Work overview

Section 03 of 08

RESULTS

Shotgun metaproteomics reveals habitat-specific antimicrobial resistance-associated proteins of Escherichia spp. and Salmonella spp. in the gut resistome of free-living long-tailed macaques in Thailand

Wirasak Fungfuang, Daraka Tongthainan, Sawanya Charoenlappanit, Narumon Phaonakrop, Sittiruk Roytrakul, and Kongphop Parunyakul · 2026

Contents

Section 03 of 08

  1. 01INTRODUCTION
  2. 02MATERIALS AND METHODS
  3. 03RESULTS
  4. 04DISCUSSION
  5. 05CONCLUSION
  6. 06DATA AVAILABILITY
  7. 07GENERATIVE AI DECLARATION
  8. 08AUTHORS’ CONTRIBUTIONS
Text size
Work overview

Section 3 of 8

RESULTS

Wirasak Fungfuang, Daraka Tongthainan, Sawanya Charoenlappanit, Narumon Phaonakrop, Sittiruk Roytrakul, and Kongphop Parunyakul · about 16 minutes

Protein expression patterns of Escherichia spp. and Salmonella spp. in locations P and S

Hierarchical clustering analysis was performed using MetaboAnalyst based on Differentially expressed proteins (DEPs) identified in Escherichia spp. (8,989 proteins) and Salmonella spp. (1,029 proteins) from the fecal microbiomes of free-living long-tailed macaques (Figures 2A and 3A, respectively). The dendrograms were constructed according to the degree of similarity among expressed proteins to provide an overall visualization of the proteomic profiles. Samples collected from the same location exhibited similar protein expression patterns, whereas clear differences were observed between locations P and S. These findings indicate that the DEP profiles of both Escherichia spp. and Salmonella spp. can effectively discriminate between the two habitats.

To further compare protein expression patterns between locations P and S, PCA was performed. The PCA score plots demonstrated a clear separation between the two locations along the first principal component (Figure 2B and Figure 3B). To maximize group discrimination, PLS-DA was subsequently performed (Figures 2C and 3C). The PLS-DA results were consistent with the PCA findings and further confirmed distinct proteomic signatures associated with each habitat.

Functional annotation of DEPs in ** Escherichia ** spp.

Volcano plot analysis using significance thresholds of fold change >2 and p < 0.05 demonstrated substantial differences in protein expression between locations P and S (Figure 4A). A total of 8,989 proteins were detected across both locations, of which 1,299 proteins were differentially expressed. Among these DEPs, 268 proteins were upregulated and 1,031 proteins were downregulated in location P compared with location S. The Venn diagram revealed 2,390 proteins that were differentially expressed across the comparison groups (Figure 4C).

GO analysis of upregulated proteins in ** Escherichia ** spp.

GO classification demonstrated that the upregulated proteins in location P were primarily associated with unclassified proteins (77 proteins), translation (14 proteins), regulation of DNA-templated transcription (8 proteins), proteolysis (8 proteins), and response to antibiotics (6 proteins) (Figure 5A).

Among molecular function categories, excluding unclassified proteins, the most abundant functions included DNA binding (44 proteins), adenosine triphosphate (ATP) binding (29 proteins), metal ion binding (20 proteins), structural constituent of ribosome (14 proteins), and magnesium ion binding (14 proteins) (Figure 5B).

Among cellular components, the majority of proteins were localized to the cytoplasm (53 proteins), followed by the plasma membrane (41 proteins), the cytosol (21 proteins), ribosomes (9 proteins), and the extracellular region (9 proteins) (Figure 5C).

GO analysis of downregulated proteins in ** Escherichia ** spp.

Among the downregulated proteins identified in location P relative to location S, the predominant biological process categories included unclassified proteins (246 proteins), regulation of DNA-templated transcription (36 proteins), proteolysis (28 proteins), DNA replication (26 proteins), defense response to virus (25 proteins), and response to antibiotics (24 proteins) (Figure 5D).

Within the molecular function category, ATP-binding (n = 160), metal ion binding (n = 137), DNA binding (n = 118), and ATP hydrolysis activity (n = 53) were the most represented functions (Figure 5E).

