 2D-gel electrophoresis was performed according to Balasubramanian et al. (2016) with minor modifications. A concentration of 1 mg of protein was taken in the volume range of (50–75 μl). The samples were processed to remove the non-specific biological contaminants such as lipids, nucleic acids, and salts using 2D cleanup kit (GE healthcare). The protein samples were dissolved in sample buffer containing urea and thiourea. Prior to 2DE, the sample containing total protein was loaded onto an immobilized pH gradient (IPG) strips (24 cm, pH 3–10 NL, GE Healthcare) and rehydrated for 12 h at room temperature. The rehydrated strips were subjected to isoelectric focusing (IEF) using Ettan™ IPGphor 3 isoelectric focusing system and accessories purchased from GE Healthcare. Then the strips were equilibrated in SDS equilibration buffer containing DTT (10 mg/ml) and iodoacetamide (25 mg/ml) for first and second equilibration respectively for 20 min each. The second dimensional separation of proteins was performed using 12% denaturant SDS polyacrylamide gels. After electrophoresis, the gels were stained with Coomassie Brilliant Blue (CBB) G-250.

The protein separated in SDS gels were digitized using Gel scanner-III and the scanned gel images were analyzed with the ImageMaster 2D platinum version 7 software (GE Healthcare). Briefly, after automatic spot detection, the manual editing of images like images like adding, splitting and removing of spots were done to increase the recognition of majority of spots. To match the spots across gels in each replicates, a class and match set were obtained with the available gel images. Among them, based on the gel and spot quality one gel was selected as the master gel for each time points, against which all other respective gels were matched. Spots, found in a match set member but absent in the master gel were added manually to the master gel. Then the automatic matching of spots on each gel was performed and matching across gels was manually verified. For each, match set, each gel was normalized to minimize the variability due to slight difference in protein load per gel, staining efficiency and image capture. This normalization was obtained by dividing the raw intensity of each spot in a gel by the total intensity of all spots in that gel that was included in the master gel. And all the analyses were performed on normalized quantities. The normalized values of all proteins were exported to in-built statistical tools for statistical analysis. The protein spots which had significant changes of >1.5-fold and p < 0.05 were considered as differentially regulated proteins.

The differentially regulated proteins were identified by following the procedure of Durai et al. (2014) with minor modifications. In brief, the protein spots of interest were manually exercised from the SDS PAGE gel with end-remove pipet tips to accommodate the various diameters of spots. The gel pieces in the microtube was destained and dehydrated with 30% acetonitrile (ACN) and 100 mM ammonium bicarbonate in 1:1 ratio. After washing, the wash solution was removed and the gel pieces were lyophilized. The lyophilized gel pieces were further dehydrated thrice with 50 μl of ACN and subsequently reswollen with 50 μl of 50 mM ammonium bicarbonate for 10 min. Again the protein spots were rehydrated with Acetonitrile and lyophilized. The protein containing gel spots were subjected to reduction and alkylation by adding 10 mM DTT and 50 mM iodoacetamide, respectively. Subsequently, the traces of DTT and iodoacetamide were removed from the gel spots by dehydrating and lyophilization. Then the dried proteins spots were rehydrated in a protease solution containing 2.5 ng/μl of trypsin (Sigma) solution and incubated in ice for 45 min. The mixture was incubated at 37°C for trypsinization. After 16 h of enzymatic digestion and vortexing for 2 min at high speed, the peptides were extracted by adding extraction buffer containing 50 mM ammonium bicarbonate, 50% can, and 0.1% formic acid. The step was repeated twice to increase the yield. The extracts were pooled together and concentrated by using speed Vac. The peptides were dissolved in MS grade water and were desalted using ZipTipC 18 pipet tips (Millipore). About 0.5 μl of α-Cyano-4-hydroxycinnamic acid was mixed with equal volume of desalted samples and allowed to dry for 4–6 h. The MALDI TOF/TOF analysis was performed using MALDI-TOF/TOF analyzer (AXIMA Performance, SHIMADZU BIOTECH) in a positive reflectron ion mode. With peptide mass fingerprint (PMF) and MALDI TOF/TOF, the proteins were identified against the all entries of C. elegans database using the MASCOT (Matrix Science Version 2.2) as a search engine. During search, the carbamidomethylation and oxidation was set as fixed and variable modifications, respectively. In the MASCOT search, the mass tolerance of 1.2 Da with one missed cleavage per peptide was allowed in all searches. Proteins with constant hit against different mass tolerance and significant scores were considered. All the proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Vizcaino et al., 2016) partner database with the dataset identifier PXD007151.

