Pediococcus pentosaceus M41 cultures stored at –80°C in 50% glycerol solution were revived in MRS broth (de Man Rogosa Sharpe broth, Lancashire, United Kingdom) and were incubated at 37°C for 20 h under anaerobic conditions according to Ayyash et al. (2020). Two consecutive subculturings in MRS broth were performed prior to stress treatments. Preliminary investigations were performed to determine the stress point of each stress factor. The pelleted culture was heated for 2 h at 40–65°C with 5°C interval/intervals. For cold stress, the pelleted culture was stored at 4°C for 1–5 h with 1 h intervals. The acid stress was determined by incubating the pelleted culture for 2 h under acid conditions pH 2.0 to 4.0 with 0.5 intervals. The bile stress was determined after the pelleted culture was subjected to different concentrations of a bile salt mixture (0.01–0.1%) (Sigma) for 2 h with 0.01% interval. The point achieved/achieved 1.0 log reduction or less was considered as stress point (data not shown).

Two culture tubes per stress treatment were used (Figure 1). The P. pentosaceus M41 cultures in MRS broth in late exponential phase were subjected to heat stress for 2 h at 50°C. Cold stress treatment was performed by incubating the cultures at 4°C for 2 h. For acid and bile stress treatments, the cultures were centrifuged at 5000 × g for 10 min at 4°C and the cell pellets were resuspended in phosphate buffer pH 3.0 for acid stress and 0.05% bile salts for bile stress. The cultures were incubated at 37°C for 2 h. One set of culture tubes without stress were considered as the control. After stress treatments all the tubes were centrifuged at 5000 × g for 10 min at 4°C and the pellets were washed with phosphate buffer and stored at –20°C until protein extraction.

Proteins were extracted from P. pentosaceus M41 cells exposed to different stress treatments using ReadyPrep^TM total protein extraction kit (Bio-Rad, United States) according to manufacturer instructions. The pellets were resuspended in 0.5 mL of 2-D sample buffer containing 7 M urea, 2 M thiourea, 40 mM Tris base, 1% w/v amidosulfobetaine-14 detergent and 0.001% bromophenol blue. The pellet suspension was transferred to a microcentrifuge tube and was placed on ice. The cells were lysed using an ultrasonic probe at 20 kHz (Sonifier SFX550 Model, Branson, CT, United States). The sonication was done four times using 30 s bursts. The pellet suspension was chilled on ice briefly between each ultrasonic treatment. The tubes were centrifuges at 10,000 × g for 20 min. The supernatant containing bacterial proteins was transferred to a clean microcentrifuge tube and protein concentration was estimated using the Bradford assay (Bradford, 1976).

For isoelectric focusing (IEF), 400 μg of bacterial proteins in a total volume of 300 μL in sample buffer was loaded onto 17 cm, pH 4.0–7.0, non-linear, IPG strips (Ready Strip^TM, Bio-Rad, United States) in a rehydration tray. Two mL of mineral oil was applied to each strip and kept overnight at 20 °C for passive rehydration of proteins. The strips were then transferred to an IEF tray (Protean II IEF^TM cell, Bio-Rad, United States) and protein focusing was performed at 20°C with a total 65,000 volt hour (Vh) current supply. Following IEF, the strips were applied with reduction and alkylation buffer for 20 min each and then rinsed with ultrapure water. For second dimension SDS PAGE, the strips were placed on 12% polyacrylamide gels and electrophoresis was performed in Protean II XL Cell^TM (Bio-Rad, United States) with a total current flow of 200 V at 10°C. The gels were run in duplicate to ensure reproducibility. After second dimension electrophoretic run, the gels were stained with Coomassie blue for protein visualization. The gels were scanned with a gel documentation system (Gel Doc^TM, Bio-Rad, United States). The protein spots from all the stress treatments and the control were compared for differential expression analysis using Melanie^TM Version 9.2.3 (Swiss Institute of Bioinformatics, Lausanne, Switzerland). The protein spot intensities having more than 1.5 fold and lower than 0.5 fold change were considered as upregulated and downregulated respectively as compared to control. The selected protein spots were picked in 0.2 mL PCR tubes and 100 μL of ultrapure water was added and stored at 4°C.

