Gut metatranscriptome–virome profiling reveals active antimicrobial peptides (AMPs) encoded in plasmids and phages linked to human diseases

Gut metatranscriptome–virome profiling reveals active antimicrobial peptides (AMPs) encoded in plasmids and phages linked to human diseases | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Gut metatranscriptome–virome profiling reveals active antimicrobial peptides (AMPs) encoded in plasmids and phages linked to human diseases Luigui Gallardo-Becerra, Fernanda Cornejo-Granados, Shirley Bikel, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7160447/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 28 Nov, 2025 Read the published version in Microbial Ecology → Version 1 posted 9 You are reading this latest preprint version Abstract Microbe-derived antimicrobial peptides (AMPs) play a crucial role in shaping the microbiota composition; however, their contribution to disease-associated dysbiosis remains poorly understood. Here, we assembled fecal metatranscriptomes from individuals with normal weight, obesity, and obesity plus metabolic syndrome, yielding 51,087 non-human transcripts. We screened 1,095 small open reading frames (smORFs) using AMP-prediction algorithms and identified 112 AMP candidates. Most of them were associated with bacterial homologs, predominantly Faecalibacterium prausnitzii , while twelve aligned with plasmid or bacteriophage sequences. Differential expression analysis identified nine AMPs that were overexpressed among our groups, of which five originated from chromosomes, one from a plasmid, and three from phages. The expression of these AMPs was inversely correlated with specific bacterial taxa, linking them to disease-associated shifts in microbiota. Additionally, we also examined the presence of these nine AMPs in 372 external gut metatranscriptomes, discovering that they were highly prevalent in up to 98% of the samples, suggesting their conservation within the human gut microbiome and highlighting mobile elements as an often-overlooked reservoir of active AMPs. Finally, through virome sequencing and prophage genome analyses, we suggest that mobile-derived AMPs were transcribed from phage particles. We synthesized a phage-encoded AMP and demonstrated its broad-spectrum antibacterial activity against Gram-positive and Gram-negative bacteria, with no detectable cytotoxicity toward human immune cells. These findings illustrate that the human gut harbors a conserved set of microbe-derived AMPs associated with mobile genetic elements, whose overexpression was linked to obesity and metabolic syndrome, underscoring their role as ecological regulators of the microbiota in disease. Antimicrobial peptides AMPs phages metatranscriptome virome plasmid Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Childhood obesity has emerged as a critical public health challenge [ 1 ], significantly increasing the risk of type 2 diabetes, cardiovascular diseases, and non-alcoholic fatty liver disease in adulthood [ 2 ]. Obesity can also lead to metabolic syndrome [ 3 ], characterized by high blood glucose levels, elevated triglycerides, low high-density lipoproteins (HDL) levels, and hypertension [ 4 ], [ 5 ]. This combination severely impacts quality of life and raises the risk of disability and premature death [ 2 ], [ 6 ]. The situation is particularly alarming in Mexico, where nearly one in five school-aged children is obese, many of whom may have undiagnosed metabolic syndrome [ 7 ]. This highlights the urgent need for effective preventive and therapeutic strategies. A growing body of evidence suggests that obesity is associated with alterations in gut microbial communities, as demonstrated by 16S rRNA profiling, shotgun metagenomics, and metatranscriptomics [ 8 ]. Recently, viromics highlighted the role of bacteriophages in regulating these communities, which can either maintain a healthy microbe balance (eubiosis) or induce an imbalance (dysbiosis) [ 9 ]. Understanding the interactions between microbes and phages may lead to effective microbiome-targeted strategies for preventing and treating obesity and metabolic syndrome. Bidirectional communication between hosts and their resident microbes is mainly mediated by the "secrebiome," a collection of proteins released into the environment [ 8 ]. This includes enzymes, toxins, and antimicrobial peptides (AMPs), which are produced by both the host and microbiota as a key innate defense strategy [ 10 ], [ 11 ]. Despite their small size, AMPs exhibit diverse antimicrobial activities by destabilizing microbial membranes or binding to essential intracellular targets [ 12 ]. In bacteria, AMPs can be synthesized through the proteolytic processing of larger precursor proteins [ 13 ], [ 14 ], non-ribosomal peptide synthetase (NRPS) pathways [ 15 ], and dedicated genomic genes [ 16 ]. Recent metagenomic studies have highlighted small open reading frames (smORFs), under 100 codons, as a significant source of novel microbial-derived AMPs [ 17 ], [ 18 ]. Understanding the diversity and regulation of these smORF-encoded AMPs is crucial for advancing knowledge of host-microbiome interactions in health and disease. In the human gut, host-encoded AMPs act as gatekeepers, regulating opportunistic pathogens and the composition of the resident microbiota [ 19 ], [ 20 ]. They help maintain microbial diversity, which is essential for intestinal health and they are naturally compatible with the host due to their co-evolution within the intestinal ecosystem [ 21 ], [ 22 ], [ 23 ]. However, some AMPs may increase vulnerability to viral infections, demonstrating context-dependent effects [ 24 ]. At the same time, microbial-derived AMPs foster competition among bacteria, allowing AMP-producing species to secure ecological niches and influence microbial dynamics [ 25 ]. Horizontal gene transfer via plasmids and bacteriophages (phages) enhances the functional capabilities of AMPs, aiding bacterial fitness through the dissemination of AMP-encoding genes [ 26 ], [ 27 ], [ 28 ], [ 29 ]. However, no study has established a link between AMPs and the dysbiosis seen in disorders like obesity and metabolic syndrome. Metatranscriptomics provides insights into which AMP genes are actively transcribed, highlighting those involved in community self-regulation. This study examines gut metatranscriptomes from healthy, obese, and obese plus metabolic syndrome individuals to catalog differentially expressed AMPs and explore their roles in microbiota dysbiosis. This research is among the first to connect the active AMP repertoire in the human gut to obesity and metabolic syndrome, paving the way for microbiome-targeted therapies. Materials and Methods De novo assembly and AMP prediction We analyzed RNA-seq data from two normal-weight, three obese, and three obese children with Metabolic Syndrome, as detailed in a prior study [ 8 ]. Quality assessment was performed using FastQC ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ), followed by filtering with a Q20 Phred score and trimming using Trimmomatic v0.36. We removed sequencing adapters, ambiguous bases and rRNAs using Ribopicker v0.4.3 and the SILVA rRNA database (138 release). The remaining non-rRNA sequences were aligned to the human genome and transcriptome with Kneaddata to eliminate human derived sequences ( https://github.com/biobakery/kneaddata ). The remaining sequences were then used as input for the de novo transcriptome assembly with the Trinity assembler, adjusting to conserve small transcripts. After that, the original reads were aligned to the transcriptome using Bowtie2. Expression levels were calculated by Expectation (RSEM). After, we removed the transcripts with an expression below 1 in any sample or with less than three cumulative observations across all samples. Then, we employed TransDecoder ( https://github.com/TransDecoder/ ) to identify ORF candidates within protein-coding regions of the transcript sequences, adjusting its parameters to obtain the smORFs (-m 5). The small proteins derived from these smORFs were analyzed for AMP prediction using Macrel ( https://github.com/BigDataBiology/macrel ), AxPEP ( https://sourceforge.net/projects/axpep/ ), and AMP Scanner V2 ( https://www.dveltri.com/ascan/v2/ascan.html ) with default parameters. Taxonomic classification, phylogeny, genomic synteny and prophage analyses The transcript sequences encoding AMPs were used for homology search against NCBI’s NT database using the BLASTN algorithm with the following parameters: e-value > = 0.00001, identity > = 80%, and coverage > = 80%. The associated taxonomy was determined with a Last Common Ancestor (LCA) using MEGAN6 Community Edition (v.6.25.10). The genomic context was obtained for all transcripts, using BLASTN best-hit, and the phylogenic trees were created with iTOL ( https://itol.embl.de ). Plots for genomic context were created with the AnnotationSketch drawing library ( https://genometools.org/annotationsketch.html ). The genomic synteny comparisons were conducted using Easyfig ( https://mjsull.github.io/Easyfig/ ) with the BLASTN algorithm. We highlighted genomes (accession numbers) containing the AMPs identified in this study in bold in the corresponding figure. Arrows indicate the positions of coding sequences, while shaded lines represent the degree of homology between genomic regions or genomes. We assigned functional categories based on the COG classification system using the COGclassifier tool ( https://github.com/moshi4/COGclassifier ), with each category represented by a distinct color to facilitate interpretation. We screened the bacterial genomes with VirSorter2 to determine whether the AMPs we identified were embedded within prophage regions ( https://github.com/jiarong/VirSorter2 ). Read mapping of viral DNA to phage genomes The quality-filtered phageome reads were obtained from a previously published dataset by our laboratory, which was derived from the same cohort (NCBI BioProject: PRJNA646512). We merged all R1 files into a single R1 file and all R2 files into a single R2 file to create a representative population of the entire virome. Next, we created an index for each phage genome using bowtie2-build with default parameters. We aligned the merged reads to the bacteriophage genomes using Bowtie2 with the parameters ‘—no-unal --end-to-end --very-sensitive.’ Finally, we calculated the total genome coverage and the X-fold coverage for each genome. 16S microbiota profiling . We obtained the V4 region of 16S rRNA gene for the samples from a previously published dataset by our laboratory [ 8 ]. All quality-filtered sequences were joined and analyzed using QIIME2 (v2024.5). Briefly, raw sequences were imported and dereplicated, followed by de novo clustering at a 97% identity threshold. Taxonomic classification was performed using a Naive Bayes classifier trained on the SILVA 138 reference database. Taxonomy bar plots were generated, and amplicon sequence variants (ASVs) that were present in more than half of the samples (n = 4) were retained for Spearman correlation analysis against AMP abundances. Peptide synthesis and purification ADR1 and ADR2 were chemically synthesized by a solid-phase method using the Fmoc methodology (GenScript Biotech, Piscataway, NJ) and purified. Briefly, ten milligrams of crude synthetic peptides were dissolved in one milliliter of 20% aqueous acetonitrile solution and were separated each by reverse phase HPLC on an analytic C18 column (Zorbax SB-C18, Agilent, USA). The C18 column was equilibrated in 20% aqueous acetonitrile containing 0.1% TFA, and the synthetic peptides were separated using a linear gradient of acetonitrile/0.1% TFA from 20 to 60% in 40 minutes at a flow rate of 1 mL/min. The presence of each peptide was monitored at 220 nm. The molecular mass of each peptide was determined by mass spectrometry. Antibacterial assays The assessment of bacterial growth was conducted using a broth microdilution assay according to the Clinical and Laboratory Standards Institute guidelines. The reference strains, Pseudomonas aeruginosa (ATCC 27853), Klebsiella pneumoniae (ATCC700603), Staphylococcus aureus (ATCC 29213) and Streptococcus pneumoniae (ATCC46916) were cultured in Mueller-Hinton broth (MHB) at 37ºC during overnight incubation. Following this culture period, the samples were diluted in MHB to achieve an endpoint corresponding to an absorbance between 0.08 and 0.13 units at 625 nm, followed by a further dilution of 1:100 in MHB (approximately 1 x 108 CFU/mL). An aliquot of fifty microliters from each bacterial suspension was introduced into each well of a 96-well microtiter culture plate, which contained 50 µL of MHB supplemented with varying concentrations of the synthetic peptides ADR1 and ADR2 (100, 50, 25, 12.5, 6.2, and 3.13 µg/mL). The growth of each bacterial strain was quantitatively evaluated by measuring absorbance at a wavelength of 630 nm after an incubation period of 18 hours at 37°C. Flow Cytometry Analysis of T Cell Populations Human red blood cells were collected from a healthy donor who gave verbal consent for phlebotomy. This related experiment was approved by the Ethics Committee of the Facultad de Medicina y Ciencias Biomédicas of the Universidad Autónoma de Chihuahua with registration number CI-068-19. To assess the potential cytotoxic effects of the synthetic peptides on human immune cells, peripheral blood mononuclear cells (PBMCs) were isolated from three healthy donors. The cells were cultured in RPMI 1640 medium (GIBCO, 11875-085, Paisley, SCT, UK) supplemented with 5% fetal bovine serum (FBS) (GIBCO, 26140-079, Grand Island, NE, USA) and GlutaMAX (GIBCO, 35050-061, Grand Island, NE, USA). PBMCs were seeded in 96-well plates and exposed to each AMP peptide (ADR1 and ADR2) at 20 mg/mL. Cells cultured in medium alone were used as untreated controls. The initial evaluation was performed one hour after peptide exposure to determine the immediate effects on T lymphocyte populations. Subsequently, the PBMCs were incubated overnight at 37°C in a humidified atmosphere with 5% CO₂ to assess potential delayed or sustained effects. Following incubation, cells were stained with the BD TriTest™ reagent (Becton Dickinson and Company, BD Biosciences, San Jose, CA, USA) to identify CD3⁺ T cells and their CD4⁺ and CD8⁺ subpopulations, the viability of CD3⁺, CD4⁺, and CD8⁺ T cell subpopulations was evaluated using the LIVE/DEAD™ Fixable Near-IR Dead Cell Stain Kit (Invitrogen™, Thermo Fisher Scientific, Eugene, OR, USA). The absolute number of cells was acquired using BDTM Liquid Counting Beads (BD Biosciences). Flow cytometric analysis was performed using a FACS Canto flow cytometer (BD Biosciences), and the data were analyzed with FlowJo software (BD Biosciences). Graphical representations and statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA). This experimental design allowed for the evaluation of peptide-induced changes in the distribution and viability of key T lymphocyte subsets, providing insight into the potential immunotoxicity or immunomodulatory properties of the peptides in human immune cells. Results 1. Identification and taxonomic classification of AMP-encoding smORFs of the gut microbial transcripts The metatranscriptomic data analyzed in this study were obtained from a previously characterized cohort (BioProject PRJNA600247) [ 8 ], which includes fecal samples from three metabolic phenotypes: two normal-weight children (NW), three with obesity (OB), and three with obesity plus metabolic syndrome (OMS) (Fig. 1 A). After filtering out low-quality reads and contaminating sequences, we obtained 42.97 million high-quality paired reads. Using de novo assembly with Trinity, we generated 51,087 transcripts (N50 = 1,372 bp) and 36.22 million assembled nucleotides (Supplementary Tables 1–2 and Supplementary Fig. 1). We recalculated transcript abundance as Fragments per Kilobase Million (FPKM). We applied strict filtering, retaining only transcripts with an FPKM value ≥ 1 that were expressed in at least three samples, resulting in 2,522 biologically robust transcripts (Supplementary Table 2). Notably, 1,095 of these transcripts encoded smORFs of fewer than 100 amino acids (Fig. 1 A). We explored the antimicrobial potential of 1,095 smORF-encoded peptides using three bioinformatics tools for AMP prediction: Macrel [ 33 ], AxPEP [ 34 ], and AMP Scanner V2 [ 35 ]. This combined screening identified 420 potential AMPs, which we refined based on three criteria: (i) encoded by a unique non-redundant gene situated in a genomic locus that did not overlap with another coding sequences, (ii) comprising more than 10 amino acids, and (iii) supported by at least one homolog transcript in NCBI (Fig. 1 B). This process yielded 112 high-confidence AMP candidates, which served as the basis for subsequent analyses (Supplementary Table 3). AxPEP identified 77 peptides (68.8%), AMP Scanner v2 identified 15 (13.4%), and Macrel only identified 2 (1.8%). Only three peptides (2.7%) were unanimously predicted by all tools (Supplementary Fig. 2A). We benchmarked the accuracy of these AMP prediction tools against a curated reference panel of 1,390 experimentally validated microbe-derived peptides compiled from APD3 [ 30 ], dbAMP [ 31 ], and DRAMP [ 32 ], confirming 1,332 (95.8%) as bona-fide AMPs. These findings highlight the variability among AMP predictors and underscore the importance of utilizing multiple tools for reliable candidate selection. The taxonomic classification of transcripts encoding the 112 AMP determined that 100 transcripts (89.3%) were from bacterial origin, with Faecalibacterium prausnitzii representing the majority with 48 transcripts (42.9%) (Supplementary Fig. 2B). The remaining 12 transcripts (10.7%) were linked to the gut virome, mainly comprising Caudovirales, including Myoviridae and Siphoviridae. This underscores bacteriophages as a source of AMP genes. Notably, one transcript was associated with a plasmid, highlighting the role of mobile genetic elements in the gut ecosystem. 