Based on cellular component classification, the majority of proteins were categorized as unclassified proteins (n = 397), cytoplasm (n = 180), and plasma membrane (n = 158) (Figure 5F).

Figure 2: Differential protein expression analysis of Escherichia spp. in the fecal microbiomes of long-tailed macaques from locations P and S. (A) Heatmap showing differentially expressed proteins (DEPs) based on Euclidean distance and Ward clustering. Protein abundance is displayed on a normalized scale ranging from blue (low abundance) to red (high abundance). (B) PCA score plot demonstrating separation between the two location groups at the 95% confidence interval. (C) PLS-DA score plot showing covariance between component 1 and component 2 and identifying DEPs between the two locations.

Figure 2: Differential protein expression analysis of Escherichia spp. in the fecal microbiomes of long-tailed macaques from locations P and S. (A) Heatmap showing differentially expressed proteins (DEPs) based on Euclidean distance and Ward clustering. Protein abundance is displayed on a normalized scale ranging from blue (low abundance) to red (high abundance). (B) PCA score plot demonstrating separation between the two location groups at the 95% confidence interval. (C) PLS-DA score plot showing covariance between component 1 and component 2 and identifying DEPs between the two locations.

Figure 3: Differential protein expression analysis of Salmonella spp. in the fecal microbiomes of long-tailed macaques from locations P and S. (A) Heatmap showing differentially expressed proteins (DEPs) based on Euclidean distance and Ward clustering. Protein abundance is displayed on a normalized scale ranging from blue (low abundance) to red (high abundance). (B) PCA score plot demonstrating separation between the two location groups at the 95% confidence interval. (C) PLS-DA score plot showing covariance between component 1 and component 2 and identifying DEPs between the two locations.

Figure 3: Differential protein expression analysis of Salmonella spp. in the fecal microbiomes of long-tailed macaques from locations P and S. (A) Heatmap showing differentially expressed proteins (DEPs) based on Euclidean distance and Ward clustering. Protein abundance is displayed on a normalized scale ranging from blue (low abundance) to red (high abundance). (B) PCA score plot demonstrating separation between the two location groups at the 95% confidence interval. (C) PLS-DA score plot showing covariance between component 1 and component 2 and identifying DEPs between the two locations.

Figure 4: Differential expression profiles of Escherichia spp. and Salmonella spp. in the fecal microbiomes of long-tailed macaques from locations P and S. (A) Volcano plot showing differentially expressed proteins (DEPs) of Escherichia spp. using significance thresholds of p < 0.05 and log₂ fold change (FC) ≥1 or ≤−1. (B) Volcano plot showing DEPs of Salmonella spp. using significance thresholds of p < 0.05 and log₂FC ≥1 or ≤−1. (C) Venn diagram illustrating unique and shared DEPs of Escherichia spp. between locations P and S. (D) Venn diagram illustrating unique and shared DEPs of Salmonella spp. between locations P and S.

Figure 4: Differential expression profiles of Escherichia spp. and Salmonella spp. in the fecal microbiomes of long-tailed macaques from locations P and S. (A) Volcano plot showing differentially expressed proteins (DEPs) of Escherichia spp. using significance thresholds of p < 0.05 and log₂ fold change (FC) ≥1 or ≤−1. (B) Volcano plot showing DEPs of Salmonella spp. using significance thresholds of p < 0.05 and log₂FC ≥1 or ≤−1. (C) Venn diagram illustrating unique and shared DEPs of Escherichia spp. between locations P and S. (D) Venn diagram illustrating unique and shared DEPs of Salmonella spp. between locations P and S.

Figure 5: Gene ontology (GO)-based functional annotation of differentially expressed proteins (DEPs) of Escherichia spp. in the fecal microbiomes of long-tailed macaques from location P compared with location S. (A) Biological process classification of the 268 upregulated proteins in location P. (B) Molecular function classification of the 268 upregulated proteins in location P. (C) Cellular component classification of the 268 upregulated proteins in location P. (D) Biological process classification of the 1,031 downregulated proteins in the location P versus location S comparison. (E) Molecular function classification of the 1,031 downregulated proteins in the location P versus location S comparison. (F) Cellular component classification of the 1,031 downregulated proteins in the location P versus location S comparison.

