Host Clustering of Campylobacter Species and Other Enteric Pathogens in a Longitudinal Cohort of Infants, Family Members and Livestock in Rural Eastern Ethiopia

Host Clustering of Campylobacter Species and Other Enteric Pathogens in a Longitudinal Cohort of Infants, Family Members and Livestock in Rural Eastern Ethiopia | 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 Host Clustering of Campylobacter Species and Other Enteric Pathogens in a Longitudinal Cohort of Infants, Family Members and Livestock in Rural Eastern Ethiopia Zelalem Mekuria, Loic Deblais, Amanda Ojeda, Nitya Singh, Wondwossen Gebreyes, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5736322/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 03 Nov, 2025 Read the published version in Microbiome → Version 1 posted 4 You are reading this latest preprint version Abstract Background Livestock are recognized as major reservoirs for Campylobacter species and other enteric pathogens, posing substantial infection risks to humans. High prevalence of Campylobacter during early childhood has been linked to environmental enteric dysfunction and stunting, particularly in low-resource settings. Methods A longitudinal study of 106 infants was conducted from December 2020 to June 2022. Monthly stool samples were collected from infants beginning in the first month after birth. Additional stool samples from mothers, siblings, and livestock (goats, cattle, sheep, and chickens) were collected biannually. A subset of 280 samples from Campylobacter positive households with complete metadata were analyzed by shotgun metagenomic sequencing followed by bioinformatic analysis via the CZ-ID metagenomic pipeline (Illumina mNGS Pipeline v7.1). Further statistical analyses in JMP PRO 16 explored the microbiome, emphasizing Campylobacter and other enteric pathogens. Two-way hierarchical clustering and split k-mer analysis examined host structuring, patterns of co-infections and genetic relationships. Principal component analysis was used to characterize microbiome composition across the seven sample types. Results More than 3,844 genera were detected in the 263 samples. Twenty-one dominant Campylobacter species were detected with distinct clustering patterns for humans, ruminants, and broad hosts. The generalist (broad-host) cluster included the most prevalent species, C. jejuni, C. concisus , and C. coli , present across sample types. Among C. jejuni a major cluster involving humans, chickens, and ruminants isolates, was detected, indicating potential zoonotic transmission to infants and mothers. Candidatus C. infans was only detected in human hosts. Campylobacter species from chickens showed strong positive correlations with mothers (r = 0.76), siblings (r = 0.61) and infants (r = 0.54), while no to weak correlation was observed between Campylobacter species from chickens and small ruminants (sheep and goats) with (r = 0.15, r = 0.0, respectively). Co-occurrence analysis revealed a higher likelihood (p > 0.5) of pairs such as C. jejuni with C. coli, C. concisus , and C. showae . Overall microbiome composition was strongly host driven, with two principal components accounting for 62% of the total variation. Analysis of the top 50 most abundant microbial taxa in infant stool revealed a distinct cluster uniquely present in human stool samples and absent in all livestock samples. Hierarchical clustering revealed frequent co-occurrence of C. jejuni with other enteric pathogens such as Salmonella , and Shigella , particularly in human and chicken samples. Additionally, instances of Candidatus C. infans were identified co-occurring with Salmonella and Shigella species in stool samples from infants, mothers, and siblings. Conclusions A comprehensive analysis of Campylobacter diversity in humans and livestock in a low-resource setting, revealed that infants can be exposed to multiple Campylobacter species early in life. C. jejuni is the dominant species with a propensity for co-occurrence with other notable enteric bacterial pathogens, including Salmonella , and Shigella , especially among infants. Campylobacter Enteric Pathogens Shotgun metagenomics Eastern Ethiopia Longitudinal Cohort Human Livestock Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Introduction Enteric diseases caused by bacterial, viral, or parasitic infections are the second leading cause of mortality globally, after lower respiratory tract infections, among children younger than five years ( 1 ). Humans are susceptible to these disease-causing microbes, encompassing zoonotic foodborne, waterborne, nosocomial, or community-acquired infections. The World Health Organization (WHO) estimates that, contaminated food leads to 600 million illnesses and more than 400,000 deaths annually ( 2 , 3 ). One-third of these cases affect young children in low- and middle-income countries (LMIC), contributing to malnutrition, stunting, and cognitive deficits ( 2 , 4 , 5 ). In addition, the economic impact of enteric infections is substantial, resulting in an estimated total productivity loss in LMIC of US $ 95 billion per year, with an additional predicted annual cost of treating these infections reaching US $ 15 billion ( 6 , 7 ) . In Ethiopia, where access to clean water and proper sanitation is limited, enteric infections pose a significant public health challenge ( 8 , 9 ). The WHO estimates 11 foodborne pathogen-associated deaths per 100,000 population in Ethiopia and diarrhea is the second most prevalent cause of mortality in infants ( 10 ). Bacterial agents include Campylobacter species, diarrheagenic Escherichia coli , non-typhoidal Salmonella enterica , Shigella species, while viral agents include rotavirus and norovirus, with children under the age of five being particularly vulnerable ( 4 ). In addition, Campylobacter and other enteric infections, especially in children, are increasingly recognized as contributing to child malnutrition ( 11 , 12 ). Our previous investigations, conducted in rural eastern Ethiopia within the sociodemographic context of the Haramaya district (in five kebeles), focused on characterizing the prevalence and distribution of Campylobacter species, assessing their interactions with child health, specifically environmental enteric dysfunction (EED) and stunting ( 13 ). Our cross-sectional study revealed a high prevalence of Campylobacter during early childhood. Employing metatranscriptomic shotgun sequencing approach, seven dominant Campylobacter species were identified in children's stools (aged 12 to 14 months). The observed diversity suggested Campylobacter colonization in children either occurred through multiple reservoirs or from a reservoir hosting several co-inhabiting Campylobacter species ( 13 ). Building upon these findings, our team embarked on a broader study spanning from December 2020 to June 2022 ( 14 ). This intensive longitudinal investigation included stool samples (from infants, mothers, and siblings), livestock feces (from cattle, chickens, goats, and sheep), and the environment (n = 1,644). The result showed potential routes for Campylobacter transmission in eastern Ethiopia and a correlation in the detection of Campylobacter species between infants and mothers and environmental sources ( 15 ). In the current study, we focused on a subset of human stools and livestock feces samples (n = 280; 40 per sample type) collected between December 2020 and June 2022. Utilizing shotgun metagenomics sequencing, Campylobacter diversity and co-circulation was investigated in infants, human (sibling, mother), and livestock species. This method, which does not rely on traditional culture techniques, has been shown to accurately identify Campylobacter at the species level, even in instances of low levels of genetic material in the samples ( 13 , 16 – 18 ). Further, we also investigated the presence of other enteric pathogens and explored the co-occurrence of Campylobacter species with enteric pathogens. Materials and methods Study area and design A longitudinal study involving 106 infants (ages ranging from 7 days to 13 months) was conducted from December 2020 to June 2022. Participants were selected randomly from 10 rural kebeles (synonymous with villages) of Haramaya woreda (synonymous with district), East Hararghe Zone, Oromia Region, Ethiopia. Detailed study area description has been published ( 14 , 15 ). Households in the selected kebeles maintain a variety of livestock (chickens, cattle, sheep, and goats) living in proximity with family members of the household. In some cases, the livestock co-habit with humans within the same room inside the house, often uncontained, to avoid issues with predators outdoors. Based on this information, human stool samples, livestock feces, and environmental samples were collected over time from each household to assess the prevalence and load of Campylobacter . Full details of the enrollment process and Campylobacter detection at the genus level have be previously described ( 14 ). In short, a total of 1,073 infant stool samples were collected monthly from 2–3 weeks after birth until 376 days of age. The first morning stools, minimum 10 grams but sometimes as little as 1 gram during the early months, were gathered in sterile plastic sheets within diapers that were sterilized with 12-hour UV light either on the same or preceding day. These samples were then transferred to 207-milliliter Whirl-Pak bags. For infants, diarrhea presence on sampling days and the day prior to specimen collection, along with the 24-hour loose stool count, were noted. Sample selection for shotgun metagenomics For the metagenomic analysis, we included samples from households (n = 106) that completed follow-up. We selected a total of 280 samples, 40 for each sample type (infants, mothers, siblings, cattle, goat, sheep and chickens) for metagenomic sequencing. We prioritized samples of good DNA quality, based on Nanodrop measurements (> 20 ng/ul, 260/280 > 1.8 and 260/230 < 1.5). We selected 50 households with at least one sample of each type. Further, we selected the latest available infant stool sample because our previous qPCR results ( 15 ) suggested the likelihood of detecting Campylobacter increased with age, which resulted in 40 households with balanced distribution across the 10 kebeles in the study area. The number of households per kebele varied between 3 and 5. We included one sample from each of the other types, preferring the sample closest in time to the infant sample if available. Extraction of genomic DNA and Campylobacter detection using qPCR. A QIAamp Power Fecal Pro DNA kit (Qiagen, CA, USA) was used and briefly, 0.25 g of the sample, as recommended by the manufacturer, was resuspended in 100 µL of nuclease-free water to extract genomic DNA from human stool samples and livestock feces. The quality and quantity of the DNA were analyzed using a UV5 Nanodrop spectrophotometer (Mettler Toledo, Columbus, OH). DNA with poor quality were cleaned using a Zymo genomic DNA clean and concentrator kit (Zymo Research, CA, USA) and stored at -20°C until further use. Aliquots of the DNA extracts were used for Campylobacter spp. Detection at the genus level in the human stool samples and animal feces using TaqMan real-time PCR( 19 ). Detailed steps and results are described elsewhere ( 15 ). Library preparation and metagenomic sequencing Library size and concentration were determined using the 4150 Tapestation system (Agilent, MA, USA). Limited-cycle PCR was subsequently employed to amplify the tagged DNA and introduce sequencing indexes. To facilitate a limit of detection assessment for each sample, we incorporated PhiX Control v3 (Illumina, Inc, San deigo, CA, USA) into each sample prior to library preparation. The prepared libraries were loaded onto a reagent cartridge and subjected to clustering on the NextSeq 2000 System. Subsequently, a paired-end sequencing run with 2 × 150 bp reads was executed using the NextSeq 2000 platform. The base calls generated by the NextSeq 2000 System were then transformed into FASTQ files, facilitating downstream metagenomic analysis. FASTQ assessment for sample and process evaluation We conducted an initial assessment of quality-related metrics, including cluster density, q-scores, and the percentage of passed reads as provided by the sequencer, using FastQC v.11.7 (Babraham Bioinformatics). Additionally, we conducted a more comprehensive quality checks using the CZ-ID pipeline (mNGS Pipeline v7.1). This involved evaluating total reads, the percentage of high-quality reads in the total pool, and assessing sequence diversity through duplicate compression ratios. Total reads were utilized to estimate the read count per sample, while the "passed QC" metric represented the percentage of reads retained after applying quality filtering using fastp, which eliminated low-quality bases, short reads (< 35 bp), and reads with low complexity. To evaluate sequence diversity within the metagenomic samples, the Duplicate Compression Ratio (DCR) was calculated for each sequencing run. DCR quantifies the prevalence of duplicate reads in a sample. A high DCR value, indicating numerous duplicate reads, suggests a less diverse library and account for potential technical artifacts affecting diversity assessments. Bioinformatic processing of the metagenomic data using the CZ-ID pipeline The CZ-ID pipeline (Illumina mNGS Pipeline v7.1) was used for metagenomic analysis, incorporating a series of steps for thorough data cleanup and analysis refinement. Host sequences were filtered, and ERCC sequences were excluded using STAR as the initial processing