2. Phage, plasmid, and bacteria-derived AMPs were differentially expressed between O and OMS groups. We focused our differential expression analysis on the O and OMS groups, each of which had three biological replicates. Using DESeq2, we set a threshold of fold change greater than two and a p-value of less than 0.05 for significance. Out of 112 candidate AMPs, only nine were differentially expressed: three were up-regulated in O and six in OMS (Fig. 2 A). The taxonomical classification of these nine differentially expressed AMPs revealed a diverse composition, including phage-derived, plasmid-associated, and chromosomally encoded bacterial AMPs (Fig. 2 A; Supplementary Table 4). The differential expression of these AMPs in O Vs. OMS suggests that AMP dynamics may play a role in the dysbiosis associated with the disease. 3. Widespread transcriptomic prevalence of differentially expressed AMPs supports their global distribution. We focused on nine AMPs with significant expression shifts between O and OMS groups, highlighting their potential in understanding disease mechanisms. To evaluate their biological relevance, we assessed whether these AMPs are widely distributed across diverse human gut communities or if they are specific to our cohort. Accordingly, we examined a dataset of 372 fecal metatranscriptomes from the U.S.-based Health Professionals Follow-Up Study (HPFS) (Bioproject PRJNA354235) [ 33 ]. This cohort spans a wide range of BMI values, metabolic phenotypes, ages, and dietary backgrounds, thereby offering a robust framework for assessing AMP prevalence across heterogeneous populations. We mapped HPFS RNA-seq reads to our nine AMP-encoding contigs and considered an AMP present if the coverage exceeded one transcript per million. Notably, all nine AMPs were detected in a significant portion of these samples, with prevalence rates ranging from 9.8–98.4% (Fig. 2 B; Supplementary Table 5). On average, each AMP was found in 67.93% of samples, with a median detection rate of 80.16%. This widespread distribution indicates that these AMPs are conserved elements within the gut microbiome, likely involved in inter-bacterial antagonism, community structuring, and host-microbe signaling. 4. Chromosomally Encoded AMPs Exhibit Conserved Sequence and Genomic Context Across Gut Microbial Taxa Among the nine differentially expressed AMPs, five exhibited protein homologs on bacterial chromosomes with phylogenetically diverse hosts. The AMP 3076 showed 100% sequence identity with homolog proteins in Escherichia coli (Supplementary Fig. 3A). The AMP 2526 shared 93.75% identity with four homolog proteins in Romboutsia (Supplementary Fig. 3B), while the AMP 3096 had 93.18% identity with homologs from Phocaeicola and Bacteroides (Supplementary Fig. 3C). The AMP 5865 showed 92% identity with ten homologs of Faecalibacterium (Supplementary Fig. 3D), and the AMP 2198 had 96.76% identity with three homologs in Blautia wexlerae (Supplementary Fig. 3E). The genomic contexts of these AMPs demonstrated high synteny conservation among surrounding genes, implying they are part of conserved genomic modules (Supplementary Fig. 4). This conservation in amino acid sequence and genomic architecture indicates strong purifying selection. 5. Plasmid-encoded AMPs contribute to the gut antimicrobial repertoire Among the chromosomally encoded AMPs, AMP 8200 stands out due to its dual genomic origin. BLASTP searches identified 15 highly similar protein homologs across Bacteroides and Phocaeicola , including one copy located on a Phocaeicola dorei plasmid (Fig. 3 ). The peptide sequence was 100% conserved in 13 strains, indicating strong purifying selection. Synteny analysis confirmed that the gene order is consistent among host genomes, suggesting a conserved functional cassette (Fig. 3 ). The presence of AMP 8200 in both chromosome and plasmid suggests a potential horizontal transfer event. Thus, we examined whether the 85 kb plasmid containing the AMP gene was integrated into the chromosome of P. dorei strain JR01. Whole-genome alignment revealed that only a 13 kb fragment of the plasmid was integrated, including the AMP gene, but not the entire plasmid sequence (Supplementary Fig. 5A). To differentiate between plasmid- and chromosome-derived transcripts, we re-mapped the reads to the 13 kb shared region in each sequence. Alignments to the plasmid sequence showed perfect matches, while the alignment to the 13 kb chromosomal sequence had 17 nucleotide mismatches. This suggests that transcripts mainly originate from the plasmid. Interestingly, when we mapped the RNA-seq reads to the plasmid, we found extensive transcription across the entire plasmid, with 95.5% coverage at an average depth of 61.7-fold. This confirms that the entire plasmid was highly active, underscoring the role of mobile genetic elements in disseminating antimicrobial functions within the gut microbiome. 6. AMPs encoded in active phages suggest the presence of phage-host dynamics In addition to the five chromosomally encoded peptides, three AMPs (AMP 3020; AMP 8681, and AMP 5245) show high sequence similarity to proteins from tailed bacteriophages (Fig. 4 ). BLASTP analysis demonstrated a perfect match (100% identity) for AMP 3020 with proteins from the Caudoviricetes phages ctJ1L4 and ctzDR1. It also identified two homologs in Anaerotignum strains, with 70.9–74.4% identity (Fig. 4 A). Despite these high sequence similarities, AMP 3020 displayed poor synteny conservation with the surrounding AMP gene in both phages and bacterial genomes (Fig. 5 A). The identical sequence between AMP 3020 and the homolog in phage ctJ1L4 suggests that the overexpression of AMP 3020 originated from this phage and not from Anaerotignum bacteria. Next, we investigated if the ctJ1L4 phage was present as an active viral particle previously reported in this same samples. To this end, we utilized a dataset of DNA-seq reads from viral-like particles, obtained from the same set of samples previously published by our laboratory [ 9 ]. After mapping the virome reads to the genome of ctJ1L4, we found that 70.46% of the genome was covered by reads, providing strong evidence for the physical presence of this phage as a viral particle (Supplementary Fig. 6A). Furthermore, RNA-seq mapping covered 14.7% of the phage genome, showing active transcription of several phage genes (Supplementary Fig. 6B). These results support the conclusion that ctJ1L4 was present as a viral particle with active transcription, reinforcing the likelihood that the overexpression of AMP 3020 originates from this phage. Given that AMP 3020 also shares 74.36% sequence identity with a protein from Anaerotignum sp. MB30-C6 genome, we investigated whether this similarity stemmed from the integration of the ctJ1L4 phage genome into the bacterial chromosome. We analyzed the 32.5 kb genome of phage ctJ1L4 and found only a minimal overlap of 372 nucleotides with the bacterial genome, corresponding to the AMP 3020 (Supplementary Fig. 6C), which suggests that the phage genome was not integrated into the bacterial chromosome. Additionally, a prophage prediction analysis of the Anaerotignum MB30-C6 genome did not detect any other prophages containing the AMP, suggesting a non-viral origin of this AMP in the bacterial genome. The AMP 8681 showed 100% sequence identity with proteins from two Caudoviricetes phages (ctlN07 and ctUyu4) and one Clostridiales bacterium KR001 hic 0007 (Fig. 4 B). It also shared 96.97% identity with proteins from phages ctvmC4 and ctlnE22. Despite this high sequence conservation, synteny was poorly conserved in both phage and bacterial genomes (Fig. 5 B). Given that this AMP has homologous proteins with 100% identity in both bacteria and phages, we investigated which of these two genomes could be the potential source of AMP overexpression. Mapping virome-derived reads to the ctlN07 phage genome showed low viral presence (7.18% coverage) (Supplementary Fig. 7A), while RNA-seq mapping revealed high transcriptional activity (12.82% coverage with 52.65-fold sequence depth). This suggests a high transcriptional activity of this phage (Supplementary Fig. 7B). To evaluate possible phage integration, we compared the 50 kb ctlN07 genome to the Clostridiales bacterium KR001 hic 0007 genomes. The comparison showed only a 3 kb shared region (98.10% identity) containing the AMP locus, indicating no phage integration in the bacterial genome (Supplementary Fig. 7C). Prophage prediction in the bacterial genome also did not identify any match for ctlN07, suggesting it is not lysogenized in the bacterium. Additionally, we did not detect any other prophages containing the AMP, suggesting a non-viral origin of this AMP in the bacterial genome. Overall, this indicates that AMP 8681 overexpression may originate from either the phage or bacterial genome, warranting further functional assays for clarification. The AMP 5245 exhibits a 100% amino acid sequence identity with five and three homolog proteins found in various phages and Blautia wexlerae (Fig. 4 C). The AMP showed significant conservation with two genomic structures among phages and bacterial genomes (Fig. 5 C). Given its high sequence similarity with both bacterial and phage proteins, we investigated the source of the observed AMP overexpression. Virome reads covered 88.7% of the phage genome, and RNA-seq reads accounted for 86.7% at 1,777X depth, indicating the presence of an active viral particle (Supplementary Fig. 8). To ensure the phage was not integrated into a bacterial genome, we compared the 8 kb human fecal virus clone to B. wexlerae DSM 19850, revealing only a 3 kb region with 98.35% identity, corresponding to 39.53% of the phage genome and encompassing the AMP locus (Supplementary Fig. 8C), suggesting a partial integration of phage into the bacterial genome. Prophage prediction analysis of the B. wexlerae genome did not identify any prophage elements matching the human fecal virus clone, suggesting that the phage was not integrated. Additionally, we did not detect any other prophages containing the AMP. The high coverage of the phage genome with DNA and RNA reads supports the notion that the phage was the primary source of AMP 5245 transcripts. 7. 16S–Metatranscriptome correlation analysis links AMPs expression with bacterial taxa To explore associations between the nine overexpressed AMPs and gut microbiota, we correlated their expression with 16S rRNA data (BioProject PRJNA600247; Supplementary Fig. 9). We observed significant correlations, focusing on the negative ones, which indicate that higher AMP expression corresponds with a decreased abundance of specific microbial taxa. AMP 5865 was inversely linked to Anaeroplasma , Clostridium , and Bacteroides species. The AMP 5245 negatively impacted over 20 taxa, particularly beneficial groups such as Akkermansia , Christensenellaceae, Moryella , and Oscillibacter . AMP 3076 showed an association with reduced abundances of Christensenellaceae, Bilophila , and several Lachnospiraceae lineages, while AMP 3096 was linked to decreased Desulfovibrionaceae, Bilophila wadsworthia , Oscillibacter , and A. muciniphila . Finally, AMP 2526 and AMP 2198 showed a negative correlation with A. muciniphila and Eubacterium species, bacteria typically associated with gut health. Together, these correlations suggest that the obesity-associated overexpression of AMPs may contribute to dysbiosis in obesity and metabolic syndrome by reducing the abundance of key commensal bacteria. 8. Experimental Validation of Phage-Encoded AMP 3020 Reveals Antibacterial Activity without T-Cell Toxicity To verify our in-silico predictions, we chose the AMP 3020 for experimental validation because of its confirmed phage origin. We synthesized two variants of AMP 3020 (Fig. 6 A): ADR1 retains the native N-terminal methionine, whereas ADR2 lacks this residue. ADR2 starts with valine, as it has been reported that some peptides undergo post-translational processing when the second residue of a nascent peptide is a short one; here, the second residue was valine [ 34 ]. This single amino acid difference also allowed us to test whether minimal sequence variation influences their functional activity. Both peptides significantly suppressed the growth of gram-negative ( Pseudomonas aeruginosa and Klebsiella pneumoniae ) and gram-positive ( Staphylococcus aureus and Streptococcus pneumoniae ) bacteria compared to controls (Fig. 6 B–E). ADR1 was more effective against P. aeruginosa , whereas ADR2 showed greater efficacy against K. pneumoniae . Both variants exhibited antibacterial activity against S. pneumoniae , with only ADR1 effective against S. aureus . Notably, neither peptide affected T-lymphocyte viability (Fig. 6 F), with similar cell death frequencies to those of untreated controls, ranging from 0.7–14.4% (Fig. 6 F). The absolute cell counts (Supplementary Figs. 10 and 11) confirmed no significant loss of viable T cells, while the positive control (PMA/ionomycin) showed a strong cytotoxic response. Overall, ADR1 and ADR2 demonstrated low cytotoxicity toward primary human T lymphocyte subsets. The removal of the initiator methionine may fine-tune target specificity due to changes in peptide folding or charge distribution. Discussion This study provides a framework for discovering and characterizing AMPs in the human gut microbiome using metatranscriptomics and viromics, representing a promising frontier to understand their role in host-microbiota interactions [ 10 ]. By integrating data from three AMP prediction algorithms — Macrell, AMP Scanner V2, and AxPEP — we identified 112 high-confidence AMP candidates from 1,095 expressed small open reading frames (smORFs). The limited overlap among the results of the different prediction tools highlights the methodological variability in AMP discovery. Notably, many expressed AMPs were linked to Faecalibacterium prausnitzii , a known gut commensal bacterium. Additionally, the identification of AMP-encoding transcripts from plasmids and phages expands the potential origins of AMPs beyond traditional chromosomal sources. The epithelial interface contains AMPs from host and microbial sources, which play a crucial role in shaping the surrounding microbiota [ 35 ]. Emerging evidence suggests that bioactive peptides can alter the gut microbiota [ 36 ]. This suggests the potential of AMPs to be associated with microbiota changes related to diseases, such as obesity and metabolic syndrome [ 8 ]. Both variants of phage-derived AMP 3020 (ADR1 and ADR2) demonstrated selective antibacterial activity while remaining non-cytotoxic to primary human T-cell subsets (CD3+, CD4+, and CD8+). Our synthetically phage-derived AMP 3020 variants showed antibacterial activity without being cytotoxic to human T-cell subsets (CD3+, CD4+, and CD8+). These results support their classification within the broader family of host defense peptides (HDPs), which also include bacteriophage-encoded peptides that influence inter-microbial competition and host-microbiota interactions [ 37 ], [ 38 ]. Notably, the lack of T cell cytotoxicity emphasizes the immunological neutrality of these peptides, suggesting that they can be produced by the microbiota to cause an effect on other bacteria without affecting the host These findings enhance our understanding