Figure 5: Gene ontology (GO)-based functional annotation of differentially expressed proteins (DEPs) of Escherichia spp. in the fecal microbiomes of long-tailed macaques from location P compared with location S. (A) Biological process classification of the 268 upregulated proteins in location P. (B) Molecular function classification of the 268 upregulated proteins in location P. (C) Cellular component classification of the 268 upregulated proteins in location P. (D) Biological process classification of the 1,031 downregulated proteins in the location P versus location S comparison. (E) Molecular function classification of the 1,031 downregulated proteins in the location P versus location S comparison. (F) Cellular component classification of the 1,031 downregulated proteins in the location P versus location S comparison.

Functional annotation of DEPs in ** Salmonella ** spp.

Volcano plot analysis identified 1,029 proteins in Salmonella spp. from locations P and S. Among these, 66 DEPs were upregulated and 165 were downregulated in location P compared with location S. The Venn diagram showed 184 proteins commonly differentially expressed between the two groups (Figure 4D).

GO analysis of upregulated proteins in ** Salmonella ** spp.

The upregulated proteins in location P were mainly associated with unclassified proteins (n = 13), DNA repair (n = 4), SOS response (n = 3), enterobacterial common antigen biosynthetic process (n = 3), and DNA replication (n = 3) (Figure 6A).

The predominant molecular functions included ATP-binding (n = 13), DNA binding (n = 6), ATP hydrolysis activity (n = 3), and magnesium ion binding (n = 3) (Figure 6B).

The principal cellular components associated with upregulated proteins were the plasma membrane (n = 16) and cytoplasm (n = 14) (Figure 6C).

GO analysis of downregulated proteins in ** Salmonella ** spp.

The downregulated proteins were primarily involved in unclassified proteins (32 proteins), cobalamin biosynthetic process (n = 6), methylation (n = 5), translation (n = 4), and glutamine metabolic process (n = 4) (Figure 6D).

ATP-binding (n = 33) represented the most abundant molecular function category, followed by unclassified proteins (n = 20), metal ion binding (n = 20), and tRNA binding (n = 8) (Figure 6E).

The predominant cellular components included unclassified proteins (n = 39), cytoplasm (n = 36), plasma membrane (n = 33), cytosol (n = 31), and extracellular region (n = 7) (Figure 6F).

Figure 6: Gene ontology (GO)-based functional annotation of differentially expressed proteins (DEPs) of Salmonella spp. in the fecal microbiomes of long-tailed macaques from location P compared with location S. (A) Biological process classification of the 66 upregulated proteins in location P. (B) Molecular function classification of the 66 upregulated proteins in location P. (C) Cellular component classification of the 66 upregulated proteins in location P. (D) Biological process classification of the 165 downregulated proteins in the location P versus location S comparison. (E) Molecular function classification of the 165 downregulated proteins in the location P versus location S comparison. (F) Cellular component classification of the 165 downregulated proteins in the location P versus location S comparison.

Figure 6: Gene ontology (GO)-based functional annotation of differentially expressed proteins (DEPs) of Salmonella spp. in the fecal microbiomes of long-tailed macaques from location P compared with location S. (A) Biological process classification of the 66 upregulated proteins in location P. (B) Molecular function classification of the 66 upregulated proteins in location P. (C) Cellular component classification of the 66 upregulated proteins in location P. (D) Biological process classification of the 165 downregulated proteins in the location P versus location S comparison. (E) Molecular function classification of the 165 downregulated proteins in the location P versus location S comparison. (F) Cellular component classification of the 165 downregulated proteins in the location P versus location S comparison.

Antibiotic resistance-associated proteins in ** Escherichia ** spp. and ** Salmonella ** spp.

Among the 268 upregulated DEPs of Escherichia spp. in location P, six proteins were associated with antibiotic response activity, including multidrug resistance (MDR) protein (P39386), chloramphenicol acetyltransferase 3 (P00484), outer membrane protein TolC (P02930), undecaprenyl-phosphate 4-deoxy-4-formamido-L-arabinose transferase (B7LM77), primosomal protein N′ (P17888), and pentapeptide repeat protein QnrB4 (Q2PT27).