steps. Subsequently, we trimmed sequencing adapters and illumina indexes through Trimmomatic. Additional filtering was then carried out using PriceSeq, enabling the retention of high-quality sequences using the Phred score values. Duplicate reads were identified and removed using czid-dedup. Following these steps, any residual host sequences were rigorously filtered out using Bowtie2. In cases where more than one million reads remained after this process (or two million for paired-end data), random subsampling was performed using unique or deduplicated reads. Finally, irrespective of host considerations, CZ ID filtered out human sequences using a combination of STAR, Bowtie2, and GSNAP, resulting in comprehensive data preparation and quality control pipeline. Sequences then subjected to metagenomic assembly to align and merge short sequencing reads obtained from a longer DNA sequence to reconstruct the original sequence. Initially, sequence alignment steps in CZ ID involved mapping reads to NCBI NT (nucleotide) and NR (protein) databases to obtain preliminary accession for each read and generating BLAST database containing all taxa identified as preliminary accessions. CZ ID implements SPAdes to assemble contigs. Each contig is composed of several sequencing reads and the pipeline maps quality-filtered reads back to the assembled contigs to determine which reads are associated with each contig using Bowtie2. This step involved aligning BLAST contigs against database of preliminary accessions generated in the previous steps. Finally, each read and the final accession (based on contig accession, or if the reads failed to assemble into contigs- based on the preliminary accession) mapped to taxa using NCBI accession to taxon database and generated compute statistics per each sample. We employed downstream filtering steps to eliminate potentially erroneous taxonomic assignments. Drawing from our experience in formative research ( 13 ) and adhering to established community best practices, we employed specific criteria for taxon selection. Specifically, we used "NT L ≥ 50" to retain taxa with alignments of at least 50 base pairs in the nucleotide databases, "NT R ≥ 10" to preserve taxa with a minimum of 10 reads aligning to the nucleotide databases. Furthermore, to ensure data normalization and mitigate background noise, we incorporated a Z-score filter derived from a background model constructed using control samples used during DNA extraction (“Z-score > 1”; taxa abundance is higher in the sample than in the control; “Z-score = 100” taxa not found in the set of controls; “Z-score = -100” taxa only present in the controls). The outcomes of our metagenomic pipelines were comprehensively evaluated for meeting all three filtering criteria mentioned above. This approach helped to enhance the reliability of our taxonomic assignments in the metagenomic dataset. Statistical analysis Statistical analyses were performed using JMP PRO 16 software (SAS Institute, Cary, NC, USA). Metagenomic reads across the seven sample types were evaluated for normality and a one-way ANOVA was conducted to assess the differences in mean sequencing depths between the seven sample types. The Chi-square test was employed to compare the probability of detecting Campylobacter species across various kebeles. Additionally, pairwise comparisons were conducted to estimate the likelihood of co-infection for specific pairs of Campylobacter species detections. The abundance of each bacterium in the stool samples was estimated based on normalized read count, reads per million (rpm). Bi-variate analysis was utilized to examine the relationships between Campylobacter load from qPCR ( 15 ), with metagenomic abundance, rpm. A non-parametric Wilcoxon test was performed to identify rpm abundance differences for a given member of the microbial species. Hierarchal clustering, multivariate analysis combined with Pearson pairwise correlation coefficient and principal component analysis, were employed to uncover patterns and associations within the microbiome dataset. For selected Campylobacter taxa, a pairwise distance matrix was generated to depict genetic relatedness between taxa identified from metagenomic analysis and inform potential source attribution. For this analysis, samples with at least one Campylobacter contig per individual were selected. Genetic distances between selected samples were determined by calculating k-mer-based reference-free genomic distance estimation using split k-mer analysis (SKA)( 20 , 21 ). Results Metagenomic sequencing depth and read distribution in the seven sample types Examination of reads from stool samples (infants, mothers, and siblings), and fecal samples from livestock (cattle, goats, sheep, and chickens) demonstrated a mean sequencing depth of 1.06 x 10 8 (95% CI, 1.02–1.09 x 10 8 ), with highest sequencing depth of 1.5 x 10 8 per sample. In Fig. 1 A, the distribution of total sequencing depth achieved is illustrated, with a comparison of means across different sample types. Chicken feces exhibited the highest mean sequencing depth (1.2 x 10 8 reads, 95% CI: 1.1 x 10 8 – 1.2 x 10 8 ) and the narrowest distribution, indicating consistent sequencing quality, while infant stool showed the lowest mean sequencing depth (9.8 x 10 7 reads, 95% CI: 8.4 x 10 7 – 1.0 x 10 8 ) with greater variability. However, a one-way ANOVA test (p = 0.06) did not reveal significant differences in means, the low p-values suggest a tendency for chicken feces to exhibit a higher mean sequencing depth compared to other samples. We also examined the complexity of sequencing reads in each sample as a measure of microbial biomass diversity by calculating the duplicate compression ratio (DCR), which represents the ratio of sequences before and after the removal of duplicated reads. Notably, all the samples exhibited a DCR ratio of < 2, underscoring the presence of diverse sequences and absence of biases from PCR amplification during the sequencing process ( Fig. 1 B ) . Detection of Campylobacter species in human stool and animal feces through shotgun metagenomics Overall, with the exception of 17 samples excluded from analysis due to not meeting sequence quality thresholds, Campylobacter was detected in all remaining samples (n = 263/280). For thirteen of these samples, detection was restricted to the genus level, while in 250 samples, Campylobacter was identified at the species level. Figure 2 A, shows the frequency of Campylobacter species detected in stool (infants, mothers, and siblings) and fecal samples (cattle, goats, sheep, and chickens). In total, we detected 21 dominant Campylobacter species across human and animal hosts. This included the detection of multiple Campylobacter species in infants (n = 5), mothers (n = 8), and siblings (n = 8). In ruminant feces, the diversity of Campylobacter species varied, with sheep (n = 11), cattle (n = 10), and goats (n = 8) species. Chicken samples showed three Campylobacter species. C. jejuni (n = 76, 30%) and C. concisus (exception for chicken feces, n = 49, 14%) were the most prevalent species detected, indicating their widespread presence in both human and animal hosts. Other highly prevalent species included, Candidatus Campylobacter infans (C. infans) , C . vicugnae, C. coli , and C. showae . However, other thermophilic Campylobacter species such as C. lari and C. upsaliensis , were also detected but not frequently with 1.6–2.8% prevalence. Other less frequently detected species include C. iguaniorum (infant), C. mucosalis (mother), C. rectus (sheep), C . sp. 2014D-0216 (goat), C . californiensis (sheep), C . RM16192 (sheep), C. volucris (mother) and C. pinnepideorioum (chicken) ( Fig. 2 A ). Campylobacter species in stool and fecal samples from different sources Two-way hierarchical clustering revealed distinct patterns in the detection of Campylobacter species across the seven sample types ( Fig. 2 B ) . This analysis identified three major clusters among the 21 Campylobacter species. The first cluster, labeled C1, represents a mixed-host group, including species such as C. jejuni, C. concisus , and C. coli , which were detected in both human and livestock samples. The presence of these species in different sample types indicate niche overlap for these Campylobacter species suggesting transmission across hosts. The second group is human associated cluster, labeled as C2 and includes Campylobacter species exclusively found in human samples. It included C. infans and C. upsaliensis , which are present in infant, mother and sibling stools. Other species, such as C. geochelonis , was detected in sibling and mother stools, while C. mucosalis and C. volucris were found in mother samples, and C. iguaniorum was identified only in infant samples. The third group is a ruminant associated cluster labeled as C3, containing 12 Campylobacter species, among them the recently described species C. vicugnae , and C. californiensis , ( 22 , 23 ) as well as long established species such as C. showae and C. lari (also detected in sibling stools) were detected. Unlabeled as clusters, C. fetus and C. pinnipediorum were each detected in only single instances per sample type. C. fetus was identified in a sample from a sibling and in cattle. Meanwhile, C. pinnipediorum was exclusively detected in chicken and cattle feces. Notably, as shown in Fig. 2 C, these clusters are driven by a strong positive correlation in Campylobacter species detection among ruminant samples (r > 0.34) and within human samples (infant: sibling, r = 0.88, infant: mother, r = 0.57), indicating a high degree of species overlap within these groups. Additionally, Campylobacter species from chicken samples showed a strong correlation with human samples (chicken: infant r = 0.54, chicken: sibling, r = 0.61, chicken: mother, r = 0.76) but only a moderate correlation with cattle samples (r = 0.46). There was no significant correlation observed between Campylobacter species from chickens and small ruminant samples (r = 0 and 0.15, goat and sheep, respectively), suggesting distinct Campylobacter profiles in these groups. Multiple Campylobacter species co-occur in stool and fecal samples A key advantage of the metagenomic approach is its ability to investigate co-infections across all samples and seven sample types, enabling profiling of both dominant and less abundant Campylobacter species within sample. Our findings reveal a broader range of Campylobacter species other than the 21 most dominant ones identified across the seven sample types ( Table S1 ). Complete list of Campylobacter species detected in individual samples are shown in Table S2 . For the purpose of co-infection analysis, we limited our focus to the 21 most dominant Campylobacter species described in the previous section. Interestingly the presence of distinct clusters along the y-axis indicate that Campylobacter species are commonly found together across multiple samples ( Fig. 3 A ) . Species such as C. jejuni, C. concisus , and C. coli that were widely represented, showed notable co-occurrence patterns across multiple stool and fecal samples. In contrast, species like C. vicugnae and C. sp. RM12175 demonstrate distinct co-occurrence patterns, particular to small ruminant samples. Using probability co-occurrence matrix, we identified specific pairs of Campylobacter species in various stool and fecal samples from different sources with higher likelihood of co-occurrence ( C. jejuni and C. coli ), ( C. jejuni and C. concisus ) and ( C. jejuni and C. showae ) with a co-occurrence probability of (p > 0.5) ( Fig. 3 B ) . Abundance of key Campylobacter species in stool and fecal samples and correlation with qPCR detection Campylobacter abundance across seven sample types was investigated using a standardized nucleotide count, reads per million (rpm) (Fig. 4 ). Samples with nt.rpm exceeding 1000 were exclusively derived from humans, including the infant stools Fig. 4 . C. infans was detected at higher rpm compared to the overall median in this study (p = 0.035, Kruskal-Walli’s test). In ruminants, despite the presence of diverse Campylobacter species, the average abundance (average nt. rpm) was notably lower than in human stools using nonparametric comparisons by Wilcoxon test. We found a significant difference in the median rpm between infants and the livestock samples (p < 0.05). The average count for Campylobacter -specific reads in ruminant species remained below 50 rpm. C. vicugnae (in cattle, goat, and sheep), C. jejuni , and C. curvus (in cattle) exhibited statistically higher abundance compared to other Campylobacter species (p = 0.01, non-parametric Wilcoxon/Kruskal-Walli’s test). Using sliding window analysis, we also found agreements between metagenomic read abundance (rpm) and the Ct values of genus level qPCR assay used for Campylobacter detection, which was previously reported by Delais et al 2023 ( 15 ). Figure 5 shows, distribution of Ct values for the genus level qPCR assay against the metagenomic rpm values. This inverse relationship between Ct values of 19–26, yielded statistically significant correlation (Pearson correlation coefficient of -53% (95% CI: -0.68 to -0.31), p < 0.001) with the metagenomic read abundance rpm. In those samples with Ct of 19–26, the average read count was 200 rpm with a median value of 93 rpm. Genetic relationships of selected Campylobacter species across diverse sample types We investigated genetic similarities of selected Campylobacter species, specifically focusing on most widely detected species across the seven sample types: C. jejuni, C. concisus, C. infans, and