of the gut virome and suggest that phage-derived AMPs can influence microbial communities while preserving host immune balance [ 39 ], [ 40 ]. Additionally, the obesity-related upregulation of several AMPs suggests their potential role in gut dysbiosis linked to obesity, as their expression negatively correlated with beneficial taxa like Akkermansia muciniphila , Christensenellaceae, and Desulfovibrionaceae, suggesting AMP-driven microbiota changes in disease. The differential expression analysis of AMPs in individuals with obesity compared to those with obesity and metabolic syndrome identified nine significantly overexpressed AMPs, linked to chromosomes, plasmids, and phages. This supports the idea that the host influences microbial gene expression of AMP production. Importantly, these nine AMPs were also expressed in 372 samples of an independent gut metatranscriptome dataset, indicating their commonality and ecological significance in the human gut microbiota, rather than being unique to our cohort. Their widespread presence suggests they are core features of the gut metatranscriptome and potential reliable biomarkers for disease. We identified three AMPs with homologs in bacteriophages, an underexplored source of antimicrobial compounds. The AMP 3020 was 98% identical to a protein from the Caudoviricetes phage ctJ1L4. Virome and metatranscriptome read mapping confirmed the presence of ctJ1L4 virions in our samples and demonstrated that several of their genes, including the AMP, were actively transcribed. Synthetic AMP 3020 inhibits the growth of both Gram-positive and Gram-negative bacteria, underscoring its potential as a broad-spectrum antimicrobial. Despite the presence of related proteins in two bacterial genomes, no evidence indicated ctJ1L4 integration or prophages with the AMP gene, suggesting that in vivo AMP production was primarily due to lytic phages. Removing the N-terminal Methionine from AMP 3020 changed its antibacterial activity, reflecting the evolutionary adaptability of phage-derived peptides. Methionine is essential for stabilizing protein structures and may also act as a regulatory switch through reversible redox reactions [ 41 ]. Overall, this data expands our understanding of the functional capabilities of gut bacteriophages, particularly those within the dominant Caudovirales lineage found in the human gut virome [ 9 ]. It emphasizes the importance of phages as a rich yet under-recognized source of AMPs that can change the gut microbiota. AMPs 8681 and 5245 show 100% amino acid identity between their phage- and bacterium-encoded homologs, with virome read mapping confirming their presence as free viral particles. Additionally, metatranscriptome data indicated viral transcription, suggesting that the phage was a significant contributor of AMP transcripts, although bacterial loci may also play a role. Prophage scans show that both chromosomal AMP loci were located outside predicted prophage regions, suggesting that the integration of both AMPs into the bacterial genomes was not related to a prophage. Interestingly, the expression of AMP 5245 negatively correlated with Moryella and Eubacterium species, known butyrate producers with anti-inflammatory effects, implying that this AMP may interfere with their beneficial effects [ 42 ], [ 43 ]. We also detected a plasmid containing a homolog protein to the AMP 8200, which was a described plasmid of Phocaeicola dorei [ 44 ]. The peptide exhibits 100% sequence identity and conserved synteny across additional Phocaeicola species. RNA-seq read profiling reveals robust transcription across the entire plasmid, suggesting that AMP expression likely originated from the plasmid, supporting the role of plasmids as reservoirs of antimicrobial capabilities in the gut microbiome [ 45 ]. Our findings highlight the potential for plasmids and phages to spread competitive traits like AMPs via horizontal transfer. AMPs 3076, 2526, 3096, 5865, and 2198 each share homolog proteins with a broad spectrum of gut bacterial commensals, including Escherichia coli , Faecalibacterium prausnitzii , Romboutsia spp., and Blautia wexlerae , indicating their widespread distribution in the intestinal microbiota. These AMPs demonstrate high sequence and synteny conservation across bacterial genomes. Despite their widespread bacterial distribution, each AMP has a unique physicochemical signature and genomic context, suggesting specialized functional roles. The negative correlations between these AMP expressions and beneficial microbes, such as Akkermansia muciniphila and Christensenellaceae , imply a potential link to microbial dysbiosis observed in obesity and metabolic syndrome [ 46 ], [ 47 ]. For example, AMP 5865 was overexpressed in obesity and showed significant negative correlations with A. muciniphila and Alistipes obesii , two organisms consistently associated with favorable metabolic outcomes [ 48 ]. This suggests that this AMP could be related to the low abundance of these two taxas observed in the metabolic syndrome cohort. Metatranscriptomic profiling provides a dynamic map of microbiome activity, revealing which genes are actively expressed under conditions such as obesity and metabolic syndrome. This approach differs from traditional metagenomics, which only lists gene presence. Our analysis identified actively expressed AMPs, helping differentiate between latent genetic potential and those that actively influence host microbiota and disease associations. Our findings highlight that gut-expressed AMPs were derived from diverse genomic sources, including bacteria, plasmids, and phages, indicating their significant ecological roles in gut microbial dynamics. This pilot study shows that metatranscriptomic data can uncover relevant, expressed AMPs implicated in the gut microbiome regulation. The differential expression of these AMPs, linked to disease, along with their antimicrobial properties that do not affect host immunity points to their importance in shaping the gut microbiota. Moreover, the presence of mobile genetic elements, such as plasmids and phages, reinforces the need to rethink the role of bacteriophages in gut ecology, acting not only as predators but also as regulators. This work sets the stage for future studies on AMPs and their therapeutic potential in microbiome-targeted treatments for dysbiosis related to obesity. Declarations Acknowledgments L.G.B. thanks to the Doctoral Biochemical Sciences Program at IBt UNAM and CONACyT for doctoral fellowship C.V.U.: 887285. Also, F.C.G. would thank the Estancias posdoctorales por México 2022 program (C.V.U.: 443238). This research was funded by CONACyT grant Ciencia de Frontera-2019-263986 and by DGAPA PAPIIT UNAM (IN219723) . We thank M.T.I Juan Manuel Hurtado Ramírez for informatics technical support. Also, the authors would like to thank the "Unidad Universitaria de Secuenciación Masiva y Bioinformática" of the "Laboratorio Nacional de Apoyo Tecnológico a las Ciencias Genómicas," UNAM, especially to Ricardo Alfredo Grande Cano and Lizeth A. Matías Valdez for the technical sequencing support. Funding This research was funded by CONACyT grant Ciencia de Frontera-2019-263986 and by DGAPA PAPIIT UNAM (IN219723) . L.G.B. was supported by the Doctoral Biochemical Sciences Program at IBt UNAM and CONACyT with the doctoral fellowship C.V.U.: 778192. F.C.G. was supported by the Estancias posdoctorales por México 2022 program (C.V.U.: 443238) . Disclosure statement The authors declare no competing interests. Data availability statement The data that support the findings of this study are openly available in NCBI GEO repository with accession number GSE143207 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143207) and the NCBI BioProject: PRJNA600247 (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA600247) and PRJNA646512 (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA646512). All the code used for this project was deposited in this GitHub repository: https://github.com/LuiguiGallardo/amps_microbiome. Requests for additional material should be made to the corresponding author. Authors contributions Conceived of or designed study: LGB, FCG, SB, and AOL; Analyzed Data: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL; Formal analysis: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL; Contributed new methods or models: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL; Wrote the paper: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL. Funding acquisition: FCG, SCQ, AOL. All authors read and approved the final manuscript. References Biro FM, Wien M (May 2010) Childhood obesity and adult morbidities. Am J Clin Nutr 91(5). 10.3945/ajcn.2010.28701B . pp. 1499S-1505S Franks PW, Hanson RL, Knowler WC, Sievers ML, Bennett PH, Looker HC (2010) Childhood Obesity, Other Cardiovascular Risk Factors, and Premature Death, N. Engl. J. Med. , vol. 362, no. 6, pp. 485–493, Feb. 10.1056/NEJMoa0904130 Perichart-Perera O, Balas-Nakash M, Schiffman-Selechnik E, Barbato-Dosal A, Vadillo-Ortega F (2007) Obesity Increases Metabolic Syndrome Risk Factors in School-Aged Children from an Urban School in Mexico City, J. Am. Diet. Assoc. , vol. 107, no. 1, pp. 81–91, Jan. 10.1016/j.jada.2006.10.011 Al-Hamad D, Raman V (2017) Metabolic syndrome in children and adolescents, Transl. Pediatr. , vol. 6, no. 4, Art. no. 4, Oct. 10.21037/tp.2017.10.02 de Ferranti SD, Gauvreau K, Ludwig DS, Neufeld EJ, Newburger JW, Rifai N (Oct. 2004) Prevalence of the Metabolic Syndrome in American Adolescents. Circulation 110(16):2494–2497. 10.1161/01.CIR.0000145117.40114.C7 Evia-Viscarra ML, Rodea-Montero ER, Apolinar-Jiménez E, Quintana-Vargas S (2013) Metabolic syndrome and its components among obese (BMI ≥ 95th) Mexican adolescents, Endocr. Connect. , vol. 2, no. 4, pp. 208–215, Dec. 10.1530/EC-13-0057 Shamah-Levy T et al (2023) Jun., Prevalencias de sobrepeso y obesidad en población escolar y adolescente de México. Ensanut Continua 2020–2022, Salud Pública México , vol. 65, pp. s218–s224. 10.21149/14762 Gallardo-Becerra L et al (Dec. 2020) Metatranscriptomic analysis to define the Secrebiome, and 16S rRNA profiling of the gut microbiome in obesity and metabolic syndrome of Mexican children. Microb Cell Factories 19(1):61. 10.1186/s12934-020-01319-y Bikel S et al (2021) Aug., Gut dsDNA virome shows diversity and richness alterations associated with childhood obesity and metabolic syndrome, iScience , vol. 24, no. 8, p. 102900. 10.1016/j.isci.2021.102900 Gallardo-Becerra L, Cervantes-Echeverría M, Cornejo-Granados F, Vazquez-Morado LE, Ochoa-Leyva A (Dec. 2023) Perspectives in Searching Antimicrobial Peptides (AMPs) Produced by the Microbiota. Microb Ecol 87(1):8. 10.1007/s00248-023-02313-8 Izadpanah A, Gallo RL (Mar. 2005) Antimicrobial peptides. J Am Acad Dermatol 52(3):381–390. 10.1016/j.jaad.2004.08.026 Ageitos JM, Sánchez-Pérez A, Calo-Mata P, Villa TG (Jun. 2017) Antimicrobial peptides (AMPs): Ancient compounds that represent novel weapons in the fight against bacteria. Biochem Pharmacol 133:117–138. 10.1016/j.bcp.2016.09.018 Torres MDT, Melo MCR, Flowers L, Crescenzi O, Notomista E, De La Fuente-Nunez C (2021) Mining for encrypted peptide antibiotics in the human proteome, Nat. Biomed. Eng. , vol. 6, no. 1, pp. 67–75, Nov. 10.1038/s41551-021-00801-1 Pizzo E, Cafaro V, Di Donato A, Notomista E (May 2018) Cryptic Antimicrobial Peptides: Identification Methods and Current Knowledge of their Immunomodulatory Properties. Curr Pharm Des 24(10):1054–1066. 10.2174/1381612824666180327165012 Nolan EM, Walsh CT (Jan. 2009) How Nature Morphs Peptide Scaffolds into Antibiotics. ChemBioChem 10(1):34–53. 10.1002/cbic.200800438 Singh N, Abraham J (Apr. 2014) Ribosomally synthesized peptides from natural sources. J Antibiot (Tokyo) 67(4):277–289. 10.1038/ja.2013.138 Torres MDT et al (Aug. 2024) Mining human microbiomes reveals an untapped source of peptide antibiotics. Cell 0(0). 10.1016/j.cell.2024.07.027 Santos-Júnior CD et al (Jun. 2024) Discovery of antimicrobial peptides in the global microbiome with machine learning. Cell 0(0). 10.1016/j.cell.2024.05.013 Nakatsuji T et al (Feb. 2017) Antimicrobials from human skin commensal bacteria protect against Staphylococcus aureus and are deficient in atopic dermatitis. Sci Transl Med 9(378):eaah4680. 10.1126/scitranslmed.aah4680 Ma Y et al (2022) Jun., Identification of antimicrobial peptides from the human gut microbiome using deep learning, Nat. Biotechnol. , vol. 40, no. 6, Art. no. 6. 10.1038/s41587-022-01226-0 Sun D et al (2021) Apr., Angiogenin maintains gut microbe homeostasis by balancing α-Proteobacteria and Lachnospiraceae, Gut , vol. 70, no. 4, pp. 666–676. 10.1136/gutjnl-2019-320135 González-Dávila P et al (Aug. 2022) Catestatin selects for colonization of antimicrobial-resistant gut bacterial communities. ISME J 16(8):1873–1882. 10.1038/s41396-022-01240-9 Magana M et al (Sep. 2020) The value of antimicrobial peptides in the age of resistance. Lancet Infect Dis 20(9):e216–e230. 10.1016/S1473-3099(20)30327-3 Deng Y, Yang S, Zhang L, Chen C, Cheng X, Hou C (Jan. 2024) Chronic bee paralysis virus exploits host antimicrobial peptides and alters gut microbiota composition to facilitate viral infection. ISME J 18(1):wrae051. 10.1093/ismejo/wrae051 Garcia-Gutierrez E, Mayer MJ, Cotter PD, Narbad A (Jan. 2019) Gut microbiota as a source of novel antimicrobials. Gut Microbes 10(1):1–21. 10.1080/19490976.2018.1455790 Kommineni S et al (2015) Oct., Bacteriocin production augments niche competition by enterococci in the mammalian gastrointestinal tract, Nature , vol. 526, no. 7575, pp. 719–722. 10.1038/nature15524 Nakazono K et al (Jan. 2022) Complete sequences of epidermin and nukacin encoding plasmids from oral-derived Staphylococcus epidermidis and their antibacterial activity. PLoS ONE 17(1):e0258283. 10.1371/journal.pone.0258283 Du J et al (Jul. 2023) Enhancing bacteriophage therapeutics through in situ production and release of heterologous antimicrobial effectors. Nat Commun 14(1):4337. 10.1038/s41467-023-39612-0 Tyagi JL, Gupta P, Ghate MM, Kumar D, Poluri KM (May 2024) Assessing the synergistic potential of bacteriophage endolysins and antimicrobial peptides for eradicating bacterial biofilms. Arch Microbiol 206(6):272. 10.1007/s00203-024-04003-6 Wang G, Li X, Wang Z (Jan. 2016) APD3: the antimicrobial peptide database as a tool for research and education. Nucleic Acids Res 44:D1087–D1093. 10.1093/nar/gkv1278 Yao L et al (2025) Jan., dbAMP 3.0: updated resource of antimicrobial activity and structural annotation of peptides in the post-pandemic era, Nucleic Acids Res. , vol. 53, no. D1, pp. D364–D376. 10.1093/nar/gkae1019 Ma T et al (Jan. 2025) DRAMP 4.0: an open-access data repository dedicated to the clinical translation of antimicrobial peptides. Nucleic Acids Res 53:D403–D410. no. D110.1093/nar/gkae1046 Abu-Ali GS et al (Jan. 2018) Metatranscriptome of human faecal microbial communities in a cohort of adult men. Nat Microbiol 3(3):356–366. 10.1038/s41564-017-0084-4 Ben-Bassat A, Bauer K (1987) Amino-terminal processing of proteins, Nature , vol. 326, no. 6110, pp. 315–315, Mar. 10.1038/326315a0 Zong X, Fu J, Xu B, Wang Y, Jin M (Dec. 2020) Interplay between gut microbiota and antimicrobial peptides. Anim Nutr 6(4):389–396. 10.1016/j.aninu.2020.09.002 Xu F et al (May 2016) GLP-1 receptor agonist promotes brown remodelling in mouse white adipose tissue through SIRT1. Diabetologia 59(5):1059–1069. 10.1007/s00125-016-3896-5 Haney EF, Straus SK, Hancock REW (Feb. 2019) Reassessing the Host Defense Peptide Landscape. Front Chem 7. 10.3389/fchem.2019.00043 Peng S-Y, You R-I, Lai M-J, Lin N-T, Chen L-K, Chang K-C (2017) Highly potent antimicrobial modified peptides derived from the Acinetobacter baumannii phage endolysin LysAB2, Sci. Rep. , vol. 7, no. 1, Art. no. 1, Sep. 10.1038/s41598-017-11832-7 Ritz NL et al (Feb. 2024) The gut virome is associated with stress-induced changes in behaviour and immune responses in mice. Nat Microbiol 9(2):359–376. 10.1038/s41564-023-01564-y Wei Y (Sep. 2024) Bacteriophages: a double-edged sword in the gastrointestinal tract. Front Microbiomes 3:1450523. 10.3389/frmbi.2024.1450523 Aledo JC (2019) Methionine in proteins: The Cinderella of the proteinogenic amino acids, Protein Sci. Publ. Protein Soc. , vol. 28, no. 10, pp. 1785–1796, Oct. 10.1002/pro.3698 Lu H et al (2022) Aug., Butyrate-producing Eubacterium rectale suppresses lymphomagenesis by alleviating the TNF-induced TLR4/MyD88/NF-κB axis, Cell Host Microbe , vol. 30, no. 8, pp. 1139–1150.e7. 10.1016/j.chom.2022.07.003 Sung MM et al (Nov. 2016) Improved Glucose Homeostasis in Obese Mice Treated With Resveratrol Is Associated With Alterations in the Gut Microbiome. Diabetes 66(2):418–425. 