Among the 1,031 downregulated DEPs in location P (corresponding to proteins upregulated in location S), 25 proteins were associated with antibiotic resistance, including aminoglycoside adenylyltransferase (P0AG05), penicillin-binding protein 1C (P76577), probable aminoglycoside efflux pump (P24177), putative O-antigen transporter (P37746), and penicillin G acylase (P06875). Detailed information is presented in Tables 1 and 2.

Protein ID | Protein name | Peptide sequence | Fold change | p-value | Gene ontology (GO) – Biological process
P39386 | Multidrug resistance protein MdtM | FTLFSNK | 3.4288 | 1.85 × 10⁻¹⁶ | Response to antibiotic; bile acid and bile salt transport; potassium ion export across plasma membrane; sodium ion export across plasma membrane; regulation of cellular pH; xenobiotic detoxification by transmembrane export across the plasma membrane
P00484 | Chloramphenicol acetyltransferase 3 (EC 2.3.1.28) | EHFEFYRHR | 3.0014 | 9.59 × 10⁻¹⁴ | Response to antibiotic
P02930 | Outer membrane protein TolC (multidrug efflux pump subunit TolC) | GAAGTQYDDSNMGQNK | 2.6668 | 3.38 × 10⁻¹⁰ | Response to antibiotic; bile acid and bile salt transport; enterobactin transport; monoatomic ion transmembrane transport; response to organic cyclic compound; response to toxic substance; response to xenobiotic stimulus; xenobiotic detoxification by transmembrane export across the cell outer membrane; xenobiotic detoxification by transmembrane export across the plasma membrane
B7LM77 | Undecaprenyl-phosphate 4-deoxy-4-formamido-L-arabinose transferase (EC 2.4.2.53) | ADEGYDVVGTVR | 2.6410 | 1.31 × 10⁻¹⁰ | 4-Amino-4-deoxy-α-L-arabinopyranosyl undecaprenyl-phosphate biosynthetic process; lipid A biosynthetic process; lipopolysaccharide biosynthetic process; response to antibiotic
P17888 | Primosomal protein N′ (ATP-dependent helicase PriA; replication factor Y) | AHSEQIPIILGSATPALETLCNVQQKKYR | 2.4961 | 4.76 × 10⁻¹³ | DNA recombination; DNA replication; DNA replication initiation; DNA replication, synthesis of RNA primer; DNA unwinding involved in DNA replication; DNA-templated DNA replication; double-strand break repair; plasmid maintenance; replication fork processing; response to antibiotic; response to gamma radiation
Q2PT27 | Pentapeptide repeat protein QnrB4 (plasmid-mediated quinolone resistance determinant) | DAIFKSCDLSMADFRNINALGIEIR | 2.4381 | 9.81 × 10⁻¹⁰ | Response to antibiotic

For Salmonella spp., two of the 66 upregulated DEPs in location P were associated with antibiotic response activity, namely signal transduction protein PmrD (P37589) and aminoglycoside N(6′)-acetyltransferase type 1 (Q9R381). In contrast, two downregulated proteins associated with antibiotic resistance were aminoglycoside N(3)-acetyltransferase III (P0A255) and MDR protein MdtM (Q8XFG0). Detailed results are presented in Table 3.

KEGG pathway analysis of AMR-associated proteins in ** Escherichia ** spp.

The AMR-associated proteins identified in Escherichia spp. from locations P and S were further analyzed using KEGG pathway mapping to investigate potential resistance mechanisms (Figure 7).