C. sp. vicugnae . For C. jejuni , three major clusters (C1-3) were identified, shedding light on genetic similarities of circulating species ( Fig. 6 ) . Cluster C1 includes C. jejuni exhibiting phylogenetic relationships with known reference isolates, specifically Campylobacter jejuni subsp. jejuni strain:00-1597 (Accession ID_CP010306) and Campylobacter jejuni subsp. doylei NCTC 11924 [LR134530] These isolates are associated with human enteritis, with CP010306 isolated from a human in Canada ( 24 ) and sequence submitted from a child with diarrhea in the United Kingdom. Additional samples in C1 cluster include samples primarily originating from human sources and two of the samples (1119, mother and 785, infant) were from the same household 16. C1 also includes samples from cattle and sheep sources, suggesting potential cross transmission of C. jejuni across human and ruminant hosts. Similarly, in C2, the C. jejuni cluster is predominantly constituted from chicken sources. The three reference strains identified in this cluster include C. jejuni strain CFSAN054107 (accession: CP028185.1) from unknown source and C. jejuni strain C57 and strain C220 (with accession CP059970.1 and CP059972.1 respectively), derived from the cloacal swab of a chicken host. Overall, this cluster exhibits some host structuring, with samples from chicken showing genetic relatedness as measured in the relative distance (Mash-like distances < 0.1) with each other. In addition, the C2 cluster also showed a spatial signature with four of these sample being from the same kebele (Biftu Geda) and three of the four also being from the same household. The three samples that shared household were from mother and chicken source, indicating potential cross transmission of C. jejuni between hosts. In contrast, C3 showed host structuring with all C. jejuni in the group coming from human stool including two infant sources. For C.infans , as shown in Fig. 6 B, the data reveals a cluster encompassing stool samples from infants, mothers, and siblings. This cluster is divided into two main groups, C1 and C2, with further subdivisions into subclusters C1.1 and C1.2, where household level concordance was observed for some of the samples within these clusters. In Cluster C1, samples from Households 3, 45, 84, and 111 demonstrate genetic clustering of C. infans. Specifically, in Household 3, the samples represent an infant and a mother; in Household 45 (located in Amuna Kebele, within C1.2), infant and sibling stool samples show genetic similarity. Additionally, Household 84 in Kuro Kebele contains clustered stool samples from a mother and a sibling, and Household 111 includes samples from a mother and a sibling as well. These instances of genetic and spatial clustering suggest potential household-level cross transmission of C. infans. Furthermore, in Household 135, as illustrated in Fig. 6 B, stool samples from an infant and a sibling appear genetically divergent despite the shared household environment. This finding indicates the presence of wider genetic diversity of C. infans, with strains circulating in the same spatial setting. For C. concisus two clusters, C1 and C2, were identified ( Fig. S1 A ), with notable host structuring within each. Cluster C1 comprised C. concisus isolates predominantly from small ruminants (sheep and goats), whereas cluster C2 grouping, was primarily associated with sibling and maternal stool samples. In contrast, C. vicugnae was identified only in goat and sheep sources, and genetic variation was observed within small ruminant hosts (Fig. S1 B). Comparative analysis of stool microbiome diversity and shared bacterial communities across seven sample types Following the application of metagenomic filters, our analysis revealed the detection of more than 3,844 genera. Overall, the microbial composition showed a consistent bacterial dominance in the microbiota composition, with variations in other kingdoms depending on sample type. Bacterial communities constituted over 85% of sequences across all groups, with sheep and goat feces having the highest bacterial detection (92.49% and 92.91%, respectively). Eukaryote presence varied, ranging from 4.56% in goat feces to 10.38% in chicken feces. Archaeal communities were most abundant in cattle feces (2.97%), while infant stool exhibited the lowest archaeal presence (0.26%). Viral representation (mostly phages), though minimal in most samples, was notably higher in infant stool (5.52%) and sibling stool (3.06%) compared to other groups, where viruses ranged from 0.29–2.35%. Comparisons of means for alpha diversity indices (Shannon and Simpson diversity) revealed significant differences in microbial diversity and evenness across the seven sample types (ANOVA, Shannon & Simpson, p < 0.001; Fig. 7 ). Further pairwise comparisons using Tukey’s HSD indicated that infant stool samples exhibited significantly lower microbial diversity and richness compared to all other sample types (Tukey HSD for all pairs, p < 0.05). For b-diversity, we used, principal component analysis (PCA) that differentiated the microbial communities into two main components explaining 62.5% of the total diversity (p < 0.001) being as a result of sample types (Fig. 8 A). Component 1 explained approximately 50% of the diversity in the total samples. This component consisted of 92% from the microbial community in sheep, 88% in goats, and 66% in cattle, with a minority of 0.03% in chicken. This demonstrates the substantial contribution of ruminant samples. In contrast, component 2, which explained about 13% of variability within the samples, was loaded from human stool (53% in siblings, 51% in mothers, and 11% in infants) and 44% from chicken feces. Furthermore, Fig. 8 B, explains these differences by showing relationship of individual sample based on variation in microbial community structure across the seven sample types as denoted by clustering of similar colors. Notably, this analysis highlighted similarities between the microbial profiles of infant stool and chicken samples compared to other livestock samples, with maternal stool showing the closest relationship to the infant (Fig. 8 A). Cluster analysis from the Bray Curtis dissimilarity metrics indicated that while infant stools shared a core microbiome with maternal and sibling samples, there was also a more pronounced compositional divergence. The infant microbiome is distinctly characterized by a high presence of Bifidobacterium, Collinsella, Streptococcus , and Olsenella species (Fig. 9 A ). These four genera are considerably less prevalent in the stools of mothers and siblings, and are absent in livestock samples, except for a single detection of Streptococcus spp.in sheep feces which underscores the differences in microbial composition between infants and mothers and siblings. Analysis of the top 50 most abundant microbial taxa in infant stool revealed a distinct cluster uniquely present in human stool samples and absent in all livestock samples. This cluster includes the genera Bifidobacterium, Collinsella, Veillonella, Megasphaera, Bruquia, Stenotrophomonas, Prevotella, Haemophilus, Ureaplasma, Dialister , and Faecalibacterium . Chicken feces exhibited microbial similarities with human stool samples (infant, mother, and sibling), sharing genera such as Ligilactobacillus, Rothia, Leclercia, Sutterella, Coriobacterium, Yokenella, Allisonella, Blautia, Dictyocoela , and Anaerostipes . Additionally, Euzebya was shared exclusively with infant stool, while Sanguibacter was shared with infant and sibling stools. In contrast, genera such as Shumwellia, Alcanivorax, Lonsdalea , and Alkaliphilus appeared variably across most sample types ( Fig. 9 B ). Sub-group analysis of enteric microbiome based on findings from the MAL-ED and GEMS studies. This sub-group analysis investigated the presence and co-occurrence of selected enteric pathogens across seven sample types, informed by pathogen selection at the genus level based on findings from the MAL-ED and GEMS studies and WHO FERG estimates, which highlight pathogens of concern for infant health in developing regions( 1 , 5 ). The analysis focused on common enteric pathogens, Shigella, Vibrio, Salmonella, Entamoeba, Cryptosporidium , and Giardia . Hierarchical clustering revealed distinct co-occurrence patterns among these pathogens, including C. jejuni . These bacterial co-occurrence patterns involving eight species, Salmonella enterica, Salmonella bongori, Shigella flexneri, Shigella sonnei, Shigella sp. genomosp. SF-2015, and C. jejuni showed a high tendency to co-occur across various sample types, particularly in human stool and chicken feces ( Fig. 10 A ). Based on reported cases of infant diarrhea (as reported by the mother) within the clustered samples, co-detection of three Shigella species alongside C. jejuni in one diarrheal sample was obaserved. In contrast, four additional infant samples within the same cluster had no reported diarrheal history ( Fig. 10 B ). In total, Salmonella spp. ( S. enterica and S. bongori ) was detected in 30 samples, with a detection rate of 11.5%. Both species co-occurred in 13 samples, and their presence was most notable in chicken feces, as well as in stool samples from infants and siblings. Shigella spp. were identified in 20.7% of samples (n = 54), with the four major species: S. dysenteriae, S. flexneri, S. sonnei , and S. boydii comprising 71% of detections. S. flexneri was the most prevalent species, with Shigella spp. detected across all sample types except for cattle, and the highest burden observed in infant stool (n = 31), sibling stool (n = 30), and chicken feces (n = 28) (Fig. 11 A &B ). Vibrio spp. was detected in 117 samples across all sample types, representing 56 species, with Vibrio cholerae found in a limited number of samples (n = 7), including sheep feces (n = 3), and stool samples from mothers (n = 2) and infants (n = 2). Vibrio parahaemolyticus (n = 25) was frequently detected across a broader range of sample types, though there was no co-detection of V. cholerae and V. parahaemolyticus (Fig. 11 A &B ). Among parasitic agents, Entamoeba spp. was detected in 20 samples, with species including E. histolytica, E. dispar, E. invadens, and E. nuttalli . Cryptosporidium muris was found in a mother stool, while Giardia intestinalis was detected in an infant stool sample ( Fig. 11 A &B). Viral detection was minimal overall, with Human mastadenovirus found in both infant and sibling stool samples. Phage-related viral detections were common, with notable exceptions were Parapoxvirus , Orf virus, in ruminant samples and Gallid herpesvirus 1 in chicken samples. The Orf virus had the highest overall detection (n = 19), particularly in goat feces (n = 12). Discussion Campylobacter infection is a significant cause of diarrheal illness among children under five in LMIC including Ethiopia ( 12 , 25 – 27 ). Systematic reviews and meta-analysis indicate that around 10% of children in this age group are affected by Campylobacter , with prevalence in animals reaching 14.6% compared to 9% in humans, suggesting its zoonotic nature ( 28 – 30 ). Prior research primarily focused on thermophilic Campylobacter species like C. jejuni, C. coli, C. lari , and C. upsaliensis found in humans and domestic animals, while the prevalence of non-thermophilic species, such as C. hyointestinalis, Candidatus C. infans, C. fetus, C. showae , and C. concisus , remains largely unexamined. This study leveraged samples from a cohort of 106 infants in rural eastern Ethiopia, representing households across 10 of Haramaya’s 36 rural kebeles, for shotgun metagenomic sequencing. A recent report from our group indicated detection by qPCR of Campylobacter at the genus level in 64% of infant stool samples, with prevalence increasing with age as reported previously ( 15 ). By analyzing stool samples from infants, their associated environment (mothers, siblings, and livestock feces ‘[from cattle, goats, sheep, and chickens), we captured a detailed view of Campylobacter infection dynamics and other enteric pathogens across multiple hosts, shedding light on potential pathways of early life colonization. Metagenomic sequencing identified 21 major Campylobacter species across different sample types. Among these, five species, C. jejuni, C. concisus, C. coli, C. infans , and C. upsaliensis were found to colonize infants. The result highlighted early exposure of infants to diverse Campylobacter strains from both human (mother and sibling) and livestock sources, highlighting the effectiveness of metagenomic approaches in characterizing Campylobacter diversity and colonization patterns often missed by traditional methods ( 31 , 32 ). Notably, the detection of species like C. concisus and C. infans , especially in infants, is significant; C. infans is a relatively new proposed species, while C. concisus has been linked with chronic gastrointestinal conditions such as inflammatory bowel disease and has been observed in cases of gastroenteritis particularly in young children and immunocompromised patients ( 33 – 35 ). Likewise, C. infans was identified as the second most dominant species in breastfed infants in a global enteric multicenter study and has been associated with diarrhea in humans and non-human primates ( 36 ). Additionally, C. upsaliensis is relatively rare in infants, with its isolation often associated with dogs and cats ( 37 , 38 ). However, these findings originate from developed countries, where both healthy and sick dogs and cats have been identified as sources of C. upsaliensis infection in