The protein-protein interaction among the differentially regulated proteins identified at different time points were explored and the interacting-network map was created by using STRING 10.0 with a high confidence score of 0.7. The interacting partners were identified for their functions and annotated manually (Schmutz et al., 2013). The Gene ontology (GO) and the functional enrichment was performed using DAVID and KEGG pathway analysis. In order to eliminate the redundancy of proteins and to reduce the number of terms to a smaller meaningful set, the list of identified proteins were analyzed by ReviGO tool (Supek et al., 2013; Larance et al., 2015).

The proteins isolated from C. elegans exposed to K. pneumoniae were separated by liquid-phase isoelectric focusing (IEF) with the MicroRotofor cell (Bio-Rad Laborotaries), according to their isoelectric points (pI). Approximately 2 mg of protein samples were mixed with IEF buffer [7 M urea, 2 M thiourea, 4% CHAPS and 0.24% Triton X100, 5% glycerol, and 1.6% ampholytes (pH 3–10)]. The focusing chamber of MicroRotofor cell was loaded with 2.5 ml of samples and the electrode cells were filled with appropriate electrolytes. The IEF was performed by applying a constant power of 1 W at room temperature. After 3 h of focusing, protein fractions were separated into different compartments (~200 μl) and were collected immediately to avoid the diffusion of separated fractions. For identification purpose, the collected fractions were trypsinized and analyzed using nano-RPLC coupled with an Orbitrap Elite Mass spectrometer (Thermo Scientific, USA). The digested peptides were dissolved in 2% ACN containing 0.1% formic acid and loaded to a C18 guard column (5 μm, 100 μm × 2 cm) (Thermo Scientific). The injected piptides were allowed to pass through a linear solvent system of 5–100% ACN for 80 min with a flow-rate of 300 nl/min and given voltage of 1.9 kV. The final MS data acquired were over the mass range of m/z 350–4,000 Da and visualized using Xcalibur software (version 2.2.SP1.48) (Thermo Scientific USA) (Ananthi et al., 2011). The data analysis was done using Proteome Discover software v. 1.4 based on sequest algorithm using a database downloaded from Uniprot. During search, the carbamidomethylation of cysteine was considered as fixed modification and methionine oxidation, N-terminal acetylation and phosphorylation (S, T, Y) as variable modifications. Here, the precursor mass tolerance of 10 ppm, fragment mass tolerance of 0.5 Da with 2 missed cleavages was allowed in protein identification.

For blotting, nematodes exposed to K. pneumoniae for different time-points (12, 24, and 36 h) were washed thoroughly with M9 buffer and homogenized in ice cold Phosphate buffered saline containing a protease inhibitor cocktail. The soluble fraction of lysates was separated from the mixture by centrifugation at 12, 000 rpm for 5 min at 4°C. The final protein concentration of the samples was determined with Bradford solution (Bio-Rad). For each experimental run, equal concentration of proteins were taken and mixed with a SDS-sample loading buffer. The mixture was boiled at 100°C for 2–3 min. After heat-denaturation, the samples were subjected to SDS-polyacrylamide gel electrophoresis, followed by electrophoretic transfer onto the polyvinylidene fluoride (PVDF) membranes (Amersham Biosciences). Then the membranes were transferred to the primary antibodies solution (1:2,000) and incubated for 6 h at 4°C. After required time of incubation, the membranes were washed thrice with tris-buffered saline containing 0.1% tween-20 and transferred to the secondary antibodies solution (1:1,000) conjugated with alkaline phosphatase for 4 h at 4°C. Finally, the membranes were transferred to the developing solution containing nitro-blue tetrazolium (NBT) and 5-bromo-4-chloro-3-indolylphosphate (BCIP). The membranes were allowed to develop until the intense bands were observed (Kamaldevi and Balamurugan, 2016). The antibodies used in the current study are, AKT, pAKT, mTOR, pmTOR (Cell signaling), actin (sigma), PI3-kinase, pPI3-kinase, PDI-2, anti-rabbit and anti-mouse conjugated with alkaline phosphatase (Santa Cruz Biotechnology).

Approximately 10 worms were transferred to the sterile M9 buffer containing K. pneumoniae (20%, 3 × 10⁵ cells/ml), in a sterile 24-well culture plate and incubated at 20°C. The survival of the nematodes was monitored at an interval of an hour till it attained complete mortality. The nematodes that did not show any response to touch and lack of pharyngeal pumping were considered as dead. Here, the nematodes fed on E. coli OP50 were considered as control. Each experiment was performed atleast three times independently under the similar conditions.