Protein ID | Protein name | Peptide sequence | Fold change | p-value | Gene ontology (GO) – Biological process
P0AG05 | Aminoglycoside (3'') (9) adenylyltransferase (EC 2.7.7.47) | ALINDLLETSASPGESEILR | 0.49454 | 0.000633 | Response to antibiotic
P76577 | Penicillin-binding protein 1C (PBP-1c) | ESREEPIWLAPR | 0.49436 | 0.000174 | Cell wall organization; peptidoglycan biosynthetic process; positive regulation of cell division; proteolysis; regulation of cell shape; response to antibiotic
P24177 | Probable aminoglycoside efflux pump (acriflavine resistance protein D) | DGGMVPFSAFATSR | 0.49019 | 0.000805 | Bile acid and bile salt transport; response to antibiotic; response to toxic substance; xenobiotic detoxification by transmembrane export across the cell outer membrane; xenobiotic transport
P37746 | Putative O-antigen transporter | GVILIKK | 0.48711 | 0.000376 | DNA damage response; O-antigen biosynthetic process; response to antibiotic
P06875 | Penicillin G acylase (EC 3.5.1.11) | EVASLLAWTHQMK | 0.48028 | 0.000264 | Antibiotic biosynthetic process; response to antibiotic
P0A1V9 | Beta-lactamase OXA-2 (EC 3.5.2.6) | AMLVFDPVR | 0.45951 | 0.000139 | Antibiotic catabolic process; response to antibiotic
P9WIE5 | Catalase-peroxidase (EC 1.11.1.21) | AAGHNITVPFTPGR | 0.45270 | 1.99 × 10⁻⁵ | Cellular response to hydrogen peroxide; hydrogen peroxide catabolic process; positive regulation of DNA repair; response to antibiotic; response to oxidative stress
P20831 | DNA gyrase subunit A (EC 5.6.2.2) | AYETGRGSIQMRSR | 0.43548 | 2.91 × 10⁻⁵ | DNA negative supercoiling; DNA topological change; DNA-templated DNA replication; response to antibiotic
Q83QT8 | Bifunctional polymyxin resistance protein ArnA | AFYGSVAHLAAER | 0.42836 | 3.86 × 10⁻⁵ | Lipid A biosynthetic process; lipopolysaccharide biosynthetic process; response to antibiotic
P00809 | Beta-lactamase 1 (EC 3.5.2.6) | EDLVDYSPVTEK | 0.42080 | 3.77 × 10⁻⁵ | Beta-lactam antibiotic catabolic process; response to antibiotic
P62577 | Chloramphenicol acetyltransferase (EC 2.3.1.28) | FYPAFIHILAR | 0.42028 | 3.24 × 10⁻⁵ | Response to antibiotic
P04382 | Dihydrofolate reductase (EC 1.5.1.3) | EMVETHWYKIDEVTTLTESVYK | 0.42018 | 3.17 × 10⁻⁵ | One-carbon metabolic process; response to antibiotic; response to methotrexate; tetrahydrofolate biosynthetic process
P0C2H3 | Macrolide export ATP-binding/permease protein MacB 2 (EC 7.6.2.-) | DDRNALQEVIIDENTR | 0.39930 | 2.33 × 10⁻⁵ | Response to antibiotic
Q05053 | Acylase ACY 1 proenzyme (includes cephalosporin acylase) | ATADMYECLSDEIGK | 0.39840 | 1.65 × 10⁻⁵ | Glutathione catabolic process; response to antibiotic
P27245 | Multiple antibiotic resistance protein MarR | LTTGGAAICEQCHQLVGQDLHQELTK | 0.39554 | 1.27 × 10⁻⁵ | Cellular response to antibiotic; negative regulation of DNA-templated transcription; regulation of DNA-templated transcription; response to heat
P35695 | Beta-lactamase OXA-7 (EC 3.5.2.6) | EVGEVRMQKYLK | 0.37483 | 4.58 × 10⁻⁶ | Antibiotic catabolic process; response to antibiotic
P13364 | DNA gyrase subunit B (EC 5.6.2.2) | AFVEYLNTNKTPVNSQVFHFSVQR | 0.36280 | 7.80 × 10⁻⁶ | DNA topological change; DNA-templated DNA replication; response to antibiotic
P0AES5 | DNA gyrase subunit A (EC 5.6.2.2) | AEVEVDAKTGR | 0.35809 | 9.17 × 10⁻⁸ | DNA negative supercoiling; DNA topological change; DNA-templated DNA replication; response to antibiotic
P20083 | DNA topoisomerase IV subunit B (EC 5.6.2.2) | CAYVLSVKMQDPQFAGQTK | 0.35549 | 3.50 × 10⁻⁸ | Chromosome organization; chromosome segregation; DNA topological change; plasmid partitioning; response to antibiotic; sister chromatid cohesion
Q83P87 | Multidrug resistance outer membrane protein MdtP | AFFGLDAIHLDTLFK | 0.34791 | 1.59 × 10⁻⁶ | Response to antibiotic
P27303 | Multidrug export protein EmrA | IISPMTGYVSR | 0.33738 | 5.41 × 10⁻⁸ | Response to antibiotic; response to toxic substance; xenobiotic detoxification by transmembrane export across the cell outer membrane; xenobiotic detoxification by transmembrane export across the plasma membrane
P33941 | ABC transporter ATP-binding/permease protein YojI | AEYVFNNLYIPDAQEYRHHIIR | 0.32348 | 1.31 × 10⁻⁷ | Microcin transport; response to antibiotic
P0AE07 | Multidrug efflux pump subunit AcrA | AIFPNPDHTLLPGMFVR | 0.31887 | 1.43 × 10⁻⁶ | Response to antibiotic; response to toxic substance
Q59397 | Dihydrofolate reductase type 9 (EC 1.5.1.3) | KTFASLPKVLPGR | 0.31394 | 2.80 × 10⁻⁸ | Glycine biosynthetic process; one-carbon metabolic process; response to antibiotic
P23874 | Serine/threonine-protein kinase toxin HipA (EC 2.7.11.1) | LANGAHTFK | 0.31063 | 3.29 × 10⁻⁸ | Dormancy process; phosphorylation; regulation of DNA-templated transcription; regulation of growth; response to antibiotic; single-species biofilm formation
Protein ID | Protein name | Peptide sequence | Fold change | p-value | Gene ontology (GO) – Biological process
Upregulated proteins in location P
P37589 | Signal transduction protein PmrD (BasR post-transcriptional activator; polymyxin resistance protein PmrD) | KSHYVKK | 3.4640 | 5.23 × 10⁻²⁰ | Response to antibiotic
Q9R381 | Aminoglycoside N(6')-acetyltransferase type 1 (EC 2.3.1.82) (AAC(6')-Iy) | DIRQMNK | 2.9683 | 1.04 × 10⁻¹⁵ | Acetyl-CoA metabolic process; response to antibiotic
Downregulated proteins in location P
P0A255 | Aminoglycoside N(3)-acetyltransferase III (EC 2.3.1.81) (AAC(3)-III) | LDDKARR | 0.46266 | 3.90 × 10⁻⁴ | Response to antibiotic
Q8XFG0 | Multidrug resistance protein MdtM (MDR efflux pump) | FTLFSNNLPK | 0.43304 | 7.45 × 10⁻⁵ | Response to antibiotic