humans ( 39 ) In African contexts, research on C. upsaliensis is limited. A study in a rural South African setting did report the presence of C. upsaliensis in dogs, indicating that pets could similarly act as reservoirs in African regions ( 40 ). However, overall, data supporting C. upsaliensis reservoir hosts in rural Africa setting remains sparse. The study also found host specific and mixed host infection patterns among Campylobacter species. The detection of C. jejuni across all sample types underscores its ecological versatility and adaptive potential, aligning with previous findings on the zoonotic nature of this species( 41 , 42 ). C. jejuni is well recognized for its host diversity and environmental resilience, which contribute to its status as one of the most prevalent causes of bacterial gastroenteritis worldwide ( 31 , 43 ). The mixed host clustering of C. coli and C. concisus also highlights their potential for cross-host transmission. Although C. coli is primarily associated with livestock, particularly pigs and poultry, it has been increasingly detected in humans and other animal hosts, indicating its adaptability and potential zoonotic threat ( 44 , 45 ). In this study, C. coli was detected in both infant stool and goat feces, reflecting its potential for mixed-host infection where direct or indirect contact between humans and livestock may have facilitated transmission. On the other hand, C. concisus was detected in all sample types except chicken feces, pointing to its relatively broad host range with a preference for mammalian hosts. A major finding in this study is also the co-infection patterns among Campylobacter species revealed significant associations, particularly for C. jejuni , which demonstrated a high probability (P > 0.5) of co-occurrence with C. coli, C. concisus , and C. showae . These observations indicate the potential for interactions within the host environment, where multiple Campylobacter species may simultaneously colonize, potentially leading to complex infection dynamics influencing disease conditions in infants. These co-infections are likely due to synergistic effects or shared virulence mechanisms that facilitate persistence and survival within host microbiomes ( 46 – 48 ). High co-occurrence probability of these pairs highlights the need for further investigation into the implications of such mixed Campylobacter infections in infant health. In addition to the Campylobacter species, our study showed that C. jejuni exhibited strong clustering with other enteric pathogens, including Salmonella enterica, Salmonella bongori , and multiple Shigella species ( S. flexneri, S. sonnei, and Shigella sp. genomosp. SF-2015 ). This clustering was particularly pronounced in human stool and chicken feces samples, suggesting shared environmental or dietary transmission routes. The frequent co-detection of C. jejuni with these pathogens aligns with other studies that have observed overlapping transmission pathways and reservoirs, particularly in resource-limited settings where sanitation and food safety practices are less stringent ( 49 , 50 ). Even though the metagenomic investigation in this study was centered on Campylobacter and key enteric pathogens, it was also complemented by a global microbiome analysis that revealed two distinct microbiota clusters. The first cluster (principal component) associated with ruminants and another consisting of human and chicken samples. Using PCA, we demonstrated that the chicken microbiome shares organisms including enteric pathogens with the infant microbiome while the ruminant microbiome is more distinct. Using multivariate analysis, we found that Campylobacter species from chicken samples show a strong correlation with human stool samples (r > 0.54) but only a moderate correlation with cattle (r = 0.46). Further analysis of C. jejuni sub-clusters revealed that human and chicken sources frequently grouped together, providing evidence of cross-host transmission between chickens and infants. In conclusion, the study provides evidence for cross-host zoonotic transmission potential of some Campylobacter species while other species appeared to be host-restricted. There was significant correlation among the Campylobacter species detected in human stools, suggesting a potential pathway for infant Campylobacter colonization through exposure via sibling and mother. Also, strong correlation with chickens indicates they may be a direct source of Campylobacter infection for humans, whereas cattle may also contribute to the environmental pool of Campylobacter in multi-host environments. Declarations Ethics approval and consent to participate Ethical approval was obtained from the University of Florida Internal Review Board (IRB201903141); the Haramaya University Institutional Health Research Ethics Committee (COHMS/1010/3796/20), and the Ethiopia National Research Ethics Review Committee (SM/14.1/1059/20). Written informed consent was obtained from all participating households (husband and wife) using a form in the local language (Afan Oromo). Consent for publication Not applicable. Availability of data The metagenomics data is submitted to NCBI under Bioproject number PRJNA1127034 and will be made available once the manuscript is accepted. Competing interests The authors declare that they have no competing interests. Funding The Bill and Melinda Gates Foundation funded the CAGED project to address food insecurity issues in Ethiopia and Burkina Faso through the project Equip—Strengthening Smallholder Livestock Systems for the Future (grant number OPP11755487). These funds are administered by the Feed the Future Innovation Lab for Livestock Systems, established with funding from the U.S. Agency for International Development (USAID) and co-led by the University of Florida’s Institute of Food and Agricultural Sciences and the International Livestock Research Institute. Support for the Feed the Future Innovation Lab for Livestock Systems is made possible by the generous support of the American people through USAID. The contents are the authors' responsibility and do not necessarily reflect the views of USAID or the U.S. Government. REDCap is hosted at the University of Florida Clinical and Translational Science Institute (CTSI) and supported by NIH National Center for Advancing Translational Sciences grant UL1TR000064. This project is funded by the U.S. Agency for International Development Bureau for Food Security under agreement number AID-OAA-L-15-00003 as part of Feed the Future Innovation Lab for Livestock Systems. Any opinions, findings, conclusions, or recommendations expressed here are those of the authors alone. Research reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, partly funded by the NIH National Center for Advancing Translational Sciences under award number UL1TR001427. Author Contributions Conceptualization: AHH, GR, ZM; Data curation: ZM, AO, LD, Formal Analysis: ZM, AO, DC, ; Funding acquisition: AHH, GR; Investigation: ZM, AO, BM, LD, NS; Methodology: ZM, LD, AO, NS, GR, AHH; Project Administration: ZM, GR, AHH, WG; Resources: WG, GR, AHH, Supervision: WG, GR, AHH; Writing – original: ZM, AHH, GR; Writing – review & editing: ZM, AO, LD, BM, NS, WG, AHH, GR. All authors approved the final version for submission. Acknowledgments This work is a result of the CAGED Research Team, whose members include: Abadir Jemal Seran, Abdulmuen Mohammed Ibrahim, Bahar Mummed Hassen, Belisa Usmael Ahmedo, Cyrus Saleem, Dehao Chen, Efrah Ali Yusuf, Getnet Yimer, Ibsa Abdusemed Ahmed, Ibsa Aliyi Usmane, Jafer Kedir Amin, Jemal Yusuf Hassen, Kedir Abdi Hassen, Kunuza Adem Umer, Karah Mechlowitz, Kedir Teji Roba, Loic Deblais, Mussie Bhrane, Mark J. Manary, Mawardi M. Dawid, Mahammad Mahammad Usmail, Nigel P. French, Nur Shaikh, Sarah L. McKune, Xiaolong Li, Yenenesh Demisie Weldesenbet, Yang Yang. This study would not have been possible without cooperation of study communities and local administration of the study kebeles. We want to express our appreciation to the study households, the Community Advisory Board and all who supported the study directly or otherwise. Research reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, which was supported in part by the NIH National Center for Advancing Translational Sciences under award number UL1TR001427. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. References Troeger CE, Khalil IA, Blacker BF, Biehl MH, Albertson SB, Zimsen SRM, et al. Quantifying risks and interventions that have affected the burden of diarrhoea among children younger than 5 years: an analysis of the Global Burden of Disease Study 2017. The Lancet Infectious Diseases. 2020;20(1):37–59. WHO 2015. Estimates of the global burden of foodborne diseases: foodborne disease burden epidemiology reference group 2007–2015 https://www.foodbornediseaseburden.org/ferg/estimates : World Health Organization; [ Havelaar AH, Kirk MD, Torgerson PR, Gibb HJ, Hald T, Lake RJ, et al. World Health Organization global estimates and regional comparisons of the burden of foodborne disease in 2010. PLoS medicine. 2015;12(12):e1001923. Kirk MD, Pires SM, Black RE, Caipo M, Crump JA, Devleesschauwer B, et al. World Health Organization Estimates of the Global and Regional Disease Burden of 22 Foodborne Bacterial, Protozoal, and Viral Diseases, 2010: A Data Synthesis. PLOS Medicine. 2015;12(12):e1001921. The MAL-ED Network Investigators. The MAL-ED Study: A Multinational and Multidisciplinary Approach to Understand the Relationship Between Enteric Pathogens, Malnutrition, Gut Physiology, Physical Growth, Cognitive Development, and Immune Responses in Infants and Children Up to 2 Years of Age in Resource-Poor Environments. Clinical Infectious Diseases2014. p. S193-S206. World Bank 2017. Drug-Resistant Infections: A Threat to Our Economic Future. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO. Jaffee S, Henson S, Unnevehr L, Grace D, Cassou E. The safe food imperative: Accelerating progress in low-and middle-income countries: World Bank Publications; 2018. Fenta SM, Nigussie TZ. Factors associated with childhood diarrheal in Ethiopia; a multilevel analysis. Archives of Public Health. 2021;79(1):123. Alemayehu K, Oljira L, Demena M, Birhanu A, Workineh D. Prevalence and Determinants of Diarrheal Diseases among Under-Five Children in Horo Guduru Wollega Zone, Oromia Region, Western Ethiopia: A Community-Based Cross-Sectional Study. Canadian Journal of Infectious Diseases and Medical Microbiology. 2021;2021:5547742. WHO. WHO estimates of the global burden of foodborne diseases: foodborne disease burden epidemiology reference group 2007–2015: World Health Organization; 2015. Platts-Mills JA, Kosek M. Update on the burden of Campylobacter in developing countries. Current opinion in infectious diseases. 2014;27(5):444. Zenebe T, Zegeye N, Eguale T. Prevalence of Campylobacter species in human, animal and food of animal origin and their antimicrobial susceptibility in Ethiopia: a systematic review and meta-analysis. Annals of clinical microbiology and antimicrobials. 2020;19(1):1–11. Terefe Y, Deblais L, Ghanem M, Helmy YA, Mummed B, Chen D, et al. Co-occurrence of Campylobacter Species in Children From Eastern Ethiopia, and Their Association With Environmental Enteric Dysfunction, Diarrhea, and Host Microbiome. Frontiers in Public Health. 2020;8. Havelaar AH, Brhane M, Ahmed IA, Kedir J, Chen D, Deblais L, et al. Unravelling the reservoirs for colonisation of infants with Campylobacter spp. in rural Ethiopia: protocol for a longitudinal study during a global pandemic and political tensions. BMJ Open. 2022;12(10):e061311. Deblais L, Ojeda A, Brhane M, Mummed B, Hassen KA, Ahmedo BU, et al. Prevalence and Load of the Campylobacter Genus in Infants and Associated Household Contacts in Rural Eastern Ethiopia: a Longitudinal Study from the Campylobacter Genomics and Environmental Enteric Dysfunction (CAGED) Project. Appl Environ Microbiol. 2023;89(7):e0042423. Parker CT, Schiaffino F, Huynh S, Paredes Olortegui M, Peñataro Yori P, Garcia Bardales PF, et al. Shotgun metagenomics of fecal samples from children in Peru reveals frequent complex co-infections with multiple Campylobacter species. PLoS Negl Trop Dis. 2022;16(10):e0010815. Andersen SC, Kiil K, Harder CB, Josefsen MH, Persson S, Nielsen EM, et al. Towards diagnostic metagenomics of Campylobacter in fecal samples. BMC Microbiol. 2017;17(1):133. Haverkamp TH, Spilsberg B, Johannessen GS, Torp M, Sekse C. Detection of Campylobacter in air samples from poultry houses using shot-gun metagenomics–a pilot study. bioRxiv. 2021:2021.05. 17.444449. Platts-Mills JA, Liu J, Gratz J, Mduma E, Amour C, Swai N, et al. Detection of Campylobacter in stool and determination of significance by culture, enzyme immunoassay, and PCR in developing countries. J Clin Microbiol. 2014;52(4):1074–80. Derelle R, von Wachsmann J, Mäklin T, Hellewell J, Russell T, Lalvani A, et al. Seamless, rapid and accurate analyses of outbreak genomic data using Split K-mer Analysis (SKA). bioRxiv. 2024:2024.03.25.586631. Harris SR. SKA: Split Kmer Analysis Toolkit for Bacterial Genomic Epidemiology. bioRxiv. 2018:453142. Miller WG, Chapman MH, Williams TG, Wood DF, Bono JL, Kelly DJ. Campylobacter californiensis sp. nov., isolated from cattle and feral swine. International Journal of Systematic and Evolutionary Microbiology. 2024;74(10). Miller WG, Lopes BS, Ramjee M, Jay-Russell MT, Chapman MH, Williams TG, et al. Campylobacter devanensis sp. nov., Campylobacter porcelli sp. nov., and Campylobacter vicugnae sp. nov., three novel Campylobacter lanienae-like species recovered from swine, small ruminants, and camelids. International Journal of Systematic and Evolutionary Microbiology. 