10.2337/db16-0680 Kijner S, Ennis D, Shmorak S, Florentin A, Yassour M (Jan. 2024) CRISPR-Cas-based identification of a sialylated human milk oligosaccharides utilization cluster in the infant gut commensal Bacteroides dorei. Nat Commun 15(1):105. 10.1038/s41467-023-44437-y Fernández-Pato A et al (Jun. 2025) Early-life development of the gut virome and plasmidome: A longitudinal study in cesarean-born infants. Cell Rep 44(6). 10.1016/j.celrep.2025.115731 Waters JL, Ley RE (Oct. 2019) The human gut bacteria Christensenellaceae are widespread, heritable, and associated with health. BMC Biol 17:83. 10.1186/s12915-019-0699-4 Everard A et al (2013) May., Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity, Proc. Natl. Acad. Sci. , vol. 110, no. 22, pp. 9066–9071. 10.1073/pnas.1219451110 Jie Z et al (2021) May., The Baseline Gut Microbiota Directs Dieting-Induced Weight Loss Trajectories, Gastroenterology , vol. 160, no. 6, pp. 2029–2042.e16. 10.1053/j.gastro.2021.01.029 Additional Declarations No competing interests reported. Supplementary Files supplementarytablesmicrobialecology.xlsx supplementaryfiguresmicrobialecology.docx Cite Share Download PDF Status: Published Journal Publication published 28 Nov, 2025 Read the published version in Microbial Ecology → Version 1 posted Editorial decision: Revision requested 25 Aug, 2025 Reviews received at journal 25 Aug, 2025 Reviews received at journal 22 Aug, 2025 Reviewers agreed at journal 06 Aug, 2025 Reviewers agreed at journal 28 Jul, 2025 Reviewers invited by journal 22 Jul, 2025 Editor assigned by journal 22 Jul, 2025 Submission checks completed at journal 22 Jul, 2025 First submitted to journal 18 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7160447","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":490088933,"identity":"f7be0c45-59b6-4876-a6c1-d3c583c000fb","order_by":0,"name":"Luigui Gallardo-Becerra","email":"","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":false,"prefix":"","firstName":"Luigui","middleName":"","lastName":"Gallardo-Becerra","suffix":""},{"id":490088934,"identity":"f274aea9-d4d8-4567-92c3-42e8c2a3d0b4","order_by":1,"name":"Fernanda Cornejo-Granados","email":"","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":false,"prefix":"","firstName":"Fernanda","middleName":"","lastName":"Cornejo-Granados","suffix":""},{"id":490088935,"identity":"f078d0af-e116-45f8-9716-0df128859684","order_by":2,"name":"Shirley Bikel","email":"","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":false,"prefix":"","firstName":"Shirley","middleName":"","lastName":"Bikel","suffix":""},{"id":490088936,"identity":"dd393eee-80d7-47a3-a26f-bdff40f6cf31","order_by":3,"name":"Iván Arenas","email":"","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":false,"prefix":"","firstName":"Iván","middleName":"","lastName":"Arenas","suffix":""},{"id":490088937,"identity":"b971363e-d66c-4b6a-bd56-03fdd131815c","order_by":4,"name":"Gamaliel López-Leal","email":"","orcid":"","institution":"Universidad Autónoma del Estado de Morelos","correspondingAuthor":false,"prefix":"","firstName":"Gamaliel","middleName":"","lastName":"López-Leal","suffix":""},{"id":490088938,"identity":"208cf051-2ba8-409d-b0b5-7a283e790b0e","order_by":5,"name":"Carolina Alvarado-Gonzalez","email":"","orcid":"","institution":"Autonomous University of Chihuahua","correspondingAuthor":false,"prefix":"","firstName":"Carolina","middleName":"","lastName":"Alvarado-Gonzalez","suffix":""},{"id":490088941,"identity":"ed16a7f1-f9bf-41fd-9637-bfbdaae58e13","order_by":6,"name":"Filiberto Sánchez","email":"","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":false,"prefix":"","firstName":"Filiberto","middleName":"","lastName":"Sánchez","suffix":""},{"id":490088942,"identity":"c890860b-222c-464d-831f-42b5c60e3c76","order_by":7,"name":"Gerardo Corzo","email":"","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":false,"prefix":"","firstName":"Gerardo","middleName":"","lastName":"Corzo","suffix":""},{"id":490088944,"identity":"2717c31f-b53f-49a8-ab0a-5cd8edbc1b0d","order_by":8,"name":"Gerardo P. Espino-Solis","email":"","orcid":"","institution":"Autonomous University of Chihuahua","correspondingAuthor":false,"prefix":"","firstName":"Gerardo","middleName":"P.","lastName":"Espino-Solis","suffix":""},{"id":490088946,"identity":"93f6df20-6953-4f2b-a38e-41708a7e7d77","order_by":9,"name":"Samuel Canizales-Quinteros","email":"","orcid":"","institution":"UNAM/Instituto Nacional de Medicina Genómica (INMEGEN)","correspondingAuthor":false,"prefix":"","firstName":"Samuel","middleName":"","lastName":"Canizales-Quinteros","suffix":""},{"id":490088947,"identity":"9468affe-061a-4bb0-b5d8-706460adf9e0","order_by":10,"name":"Adrian Ochoa-Leyva","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYFCCBIYDYFoCiD8YHCBRC+MMgwMSRGlhgGlh5mEgQgt/e+7DAz93MMjxz24++Nmm4E4dg0TuwU83GLYlNuDQInHmucHB3jMMxhJ3jiVL5xg8k2CQyAMyGG4b47TmRhrDAd42hsQNEjkGQC2HgVpADIbbcrh0yAO1HPwL0WL82wKixfg3UAsPLi0GQC2HobaYSTNAtJjhtcXwzDOGw7JtEsYSN9LSLHsMnkm28bwxs84xwO0XueNpzB/fttnI8c9IPnzjx587/PzsOca3cypu4wwxKECKDTaIg/GrHwWjYBSMglGAHwAA0vxTjFJ8FnYAAAAASUVORK5CYII=","orcid":"","institution":"Universidad Nacional Autónoma de México","correspondingAuthor":true,"prefix":"","firstName":"Adrian","middleName":"","lastName":"Ochoa-Leyva","suffix":""}],"badges":[],"createdAt":"2025-07-18 20:08:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7160447/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7160447/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00248-025-02620-2","type":"published","date":"2025-11-28T15:58:36+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":87604291,"identity":"1f93a353-1caa-4bd3-a4aa-4707184d5970","added_by":"auto","created_at":"2025-07-25 17:42:06","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":505103,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow for Discovering Gut-Expressed AMPs from Fecal Metatranscriptomes. A) Data Processing: Raw Illumina reads undergo quality filtering to remove host and rRNA sequences. The remaining reads were then assembled de novo, and smORFs were extracted. B) AMP Discovery: The extracted smORFs were screened using three independent predictors Macrel, AMP-Scanner v2, and AxPEP. Candidate peptides undergo manual curation based on factors such as length, genomic context, transcript support, and protein homolog similarity. C) Differential Expression: The curated AMPs were quantified and compared between the two groups, O and OMS. This analysis results in the final set of AMPs associated with the disease.\u003c/p\u003e","description":"","filename":"11.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/9b1825d59a1ded76221eb5b2.jpg"},{"id":87604425,"identity":"075f93fa-3e0d-4847-b557-54e8718e952c","added_by":"auto","created_at":"2025-07-25 17:50:06","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":462284,"visible":true,"origin":"","legend":"\u003cp\u003eDisease-Linked over-expressed AMPs and Their Prevalence in External Datasets. A) Differential Expression Heatmap: Row sidebars indicate the overexpressed group (O or OMS) and predicted genomic origin (chromosome, plasmid or virus) and the lowest-level taxonomic assignment. B) Global Prevalence of differentially expressed AMPs were analyzed in 372 independent gut metatranscriptomes (BioProject PRJNA354235).\u003c/p\u003e","description":"","filename":"12.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/c56405b6cf5f6d4e1f48f6f6.jpg"},{"id":87604295,"identity":"a414e90d-ecd5-4d17-8605-2de7275e6c7d","added_by":"auto","created_at":"2025-07-25 17:42:06","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":516284,"visible":true,"origin":"","legend":"\u003cp\u003eSequence conservation, molecular phylogeny, and conserved synteny of plasmid-derived AMP 8200. The colors for functional categories were as follows: translation, ribosomal structure and biogenesis (blue); RNA processing and modification (forest green); transcription (dark red); replication, recombination and repair (orange); chromatin structure and dynamics (purple); cell cycle control, cell division, and chromosome partitioning (golden yellow); nuclear structure (steel blue); defense mechanisms (firebrick red); signal transduction mechanisms (lime green); cell wall/membrane/envelope biogenesis (dark brown); cell motility (cyan); cytoskeleton (royal blue); extracellular structures (slate gray); intracellular trafficking, secretion, and vesicular transport (magenta); posttranslational modification, protein turnover, and chaperones (dark orange); mobilome elements such as prophages and transposons (olive green); energy production and conversion (tomato red); carbohydrate transport and metabolism (turquoise); amino acid transport and metabolism (deep pink); nucleotide transport and metabolism (violet); coenzyme transport and metabolism (sea green); lipid transport and metabolism (khaki); inorganic ion transport and metabolism (cobalt blue); secondary metabolite biosynthesis, transport, and catabolism (indian red); general function prediction only (dark gray); and unknown function (black).\u003c/p\u003e","description":"","filename":"13.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/fe14acc08a0874aef7d98f35.jpg"},{"id":87604297,"identity":"5a2a1be9-3496-4f6c-9370-b798cb736ac1","added_by":"auto","created_at":"2025-07-25 17:42:06","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":579203,"visible":true,"origin":"","legend":"\u003cp\u003eMolecular phylogeny and sequence conservation of phage-derived AMPs. A) AMP 3020, B) AMP 8681, and C) AMP 5245. The colors were represented as detailed in the legend of figure 3.\u003c/p\u003e","description":"","filename":"14.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/6545237cbdde2c051b261b8b.jpg"},{"id":87604293,"identity":"945607fb-4630-4612-b729-ad711cc102fc","added_by":"auto","created_at":"2025-07-25 17:42:06","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":454508,"visible":true,"origin":"","legend":"\u003cp\u003eConserved synteny of phage-derived AMPs. A) AMP 3020, B) AMP 8681, and C) AMP 5245. The colors were represented as detailed in the legend of figure 3.\u003c/p\u003e","description":"","filename":"15.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/1a0629e15d0d6668a5f92605.jpg"},{"id":87604296,"identity":"a68e1902-a6ca-423c-b005-df206c6d6d6e","added_by":"auto","created_at":"2025-07-25 17:42:06","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":316364,"visible":true,"origin":"","legend":"\u003cp\u003eIn-vitro Characterization of Phage-Encoded AMP 3020. A) Amino acid sequences of AMP 3020 (ADR1) and its variants (ADR2). \u003cstrong\u003eAntimicrobial activity over time against B) \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePseudomonas aeruginosa\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e, C) \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eKlebsiella pneumoniae\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e, D) \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eStaphylococcus aureus\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e, and E) \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eStreptococcus pneumoniae\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. Viability analysis of T cell subpopulations after 24 hours of exposure to 20 µg of ADR1 or ADR2 (\u003c/strong\u003eThe bars represent the mean ± standard deviation. Statistical significance was assessed using two-sided t-tests: ns = not significant, p \u0026lt; 0.05, p \u0026lt; 0.01, p \u0026lt; 0.001, p \u0026lt; 0.0001).\u003c/p\u003e","description":"","filename":"16.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/22865be6b28aeaac17b110d1.jpg"},{"id":97178624,"identity":"2fc59c10-c8a3-4843-bb9e-42a9d607b7c4","added_by":"auto","created_at":"2025-12-01 16:11:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3962850,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/5cd25851-4501-48d9-a5ce-5cffdae66245.pdf"},{"id":87604426,"identity":"04bafb79-6434-47fe-907a-90c6c31919fe","added_by":"auto","created_at":"2025-07-25 17:50:06","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":37017,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarytablesmicrobialecology.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/56f817cb3d1849ce959d2a4f.xlsx"},{"id":87604312,"identity":"90253afe-0b05-45d0-ab84-ec373d4b1e04","added_by":"auto","created_at":"2025-07-25 17:42:07","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":9495980,"visible":true,"origin":"","legend":"","description":"","filename":"supplementaryfiguresmicrobialecology.docx","url":"https://assets-eu.researchsquare.com/files/rs-7160447/v1/2f83dbae862bb01abfe01791.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Gut metatranscriptome–virome profiling reveals active antimicrobial peptides (AMPs) encoded in plasmids and phages linked to human diseases","fulltext":[{"header":"Introduction","content":"\u003cp\u003eChildhood obesity has emerged as a critical public health challenge [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], significantly increasing the risk of type 2 diabetes, cardiovascular diseases, and non-alcoholic fatty liver disease in adulthood [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Obesity can also lead to metabolic syndrome [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], characterized by high blood glucose levels, elevated triglycerides, low high-density lipoproteins (HDL) levels, and hypertension [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. This combination severely impacts quality of life and raises the risk of disability and premature death [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. The situation is particularly alarming in Mexico, where nearly one in five school-aged children is obese, many of whom may have undiagnosed metabolic syndrome [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. This highlights the urgent need for effective preventive and therapeutic strategies.\u003c/p\u003e\u003cp\u003eA growing body of evidence suggests that obesity is associated with alterations in gut microbial communities, as demonstrated by 16S rRNA profiling, shotgun metagenomics, and metatranscriptomics [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Recently, viromics highlighted the role of bacteriophages in regulating these communities, which can either maintain a healthy microbe balance (eubiosis) or induce an imbalance (dysbiosis) [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Understanding the interactions between microbes and phages may lead to effective microbiome-targeted strategies for preventing and treating obesity and metabolic syndrome.\u003c/p\u003e\u003cp\u003eBidirectional communication between hosts and their resident microbes is mainly mediated by the \"secrebiome,\" a collection of proteins released into the environment [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. This includes enzymes, toxins, and antimicrobial peptides (AMPs), which are produced by both the host and microbiota as a key innate defense strategy [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Despite their small size, AMPs exhibit diverse antimicrobial activities by destabilizing microbial membranes or binding to essential intracellular targets [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. In bacteria, AMPs can be synthesized through the proteolytic processing of larger precursor proteins [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], non-ribosomal peptide synthetase (NRPS) pathways [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], and dedicated genomic genes [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Recent metagenomic studies have highlighted small open reading frames (smORFs), under 100 codons, as a significant source of novel microbial-derived AMPs [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Understanding the diversity and regulation of these smORF-encoded AMPs is crucial for advancing knowledge of host-microbiome interactions in health and disease.