Proteins associated with AMR were enriched in the β-lactam resistance pathway. Outer membrane protein TolC from location P was upregulated 2.67-fold and was associated with the resistance-nodulation-division (RND) efflux pump system. KEGG mapping indicated that TolC functions as a MDR-associated protein involved in active antibiotic efflux.

In contrast, proteins enriched in location S included multidrug efflux pump-associated proteins and class D β-lactamases, including AcrA, β-lactamase 1, β-lactamase OXA-7, and β-lactamase OXA-2. These proteins were associated with multidrug efflux mechanisms and β-lactamase-mediated antibiotic degradation.

Notably, habitat-dependent expression of specific resistance-associated proteins was observed. TolC-mediated RND efflux activity predominated in the natural ecotourism habitat (location P), whereas multiple β-lactamases and AcrA-associated efflux mechanisms predominated in the urban-proximate habitat (location S), suggesting distinct AMR strategies across habitats.

Figure 7: KEGG pathway analysis of antimicrobial resistance-associated proteins of Escherichia spp. in the fecal microbiomes of long-tailed macaques. Purple and green indicate proteins upregulated in locations P and S, respectively. The analysis highlights proteins enriched in the β-lactam resistance pathway and associated antimicrobial resistance mechanisms.

Figure 7: KEGG pathway analysis of antimicrobial resistance-associated proteins of Escherichia spp. in the fecal microbiomes of long-tailed macaques. Purple and green indicate proteins upregulated in locations P and S, respectively. The analysis highlights proteins enriched in the β-lactam resistance pathway and associated antimicrobial resistance mechanisms.