The protein identification process was performed by BGI Genomics (Beijing, China). The dye from the spots was removed by washing with 1mL of a decoloring solution (50% acetonitrile + 25 mM ammonium bicarbonate solution) several times until the gel spot was decolorized. After cleaning, 500 μl of acetonitrile was added to gel spots for dehydration. The supernatant was removed and the gel spot was dried in air. The gel spots were added with 10 mM DTT (dithiothreitol) at 56°C for 1 h to reduce the disulfide bond. The supernatant was removed and 55 mM IAM (iodoacetamide) was added in a dark room for 45 min to perform alkylation of cysteine. The supernatant was removed and the gel spots were dried in air. An aliquot of 0.1 μg/μl trypsin solution was added and the tubes were kept on ice for 30 min, and 25 mM ammonium bicarbonate solution was added to cover the gel spots and digested overnight at 37°C. The remaining liquid outside the gel spot was transferred to a new centrifuge tube. The peptide was extracted once from the gel spot with 50% acetonitrile-water and again with 100% acetonitrile. The two extracted solutions were combined with the tryptic digested solution and lyophilized.

The lyophilized peptide samples were reconstituted with mobile phase A (2% ACN, 0.1% FA), shaken and centrifuged, and the supernatant was taken. The samples were separated on a Shimadzu LC-20AD Nano nanoliter liquid chromatograph. The sample was first loaded to a trap column for enrichment, and then the sample was eluted with a sharp gradient at a flow rate of 300 nl/min with a self-packed C18 column (75 μm × 15 cm, 3 μm, 120 Å). The nanoliter liquid chromatograph end was directly connected to a mass spectrometer LTQ Orbitrap Velos (Thermo Fisher, United States), and the ion source was nanoESI. In the data acquisition, the positive ion scanning mode was adopted, the resolution of the MS1 was 30,000, and the scanning mass range was 350–1,500 m/z. Based on the MS1 scanning information, according to the ion intensity in the order MS1 from high to low, the top six were selected for fragmentation and the MS2 information was scanned in Orbitrap analyzer. The resolution of the MS2 was 7,500, and the fragmentation mode was HCD.

The spot intensities were quantitatively detected by Melanie^TM Version 9.2.3 (Swiss Institute of Bioinformatics, Lausanne, Switzerland). For differentially expressed proteins the spot intensities were statistically analyzed by using Metaboanalyst 5.0 software (Chong et al., 2018). The spot intensities were log2 transformed for correlation analysis, principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA) and heatmap. The analysis was performed on two biological replicates using one-way analysis of variance (ANOVA). All the values are presented as the mean ± standard error (SE).

A total of 60 protein spots were quantitatively detected under control, heat, cold, acid, and bile stress treatments (Figures 2A–E), out of which 29 proteins under acid stress (21 upregulated and 8 downregulated), 33 proteins under bile stress (25 upregulated and 8 downregulated), 31 proteins under heat stress (28 upregulated and 3 downregulated) and 21proteins under cold stress (18 upregulated and 3 downregulated) were differentially expressed compared to the control. Seven upregulated proteins were common among all the stress treatments which were spot M12, spot M25, spot M31, spot M40, spot M41, spot M43, and spot M53. One downregulated protein (spot M22) was common among all the stress treatments (Table 1).

For functional categorization, the identified protein ID’s were searched against UniProt database and the proteins were assigned to different cellular functions (Figure 3). Maximum 31% of identified proteins were found to be associated with protein metabolism followed by 18% of carbohydrate and nucleotide metabolism each. In addition, 13% proteins were found to be associated with stress responses, 8% proteins were involved in oxidation reduction, 7% proteins with functions in amino acid metabolism and 5% proteins were involved in cell wall synthesis.

For analyzing correlation between protein profiles and changes between control, heat, cold, acid and bile treatments, multivariate data analysis was performed by using MetaboAnalyst software (Figures 4A–D). Before analysis, the differences in spot intensities for all the proteins were normalized by log transformation.

The exposure of LAB to different environmental stresses could be a great challenge for ever growing food industry. Different stresses such as high or low temperature, low pH, osmotic pressure due to food processing and gut environment can impair or reduce the biological properties of LAB. Several studies reported adaptation of LAB to different environmental stresses which could be due to underlying changes in gene transcription and protein expression (De Angelis and Gobbetti, 2004).

In the food industry, the processing of food products requires heating at different temperatures. Therefore, the LAB are frequently exposed to heat stress in LAB fortified foods. Several probiotic bacteria were studied by other researchers under heat stress conditions. LAB which can survive high temperatures were then identified. The mechanism of adaptation of LAB under stress conditions involves upregulation of different stress responsive proteins. Lactobacillus kefiranofaciens M1 adapts itself under stress conditions by increasing the expression of heat shock proteins (HSPs), enzymes such as chaperone protein DnaK prolyl-tRNA synthase, phosphoenolpyruvate-protein phosphotransferase, cofactors (GroES) and chaperonins (GroEL) (Hernández-Alcántara et al., 2018). These proteins promote proper folding and translocation of newly synthesized polypeptides (Hernández-Alcántara et al., 2018). Different proteins were identified in LAB exposed to heat stress such as DnaK, GroEL, and GroES in Lb. plantarum DPC2739 and DPC2741 (De Angelis and Gobbetti, 2004), enolase, and glyceraldehyde-3-phosphate dehydrogenase in Lb. helveticus PR4 (Di Cagno et al., 2006), small heat shock proteins (HSP1 and HSP3) and chaperonins (DnaK, GrpE, GroEL, and GroES) in Lb. plantarum (Lp 813 and Lp 998) and Lb. casei (ATCC393) (Haddaji et al., 2015; Ferrando et al., 2016), small heat shock proteins (HSP27 and HSP72) in Lb. rhamnosus GG (ATCC 53103), Lb. johnsonii P47-HY, and Lb. reuteri P43-HUV (Liu et al., 2015).