\u003c/p\u003e\u003cp\u003eIn the human gut, host-encoded AMPs act as gatekeepers, regulating opportunistic pathogens and the composition of the resident microbiota [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. They help maintain microbial diversity, which is essential for intestinal health and they are naturally compatible with the host due to their co-evolution within the intestinal ecosystem [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. However, some AMPs may increase vulnerability to viral infections, demonstrating context-dependent effects [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. At the same time, microbial-derived AMPs foster competition among bacteria, allowing AMP-producing species to secure ecological niches and influence microbial dynamics [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Horizontal gene transfer via plasmids and bacteriophages (phages) enhances the functional capabilities of AMPs, aiding bacterial fitness through the dissemination of AMP-encoding genes [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. However, no study has established a link between AMPs and the dysbiosis seen in disorders like obesity and metabolic syndrome.\u003c/p\u003e\u003cp\u003eMetatranscriptomics provides insights into which AMP genes are actively transcribed, highlighting those involved in community self-regulation. This study examines gut metatranscriptomes from healthy, obese, and obese plus metabolic syndrome individuals to catalog differentially expressed AMPs and explore their roles in microbiota dysbiosis. This research is among the first to connect the active AMP repertoire in the human gut to obesity and metabolic syndrome, paving the way for microbiome-targeted therapies.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cb\u003eDe novo assembly and AMP prediction\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe analyzed RNA-seq data from two normal-weight, three obese, and three obese children with Metabolic Syndrome, as detailed in a prior study [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Quality assessment was performed using FastQC (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.bioinformatics.babraham.ac.uk/projects/fastqc/\u003c/span\u003e\u003cspan address=\"https://www.bioinformatics.babraham.ac.uk/projects/fastqc/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), followed by filtering with a Q20 Phred score and trimming using Trimmomatic v0.36. We removed sequencing adapters, ambiguous bases and rRNAs using Ribopicker v0.4.3 and the SILVA rRNA database (138 release). The remaining non-rRNA sequences were aligned to the human genome and transcriptome with Kneaddata to eliminate human derived sequences (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/biobakery/kneaddata\u003c/span\u003e\u003cspan address=\"https://github.com/biobakery/kneaddata\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The remaining sequences were then used as input for the de novo transcriptome assembly with the Trinity assembler, adjusting to conserve small transcripts. After that, the original reads were aligned to the transcriptome using Bowtie2. Expression levels were calculated by Expectation (RSEM). After, we removed the transcripts with an expression below 1 in any sample or with less than three cumulative observations across all samples. Then, we employed TransDecoder (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/TransDecoder/\u003c/span\u003e\u003cspan address=\"https://github.com/TransDecoder/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to identify ORF candidates within protein-coding regions of the transcript sequences, adjusting its parameters to obtain the smORFs (-m 5). The small proteins derived from these smORFs were analyzed for AMP prediction using Macrel (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/BigDataBiology/macrel\u003c/span\u003e\u003cspan address=\"https://github.com/BigDataBiology/macrel\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), AxPEP (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://sourceforge.net/projects/axpep/\u003c/span\u003e\u003cspan address=\"https://sourceforge.net/projects/axpep/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and AMP Scanner V2 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.dveltri.com/ascan/v2/ascan.html\u003c/span\u003e\u003cspan address=\"https://www.dveltri.com/ascan/v2/ascan.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) with default parameters.\u003c/p\u003e\u003cp\u003e\u003cb\u003eTaxonomic classification, phylogeny, genomic synteny and prophage analyses\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe transcript sequences encoding AMPs were used for homology search against NCBI\u0026rsquo;s NT database using the BLASTN algorithm with the following parameters: e-value\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;0.00001, identity\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;80%, and coverage\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;80%. The associated taxonomy was determined with a Last Common Ancestor (LCA) using MEGAN6 Community Edition (v.6.25.10). The genomic context was obtained for all transcripts, using BLASTN best-hit, and the phylogenic trees were created with iTOL (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://itol.embl.de\u003c/span\u003e\u003cspan address=\"https://itol.embl.de\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Plots for genomic context were created with the AnnotationSketch drawing library (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://genometools.org/annotationsketch.html\u003c/span\u003e\u003cspan address=\"https://genometools.org/annotationsketch.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The genomic synteny comparisons were conducted using Easyfig (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mjsull.github.io/Easyfig/\u003c/span\u003e\u003cspan address=\"https://mjsull.github.io/Easyfig/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) with the BLASTN algorithm. We highlighted genomes (accession numbers) containing the AMPs identified in this study in bold in the corresponding figure. Arrows indicate the positions of coding sequences, while shaded lines represent the degree of homology between genomic regions or genomes. We assigned functional categories based on the COG classification system using the COGclassifier tool (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/moshi4/COGclassifier\u003c/span\u003e\u003cspan address=\"https://github.com/moshi4/COGclassifier\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), with each category represented by a distinct color to facilitate interpretation. We screened the bacterial genomes with VirSorter2 to determine whether the AMPs we identified were embedded within prophage regions (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/jiarong/VirSorter2\u003c/span\u003e\u003cspan address=\"https://github.com/jiarong/VirSorter2\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cb\u003eRead mapping of viral DNA to phage genomes\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe quality-filtered phageome reads were obtained from a previously published dataset by our laboratory, which was derived from the same cohort (NCBI BioProject: PRJNA646512). We merged all R1 files into a single R1 file and all R2 files into a single R2 file to create a representative population of the entire virome. Next, we created an index for each phage genome using bowtie2-build with default parameters. We aligned the merged reads to the bacteriophage genomes using Bowtie2 with the parameters \u0026lsquo;\u0026mdash;no-unal --end-to-end --very-sensitive.\u0026rsquo; Finally, we calculated the total genome coverage and the X-fold coverage for each genome.\u003c/p\u003e\n\u003cdiv class=\"Heading\"\u003e\u003cb\u003e16S microbiota profiling\u003c/b\u003e.\u003c/div\u003e\u003cp\u003eWe obtained the V4 region of 16S rRNA gene for the samples from a previously published dataset by our laboratory [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. All quality-filtered sequences were joined and analyzed using QIIME2 (v2024.5). Briefly, raw sequences were imported and dereplicated, followed by de novo clustering at a 97% identity threshold. Taxonomic classification was performed using a Naive Bayes classifier trained on the SILVA 138 reference database. Taxonomy bar plots were generated, and amplicon sequence variants (ASVs) that were present in more than half of the samples (n\u0026thinsp;=\u0026thinsp;4) were retained for Spearman correlation analysis against AMP abundances.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePeptide synthesis and purification\u003c/b\u003e\u003c/p\u003e\u003cp\u003eADR1 and ADR2 were chemically synthesized by a solid-phase method using the Fmoc methodology (GenScript Biotech, Piscataway, NJ) and purified. Briefly, ten milligrams of crude synthetic peptides were dissolved in one milliliter of 20% aqueous acetonitrile solution and were separated each by reverse phase HPLC on an analytic C18 column (Zorbax SB-C18, Agilent, USA). The C18 column was equilibrated in 20% aqueous acetonitrile containing 0.1% TFA, and the synthetic peptides were separated using a linear gradient of acetonitrile/0.1% TFA from 20 to 60% in 40 minutes at a flow rate of 1 mL/min. The presence of each peptide was monitored at 220 nm. The molecular mass of each peptide was determined by mass spectrometry.\u003c/p\u003e\u003cp\u003e\u003cb\u003eAntibacterial assays\u003c/b\u003e\u003c/p\u003e\u003cp\u003e The assessment of bacterial growth was conducted using a broth microdilution assay according to the Clinical and Laboratory Standards Institute guidelines. The reference strains, \u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e (ATCC 27853), \u003cem\u003eKlebsiella pneumoniae\u003c/em\u003e (ATCC700603), \u003cem\u003eStaphylococcus aureus\u003c/em\u003e (ATCC 29213) and \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e (ATCC46916) were cultured in Mueller-Hinton broth (MHB) at 37\u0026ordm;C during overnight incubation. Following this culture period, the samples were diluted in MHB to achieve an endpoint corresponding to an absorbance between 0.08 and 0.13 units at 625 nm, followed by a further dilution of 1:100 in MHB (approximately 1 x 108 CFU/mL). An aliquot of fifty microliters from each bacterial suspension was introduced into each well of a 96-well microtiter culture plate, which contained 50 \u0026micro;L of MHB supplemented with varying concentrations of the synthetic peptides ADR1 and ADR2 (100, 50, 25, 12.5, 6.2, and 3.13 \u0026micro;g/mL). The growth of each bacterial strain was quantitatively evaluated by measuring absorbance at a wavelength of 630 nm after an incubation period of 18 hours at 37\u0026deg;C.\u003c/p\u003e\u003cp\u003e\u003cb\u003eFlow Cytometry Analysis of T Cell Populations\u003c/b\u003e\u003c/p\u003e\u003cp\u003eHuman red blood cells were collected from a healthy donor who gave verbal consent for phlebotomy. This related experiment was approved by the Ethics Committee of the Facultad de Medicina y Ciencias Biom\u0026eacute;dicas of the Universidad Aut\u0026oacute;noma de Chihuahua with registration number CI-068-19. To assess the potential cytotoxic effects of the synthetic peptides on human immune cells, peripheral blood mononuclear cells (PBMCs) were isolated from three healthy donors. The cells were cultured in RPMI 1640 medium (GIBCO, 11875-085, Paisley, SCT, UK) supplemented with 5% fetal bovine serum (FBS) (GIBCO, 26140-079, Grand Island, NE, USA) and GlutaMAX (GIBCO, 35050-061, Grand Island, NE, USA). PBMCs were seeded in 96-well plates and exposed to each AMP peptide (ADR1 and ADR2) at 20 mg/mL. Cells cultured in medium alone were used as untreated controls. The initial evaluation was performed one hour after peptide exposure to determine the immediate effects on T lymphocyte populations. Subsequently, the PBMCs were incubated overnight at 37\u0026deg;C in a humidified atmosphere with 5% CO₂ to assess potential delayed or sustained effects. Following incubation, cells were stained with the BD TriTest\u0026trade; reagent (Becton Dickinson and Company, BD Biosciences, San Jose, CA, USA) to identify CD3⁺ T cells and their CD4⁺ and CD8⁺ subpopulations, the viability of CD3⁺, CD4⁺, and CD8⁺ T cell subpopulations was evaluated using the LIVE/DEAD\u0026trade; Fixable Near-IR Dead Cell Stain Kit (Invitrogen\u0026trade;, Thermo Fisher Scientific, Eugene, OR, USA). The absolute number of cells was acquired using BDTM Liquid Counting Beads (BD Biosciences). Flow cytometric analysis was performed using a FACS Canto flow cytometer (BD Biosciences), and the data were analyzed with FlowJo software (BD Biosciences). Graphical representations and statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA). This experimental design allowed for the evaluation of peptide-induced changes in the distribution and viability of key T lymphocyte subsets, providing insight into the potential immunotoxicity or immunomodulatory properties of the peptides in human immune cells.\u003c/p\u003e"},{"header":"Results","content":"\n\u003ch3\u003e1. Identification and taxonomic classification of AMP-encoding smORFs of the gut microbial transcripts\u003c/h3\u003e\n\u003cp\u003eThe metatranscriptomic data analyzed in this study were obtained from a previously characterized cohort (BioProject PRJNA600247) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], which includes fecal samples from three metabolic phenotypes: two normal-weight children (NW), three with obesity (OB), and three with obesity plus metabolic syndrome (OMS) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). After filtering out low-quality reads and contaminating sequences, we obtained 42.97\u0026nbsp;million high-quality paired reads. Using de novo assembly with Trinity, we generated 51,087 transcripts (N50\u0026thinsp;=\u0026thinsp;1,372 bp) and 36.22\u0026nbsp;million assembled nucleotides (Supplementary Tables\u0026nbsp;1\u0026ndash;2 and Supplementary Fig.\u0026nbsp;1). We recalculated transcript abundance as Fragments per Kilobase Million (FPKM). We applied strict filtering, retaining only transcripts with an FPKM value\u0026thinsp;\u0026ge;\u0026thinsp;1 that were expressed in at least three samples, resulting in 2,522 biologically robust transcripts (Supplementary Table\u0026nbsp;2). Notably, 1,095 of these transcripts encoded smORFs of fewer than 100 amino acids (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe explored the antimicrobial potential of 1,095 smORF-encoded peptides using three bioinformatics tools for AMP prediction: Macrel [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], AxPEP [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], and AMP Scanner V2 [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. This combined screening identified 420 potential AMPs, which we refined based on three criteria: (i) encoded by a unique non-redundant gene situated in a genomic locus that did not overlap with another coding sequences, (ii) comprising more than 10 amino acids, and (iii) supported by at least one homolog transcript in NCBI (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). This process yielded 112 high-confidence AMP candidates, which served as the basis for subsequent analyses (Supplementary Table\u0026nbsp;3). AxPEP identified 77 peptides (68.8%), AMP Scanner v2 identified 15 (13.4%), and Macrel only identified 2 (1.8%). Only three peptides (2.7%) were unanimously predicted by all tools (Supplementary Fig.\u0026nbsp;2A). We benchmarked the accuracy of these AMP prediction tools against a curated reference panel of 1,390 experimentally validated microbe-derived peptides compiled from APD3 [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], dbAMP [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], and DRAMP [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e], confirming 1,332 (95.8%) as bona-fide AMPs. These findings highlight the variability among AMP predictors and underscore the importance of utilizing multiple tools for reliable candidate selection.\u003c/p\u003e\u003cp\u003eThe taxonomic classification of transcripts encoding the 112 AMP determined that 100 transcripts (89.3%) were from bacterial origin, with \u003cem\u003eFaecalibacterium prausnitzii\u003c/em\u003e representing the majority with 48 transcripts (42.9%) (Supplementary Fig.\u0026nbsp;2B). The remaining 12 transcripts (10.7%) were linked to the gut virome, mainly comprising Caudovirales, including Myoviridae and Siphoviridae. This underscores bacteriophages as a source of AMP genes. Notably, one transcript was associated with a plasmid, highlighting the role of mobile genetic elements in the gut ecosystem.\u003c/p\u003e\n\u003ch3\u003e2. Phage, plasmid, and bacteria-derived AMPs were differentially expressed between O and OMS groups.