2024;74(6). Clark CG, Chen C-y, Berry C, Walker M, McCorrister SJ, Chong PM, et al. Comparison of genomes and proteomes of four whole genome-sequenced Campylobacter jejuni from different phylogenetic backgrounds. PLOS ONE. 2018;13(1):e0190836. Lengerh A, Moges F, Unakal C, Anagaw B. Prevalence, associated risk factors and antimicrobial susceptibility pattern of Campylobacter species among under five diarrheic children at Gondar University Hospital, Northwest Ethiopia. BMC pediatrics. 2013;13:1–9. Abamecha A, Assebe G, Tafa B, Wondafrash B. Prevalence of thermophilic Campylobacter and their antimicrobial resistance profile in food animals in Lare District, Nuer Zone, Gambella, Ethiopia. J Drug Res Dev. 2015;1(2):2470–1009. Diriba K, Awulachew E, Anja A. Prevalence and associated factor of Campylobacter species among less than 5-year-old children in Ethiopia: a systematic review and meta-analysis. European Journal of Medical Research. 2021;26(1):1–10. Diriba K, Awulachew E, Anja A. Prevalence and associated factor of Campylobacter species among less than 5-year-old children in Ethiopia: a systematic review and meta-analysis. European Journal of Medical Research. 2021;26(1):2. Zenebe T, Zegeye N, Eguale T. Prevalence of Campylobacter species in human, animal and food of animal origin and their antimicrobial susceptibility in Ethiopia: a systematic review and meta-analysis. Annals of Clinical Microbiology and Antimicrobials. 2020;19(1):61. Belina D, Gobena T, Kebede A, Chimdessa M, Mummed B, Thystrup CAN, et al. Occurrence and diversity of Campylobacter species in diarrheic children and their exposure environments in Ethiopia. PLOS Global Public Health. 2024;4(10):e0003885. Kaakoush NO, Castaño-Rodríguez N, Mitchell HM, Man SM. Global Epidemiology of Campylobacter Infection. Clin Microbiol Rev. 2015;28(3):687–720. Quince C, Walker AW, Simpson JT, Loman NJ, Segata N. Shotgun metagenomics, from sampling to analysis. Nature Biotechnology. 2017;35(9):833–44. Bian X, Garber Jolene M, Cooper Kerry K, Huynh S, Jones J, Mills Michael K, et al. Campylobacter Abundance in Breastfed Infants and Identification of a New Species in the Global Enterics Multicenter Study. mSphere. 2020;5(1): 10.1128/msphere.00735 – 19 . Kaakoush NO, Mitchell HM. Campylobacter concisus–a new player in intestinal disease. Frontiers in cellular and infection microbiology. 2012;2:4. Istivan T, Huq M. Biofilms of Campylobacter concisus: a potential survival mechanism in the oral cavity. Microbiology Australia. 2023;44(2):100–3. Napit R, Manandhar P, Poudel A, Rajbhandari PG, Watson S, Shakya S, et al. Novel strains of Campylobacter cause diarrheal outbreak in Rhesus macaques (Macaca mulatta) of Kathmandu Valley. PLOS ONE. 2023;18(3):e0270778. Carbonero A, Torralbo A, Borge C, García-Bocanegra I, Arenas A, Perea A. Campylobacter spp., C. jejuni and C. upsaliensis infection-associated factors in healthy and ill dogs from clinics in Cordoba, Spain. Screening tests for antimicrobial susceptibility. Comparative immunology, microbiology and infectious diseases. 2012;35(6):505–12. Parsons B, Porter C, Stavisky J, Williams N, Birtles R, Miller W, et al. Multilocus sequence typing of human and canine C. upsaliensis isolates. Veterinary microbiology. 2012;157(3–4):391–7. Jaime AL, Joan S, Lee B, Nancy S, Sydney MH, Eleanor L, et al. Campylobacter upsaliensis: Another Pathogen for Consideration in the United States. Clinical Infectious Diseases. 2002;34(11):e59-e60. Karama M, Etter E, McCrindle C, El-Ashram S, Prosperi A, Ombui JN, et al. Prevalence and risk factors associated with Campylobacter spp. occurrence in healthy dogs visiting four rural community veterinary clinics in South Africa. Onderstepoort Journal of Veterinary Research. 2019;86(1):1–6. Duong T, Konkel ME. Comparative studies of Campylobacter jejuni genomic diversity reveal the importance of core and dispensable genes in the biology of this enigmatic food-borne pathogen. Curr Opin Biotechnol. 2009;20(2):158–65. Hermans D, Pasmans F, Messens W, Martel A, Van Immerseel F, Rasschaert G, et al. Poultry as a host for the zoonotic pathogen Campylobacter jejuni. Vector-Borne and Zoonotic Diseases. 2012;12(2):89–98. Gripp E, Hlahla D, Didelot X, Kops F, Maurischat S, Tedin K, et al. Closely related Campylobacter jejuni strains from different sources reveal a generalist rather than a specialist lifestyle. BMC genomics. 2011;12:1–21. Horrocks S, Anderson R, Nisbet D, Ricke S. Incidence and ecology of Campylobacter jejuni and coli in animals. Anaerobe. 2009;15(1–2):18–25. Humphrey T, O'Brien S, Madsen M. Campylobacters as zoonotic pathogens: a food production perspective. International journal of food microbiology. 2007;117(3):237–57. Godschalk PC, Gilbert M, Jacobs BC, Kramers T, Tio-Gillen AP, Ang CW, et al. Co-infection with two different Campylobacter jejuni strains in a patient with the Guillain-Barré syndrome. Microbes Infect. 2006;8(1):248–53. Castaño-Rodríguez N, Kaakoush NO, Lee WS, Mitchell HM. Dual role of Helicobacter and Campylobacter species in IBD: a systematic review and meta-analysis. Gut. 2017;66(2):235–49. Wang G, He Y, Jin X, Zhou Y, Chen X, Zhao J, et al. The effect of co-infection of food-borne pathogenic bacteria on the progression of Campylobacter jejuni infection in mice. Frontiers in microbiology. 2018;9:1977. Nataro James P, Kaper James B. Diarrheagenic Escherichia coli. Clinical Microbiology Reviews. 1998;11(1):142–201. Bhattarai V, Sharma S, Rijal KR, Banjara MR. Co-infection with Campylobacter and rotavirus in less than 5 year old children with acute gastroenteritis in Nepal during 2017–2018. BMC pediatrics. 2020;20:1–8. Additional Declarations No competing interests reported. Supplementary Files FigureS1.pdf TableS1.xlsx TableS2.xlsx Cite Share Download PDF Status: Published Journal Publication published 03 Nov, 2025 Read the published version in Microbiome → Version 1 posted Editorial decision: Revision requested 13 Jan, 2025 Editor assigned by journal 13 Jan, 2025 Submission checks completed at journal 02 Jan, 2025 First submitted to journal 30 Dec, 2024 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-5736322","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":401740243,"identity":"a4fe9d31-e9a1-463a-a09d-c019c46ab454","order_by":0,"name":"Zelalem Mekuria","email":"","orcid":"","institution":"The Ohio State University","correspondingAuthor":false,"prefix":"","firstName":"Zelalem","middleName":"","lastName":"Mekuria","suffix":""},{"id":401740244,"identity":"f12ebd1e-f1b6-4b66-90ce-30eeaa449fb6","order_by":1,"name":"Loic Deblais","email":"","orcid":"","institution":"The Ohio State University","correspondingAuthor":false,"prefix":"","firstName":"Loic","middleName":"","lastName":"Deblais","suffix":""},{"id":401740245,"identity":"5feee0ee-1460-4fca-9f74-3460769eecd5","order_by":2,"name":"Amanda Ojeda","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Amanda","middleName":"","lastName":"Ojeda","suffix":""},{"id":401740246,"identity":"facc5bc7-b8ea-4a44-9baf-dd1783a58fb0","order_by":3,"name":"Nitya Singh","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Nitya","middleName":"","lastName":"Singh","suffix":""},{"id":401740247,"identity":"e413d838-ace3-43bc-98b7-5c2330970941","order_by":4,"name":"Wondwossen Gebreyes","email":"","orcid":"","institution":"The Ohio State University","correspondingAuthor":false,"prefix":"","firstName":"Wondwossen","middleName":"","lastName":"Gebreyes","suffix":""},{"id":401740249,"identity":"6dd4064e-8775-45c3-9633-c406798192bd","order_by":5,"name":"Arie Havelaar","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Arie","middleName":"","lastName":"Havelaar","suffix":""},{"id":401740251,"identity":"88c758cb-f524-4a87-9faf-ca6a09925fef","order_by":6,"name":"Gireesh Rajashekara","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIie3PMQrCMBSA4ZQHzyXiWlHMFSIFQRQcvYYieA0jhTpJZ8FDCC6OkQe6eABBB7s4OTiJYgerrZMQHR3yD4GEfLyEMZvtH4PXqisFgKHuMMafW/cX4hV9R/1I0nR3phPy3hqJaOejwzneO/N1QqJFsywULHfcQKqU86qT4Ag1Skh30+dSY69hJD5iKa8IUxIQl4zXSl9I7h7HxD0/I0IVLkYiABEYkishI0xzNBIJCMVxQNJNH5b8hdCrT01TwpVzvsY0CENaRreg2RIjP9qeTFP0xxEYrr+mqC8XbDabzcYeEoNNA6Glj5EAAAAASUVORK5CYII=","orcid":"","institution":"University of Illinois Urbana-Champaign","correspondingAuthor":true,"prefix":"","firstName":"Gireesh","middleName":"","lastName":"Rajashekara","suffix":""},{"id":401740252,"identity":"e36b827f-1584-423a-9fee-00b3c1b458cd","order_by":7,"name":"Bahar Mummed","email":"","orcid":"","institution":"Haramaya University","correspondingAuthor":false,"prefix":"","firstName":"Bahar","middleName":"","lastName":"Mummed","suffix":""}],"badges":[],"createdAt":"2024-12-30 14:38:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5736322/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5736322/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s40168-025-02203-w","type":"published","date":"2025-11-03T15:56:55+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":78751175,"identity":"458f6d1a-dc99-4146-bffd-0862361c9bf7","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":467073,"visible":true,"origin":"","legend":"\u003cp\u003eMetagenomic sequencing depth and read distribution in seven sample types. \u003cstrong\u003eA)\u003c/strong\u003e Density plots showing the distribution of total sequence reads across different sample types. Each curve represents the distribution of sequence read counts for each sample type, with mean values, standard deviations, and 95% confidence intervals (CI). \u003cstrong\u003eB)\u003c/strong\u003e The duplicate compression ratio, denoting the ratio of sequences before and after duplicate sequence removal. Values less than 2 conforms the metagenomic sequencing criteria, affirming the absence\u003cstrong\u003e \u003c/strong\u003eof sequencing biases.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/b898b212f81a8560d244b454.jpg"},{"id":78751171,"identity":"515bfa2f-4dcf-45f8-b2ed-bf518094dddd","added_by":"auto","created_at":"2025-03-18 11:52:50","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":925749,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eCampylobacter\u003c/em\u003e diversity in stool and fecal samples from seven sample types. \u003cstrong\u003eA)\u003c/strong\u003e Bar plot showing diversity of \u003cem\u003eCampylobacter\u003c/em\u003especies detected in human stool and livestock feces. Multiple \u003cem\u003eCampylobacter\u003c/em\u003especies were detected in infants (n=5), mothers (n=8), siblings (n=8), sheep (n=11), cattle (n=10), goats (n=8) and chickens (n=3). \u003cstrong\u003eB)\u003c/strong\u003e Two-way hierarchical clustering revealed distinct clustering of \u003cem\u003eCampylobacter\u003c/em\u003especies across the seven sample types. Cluster C1, represents a broad-host range species, Cluster C2, human associated species and cluster C3, ruminant associated species. Red cells represent detection and blue cells represent absence of detection) \u003cstrong\u003eC) \u003c/strong\u003eMultivariate analysis of the detection frequencies of each \u003cem\u003eCampylobacter\u003c/em\u003e species in different host types. Significant correlations were observed among \u003cem\u003eCampylobacter\u003c/em\u003e species detected within human (infant, sibling and mother r \u0026gt; 0.64). Strong positive correlation between infant and chicken (r =0.54), no to low correlation between infants with small ruminants (r = 0 – 0.15) and cattle (r = 0.19) respectively.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/425de2e6d9c87bd057c6b569.jpg"},{"id":78753382,"identity":"6c7a0e08-5485-4b63-ac9d-ede5d84bc2d6","added_by":"auto","created_at":"2025-03-18 12:16:51","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":955917,"visible":true,"origin":"","legend":"\u003cp\u003eHierarchical clustering heatmap displaying co-occurrence relationships among various \u003cem\u003eCampylobacter\u003c/em\u003e species across stool and fecal samples from seven sample sources. \u003cstrong\u003eA)\u003c/strong\u003e Columns represent individual samples, and rows correspond to different \u003cem\u003eCampylobacter\u003c/em\u003e species. \u003cem\u003eC. jejuni, C. concisus\u003c/em\u003e, and \u003cem\u003eC. coli\u003c/em\u003e, showing notable co-occurrence patterns across multiple stool and fecal samples. In contrast, species like \u003cem\u003eC. vicunae\u003c/em\u003e and \u003cem\u003eC. sp. RM12175 \u003c/em\u003edemonstrate distinct co-occurrence restricted to small ruminant samples. \u003cstrong\u003eB)\u003c/strong\u003e Pairwise comparisons revealed a co-occurrence likelihood exceeding 0.5 in certain species pairs (\u003cem\u003eC. jejuni and C. coli\u003c/em\u003e), (\u003cem\u003eC. jejuni and C. concisus\u003c/em\u003e), and (\u003cem\u003eC. jejuni \u003c/em\u003eand\u003cem\u003eC. showae\u003c/em\u003e).\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/11a429b35ead2895d4f22a79.jpg"},{"id":78751190,"identity":"73fec8bd-809b-44ff-86f3-0083c84a3d29","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":589831,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eCampylobacter\u003c/em\u003e abundance measured as reads per million (rpm) across the seven sample types. Graphs depicts on x-axis the 21 \u003cem\u003eCampylobacter\u003c/em\u003e species, with abundance measurement values (rpm) plotted on the y-axis. Samples with 10\u003csup\u003e3\u003c/sup\u003e rpm were exclusively derived from humans, prominently in the infant stools. \u003cem\u003eCandidatus \u003c/em\u003eCampylobacter infans was detected at higher rpm (indicate range) compared to the overall average (indicate number) in this study (p = 0.035, Kruskal-Walli’s test).