The defect caused by the K. pneumoniae in the nematodes' intestine was monitored under microscope. The infected nematodes were washed thrice with M9 buffer containing 1 mM sodium azide (anesthetizing agent) and placed on a 2% agarose pad. The physiological changes with the exposed C. elegans were examined under inverted fluorescence microscope (Nikon, Japan).

The reactive oxygen species (ROS) production in nematodes during K. pneumoniae infection was quantitatively analyzed by confocal laser scanning microscopy (CLSM). The nematodes exposed to pathogen were washed thrice to remove the surface bacteria and transferred to the liquid M9 medium containing 5 μg/ml of 2′, 7′-dichlorofluoresceine-diacetate (DCFH-DA). The content was incubated in dark for 10 min at room temperature. Then the nematodes were washed thoroughly with PBS to remove the excess DCFH-DA and placed the worms on an agar pad (2%) containing an anesthetizer (30 mM sodium azide). Finally, the slides were subjected to microscopic observations (Carl zeiss, Germany). The relative fluorescent intensity of 2′, 7′-dichlorofluorescein (DCF) in the nematodes represents the amount of ROS production in nematodes (Kamaladevi et al., 2013).

The quantitative measurement of reactive oxygen species was performed by (DCFH-DA) method (Scherz-Shouval et al., 2007). The nematodes exposed to K. pneumoniae for different time-points (12, 24, 36 h) were washed thoroughly with M9 buffer. The infected nematodes were homogenized and the concentration of proteins was determined by Bradford's method. An equal concentration of protein was mixed with 5 μM DCFH-DA in a final volume of 2 ml of Tris-HCl. The mixture was incubated at room temperature for 45 min in dark condition. The level of DCF produced by ROS mediated oxidation of DCFH-DA was measured spectroscopically at 530 nm, following excitation at 485 nm and the level of DCF was quantified against DCF standard. Results were expressed as nmol of DCF/mg of protein.

The total RNA was isolated from N2 wild-type and mutants exposed to K. pneumoniae for different time-points (12, 24, and 36 h) by TRIzol method. For RT-PCR, the oligo-dT method was followed to synthesis a cDNA strands. According to instruction of SuperScript III kit (Invitrogen Inc., USA), the CDNA template was generated from 100 ng of RNA isolated from nematodes. PCR was carried out to amplify from the cDNA by using forward and reverse primers. Quantitative real time (qRT-PCR) was performed by using the Applied Biosystem SYBR Green fluorescence dye. The expression of the candidate genes were normalized by act-2. The relative gene expression was determined by calculating 2^−ΔΔct.

The 2D-gel electrophoresis was performed at least three times from the biologically different samples for consistency of differentially regulated proteins and statistical significance of fold-changes. Real-time analysis and quantification of ROS were conducted in triplicates and the statistical analysis was performed by one-way ANOVA using SPAA software 10.0. Significant differences between the groups were calculated by Dunnet's multiple range test (P < 0.05).

In our previous study Kamaladevi and Balamurugan (2015), we have reported that K. pneumoniae required 48 h for complete killing of nematodes. Hence, to understand the host response against K. pneumoniae pathogenesis at their proteome level, 12, 24, and 36 h were considered. Using conventional proteomic approach, infection induced host response against K. pneumoniae was monitored kinetically. The CBB stained gel images of respective time-points were compared and approximately > 300 protein spots were detected in each time-points. Figure 1 showed the CBB stained gel images of 12 (Figure 1A), 24 (Figure 1B), and 36 h (Figure 1C). The protein spots matched between the OP50 control and K. pneumoniae infected at 12, 24, and 36 h were 381, 630, and 742, respectively. We performed three biologically independent experiments and used the stringent criteria of a 1.2 Da mass tolerance and more than 1.5-fold regulation. This resulted in an identification of a total of 53, 82, and 43 regulatory proteins at 12 Supplementary Table 1, 24 Supplementary Table 2, and 36 h Supplementary Table 3, respectively.