Dairy-based food products such as cheese, yogurt and ice cream require storage at low temperatures which expose LAB to cold stress. Some LAB can survive at low temperatures which is important for their commercial use because of their storage and transport at chilled to subzero temperatures (Fonseca et al., 2015; Zhang et al., 2016). Cold stress can alter the protein expression in LAB. Cold-shock and anti-freeze proteins protect LAB under low temperature stress. Cold-stress induced proteins were identified in LAB such as cold-shock proteins (cspC, cspL, and cspP) and ATPase in Lb. plantarum L67 (Song et al., 2014). Enzymes that activate membrane fluidity and membrane phospholipids were identified in Lb. bulgaricus (ATCC 11842 and CFL1) (Meneghel et al., 2017). Anti-freeze and ice nucleation proteins were found in Lb. paracasei and Lb. mali (Polo et al., 2017).

During intestinal digestion LAB are exposed to acidic pH which can affect their function and survival. Food products with acidic pH such as dairy-based fermented foods and pickles can also exposes LAB to acid stress. Some LAB can survive under acidic environments due to the presence of molecules such as proton-translocating ATPase which protect the LAB by stabilizing the intracellular pH (Šeme et al., 2015; Pérez Montoro et al., 2018; Wang et al., 2018). Some acid stress induced proteins identified in LAB include glyceraldehyde-3-phosphate dehydrogenase and DnaK in Lb. casei Zhang (Wu et al., 2011); GrpE, methionine synthase, and 30S ribosomal protein S2 in Lb. plantarum LC 804, CIP A159, and CECT 4185 (Hamon et al., 2011); 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase 2 (PGAM-d); elongation factor G, and DnaK in Lb. pentosus AP2-15, AP2-18, and LP-1 (Pérez Montoro et al., 2018). Proteins related to purine biosynthesis, galactose metabolism, and fatty acid biosynthesis were also identified in Lb. rhamnosus GG (ATCC 53103) (Koskenniemi et al., 2009).

The probiotic action of LAB depends on its ability to survive under gastrointestinal conditions where they are exposed to acidic environments and bile salts. Bile salts can cause DNA damage as well as changes in protein conformation and oxidative stress (Merritt and Donaldson, 2009). Several studies reported the responses of LAB under bile stress. Induction of stress responsive proteins was observed in different LAB species when exposed to bile stress (Alcántara and Zúñiga, 2012). Proteins identified in LAB upon exposure to bile stress were glutathione reductases and cyclopropane fatty acyl phospholipid synthase in Lb. plantarum 299 V, LC 804, and LC 56 (Hamon et al., 2011), in addition to bile salt hydrolases and F0F1-ATPase (+) in Lactobacillus and Bifidobacterium (Ruiz et al., 2013).

The extensive applications of LAB in the food industry make it imperative to link their important characteristics with their corresponding genes and proteins (Sharma et al., 2020). Stress responsive genes and proteins could be helpful in marker-assisted selection of LAB which are more resistant to adverse environments.

In our study, P. pentosaceus M41 responded to different stresses by maximum increases in expression of proteins related to protein and nucleotide metabolism which indicate the tolerance of P. pentosaceus M41 to these stresses through growth and repair mechanisms. The proteome of P. pentosaceus M41 under heat, cold, acid and bile stresses showed several proteins involved in stress response and tolerance. The maximum number of proteins identified was involved in protein metabolism which mainly regulates transcription and translation such as ribosomal RNA small subunit methyltransferase, ribosome maturation factor RimM, elongation factor Tu, 50S ribosomal protein L2, elongation factor Ts, 30S ribosomal protein S2, ribosomal RNA large subunit methyltransferase, 30S ribosomal protein S18 and 50S ribosomal protein L10. The ribosomal proteins regulate RNA synthesis and proteins involved in ribosome assembly. They are ubiquitous proteins primarily involved in extra-ribosomal functions (Warner and McIntosh, 2009). 50S ribosomal protein L2 (spot M31) and 30S ribosomal protein S2 (spot M38) were upregulated under all the stress treatments.