\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/7844a93718d43a9098edb23b.jpg"},{"id":78751922,"identity":"26997b5e-5ea6-4c5e-ac9a-2dbaa6eb9204","added_by":"auto","created_at":"2025-03-18 12:00:51","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":300671,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between \u003cem\u003eCampylobacter\u003c/em\u003e metagenomic relative abundance and their detection through qPCR. Sliding window analysis depicts the correlation between metagenomic read per million with the genus-level qPCR Ct values. The analysis revealed a Pearson correlation coefficient of -53% (95% CI: -0.68 to -0.31), demonstrating agreement between the metagenomic relative abundance and genus-level qPCR Ct values.\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/0a985a2373e5d1f24b6c0998.jpg"},{"id":78751186,"identity":"b5fe86ff-617a-4022-b589-b0c2c9d252f0","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":796999,"visible":true,"origin":"","legend":"\u003cp\u003ePairwise Mash-like distances for \u003cem\u003eCampylobacter\u003c/em\u003e species in a matrix format. \u003cstrong\u003eA)\u003c/strong\u003e \u003cem\u003eCampylobacter jejuni; \u003c/em\u003e\u003cstrong\u003eB)\u003c/strong\u003e \u003cem\u003eCandidatus \u003c/em\u003eCampylobacter infans. Each sample in the dataset was compared to itself and all other samples. The colors in the matrix represent the range of Mash-like distances (refer to the scale). The diagonal line compares each sample to itself, resulting in dark red squares, indicating zero differences between the sequences. Black boxes are used to show clustering of samples (labeled as C1-3) and on the right side, sample metadata, including host, Kebele, and household, utilized to support the genetic similarities observed in the metagenomic dataset. Asterisks (*) denote samples collected from the same\u003cstrong\u003e \u003c/strong\u003ehousehold\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/ac1b716b0e87f5fe221d0a6d.jpg"},{"id":78751193,"identity":"5f8bc6d2-9b56-49b8-9fca-0621468c6f08","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":358614,"visible":true,"origin":"","legend":"\u003cp\u003eAlpha diversity of microbial communities across seven sample types, measured by the Simpson \u003cstrong\u003e(A)\u003c/strong\u003e and Shannon \u003cstrong\u003e(B)\u003c/strong\u003e diversity indices\u003cstrong\u003e. \u003c/strong\u003eLivestock fecal samples, especially from cattle, goats, and sheep, display higher microbial diversity and evenness, indicated by broader distributions and higher mean, median values in both Simpson and Shannon indices (ANOVA, Shannon \u0026amp; Simpson, p \u0026lt; 0.001). Tukey’s HSD indicated that infant stool samples exhibited significantly lower microbial diversity and richness compared to all other sample types (Tukey HSD for all pairs, p \u0026lt; 0.05)\u003c/p\u003e","description":"","filename":"Figure7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/11fa0e14834a2ad8fddf0ada.jpg"},{"id":78752967,"identity":"bddc0116-486c-42e8-8d55-1e6e16fbc722","added_by":"auto","created_at":"2025-03-18 12:08:51","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":481882,"visible":true,"origin":"","legend":"\u003cp\u003ePrincipal Component Analysis (PCA) based on microbial community composition: The plots show clustering of microbial communities from different sample types based on beta diversity. \u003cstrong\u003eA)\u003c/strong\u003e Microbiota diversity was explored through principal component analysis (PCA), revealing two major components explaining 62.5% of the microbial diversity, attributed to sample types. PCA highlighted distinct microbiota clustering within ruminant samples, \u003cstrong\u003eB)\u003c/strong\u003eEach point represents a sample, color-coded by sample type as indicated in the legend. The axes represent the principal coordinates (PCoA1 and PCoA2), which explain the variation in microbial community structure between samples. Samples that are closer together share more similar microbial compositions, while those farther apart are more dissimilar. The distinct clustering of colors reflects differences in microbial communities across sample types.\u003c/p\u003e","description":"","filename":"Figure8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/cc69296382979577bbd2c2d2.jpg"},{"id":78751191,"identity":"a7d6e863-06ab-42d4-9c73-847621bd88ee","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":638464,"visible":true,"origin":"","legend":"\u003cp\u003eTop 50 abundant organisms in infant stool. \u003cstrong\u003eA) \u003c/strong\u003eHeat map with\u003cstrong\u003e \u003c/strong\u003ecolor gradient represents relative abundance of each microbial genus, with darker red indicating higher abundance and darker blue representing lower abundance or absence. The dendrograms on the right and bottom illustrate hierarchical clustering based on the similarity of microbial composition across taxa and sample types, respectively. \u003cstrong\u003eB) \u003c/strong\u003eHeat map displaying the hierarchical clustering based on presence (red) or absence (blue) of organisms across different sample types. The dendrograms on right and bottom represent the similarity between microbial taxa and, sample type respectively, with more closely related clusters grouped together.\u003c/p\u003e","description":"","filename":"Figure9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/6ea2fefae524cef7adb20f37.jpg"},{"id":78751188,"identity":"cffa4807-64f4-402a-9409-053eca2e79b2","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":1094907,"visible":true,"origin":"","legend":"\u003cp\u003eHierarchical clustering heat maps showing the presence of selected enteric pathogens across samples. \u003cstrong\u003eA)\u003c/strong\u003e Heat map showing the presence of enteric species across different samples (columns represents individual samples, red cells indicating the presence of a pathogen and dark blue cells representing its absence. \u003cstrong\u003eB)\u003c/strong\u003e Hierarchical clustering analysis identifies groups of organisms that tend to co-occur mostly across human and chicken samples. Notably, \u003cem\u003eEscherichia coli, Salmonella enterica, Salmonella bongori, three species of Shigella (S. flexneri, S. sonnei, and Shigella sp. genomosp. SF-2015), \u003c/em\u003eand\u003cem\u003e Campylobacter jejuni\u003c/em\u003e form a cluster of species with higher co-occurrence patterns compared to other pathogens. Additionally, one infant sample with a history of diarrhea was found to harbor three species of \u003cem\u003eShigella\u003c/em\u003eas well as \u003cem\u003eCampylobacter jejuni\u003c/em\u003e, whereas four other infant samples clustered in this group showed no history of diarrheal illness.\u003c/p\u003e","description":"","filename":"Figure10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/63bb27895924ebb904094321.jpg"},{"id":78751178,"identity":"a7b0e2ed-f994-4efb-8ff2-28ea776f02dc","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"jpg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":659873,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of selected enteric pathogens identified from various sample types, \u003cstrong\u003eA)\u003c/strong\u003e human stools (infants, mothers, and siblings), animal fecal samples (cattle, goats, sheep, and chickens),\u003cstrong\u003e B)\u003c/strong\u003e distribution of enteric pathogens across the study Kebeles. The numbers in each bar represent the number of positive detections of the pathogen in the corresponding sample type (A) and Kebele in (B).\u003c/p\u003e","description":"","filename":"Figure11.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/20697fc71a2be442f8cf7e05.jpg"},{"id":95563899,"identity":"f0ccb0b6-66e3-4632-99e6-702d15364fa2","added_by":"auto","created_at":"2025-11-10 16:01:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8571689,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/808ca5f0-de27-4a9c-9180-a5f16aaada4b.pdf"},{"id":78751172,"identity":"7cf2572d-fb95-4f69-872b-86b8dff4996b","added_by":"auto","created_at":"2025-03-18 11:52:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":241213,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/fc94fd8d58780188c67c4016.pdf"},{"id":78751920,"identity":"baff5c0c-2a12-424d-b95e-40bfc38e0a25","added_by":"auto","created_at":"2025-03-18 12:00:50","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12397,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/d058186021ca40ccc9e6318a.xlsx"},{"id":78751168,"identity":"92f45661-f882-4268-b48d-f82136475056","added_by":"auto","created_at":"2025-03-18 11:52:50","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":11455,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5736322/v1/f3bb5aadaa066313096a32c1.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Host Clustering of Campylobacter Species and Other Enteric Pathogens in a Longitudinal Cohort of Infants, Family Members and Livestock in Rural Eastern Ethiopia","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEnteric diseases caused by bacterial, viral, or parasitic infections are the second leading cause of mortality globally, after lower respiratory tract infections, among children younger than five years (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Humans are susceptible to these disease-causing microbes, encompassing zoonotic foodborne, waterborne, nosocomial, or community-acquired infections. The World Health Organization (WHO) estimates that, contaminated food leads to 600\u0026nbsp;million illnesses and more than 400,000 deaths annually (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). One-third of these cases affect young children in low- and middle-income countries (LMIC), contributing to malnutrition, stunting, and cognitive deficits (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). In addition, the economic impact of enteric infections is substantial, resulting in an estimated total productivity loss in LMIC of US\u003cspan\u003e$\u003c/span\u003e95\u0026nbsp;billion per year, with an additional predicted annual cost of treating these infections reaching US\u003cspan\u003e$\u003c/span\u003e15\u0026nbsp;billion (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e) .\u003c/p\u003e \u003cp\u003eIn Ethiopia, where access to clean water and proper sanitation is limited, enteric infections pose a significant public health challenge (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). The WHO estimates 11 foodborne pathogen-associated deaths per 100,000 population in Ethiopia and diarrhea is the second most prevalent cause of mortality in infants (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Bacterial agents include \u003cem\u003eCampylobacter\u003c/em\u003e species, diarrheagenic \u003cem\u003eEscherichia coli\u003c/em\u003e, non-typhoidal \u003cem\u003eSalmonella enterica\u003c/em\u003e, \u003cem\u003eShigella\u003c/em\u003e species, while viral agents include rotavirus and norovirus, with children under the age of five being particularly vulnerable (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). In addition, \u003cem\u003eCampylobacter\u003c/em\u003e and other enteric infections, especially in children, are increasingly recognized as contributing to child malnutrition (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOur previous investigations, conducted in rural eastern Ethiopia within the sociodemographic context of the Haramaya district (in five kebeles), focused on characterizing the prevalence and distribution of \u003cem\u003eCampylobacter\u003c/em\u003e species, assessing their interactions with child health, specifically environmental enteric dysfunction (EED) and stunting (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). Our cross-sectional study revealed a high prevalence of \u003cem\u003eCampylobacter\u003c/em\u003e during early childhood. Employing metatranscriptomic shotgun sequencing approach, seven dominant \u003cem\u003eCampylobacter\u003c/em\u003e species were identified in children's stools (aged 12 to 14 months). The observed diversity suggested \u003cem\u003eCampylobacter\u003c/em\u003e colonization in children either occurred through multiple reservoirs or from a reservoir hosting several co-inhabiting \u003cem\u003eCampylobacter\u003c/em\u003e species (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBuilding upon these findings, our team embarked on a broader study spanning from December 2020 to June 2022 (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). This intensive longitudinal investigation included stool samples (from infants, mothers, and siblings), livestock feces (from cattle, chickens, goats, and sheep), and the environment (n\u0026thinsp;=\u0026thinsp;1,644). The result showed potential routes for \u003cem\u003eCampylobacter\u003c/em\u003e transmission in eastern Ethiopia and a correlation in the detection of \u003cem\u003eCampylobacter\u003c/em\u003e species between infants and mothers and environmental sources (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the current study, we focused on a subset of human stools and livestock feces samples (n\u0026thinsp;=\u0026thinsp;280; 40 per sample type) collected between December 2020 and June 2022. Utilizing shotgun metagenomics sequencing, \u003cem\u003eCampylobacter\u003c/em\u003e diversity and co-circulation was investigated in infants, human (sibling, mother), and livestock species. This method, which does not rely on traditional culture techniques, has been shown to accurately identify \u003cem\u003eCampylobacter\u003c/em\u003e at the species level, even in instances of low levels of genetic material in the samples (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Further, we also investigated the presence of other enteric pathogens and explored the co-occurrence of \u003cem\u003eCampylobacter\u003c/em\u003e species with enteric pathogens.