To decipher the interaction between the identified regulatory proteins and their role in biological activities during host defense, STRING 10.0 analysis was performed. The interactions between the identified proteins were analyzed based on the literature with a high confidence score of 0.7. The interaction between the proteins identified at 12 Figure 2A https://string-db.org/cgi/network.pl?taskId=2ewoymMJQSou), 24 Figure 2B https://string-db.org/cgi/network.pl?taskId=KS9tVO9Zw468), and 36 h (Figure 2C https://string-db.org/cgi/network.pl?taskId=iBw1EvsLlGVD) were analyzed separately and their biological functions were determined by Gene ontology (GO) database. The GO results revealed that the major group of identified proteins at different time points were predominantly encodes for metabolism, development, translation, reproduction, protein unfolding, and apoptosis (Figures 2D–F). Concomitantly, in order to investigate the interactions between the proteins identified at 12, 24, and 36 h and their biological significance in elucidating the host defense against K. pneumoniae infection, STRING analysis was performed with the total list of identified proteins and they were manually annotated (Figure 3A https://string-db.org/cgi/network.pl?taskId=I10REy8oLgGn). Based on their biological functions, the identified proteins and their interacting partners were classified into six major classes such as metabolism, translation, dauer formation, signal transduction, endocytosis, apoptosis, and protein folding (Figures 3A,B).

In order to separate the interacting partners of regulatory proteins, a liquid IEF was performed. Proteins isolated from the control and experimental C. elegans homogenate were separated on the basis of their pI (pH 3–10) in solution. After the IEF separation, SDS-PAGE was performed to determine the separation of the proteins in all the 10 fractions Figure 4. Since most of the interacting proteins were identified to be in the pI ranges 5–9, the fraction numbers 5–8 were pooled together and enzymatically digested with trypsin. The peptides were analyzed by nano LC-MS/MS. A total of 88 proteins were identified in the IEF fractionations. The list of identified proteins was given in Supplementary Table 4.

To identify the protein that have major role in activating the host response against K. pneumoniae infection, several bioinformatics analysis were performed. The total of 266 proteins were enriched using the MetaCore software (DAVID) (Figure 5A) following enrichment analysis as described in the methods (Bessarabova et al., 2015). In proportion with the large number of regulated proteins in response to K. pneumoniae infection, more enriched GO categories were identified. In order to reduce the number of GO terms, enriched GO categories with false discovery rate (FDR) < 0.05 from DAVID analysis were submitted to the REVIGO tool (Supek et al., 2013). This tool uses the Uniprot as background database and has default semantic similarity measure (simrel). Investigation by this tool clearly showed that the biological process associated with metabolism, translation, protein metabolism, cellular homeostasis, protein folding, reproduction etc. were significantly over-represented among the proteins regulated in C. elegans during K. pneumoniae infection (Figure 5B). The lists of top 10 GO processes for these proteins in the host functions are provided in the Figures 2D–F. In addition, KEGG pathway analysis revealed that most of the identified proteins were regulated for metabolic pathway (Figure 5C). Concerning the GO process, the most regulated proteins were involved in metabolism, translation, apoptosis, development, immunity, and response to stimuli. Further, to identify the contribution of each protein in the different biological functions like metabolic process, translation, protein folding or oxidative stress, apoptosis and dauer formation, a Venn diagram were drawn with the list of proteins contributing to each function. The results revealed several overlapping proteins among the identified functions. Interestingly, a protein mTOR (mammalian target of Rapamycin) was found to have a contribution in all the biological functions (Figure 5D) and hence, was selected for further characterization studies.

mTOR, a mammalian Target of Rapamycin regulates several cellular process in normal system. However, the impact of bacterial infection on mTOR signaling remains unclear. In the present study, the proteomics approach coupled with bioinformatics analysis revealed mTOR has appeared to be involving in regulating the different biological functions. So we examined the regulation of mTOR in host during infection with K. pneumoniae in 2D gel analysis. Interestingly, we found that mTOR was significantly (P < 0.05) down regulated in 12 h when compared to the control. The downregulation of mTOR was showed in the 3D images Figure 6. The down regulation of mTOR was believed to induce a metabolic stress in host (Jia et al., 2004). mTOR is a major component for regulating the metabolic process (Tattoli et al., 2012). In C. elegans, metabolic stress acts as a double edged-sword (Kenific et al., 2010; Tattoli et al., 2012).

Intestinal atrophy in a biological system was usually determined by the level of autophagy. It is well-established that autophagy is induced following the inhibition of mTOR in multicellular systems (Mortimore and Schworer, 1977; Chakrabarti et al., 2012; Kapuy et al., 2014). To examine the mTOR mediated autophagy, we monitored the physiological alteration in C. elegans intestine and marker for autophagy. The microscopic analysis of C. elegans exposed to K. pneumoniae displayed a distended intestine Figure 7. Due to the defective intestine, the infected nematodes could not absorb the nutrients from the ingested food. This promoted a dauer formation in infected nematodes. The results of short-time exposure assay indicated the presence of dauer nematodes during the course of infection (Table 1). This explicates that the infection with K. pneumoniae induced an abnormal dauer formation in nematodes. Interestingly, the dauer formation due to the deficient mTOR shares few unique and common features of normally formed dauer due to food starvation where it is not having severe intestinal atrophy (Long et al., 2002). Thus, the dauer formation along with intestinal atrophy in infected nematodes suggested that might be due to the downregulation of mTOR during infection. Furthermore, the increased autophagy was also confirmed at the molecular level by measuring the level of bec-1, a gene responsible for autophagy (Figure 8A). The upregulation of bec-1 during infection suggested the activation of autophagic process during K. pneumoniae infection.