\u003c/h3\u003e\n\u003cp\u003eWe focused our differential expression analysis on the O and OMS groups, each of which had three biological replicates. Using DESeq2, we set a threshold of fold change greater than two and a p-value of less than 0.05 for significance. Out of 112 candidate AMPs, only nine were differentially expressed: three were up-regulated in O and six in OMS (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The taxonomical classification of these nine differentially expressed AMPs revealed a diverse composition, including phage-derived, plasmid-associated, and chromosomally encoded bacterial AMPs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA; Supplementary Table\u0026nbsp;4). The differential expression of these AMPs in O Vs. OMS suggests that AMP dynamics may play a role in the dysbiosis associated with the disease.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003e3. Widespread transcriptomic prevalence of differentially expressed AMPs supports their global distribution.\u003c/h3\u003e\n\u003cp\u003eWe focused on nine AMPs with significant expression shifts between O and OMS groups, highlighting their potential in understanding disease mechanisms. To evaluate their biological relevance, we assessed whether these AMPs are widely distributed across diverse human gut communities or if they are specific to our cohort. Accordingly, we examined a dataset of 372 fecal metatranscriptomes from the U.S.-based Health Professionals Follow-Up Study (HPFS) (Bioproject PRJNA354235) [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. This cohort spans a wide range of BMI values, metabolic phenotypes, ages, and dietary backgrounds, thereby offering a robust framework for assessing AMP prevalence across heterogeneous populations. We mapped HPFS RNA-seq reads to our nine AMP-encoding contigs and considered an AMP present if the coverage exceeded one transcript per million. Notably, all nine AMPs were detected in a significant portion of these samples, with prevalence rates ranging from 9.8\u0026ndash;98.4% (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB; Supplementary Table\u0026nbsp;5). On average, each AMP was found in 67.93% of samples, with a median detection rate of 80.16%. This widespread distribution indicates that these AMPs are conserved elements within the gut microbiome, likely involved in inter-bacterial antagonism, community structuring, and host-microbe signaling.\u003c/p\u003e\n\u003ch3\u003e4. Chromosomally Encoded AMPs Exhibit Conserved Sequence and Genomic Context Across Gut Microbial Taxa\u003c/h3\u003e\n\u003cp\u003eAmong the nine differentially expressed AMPs, five exhibited protein homologs on bacterial chromosomes with phylogenetically diverse hosts. The AMP 3076 showed 100% sequence identity with homolog proteins in \u003cem\u003eEscherichia coli\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3A). The AMP 2526 shared 93.75% identity with four homolog proteins in \u003cem\u003eRomboutsia\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3B), while the AMP 3096 had 93.18% identity with homologs from \u003cem\u003ePhocaeicola\u003c/em\u003e and \u003cem\u003eBacteroides\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3C). The AMP 5865 showed 92% identity with ten homologs of \u003cem\u003eFaecalibacterium\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3D), and the AMP 2198 had 96.76% identity with three homologs in \u003cem\u003eBlautia wexlerae\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3E). The genomic contexts of these AMPs demonstrated high synteny conservation among surrounding genes, implying they are part of conserved genomic modules (Supplementary Fig.\u0026nbsp;4). This conservation in amino acid sequence and genomic architecture indicates strong purifying selection.\u003c/p\u003e\n\u003ch3\u003e5. Plasmid-encoded AMPs contribute to the gut antimicrobial repertoire\u003c/h3\u003e\n\u003cp\u003eAmong the chromosomally encoded AMPs, AMP 8200 stands out due to its dual genomic origin. BLASTP searches identified 15 highly similar protein homologs across \u003cem\u003eBacteroides\u003c/em\u003e and \u003cem\u003ePhocaeicola\u003c/em\u003e, including one copy located on a \u003cem\u003ePhocaeicola dorei\u003c/em\u003e plasmid (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The peptide sequence was 100% conserved in 13 strains, indicating strong purifying selection. Synteny analysis confirmed that the gene order is consistent among host genomes, suggesting a conserved functional cassette (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The presence of AMP 8200 in both chromosome and plasmid suggests a potential horizontal transfer event. Thus, we examined whether the 85 kb plasmid containing the AMP gene was integrated into the chromosome of \u003cem\u003eP. dorei\u003c/em\u003e strain JR01. Whole-genome alignment revealed that only a 13 kb fragment of the plasmid was integrated, including the AMP gene, but not the entire plasmid sequence (Supplementary Fig.\u0026nbsp;5A).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo differentiate between plasmid- and chromosome-derived transcripts, we re-mapped the reads to the 13 kb shared region in each sequence. Alignments to the plasmid sequence showed perfect matches, while the alignment to the 13 kb chromosomal sequence had 17 nucleotide mismatches. This suggests that transcripts mainly originate from the plasmid. Interestingly, when we mapped the RNA-seq reads to the plasmid, we found extensive transcription across the entire plasmid, with 95.5% coverage at an average depth of 61.7-fold. This confirms that the entire plasmid was highly active, underscoring the role of mobile genetic elements in disseminating antimicrobial functions within the gut microbiome.\u003c/p\u003e\n\u003ch3\u003e6. AMPs encoded in active phages suggest the presence of phage-host dynamics\u003c/h3\u003e\n\u003cp\u003eIn addition to the five chromosomally encoded peptides, three AMPs (AMP 3020; AMP 8681, and AMP 5245) show high sequence similarity to proteins from tailed bacteriophages (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). BLASTP analysis demonstrated a perfect match (100% identity) for AMP 3020 with proteins from the Caudoviricetes phages ctJ1L4 and ctzDR1. It also identified two homologs in \u003cem\u003eAnaerotignum\u003c/em\u003e strains, with 70.9\u0026ndash;74.4% identity (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Despite these high sequence similarities, AMP 3020 displayed poor synteny conservation with the surrounding AMP gene in both phages and bacterial genomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe identical sequence between AMP 3020 and the homolog in phage ctJ1L4 suggests that the overexpression of AMP 3020 originated from this phage and not from \u003cem\u003eAnaerotignum\u003c/em\u003e bacteria. Next, we investigated if the ctJ1L4 phage was present as an active viral particle previously reported in this same samples. To this end, we utilized a dataset of DNA-seq reads from viral-like particles, obtained from the same set of samples previously published by our laboratory [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. After mapping the virome reads to the genome of ctJ1L4, we found that 70.46% of the genome was covered by reads, providing strong evidence for the physical presence of this phage as a viral particle (Supplementary Fig.\u0026nbsp;6A). Furthermore, RNA-seq mapping covered 14.7% of the phage genome, showing active transcription of several phage genes (Supplementary Fig.\u0026nbsp;6B). These results support the conclusion that ctJ1L4 was present as a viral particle with active transcription, reinforcing the likelihood that the overexpression of AMP 3020 originates from this phage.\u003c/p\u003e\u003cp\u003eGiven that AMP 3020 also shares 74.36% sequence identity with a protein from \u003cem\u003eAnaerotignum\u003c/em\u003e sp. MB30-C6 genome, we investigated whether this similarity stemmed from the integration of the ctJ1L4 phage genome into the bacterial chromosome. We analyzed the 32.5 kb genome of phage ctJ1L4 and found only a minimal overlap of 372 nucleotides with the bacterial genome, corresponding to the AMP 3020 (Supplementary Fig.\u0026nbsp;6C), which suggests that the phage genome was not integrated into the bacterial chromosome. Additionally, a prophage prediction analysis of the \u003cem\u003eAnaerotignum\u003c/em\u003e MB30-C6 genome did not detect any other prophages containing the AMP, suggesting a non-viral origin of this AMP in the bacterial genome.\u003c/p\u003e\u003cp\u003eThe AMP 8681 showed 100% sequence identity with proteins from two Caudoviricetes phages (ctlN07 and ctUyu4) and one Clostridiales bacterium KR001 hic 0007 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). It also shared 96.97% identity with proteins from phages ctvmC4 and ctlnE22. Despite this high sequence conservation, synteny was poorly conserved in both phage and bacterial genomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Given that this AMP has homologous proteins with 100% identity in both bacteria and phages, we investigated which of these two genomes could be the potential source of AMP overexpression. Mapping virome-derived reads to the ctlN07 phage genome showed low viral presence (7.18% coverage) (Supplementary Fig.\u0026nbsp;7A), while RNA-seq mapping revealed high transcriptional activity (12.82% coverage with 52.65-fold sequence depth). This suggests a high transcriptional activity of this phage (Supplementary Fig.\u0026nbsp;7B). To evaluate possible phage integration, we compared the 50 kb \u003cem\u003ectlN07\u003c/em\u003e genome to the Clostridiales bacterium KR001 hic 0007 genomes. The comparison showed only a 3 kb shared region (98.10% identity) containing the AMP locus, indicating no phage integration in the bacterial genome (Supplementary Fig.\u0026nbsp;7C). Prophage prediction in the bacterial genome also did not identify any match for ctlN07, suggesting it is not lysogenized in the bacterium. Additionally, we did not detect any other prophages containing the AMP, suggesting a non-viral origin of this AMP in the bacterial genome. Overall, this indicates that AMP 8681 overexpression may originate from either the phage or bacterial genome, warranting further functional assays for clarification.\u003c/p\u003e\u003cp\u003eThe AMP 5245 exhibits a 100% amino acid sequence identity with five and three homolog proteins found in various phages and \u003cem\u003eBlautia wexlerae\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). The AMP showed significant conservation with two genomic structures among phages and bacterial genomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Given its high sequence similarity with both bacterial and phage proteins, we investigated the source of the observed AMP overexpression. Virome reads covered 88.7% of the phage genome, and RNA-seq reads accounted for 86.7% at 1,777X depth, indicating the presence of an active viral particle (Supplementary Fig.\u0026nbsp;8).\u003c/p\u003e\u003cp\u003eTo ensure the phage was not integrated into a bacterial genome, we compared the 8 kb human fecal virus clone to \u003cem\u003eB. wexlerae\u003c/em\u003e DSM 19850, revealing only a 3 kb region with 98.35% identity, corresponding to 39.53% of the phage genome and encompassing the AMP locus (Supplementary Fig.\u0026nbsp;8C), suggesting a partial integration of phage into the bacterial genome. Prophage prediction analysis of the \u003cem\u003eB. wexlerae\u003c/em\u003e genome did not identify any prophage elements matching the human fecal virus clone, suggesting that the phage was not integrated. Additionally, we did not detect any other prophages containing the AMP. The high coverage of the phage genome with DNA and RNA reads supports the notion that the phage was the primary source of AMP 5245 transcripts.\u003c/p\u003e\n\u003ch3\u003e7. 16S–Metatranscriptome correlation analysis links AMPs expression with bacterial taxa\u003c/h3\u003e\n\u003cp\u003eTo explore associations between the nine overexpressed AMPs and gut microbiota, we correlated their expression with 16S rRNA data (BioProject PRJNA600247; Supplementary Fig.\u0026nbsp;9). We observed significant correlations, focusing on the negative ones, which indicate that higher AMP expression corresponds with a decreased abundance of specific microbial taxa. AMP 5865 was inversely linked to \u003cem\u003eAnaeroplasma\u003c/em\u003e, \u003cem\u003eClostridium\u003c/em\u003e, and \u003cem\u003eBacteroides\u003c/em\u003e species. The AMP 5245 negatively impacted over 20 taxa, particularly beneficial groups such as \u003cem\u003eAkkermansia\u003c/em\u003e, Christensenellaceae, \u003cem\u003eMoryella\u003c/em\u003e, and \u003cem\u003eOscillibacter\u003c/em\u003e. AMP 3076 showed an association with reduced abundances of Christensenellaceae, \u003cem\u003eBilophila\u003c/em\u003e, and several Lachnospiraceae lineages, while AMP 3096 was linked to decreased Desulfovibrionaceae, \u003cem\u003eBilophila wadsworthia\u003c/em\u003e, \u003cem\u003eOscillibacter\u003c/em\u003e, and \u003cem\u003eA. muciniphila\u003c/em\u003e. Finally, AMP 2526 and AMP 2198 showed a negative correlation with \u003cem\u003eA. muciniphila\u003c/em\u003e and \u003cem\u003eEubacterium\u003c/em\u003e species, bacteria typically associated with gut health. Together, these correlations suggest that the obesity-associated overexpression of AMPs may contribute to dysbiosis in obesity and metabolic syndrome by reducing the abundance of key commensal bacteria.\u003c/p\u003e\n\u003ch3\u003e8. Experimental Validation of Phage-Encoded AMP 3020 Reveals Antibacterial Activity without T-Cell Toxicity\u003c/h3\u003e\n\u003cp\u003eTo verify our in-silico predictions, we chose the AMP 3020 for experimental validation because of its confirmed phage origin. We synthesized two variants of AMP 3020 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA): ADR1 retains the native N-terminal methionine, whereas ADR2 lacks this residue. ADR2 starts with valine, as it has been reported that some peptides undergo post-translational processing when the second residue of a nascent peptide is a short one; here, the second residue was valine [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. This single amino acid difference also allowed us to test whether minimal sequence variation influences their functional activity. Both peptides significantly suppressed the growth of gram-negative (\u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e and \u003cem\u003eKlebsiella pneumoniae\u003c/em\u003e) and gram-positive (\u003cem\u003eStaphylococcus aureus\u003c/em\u003e and \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e) bacteria compared to controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB\u0026ndash;E). ADR1 was more effective against \u003cem\u003eP. aeruginosa\u003c/em\u003e, whereas ADR2 showed greater efficacy against \u003cem\u003eK. pneumoniae\u003c/em\u003e. Both variants exhibited antibacterial activity against \u003cem\u003eS. pneumoniae\u003c/em\u003e, with only ADR1 effective against \u003cem\u003eS. aureus\u003c/em\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNotably, neither peptide affected T-lymphocyte viability (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF), with similar cell death frequencies to those of untreated controls, ranging from 0.7\u0026ndash;14.4% (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF). The absolute cell counts (Supplementary Figs.\u0026nbsp;10 and 11) confirmed no significant loss of viable T cells, while the positive control (PMA/ionomycin) showed a strong cytotoxic response. Overall, ADR1 and ADR2 demonstrated low cytotoxicity toward primary human T lymphocyte subsets. The removal of the initiator methionine may fine-tune target specificity due to changes in peptide folding or charge distribution.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study provides a framework for discovering and characterizing AMPs in the human gut microbiome using metatranscriptomics and viromics, representing a promising frontier to understand their role in host-microbiota interactions [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. By integrating data from three AMP prediction algorithms \u0026mdash; Macrell, AMP Scanner V2, and AxPEP \u0026mdash; we identified 112 high-confidence AMP candidates from 1,095 expressed small open reading frames (smORFs). The limited overlap among the results of the different prediction tools highlights the methodological variability in AMP discovery. Notably, many expressed AMPs were linked to \u003cem\u003eFaecalibacterium prausnitzii\u003c/em\u003e, a known gut commensal bacterium. Additionally, the identification of AMP-encoding transcripts from plasmids and phages expands the potential origins of AMPs beyond traditional chromosomal sources.