An inhibition of translation machinery is the other negative consequence of the inhibition of mTOR. The results of 2D gel electrophoresis also suggested that the infection with K. pneumoniae in C. elegans regulated several proteins that are encoded for translation. Hence, to investigate the status of translational regulatory gene in the C. elegans during infection, the mRNA level of eef-2 was examined. The eef-2 is the downstream regulator of mTOR, which is also an essential component for translation. The regulation of eef-2 in the nematodes infected with K. pneumoniae was significantly downregulated during course of infection (Figure 8B). This result indicated that the downregulation of mTOR influences the regulation of eef-2.

To elucidate the role of mTOR signaling in host defense against K. pneumoniae infection, the upstream regulators of mTOR were also examined for their regulation at their protein level using western blot analysis. The PI3-kinase and AKT were found to be the upstream regulators of mTOR. Hence, the level of expression of PI3-kinase, AKT, and mTOR were investigated. The PI3-Kinase activates the AKT and induced the phosphorylation of mTOR in normal conditions (Schmitz et al., 2008; Rafii et al., 2015). Our results showed that the level of PI3-kinase and AKT was inhibited in infected nematodes Figures 9A,B. This in turn inhibited the phosphorylation of mTOR Figures 9A,B. The inhibition of PI3-kinase/AKT/mTOR signaling pathway appeared to be increased the risk of infection (Figure 9C) (Rafii et al., 2015). Additionally, PDI-2, the protein encoded for protein folding was also examined for their regulation. The PDI is a protein-di-isomerase, which catalyzes the protein folding. The downregulation of PDI-2 (Figures 9A,B) indicated that the K. pneumoniae infection not favors the protein not to fold properly at 24 h in the host system. The inhibition of PDI-2 in the K. pneumoniae infected samples in western blot analysis suggested the possibilities of important regulatory proteins were not in fully or properly folded state in C. elegans. The result of the present study goes well with the previous study wherein the similar kind of regulatory event was observed in C. elegans during infection with V. alginolyticus (Durai et al., 2014).

To further confirm the role of PI3-kinase/AKT/mTOR signaling pathway and PDI-2 in host defense against K. pneumoniae infection, the C. elegans with specific mutation in AKT, mTOR and PDI-2 were examined for their survival during K. pneumoniae infection. The mutants of akt (Figure 10A), mTOR (Figure 10B), and pdi-2 (Figure 10C) exposed with K. pneumoniae exhibited a significant (P < 0.05) shorter life span (mean lifespan of 27 ± 5, 20 ± 2, and 12 ± 7 h for AKT, mTOR, and PDI-2 mutants, respectively) than the controls (N2 mean lifespan of 48 ± 5 h) (Figure 10). The susceptibility of AKT, mTOR, and PDI-2 mutants to K. pneumoniae suggested the role and importance of PI3-kinase/AKT/mTOR pathway and unresponsive protein folding in C. elegans during host defense.

In our proteomic analysis, we identified several stress responsive, detoxification, and protein mis-folding genes strongly induced after K. pneumoniae infection. The process of increased autophagy and translation blockage is also another source of oxidative burst (Chavez et al., 2007; Schmitz et al., 2008; Chakrabarti et al., 2012). The increased ROS is an alternate part of innate immune defense mechanism (Chavez et al., 2007). A significant increase in the level of intracellular ROS in the C. elegans exposed to K. pneumoniae than the control revealed the oxidative burst in host during infection (Figure 11A). Concomitantly, a high level of DCF fluorescence in the C. elegans with the increase in an infection time corroborated the generation of ROS due to infection mediated oxidative stress in the host (Figure 11B). Furthermore, in our 2D analysis, several proteins that encoded for antioxidants like gst-8, sod-3, catalase (cat-1, cat-2, and cat-3) were also found to be regulated. To validate these regulations, the mRNA levels of the identified proteins were determined by qPCR using specific primers. The mRNA levels of all the selected antioxidant enzymes were significantly (P < 0.05) upregulated during course of infection except cat-2 (Figure 12). It appears that the levels of antioxidant enzymes increased to combat the elevated ROS during the K. pneumoniae infection in the host system.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.