\u003c/p\u003e\u003cp\u003eThe epithelial interface contains AMPs from host and microbial sources, which play a crucial role in shaping the surrounding microbiota [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Emerging evidence suggests that bioactive peptides can alter the gut microbiota [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. This suggests the potential of AMPs to be associated with microbiota changes related to diseases, such as obesity and metabolic syndrome [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Both variants of phage-derived AMP 3020 (ADR1 and ADR2) demonstrated selective antibacterial activity while remaining non-cytotoxic to primary human T-cell subsets (CD3+, CD4+, and CD8+). Our synthetically phage-derived AMP 3020 variants showed antibacterial activity without being cytotoxic to human T-cell subsets (CD3+, CD4+, and CD8+). These results support their classification within the broader family of host defense peptides (HDPs), which also include bacteriophage-encoded peptides that influence inter-microbial competition and host-microbiota interactions [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e], [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Notably, the lack of T cell cytotoxicity emphasizes the immunological neutrality of these peptides, suggesting that they can be produced by the microbiota to cause an effect on other bacteria without affecting the host These findings enhance our understanding of the gut virome and suggest that phage-derived AMPs can influence microbial communities while preserving host immune balance [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Additionally, the obesity-related upregulation of several AMPs suggests their potential role in gut dysbiosis linked to obesity, as their expression negatively correlated with beneficial taxa like \u003cem\u003eAkkermansia muciniphila\u003c/em\u003e, Christensenellaceae, and Desulfovibrionaceae, suggesting AMP-driven microbiota changes in disease.\u003c/p\u003e\u003cp\u003eThe differential expression analysis of AMPs in individuals with obesity compared to those with obesity and metabolic syndrome identified nine significantly overexpressed AMPs, linked to chromosomes, plasmids, and phages. This supports the idea that the host influences microbial gene expression of AMP production. Importantly, these nine AMPs were also expressed in 372 samples of an independent gut metatranscriptome dataset, indicating their commonality and ecological significance in the human gut microbiota, rather than being unique to our cohort. Their widespread presence suggests they are core features of the gut metatranscriptome and potential reliable biomarkers for disease.\u003c/p\u003e\u003cp\u003eWe identified three AMPs with homologs in bacteriophages, an underexplored source of antimicrobial compounds. The AMP 3020 was 98% identical to a protein from the Caudoviricetes phage ctJ1L4. Virome and metatranscriptome read mapping confirmed the presence of ctJ1L4 virions in our samples and demonstrated that several of their genes, including the AMP, were actively transcribed. Synthetic AMP 3020 inhibits the growth of both Gram-positive and Gram-negative bacteria, underscoring its potential as a broad-spectrum antimicrobial. Despite the presence of related proteins in two bacterial genomes, no evidence indicated ctJ1L4 integration or prophages with the AMP gene, suggesting that in vivo AMP production was primarily due to lytic phages. Removing the N-terminal Methionine from AMP 3020 changed its antibacterial activity, reflecting the evolutionary adaptability of phage-derived peptides. Methionine is essential for stabilizing protein structures and may also act as a regulatory switch through reversible redox reactions [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Overall, this data expands our understanding of the functional capabilities of gut bacteriophages, particularly those within the dominant Caudovirales lineage found in the human gut virome [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. It emphasizes the importance of phages as a rich yet under-recognized source of AMPs that can change the gut microbiota.\u003c/p\u003e\u003cp\u003eAMPs 8681 and 5245 show 100% amino acid identity between their phage- and bacterium-encoded homologs, with virome read mapping confirming their presence as free viral particles. Additionally, metatranscriptome data indicated viral transcription, suggesting that the phage was a significant contributor of AMP transcripts, although bacterial loci may also play a role. Prophage scans show that both chromosomal AMP loci were located outside predicted prophage regions, suggesting that the integration of both AMPs into the bacterial genomes was not related to a prophage. Interestingly, the expression of AMP 5245 negatively correlated with Moryella and Eubacterium species, known butyrate producers with anti-inflammatory effects, implying that this AMP may interfere with their beneficial effects [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e], [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eWe also detected a plasmid containing a homolog protein to the AMP 8200, which was a described plasmid of \u003cem\u003ePhocaeicola dorei\u003c/em\u003e [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. The peptide exhibits 100% sequence identity and conserved synteny across additional \u003cem\u003ePhocaeicola\u003c/em\u003e species. RNA-seq read profiling reveals robust transcription across the entire plasmid, suggesting that AMP expression likely originated from the plasmid, supporting the role of plasmids as reservoirs of antimicrobial capabilities in the gut microbiome [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Our findings highlight the potential for plasmids and phages to spread competitive traits like AMPs via horizontal transfer.\u003c/p\u003e\u003cp\u003eAMPs 3076, 2526, 3096, 5865, and 2198 each share homolog proteins with a broad spectrum of gut bacterial commensals, including \u003cem\u003eEscherichia coli\u003c/em\u003e, \u003cem\u003eFaecalibacterium prausnitzii\u003c/em\u003e, \u003cem\u003eRomboutsia\u003c/em\u003e spp., and \u003cem\u003eBlautia wexlerae\u003c/em\u003e, indicating their widespread distribution in the intestinal microbiota. These AMPs demonstrate high sequence and synteny conservation across bacterial genomes. Despite their widespread bacterial distribution, each AMP has a unique physicochemical signature and genomic context, suggesting specialized functional roles. The negative correlations between these AMP expressions and beneficial microbes, such as \u003cem\u003eAkkermansia muciniphila and Christensenellaceae\u003c/em\u003e, imply a potential link to microbial dysbiosis observed in obesity and metabolic syndrome [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e], [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. For example, AMP 5865 was overexpressed in obesity and showed significant negative correlations with \u003cem\u003eA. muciniphila\u003c/em\u003e and \u003cem\u003eAlistipes obesii\u003c/em\u003e, two organisms consistently associated with favorable metabolic outcomes [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. This suggests that this AMP could be related to the low abundance of these two taxas observed in the metabolic syndrome cohort.\u003c/p\u003e\u003cp\u003eMetatranscriptomic profiling provides a dynamic map of microbiome activity, revealing which genes are actively expressed under conditions such as obesity and metabolic syndrome. This approach differs from traditional metagenomics, which only lists gene presence. Our analysis identified actively expressed AMPs, helping differentiate between latent genetic potential and those that actively influence host microbiota and disease associations. Our findings highlight that gut-expressed AMPs were derived from diverse genomic sources, including bacteria, plasmids, and phages, indicating their significant ecological roles in gut microbial dynamics. This pilot study shows that metatranscriptomic data can uncover relevant, expressed AMPs implicated in the gut microbiome regulation. The differential expression of these AMPs, linked to disease, along with their antimicrobial properties that do not affect host immunity points to their importance in shaping the gut microbiota. Moreover, the presence of mobile genetic elements, such as plasmids and phages, reinforces the need to rethink the role of bacteriophages in gut ecology, acting not only as predators but also as regulators. This work sets the stage for future studies on AMPs and their therapeutic potential in microbiome-targeted treatments for dysbiosis related to obesity.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eL.G.B. thanks to the Doctoral Biochemical Sciences Program at IBt UNAM and CONACyT for doctoral fellowship C.V.U.: 887285. Also, F.C.G. would thank the \u003cem\u003eEstancias posdoctorales por M\u0026eacute;xico 2022\u003c/em\u003e program (C.V.U.: 443238). This research was funded by \u003cem\u003eCONACyT grant Ciencia de Frontera-2019-263986\u0026nbsp;\u003c/em\u003eand by \u003cem\u003eDGAPA PAPIIT UNAM (IN219723)\u003c/em\u003e. We thank M.T.I Juan Manuel Hurtado Ram\u0026iacute;rez for informatics technical support. Also, the authors would like to thank the \u0026quot;Unidad Universitaria de Secuenciaci\u0026oacute;n Masiva y Bioinform\u0026aacute;tica\u0026quot; of the \u0026quot;Laboratorio Nacional de Apoyo Tecnol\u0026oacute;gico a las Ciencias Gen\u0026oacute;micas,\u0026quot; UNAM, especially to Ricardo Alfredo Grande Cano and\u0026nbsp;Lizeth A. Mat\u0026iacute;as Valdez for the technical sequencing support.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by \u003cem\u003eCONACyT grant Ciencia de Frontera-2019-263986 and by DGAPA PAPIIT UNAM (IN219723)\u003c/em\u003e. L.G.B. was supported by the Doctoral Biochemical Sciences Program at IBt UNAM and CONACyT with the doctoral fellowship C.V.U.: 778192. F.C.G. was supported by the \u003cem\u003eEstancias posdoctorales por M\u0026eacute;xico 2022 program (C.V.U.: 443238)\u003c/em\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclosure statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are openly available in NCBI GEO repository with accession number GSE143207 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143207) and the NCBI BioProject: PRJNA600247 (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA600247) and PRJNA646512 (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA646512). All the code used for this project was deposited in this GitHub repository: https://github.com/LuiguiGallardo/amps_microbiome. Requests for additional material should be made to the corresponding author.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceived of or designed study: LGB, FCG, SB, and AOL; Analyzed Data: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL; Formal analysis: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL; Contributed new methods or models: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL; Wrote the paper: LGB, FCG, AA, GLL, CAG, FS, GC, GPE, SCQ, and AOL. Funding acquisition: FCG, SCQ, AOL. All authors read and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBiro FM, Wien M (May 2010) Childhood obesity and adult morbidities. Am J Clin Nutr 91(5). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3945/ajcn.2010.28701B\u003c/span\u003e\u003cspan address=\"10.3945/ajcn.2010.28701B\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. pp. 1499S-1505S\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFranks PW, Hanson RL, Knowler WC, Sievers ML, Bennett PH, Looker HC (2010) Childhood Obesity, Other Cardiovascular Risk Factors, and Premature Death, \u003cem\u003eN. Engl. J. Med.\u003c/em\u003e, vol. 362, no. 6, pp. 485\u0026ndash;493, Feb. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/NEJMoa0904130\u003c/span\u003e\u003cspan address=\"10.1056/NEJMoa0904130\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerichart-Perera O, Balas-Nakash M, Schiffman-Selechnik E, Barbato-Dosal A, Vadillo-Ortega F (2007) Obesity Increases Metabolic Syndrome Risk Factors in School-Aged Children from an Urban School in Mexico City, \u003cem\u003eJ. Am. Diet. Assoc.\u003c/em\u003e, vol. 107, no. 1, pp. 81\u0026ndash;91, Jan. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jada.2006.10.011\u003c/span\u003e\u003cspan address=\"10.1016/j.jada.2006.10.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl-Hamad D, Raman V (2017) Metabolic syndrome in children and adolescents, \u003cem\u003eTransl. Pediatr.\u003c/em\u003e, vol. 6, no. 4, Art. no. 4, Oct. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21037/tp.2017.10.02\u003c/span\u003e\u003cspan address=\"10.21037/tp.2017.10.02\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ede Ferranti SD, Gauvreau K, Ludwig DS, Neufeld EJ, Newburger JW, Rifai N (Oct. 2004) Prevalence of the Metabolic Syndrome in American Adolescents. Circulation 110(16):2494\u0026ndash;2497. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/01.CIR.0000145117.40114.C7\u003c/span\u003e\u003cspan address=\"10.1161/01.CIR.0000145117.40114.C7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEvia-Viscarra ML, Rodea-Montero ER, Apolinar-Jim\u0026eacute;nez E, Quintana-Vargas S (2013) Metabolic syndrome and its components among obese (BMI\u0026thinsp;\u0026ge;\u0026thinsp;95th) Mexican adolescents, \u003cem\u003eEndocr. Connect.\u003c/em\u003e, vol. 2, no. 4, pp. 208\u0026ndash;215, Dec. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1530/EC-13-0057\u003c/span\u003e\u003cspan address=\"10.1530/EC-13-0057\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShamah-Levy T et al (2023) Jun., Prevalencias de sobrepeso y obesidad en poblaci\u0026oacute;n escolar y adolescente de M\u0026eacute;xico. Ensanut Continua 2020\u0026ndash;2022, \u003cem\u003eSalud P\u0026uacute;blica M\u0026eacute;xico\u003c/em\u003e, vol. 65, pp. s218\u0026ndash;s224. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21149/14762\u003c/span\u003e\u003cspan address=\"10.21149/14762\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGallardo-Becerra L et al (Dec. 2020) Metatranscriptomic analysis to define the Secrebiome, and 16S rRNA profiling of the gut microbiome in obesity and metabolic syndrome of Mexican children. Microb Cell Factories 19(1):61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12934-020-01319-y\u003c/span\u003e\u003cspan address=\"10.1186/s12934-020-01319-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBikel S et al (2021) Aug., Gut dsDNA virome shows diversity and richness alterations associated with childhood obesity and metabolic syndrome, \u003cem\u003eiScience\u003c/em\u003e, vol. 24, no. 8, p. 102900. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.isci.2021.102900\u003c/span\u003e\u003cspan address=\"10.1016/j.isci.2021.102900\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGallardo-Becerra L, Cervantes-Echeverr\u0026iacute;a M, Cornejo-Granados F, Vazquez-Morado LE, Ochoa-Leyva A (Dec. 2023) Perspectives in Searching Antimicrobial Peptides (AMPs) Produced by the Microbiota. Microb Ecol 87(1):8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00248-023-02313-8\u003c/span\u003e\u003cspan address=\"10.1007/s00248-023-02313-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIzadpanah A, Gallo RL (Mar. 2005) Antimicrobial peptides. J Am Acad Dermatol 52(3):381\u0026ndash;390. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaad.2004.08.026\u003c/span\u003e\u003cspan address=\"10.1016/j.jaad.2004.08.026\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAgeitos JM, S\u0026aacute;nchez-P\u0026eacute;rez A, Calo-Mata P, Villa TG (Jun. 2017) Antimicrobial peptides (AMPs): Ancient compounds that represent novel weapons in the fight against bacteria. Biochem Pharmacol 133:117\u0026ndash;138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.bcp.2016.09.018\u003c/span\u003e\u003cspan address=\"10.1016/j.bcp.2016.09.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTorres MDT, Melo MCR, Flowers L, Crescenzi O, Notomista E, De La Fuente-Nunez C (2021) Mining for encrypted peptide antibiotics in the human proteome, \u003cem\u003eNat. Biomed. Eng.\u003c/em\u003e, vol. 6, no. 1, pp. 67\u0026ndash;75, Nov. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41551-021-00801-1\u003c/span\u003e\u003cspan address=\"10.1038/s41551-021-00801-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePizzo E, Cafaro V, Di Donato A, Notomista E (May 2018) Cryptic Antimicrobial Peptides: Identification Methods and Current Knowledge of their Immunomodulatory Properties. Curr Pharm Des 24(10):1054\u0026ndash;1066. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2174/1381612824666180327165012\u003c/span\u003e\u003cspan address=\"10.2174/1381612824666180327165012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNolan EM, Walsh CT (Jan. 2009) How Nature Morphs Peptide Scaffolds into Antibiotics. ChemBioChem 10(1):34\u0026ndash;53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/cbic.200800438\u003c/span\u003e\u003cspan address=\"10.1002/cbic.200800438\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSingh N, Abraham J (Apr. 2014) Ribosomally synthesized peptides from natural sources. J Antibiot (Tokyo) 67(4):277\u0026ndash;289. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ja.2013.138\u003c/span\u003e\u003cspan address=\"10.1038/ja.2013.138\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTorres MDT et al (Aug. 2024) Mining human microbiomes reveals an untapped source of peptide antibiotics. Cell 0(0). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2024.07.027\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2024.07.027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSantos-J\u0026uacute;nior CD et al (Jun. 2024) Discovery of antimicrobial peptides in the global microbiome with machine learning. Cell 0(0). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2024.05.013\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2024.05.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNakatsuji T et al (Feb. 2017) Antimicrobials from human skin commensal bacteria protect against \u003cem\u003eStaphylococcus aureus\u003c/em\u003e and are deficient in atopic dermatitis. Sci Transl Med 9(378):eaah4680. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/scitranslmed.aah4680\u003c/span\u003e\u003cspan address=\"10.1126/scitranslmed.aah4680\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMa Y et al (2022) Jun., Identification of antimicrobial peptides from the human gut microbiome using deep learning, \u003cem\u003eNat. Biotechnol.\u003c/em\u003e, vol. 40, no. 6, Art. no. 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41587-022-01226-0\u003c/span\u003e\u003cspan address=\"10.1038/s41587-022-01226-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun D et al (2021) Apr., Angiogenin maintains gut microbe homeostasis by balancing α-Proteobacteria and Lachnospiraceae, \u003cem\u003eGut\u003c/em\u003e, vol. 70, no. 4, pp. 666\u0026ndash;676. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/gutjnl-2019-320135\u003c/span\u003e\u003cspan address=\"10.1136/gutjnl-2019-320135\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez-D\u0026aacute;vila P et al (Aug. 2022) Catestatin selects for colonization of antimicrobial-resistant gut bacterial communities. ISME J 16(8):1873\u0026ndash;1882. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41396-022-01240-9\u003c/span\u003e\u003cspan address=\"10.1038/s41396-022-01240-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMagana M et al (Sep. 2020) The value of antimicrobial peptides in the age of resistance. Lancet Infect Dis 20(9):e216\u0026ndash;e230. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S1473-3099(20)30327-3\u003c/span\u003e\u003cspan address=\"10.1016/S1473-3099(20)30327-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng Y, Yang S, Zhang L, Chen C, Cheng X, Hou C (Jan. 2024) Chronic bee paralysis virus exploits host antimicrobial peptides and alters gut microbiota composition to facilitate viral infection. ISME J 18(1):wrae051. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/ismejo/wrae051\u003c/span\u003e\u003cspan address=\"10.1093/ismejo/wrae051\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarcia-Gutierrez E, Mayer MJ, Cotter PD, Narbad A (Jan. 2019) Gut microbiota as a source of novel antimicrobials. Gut Microbes 10(1):1\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/19490976.2018.1455790\u003c/span\u003e\u003cspan address=\"10.1080/19490976.2018.1455790\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKommineni S et al (2015) Oct., Bacteriocin production augments niche competition by enterococci in the mammalian gastrointestinal tract, \u003cem\u003eNature\u003c/em\u003e, vol. 526, no. 7575, pp. 719\u0026ndash;722. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature15524\u003c/span\u003e\u003cspan address=\"10.1038/nature15524\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNakazono K et al (Jan. 2022) Complete sequences of epidermin and nukacin encoding plasmids from oral-derived Staphylococcus epidermidis and their antibacterial activity. PLoS ONE 17(1):e0258283. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0258283\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0258283\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDu J et al (Jul. 2023) Enhancing bacteriophage therapeutics through in situ production and release of heterologous antimicrobial effectors. Nat Commun 14(1):4337. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-023-39612-0\u003c/span\u003e\u003cspan address=\"10.1038/s41467-023-39612-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTyagi JL, Gupta P, Ghate MM, Kumar D, Poluri KM (May 2024) Assessing the synergistic potential of bacteriophage endolysins and antimicrobial peptides for eradicating bacterial biofilms. Arch Microbiol 206(6):272. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00203-024-04003-6\u003c/span\u003e\u003cspan address=\"10.1007/s00203-024-04003-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang G, Li X, Wang Z (Jan. 2016) APD3: the antimicrobial peptide database as a tool for research and education. Nucleic Acids Res 44:D1087\u0026ndash;D1093. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkv1278\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkv1278\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYao L et al (2025) Jan., dbAMP 3.0: updated resource of antimicrobial activity and structural annotation of peptides in the post-pandemic era, \u003cem\u003eNucleic Acids Res.\u003c/em\u003e, vol. 53, no. D1, pp. D364\u0026ndash;D376. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkae1019\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkae1019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMa T et al (Jan. 2025) DRAMP 4.0: an open-access data repository dedicated to the clinical translation of antimicrobial peptides. Nucleic Acids Res 53:D403\u0026ndash;D410. no. D110.1093/nar/gkae1046\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbu-Ali GS et al (Jan. 2018) Metatranscriptome of human faecal microbial communities in a cohort of adult men. Nat Microbiol 3(3):356\u0026ndash;366. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41564-017-0084-4\u003c/span\u003e\u003cspan address=\"10.1038/s41564-017-0084-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBen-Bassat A, Bauer K (1987) Amino-terminal processing of proteins, \u003cem\u003eNature\u003c/em\u003e, vol. 326, no. 6110, pp. 315\u0026ndash;315, Mar. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/326315a0\u003c/span\u003e\u003cspan address=\"10.1038/326315a0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZong X, Fu J, Xu B, Wang Y, Jin M (Dec. 2020) Interplay between gut microbiota and antimicrobial peptides. Anim Nutr 6(4):389\u0026ndash;396. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.aninu.2020.09.002\u003c/span\u003e\u003cspan address=\"10.1016/j.aninu.2020.09.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu F et al (May 2016) GLP-1 receptor agonist promotes brown remodelling in mouse white adipose tissue through SIRT1. Diabetologia 59(5):1059\u0026ndash;1069. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00125-016-3896-5\u003c/span\u003e\u003cspan address=\"10.1007/s00125-016-3896-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaney EF, Straus SK, Hancock REW (Feb. 2019) Reassessing the Host Defense Peptide Landscape. Front Chem 7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fchem.2019.00043\u003c/span\u003e\u003cspan address=\"10.3389/fchem.2019.00043\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeng S-Y, You R-I, Lai M-J, Lin N-T, Chen L-K, Chang K-C (2017) Highly potent antimicrobial modified peptides derived from the Acinetobacter baumannii phage endolysin LysAB2, \u003cem\u003eSci. Rep.\u003c/em\u003e, vol. 7, no. 1, Art. no. 1, Sep. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-017-11832-7\u003c/span\u003e\u003cspan address=\"10.1038/s41598-017-11832-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRitz NL et al (Feb. 2024) The gut virome is associated with stress-induced changes in behaviour and immune responses in mice. Nat Microbiol 9(2):359\u0026ndash;376. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41564-023-01564-y\u003c/span\u003e\u003cspan address=\"10.1038/s41564-023-01564-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWei Y (Sep. 2024) Bacteriophages: a double-edged sword in the gastrointestinal tract. Front Microbiomes 3:1450523. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/frmbi.2024.1450523\u003c/span\u003e\u003cspan address=\"10.3389/frmbi.2024.1450523\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAledo JC (2019) Methionine in proteins: The Cinderella of the proteinogenic amino acids, \u003cem\u003eProtein Sci. Publ. Protein Soc.\u003c/em\u003e, vol. 28, no. 10, pp. 1785\u0026ndash;1796, Oct. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/pro.3698\u003c/span\u003e\u003cspan address=\"10.1002/pro.3698\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu H et al (2022) Aug., Butyrate-producing Eubacterium rectale suppresses lymphomagenesis by alleviating the TNF-induced TLR4/MyD88/NF-κB axis, \u003cem\u003eCell Host Microbe\u003c/em\u003e, vol. 30, no. 8, pp. 1139\u0026ndash;1150.e7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.chom.2022.07.003\u003c/span\u003e\u003cspan address=\"10.1016/j.chom.2022.07.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSung MM et al (Nov. 2016) Improved Glucose Homeostasis in Obese Mice Treated With Resveratrol Is Associated With Alterations in the Gut Microbiome. Diabetes 66(2):418\u0026ndash;425. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2337/db16-0680\u003c/span\u003e\u003cspan address=\"10.2337/db16-0680\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKijner S, Ennis D, Shmorak S, Florentin A, Yassour M (Jan. 2024) CRISPR-Cas-based identification of a sialylated human milk oligosaccharides utilization cluster in the infant gut commensal Bacteroides dorei. Nat Commun 15(1):105. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-023-44437-y\u003c/span\u003e\u003cspan address=\"10.1038/s41467-023-44437-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFern\u0026aacute;ndez-Pato A et al (Jun. 2025) Early-life development of the gut virome and plasmidome: A longitudinal study in cesarean-born infants. Cell Rep 44(6). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.celrep.2025.115731\u003c/span\u003e\u003cspan address=\"10.1016/j.celrep.2025.115731\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWaters JL, Ley RE (Oct. 2019) The human gut bacteria Christensenellaceae are widespread, heritable, and associated with health. BMC Biol 17:83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12915-019-0699-4\u003c/span\u003e\u003cspan address=\"10.1186/s12915-019-0699-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEverard A et al (2013) May., Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity, \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e, vol. 110, no. 22, pp. 9066\u0026ndash;9071. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.1219451110\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1219451110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJie Z et al (2021) May., The Baseline Gut Microbiota Directs Dieting-Induced Weight Loss Trajectories, \u003cem\u003eGastroenterology\u003c/em\u003e, vol. 160, no. 6, pp. 2029\u0026ndash;2042.e16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2021.01.029\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2021.01.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"microbial-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"meco","sideBox":"Learn more about [Microbial Ecology](https://www.springer.com/journal/248)","snPcode":"248","submissionUrl":"https://submission.nature.com/new-submission/248/3","title":"Microbial Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Antimicrobial peptides, AMPs, phages, metatranscriptome, virome, plasmid","lastPublishedDoi":"10.21203/rs.3.rs-7160447/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7160447/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMicrobe-derived antimicrobial peptides (AMPs) play a crucial role in shaping the microbiota composition; however, their contribution to disease-associated dysbiosis remains poorly understood. Here, we assembled fecal metatranscriptomes from individuals with normal weight, obesity, and obesity plus metabolic syndrome, yielding 51,087 non-human transcripts. We screened 1,095 small open reading frames (smORFs) using AMP-prediction algorithms and identified 112 AMP candidates. Most of them were associated with bacterial homologs, predominantly \u003cem\u003eFaecalibacterium prausnitzii\u003c/em\u003e, while twelve aligned with plasmid or bacteriophage sequences. Differential expression analysis identified nine AMPs that were overexpressed among our groups, of which five originated from chromosomes, one from a plasmid, and three from phages. The expression of these AMPs was inversely correlated with specific bacterial taxa, linking them to disease-associated shifts in microbiota. Additionally, we also examined the presence of these nine AMPs in 372 external gut metatranscriptomes, discovering that they were highly prevalent in up to 98% of the samples, suggesting their conservation within the human gut microbiome and highlighting mobile elements as an often-overlooked reservoir of active AMPs. Finally, through virome sequencing and prophage genome analyses, we suggest that mobile-derived AMPs were transcribed from phage particles. We synthesized a phage-encoded AMP and demonstrated its broad-spectrum antibacterial activity against Gram-positive and Gram-negative bacteria, with no detectable cytotoxicity toward human immune cells. These findings illustrate that the human gut harbors a conserved set of microbe-derived AMPs associated with mobile genetic elements, whose overexpression was linked to obesity and metabolic syndrome, underscoring their role as ecological regulators of the microbiota in disease.\u003c/p\u003e","manuscriptTitle":"Gut metatranscriptome–virome profiling reveals active antimicrobial peptides (AMPs) encoded in plasmids and phages linked to human diseases","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-25 17:42:02","doi":"10.21203/rs.3.rs-7160447/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-25T09:20:08+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-25T04:11:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-22T07:46:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"958376719817442170962041356465362318","date":"2025-08-06T18:22:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"335549207443100930614670917985974608876","date":"2025-07-28T07:56:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-23T01:45:44+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-22T13:49:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-22T13:47:31+00:00","index":"","fulltext":""},{"type":"submitted","content":"Microbial Ecology","date":"2025-07-18T20:06:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"microbial-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"meco","sideBox":"Learn more about [Microbial Ecology](https://www.springer.com/journal/248)","snPcode":"248","submissionUrl":"https://submission.nature.com/new-submission/248/3","title":"Microbial Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"31333d69-2559-4a74-b295-b30bcfa38429","owner":[],"postedDate":"July 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-01T16:05:08+00:00","versionOfRecord":{"articleIdentity":"rs-7160447","link":"https://doi.org/10.1007/s00248-025-02620-2","journal":{"identity":"microbial-ecology","isVorOnly":false,"title":"Microbial Ecology"},"publishedOn":"2025-11-28 15:58:36","publishedOnDateReadable":"November 28th, 2025"},"versionCreatedAt":"2025-07-25 17:42:02","video":"","vorDoi":"10.1007/s00248-025-02620-2","vorDoiUrl":"https://doi.org/10.1007/s00248-025-02620-2","workflowStages":[]},"version":"v1","identity":"rs-7160447","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7160447","identity":"rs-7160447","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}