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy area and design\u003c/h2\u003e \u003cp\u003eA longitudinal study involving 106 infants (ages ranging from 7 days to 13 months) was conducted from December 2020 to June 2022. Participants were selected randomly from 10 rural kebeles (synonymous with villages) of Haramaya woreda (synonymous with district), East Hararghe Zone, Oromia Region, Ethiopia. Detailed study area description has been published (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Households in the selected kebeles maintain a variety of livestock (chickens, cattle, sheep, and goats) living in proximity with family members of the household. In some cases, the livestock co-habit with humans within the same room inside the house, often uncontained, to avoid issues with predators outdoors. Based on this information, human stool samples, livestock feces, and environmental samples were collected over time from each household to assess the prevalence and load of \u003cem\u003eCampylobacter\u003c/em\u003e. Full details of the enrollment process and \u003cem\u003eCampylobacter\u003c/em\u003e detection at the genus level have be previously described (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). In short, a total of 1,073 infant stool samples were collected monthly from 2\u0026ndash;3 weeks after birth until 376 days of age. The first morning stools, minimum 10 grams but sometimes as little as 1 gram during the early months, were gathered in sterile plastic sheets within diapers that were sterilized with 12-hour UV light either on the same or preceding day. These samples were then transferred to 207-milliliter Whirl-Pak bags. For infants, diarrhea presence on sampling days and the day prior to specimen collection, along with the 24-hour loose stool count, were noted.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSample selection for shotgun metagenomics\u003c/h3\u003e\n\u003cp\u003eFor the metagenomic analysis, we included samples from households (n\u0026thinsp;=\u0026thinsp;106) that completed follow-up. We selected a total of 280 samples, 40 for each sample type (infants, mothers, siblings, cattle, goat, sheep and chickens) for metagenomic sequencing. We prioritized samples of good DNA quality, based on Nanodrop measurements (\u0026gt;\u0026thinsp;20 ng/ul, 260/280\u0026thinsp;\u0026gt;\u0026thinsp;1.8 and 260/230\u0026thinsp;\u0026lt;\u0026thinsp;1.5). We selected 50 households with at least one sample of each type. Further, we selected the latest available infant stool sample because our previous qPCR results (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) suggested the likelihood of detecting \u003cem\u003eCampylobacter\u003c/em\u003e increased with age, which resulted in 40 households with balanced distribution across the 10 kebeles in the study area. The number of households per kebele varied between 3 and 5. We included one sample from each of the other types, preferring the sample closest in time to the infant sample if available.\u003c/p\u003e \u003cp\u003e \u003cb\u003eExtraction of genomic DNA and\u003c/b\u003e \u003cb\u003eCampylobacter\u003c/b\u003e \u003cb\u003edetection using qPCR.\u003c/b\u003e\u003c/p\u003e \u003cp\u003eA QIAamp Power Fecal Pro DNA kit (Qiagen, CA, USA) was used and briefly, 0.25 g of the sample, as recommended by the manufacturer, was resuspended in 100 \u0026micro;L of nuclease-free water to extract genomic DNA from human stool samples and livestock feces. The quality and quantity of the DNA were analyzed using a UV5 Nanodrop spectrophotometer (Mettler Toledo, Columbus, OH). DNA with poor quality were cleaned using a Zymo genomic DNA clean and concentrator kit (Zymo Research, CA, USA) and stored at -20\u0026deg;C until further use. Aliquots of the DNA extracts were used for \u003cem\u003eCampylobacter\u003c/em\u003e spp. Detection at the genus level in the human stool samples and animal feces using TaqMan real-time PCR(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Detailed steps and results are described elsewhere (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eLibrary preparation and metagenomic sequencing\u003c/h3\u003e\n\u003cp\u003eLibrary size and concentration were determined using the 4150 Tapestation system (Agilent, MA, USA). Limited-cycle PCR was subsequently employed to amplify the tagged DNA and introduce sequencing indexes. To facilitate a limit of detection assessment for each sample, we incorporated PhiX Control v3 (Illumina, Inc, San deigo, CA, USA) into each sample prior to library preparation. The prepared libraries were loaded onto a reagent cartridge and subjected to clustering on the NextSeq 2000 System. Subsequently, a paired-end sequencing run with 2 \u0026times; 150 bp reads was executed using the NextSeq 2000 platform. The base calls generated by the NextSeq 2000 System were then transformed into FASTQ files, facilitating downstream metagenomic analysis.\u003c/p\u003e\n\u003ch3\u003eFASTQ assessment for sample and process evaluation\u003c/h3\u003e\n\u003cp\u003eWe conducted an initial assessment of quality-related metrics, including cluster density, q-scores, and the percentage of passed reads as provided by the sequencer, using FastQC v.11.7 (Babraham Bioinformatics). Additionally, we conducted a more comprehensive quality checks using the CZ-ID pipeline (mNGS Pipeline v7.1). This involved evaluating total reads, the percentage of high-quality reads in the total pool, and assessing sequence diversity through duplicate compression ratios. Total reads were utilized to estimate the read count per sample, while the \"passed QC\" metric represented the percentage of reads retained after applying quality filtering using fastp, which eliminated low-quality bases, short reads (\u0026lt;\u0026thinsp;35 bp), and reads with low complexity. To evaluate sequence diversity within the metagenomic samples, the Duplicate Compression Ratio (DCR) was calculated for each sequencing run. DCR quantifies the prevalence of duplicate reads in a sample. A high DCR value, indicating numerous duplicate reads, suggests a less diverse library and account for potential technical artifacts affecting diversity assessments.\u003c/p\u003e\n\u003ch3\u003eBioinformatic processing of the metagenomic data using the CZ-ID pipeline\u003c/h3\u003e\n\u003cp\u003eThe CZ-ID pipeline (Illumina mNGS Pipeline v7.1) was used for metagenomic analysis, incorporating a series of steps for thorough data cleanup and analysis refinement. Host sequences were filtered, and ERCC sequences were excluded using STAR as the initial processing steps. Subsequently, we trimmed sequencing adapters and illumina indexes through Trimmomatic. Additional filtering was then carried out using PriceSeq, enabling the retention of high-quality sequences using the Phred score values. Duplicate reads were identified and removed using czid-dedup. Following these steps, any residual host sequences were rigorously filtered out using Bowtie2. In cases where more than one million reads remained after this process (or two million for paired-end data), random subsampling was performed using unique or deduplicated reads. Finally, irrespective of host considerations, CZ ID filtered out human sequences using a combination of STAR, Bowtie2, and GSNAP, resulting in comprehensive data preparation and quality control pipeline. Sequences then subjected to metagenomic assembly to align and merge short sequencing reads obtained from a longer DNA sequence to reconstruct the original sequence. Initially, sequence alignment steps in CZ ID involved mapping reads to NCBI NT (nucleotide) and NR (protein) databases to obtain preliminary accession for each read and generating BLAST database containing all taxa identified as preliminary accessions. CZ ID implements SPAdes to assemble contigs. Each contig is composed of several sequencing reads and the pipeline maps quality-filtered reads back to the assembled contigs to determine which reads are associated with each contig using Bowtie2. This step involved aligning BLAST contigs against database of preliminary accessions generated in the previous steps. Finally, each read and the final accession (based on contig accession, or if the reads failed to assemble into contigs- based on the preliminary accession) mapped to taxa using NCBI accession to taxon database and generated compute statistics per each sample.\u003c/p\u003e \u003cp\u003eWe employed downstream filtering steps to eliminate potentially erroneous taxonomic assignments. Drawing from our experience in formative research (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e) and adhering to established community best practices, we employed specific criteria for taxon selection. Specifically, we used \"NT L\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026ge;\u003c/span\u003e\u0026thinsp;50\" to retain taxa with alignments of at least 50 base pairs in the nucleotide databases, \"NT R\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026ge;\u003c/span\u003e\u0026thinsp;10\" to preserve taxa with a minimum of 10 reads aligning to the nucleotide databases. Furthermore, to ensure data normalization and mitigate background noise, we incorporated a Z-score filter derived from a background model constructed using control samples used during DNA extraction (\u0026ldquo;Z-score\u0026thinsp;\u0026gt;\u0026thinsp;1\u0026rdquo;; taxa abundance is higher in the sample than in the control; \u0026ldquo;Z-score\u0026thinsp;=\u0026thinsp;100\u0026rdquo; taxa not found in the set of controls; \u0026ldquo;Z-score = -100\u0026rdquo; taxa only present in the controls). The outcomes of our metagenomic pipelines were comprehensively evaluated for meeting all three filtering criteria mentioned above. This approach helped to enhance the reliability of our taxonomic assignments in the metagenomic dataset.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses were performed using JMP PRO 16 software (SAS Institute, Cary, NC, USA). Metagenomic reads across the seven sample types were evaluated for normality and a one-way ANOVA was conducted to assess the differences in mean sequencing depths between the seven sample types. The Chi-square test was employed to compare the probability of detecting \u003cem\u003eCampylobacter\u003c/em\u003e species across various kebeles. Additionally, pairwise comparisons were conducted to estimate the likelihood of co-infection for specific pairs of \u003cem\u003eCampylobacter\u003c/em\u003e species detections. The abundance of each bacterium in the stool samples was estimated based on normalized read count, reads per million (rpm). Bi-variate analysis was utilized to examine the relationships between \u003cem\u003eCampylobacter\u003c/em\u003e load from qPCR (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e), with metagenomic abundance, rpm. A non-parametric Wilcoxon test was performed to identify rpm abundance differences for a given member of the microbial species. Hierarchal clustering, multivariate analysis combined with Pearson pairwise correlation coefficient and principal component analysis, were employed to uncover patterns and associations within the microbiome dataset. For selected \u003cem\u003eCampylobacter\u003c/em\u003e taxa, a pairwise distance matrix was generated to depict genetic relatedness between taxa identified from metagenomic analysis and inform potential source attribution. For this analysis, samples with at least one \u003cem\u003eCampylobacter\u003c/em\u003e contig per individual were selected. Genetic distances between selected samples were determined by calculating k-mer-based reference-free genomic distance estimation using split k-mer analysis (SKA)(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eMetagenomic sequencing depth and read distribution in the seven sample types\u003c/h2\u003e \u003cp\u003eExamination of reads from stool samples (infants, mothers, and siblings), and fecal samples from livestock (cattle, goats, sheep, and chickens) demonstrated a mean sequencing depth of 1.06 x 10\u003csup\u003e8\u003c/sup\u003e (95% CI, 1.02\u0026ndash;1.09 x 10\u003csup\u003e8\u003c/sup\u003e), with highest sequencing depth of 1.5 x 10\u003csup\u003e8\u003c/sup\u003e per sample. In Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA, the distribution of total sequencing depth achieved is illustrated, with a comparison of means across different sample types. Chicken feces exhibited the highest mean sequencing depth (1.2 x 10\u003csup\u003e8\u003c/sup\u003e reads, 95% CI: 1.1 x 10\u003csup\u003e8\u003c/sup\u003e \u0026ndash; 1.2 x 10\u003csup\u003e8\u003c/sup\u003e) and the narrowest distribution, indicating consistent sequencing quality, while infant stool showed the lowest mean sequencing depth (9.8 x 10\u003csup\u003e7\u003c/sup\u003e reads, 95% CI: 8.4 x 10\u003csup\u003e7\u003c/sup\u003e \u0026ndash; 1.0 x 10\u003csup\u003e8\u003c/sup\u003e) with greater variability. However, a one-way ANOVA test (p\u0026thinsp;=\u0026thinsp;0.06) did not reveal significant differences in means, the low p-values suggest a tendency for chicken feces to exhibit a higher mean sequencing depth compared to other samples. We also examined the complexity of sequencing reads in each sample as a measure of microbial biomass diversity by calculating the duplicate compression ratio (DCR), which represents the ratio of sequences before and after the removal of duplicated reads. Notably, all the samples exhibited a DCR ratio of \u0026lt;\u0026thinsp;2, underscoring the presence of diverse sequences and absence of biases from PCR amplification during the sequencing process \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eDetection of\u003c/b\u003e \u003cb\u003eCampylobacter\u003c/b\u003e \u003cb\u003especies in human stool and animal feces through shotgun metagenomics\u003c/b\u003e\u003c/p\u003e \u003cp\u003eOverall, with the exception of 17 samples excluded from analysis due to not meeting sequence quality thresholds, \u003cem\u003eCampylobacter\u003c/em\u003e was detected in all remaining samples (n\u0026thinsp;=\u0026thinsp;263/280). For thirteen of these samples, detection was restricted to the genus level, while in 250 samples, \u003cem\u003eCampylobacter\u003c/em\u003e was identified at the species level. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, shows the frequency of \u003cem\u003eCampylobacter\u003c/em\u003e species detected in stool (infants, mothers, and siblings) and fecal samples (cattle, goats, sheep, and chickens). In total, we detected 21 dominant \u003cem\u003eCampylobacter\u003c/em\u003e species across human and animal hosts. This included the detection of multiple \u003cem\u003eCampylobacter\u003c/em\u003e species in infants (n\u0026thinsp;=\u0026thinsp;5), mothers (n\u0026thinsp;=\u0026thinsp;8), and siblings (n\u0026thinsp;=\u0026thinsp;8). In ruminant feces, the diversity of \u003cem\u003eCampylobacter\u003c/em\u003e species varied, with sheep (n\u0026thinsp;=\u0026thinsp;11), cattle (n\u0026thinsp;=\u0026thinsp;10), and goats (n\u0026thinsp;=\u0026thinsp;8) species. Chicken samples showed three \u003cem\u003eCampylobacter\u003c/em\u003e species.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eC. jejuni\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;76, 30%) and \u003cem\u003eC. concisus\u003c/em\u003e (exception for chicken feces, n\u0026thinsp;=\u0026thinsp;49, 14%) were the most prevalent species detected, indicating their widespread presence in both human and animal hosts. Other highly prevalent species included, \u003cem\u003eCandidatus\u003c/em\u003e Campylobacter infans \u003cem\u003e(C. infans)\u003c/em\u003e, \u003cem\u003eC\u003c/em\u003e.\u003cem\u003evicugnae, C. coli\u003c/em\u003e, and \u003cem\u003eC. showae\u003c/em\u003e. However, other thermophilic \u003cem\u003eCampylobacter\u003c/em\u003e species such as \u003cem\u003eC. lari\u003c/em\u003e and \u003cem\u003eC. upsaliensis\u003c/em\u003e, were also detected but not frequently with 1.6\u0026ndash;2.8% prevalence. Other less frequently detected species include \u003cem\u003eC. iguaniorum\u003c/em\u003e (infant), \u003cem\u003eC. mucosalis\u003c/em\u003e (mother), \u003cem\u003eC. rectus\u003c/em\u003e (sheep), \u003cem\u003eC\u003c/em\u003e. sp. 2014D-0216 (goat), \u003cem\u003eC\u003c/em\u003e. \u003cem\u003ecaliforniensis\u003c/em\u003e (sheep), \u003cem\u003eC\u003c/em\u003e. \u003cem\u003eRM16192\u003c/em\u003e (sheep), \u003cem\u003eC. volucris\u003c/em\u003e (mother) and \u003cem\u003eC. pinnepideorioum\u003c/em\u003e (chicken) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003eCampylobacter\u003c/b\u003e \u003cb\u003especies in stool and fecal samples from different sources\u003c/b\u003e\u003c/p\u003e \u003cp\u003eTwo-way hierarchical clustering revealed distinct patterns in the detection of \u003cem\u003eCampylobacter\u003c/em\u003e species across the seven sample types \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. This analysis identified three major clusters among the 21 \u003cem\u003eCampylobacter\u003c/em\u003e species. The first cluster, labeled C1, represents a mixed-host group, including species such as \u003cem\u003eC. jejuni, C. concisus\u003c/em\u003e, and \u003cem\u003eC. coli\u003c/em\u003e, which were detected in both human and livestock samples. The presence of these species in different sample types indicate niche overlap for these \u003cem\u003eCampylobacter\u003c/em\u003e species suggesting transmission across hosts. The second group is human associated cluster, labeled as C2 and includes \u003cem\u003eCampylobacter\u003c/em\u003e species exclusively found in human samples. It included \u003cem\u003eC. infans\u003c/em\u003e and \u003cem\u003eC. upsaliensis\u003c/em\u003e, which are present in infant, mother and sibling stools. Other species, such as \u003cem\u003eC. geochelonis\u003c/em\u003e, was detected in sibling and mother stools, while \u003cem\u003eC. mucosalis\u003c/em\u003e and \u003cem\u003eC. volucris\u003c/em\u003e were found in mother samples, and \u003cem\u003eC. iguaniorum\u003c/em\u003e was identified only in infant samples. The third group is a ruminant associated cluster labeled as C3, containing 12 \u003cem\u003eCampylobacter\u003c/em\u003e species, among them the recently described species C. \u003cem\u003evicugnae\u003c/em\u003e, and \u003cem\u003eC. californiensis\u003c/em\u003e, (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) as well as long established species such as \u003cem\u003eC. showae\u003c/em\u003e and \u003cem\u003eC. lari\u003c/em\u003e (also detected in sibling stools) were detected. Unlabeled as clusters, \u003cem\u003eC. fetus\u003c/em\u003e and \u003cem\u003eC. pinnipediorum\u003c/em\u003e were each detected in only single instances per sample type. \u003cem\u003eC. fetus\u003c/em\u003e was identified in a sample from a sibling and in cattle. Meanwhile, \u003cem\u003eC. pinnipediorum\u003c/em\u003e was exclusively detected in chicken and cattle feces.\u003c/p\u003e \u003cp\u003eNotably, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, these clusters are driven by a strong positive correlation in \u003cem\u003eCampylobacter\u003c/em\u003e species detection among ruminant samples (r\u0026thinsp;\u0026gt;\u0026thinsp;0.34) and within human samples (infant: sibling, r\u0026thinsp;=\u0026thinsp;0.88, infant: mother, r\u0026thinsp;=\u0026thinsp;0.57), indicating a high degree of species overlap within these groups. Additionally, \u003cem\u003eCampylobacter\u003c/em\u003e species from chicken samples showed a strong correlation with human samples (chicken: infant r\u0026thinsp;=\u0026thinsp;0.54, chicken: sibling, r\u0026thinsp;=\u0026thinsp;0.61, chicken: mother, r\u0026thinsp;=\u0026thinsp;0.76) but only a moderate correlation with cattle samples (r\u0026thinsp;=\u0026thinsp;0.46). There was no significant correlation observed between \u003cem\u003eCampylobacter\u003c/em\u003e species from chickens and small ruminant samples (r\u0026thinsp;=\u0026thinsp;0 and 0.15, goat and sheep, respectively), suggesting distinct \u003cem\u003eCampylobacter\u003c/em\u003e profiles in these groups.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMultiple\u003c/b\u003e \u003cb\u003eCampylobacter\u003c/b\u003e \u003cb\u003especies co-occur in stool and fecal samples\u003c/b\u003e\u003c/p\u003e \u003cp\u003eA key advantage of the metagenomic approach is its ability to investigate co-infections across all samples and seven sample types, enabling profiling of both dominant and less abundant \u003cem\u003eCampylobacter\u003c/em\u003e species within sample. Our findings reveal a broader range of \u003cem\u003eCampylobacter\u003c/em\u003e species other than the 21 most dominant ones identified across the seven sample types (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/b\u003e Complete list of \u003cem\u003eCampylobacter\u003c/em\u003e species detected in individual samples are shown in \u003cb\u003eTable \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e. For the purpose of co-infection analysis, we limited our focus to the 21 most dominant \u003cem\u003eCampylobacter\u003c/em\u003e species described in the previous section. Interestingly the presence of distinct clusters along the y-axis indicate that \u003cem\u003eCampylobacter\u003c/em\u003e species are commonly found together across multiple samples \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA\u003cb\u003e)\u003c/b\u003e. Species such as \u003cem\u003eC. jejuni, C. concisus\u003c/em\u003e, and \u003cem\u003eC. coli\u003c/em\u003e that were widely represented, showed notable co-occurrence patterns across multiple stool and fecal samples. In contrast, species like \u003cem\u003eC. vicugnae\u003c/em\u003e and C. \u003cem\u003esp. RM12175\u003c/em\u003e demonstrate distinct co-occurrence patterns, particular to small ruminant samples. Using probability co-occurrence matrix, we identified specific pairs of \u003cem\u003eCampylobacter\u003c/em\u003e species in various stool and fecal samples from different sources with higher likelihood of co-occurrence (\u003cem\u003eC. jejuni and C. coli\u003c/em\u003e), (\u003cem\u003eC. jejuni and C. concisus\u003c/em\u003e) and (\u003cem\u003eC. jejuni and C. showae\u003c/em\u003e) with a co-occurrence probability of (p\u0026thinsp;\u0026gt;\u0026thinsp;0.5) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eAbundance of key\u003c/b\u003e \u003cb\u003eCampylobacter\u003c/b\u003e \u003cb\u003especies in stool and fecal samples and correlation with qPCR detection\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cem\u003eCampylobacter\u003c/em\u003e abundance across seven sample types was investigated using a standardized nucleotide count, reads per million (rpm) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e\u003cb\u003e).\u003c/b\u003e Samples with nt.rpm exceeding 1000 were exclusively derived from humans, including the infant stools Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. \u003cem\u003eC. infans\u003c/em\u003e was detected at higher rpm compared to the overall median in this study (p\u0026thinsp;=\u0026thinsp;0.035, Kruskal-Walli\u0026rsquo;s test). In ruminants, despite the presence of diverse \u003cem\u003eCampylobacter\u003c/em\u003e species, the average abundance (average nt. rpm) was notably lower than in human stools using nonparametric comparisons by Wilcoxon test. We found a significant difference in the median rpm between infants and the livestock samples (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The average count for \u003cem\u003eCampylobacter\u003c/em\u003e-specific reads in ruminant species remained below 50 rpm. \u003cem\u003eC. vicugnae\u003c/em\u003e (in cattle, goat, and sheep), \u003cem\u003eC. jejuni\u003c/em\u003e, and \u003cem\u003eC. curvus\u003c/em\u003e (in cattle) exhibited statistically higher abundance compared to other \u003cem\u003eCampylobacter\u003c/em\u003e species (p\u0026thinsp;=\u0026thinsp;0.01, non-parametric Wilcoxon/Kruskal-Walli\u0026rsquo;s test). Using sliding window analysis, we also found agreements between metagenomic read abundance (rpm) and the Ct values of genus level qPCR assay used for \u003cem\u003eCampylobacter\u003c/em\u003e detection, which was previously reported by Delais et al 2023 (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows, distribution of Ct values for the genus level qPCR assay against the metagenomic rpm values. This inverse relationship between Ct values of 19\u0026ndash;26, yielded statistically significant correlation (Pearson correlation coefficient of -53% (95% CI: -0.68 to -0.31), p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) with the metagenomic read abundance rpm. In those samples with Ct of 19\u0026ndash;26, the average read count was 200 rpm with a median value of 93 rpm.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eGenetic relationships of selected\u003c/b\u003e \u003cb\u003eCampylobacter\u003c/b\u003e \u003cb\u003especies across diverse sample types\u003c/b\u003e\u003c/p\u003e \u003cp\u003eWe investigated genetic similarities of selected \u003cem\u003eCampylobacter\u003c/em\u003e species, specifically focusing on most widely detected species across the seven sample types: \u003cem\u003eC. jejuni, C. concisus, C. infans, and C. sp. vicugnae\u003c/em\u003e. For \u003cem\u003eC. jejuni\u003c/em\u003e, three major clusters (C1-3) were identified, shedding light on genetic similarities of circulating species \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. Cluster C1 includes \u003cem\u003eC. jejuni\u003c/em\u003e exhibiting phylogenetic relationships with known reference isolates, specifically \u003cem\u003eCampylobacter jejuni subsp. jejuni\u003c/em\u003e strain:00-1597 (Accession ID_CP010306) and \u003cem\u003eCampylobacter jejuni subsp. doylei\u003c/em\u003e NCTC 11924 [LR134530] These isolates are associated with human enteritis, with CP010306 isolated from a human in Canada (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) and sequence submitted from a child with diarrhea in the United Kingdom. Additional samples in C1 cluster include samples primarily originating from human sources and two of the samples (1119, mother and 785, infant) were from the same household 16. C1 also includes samples from cattle and sheep sources, suggesting potential cross transmission of \u003cem\u003eC. jejuni\u003c/em\u003e across human and ruminant hosts.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSimilarly, in C2, the \u003cem\u003eC. jejuni\u003c/em\u003e cluster is predominantly constituted from chicken sources. The three reference strains identified in this cluster include \u003cem\u003eC. jejuni\u003c/em\u003e strain CFSAN054107 (accession: CP028185.1) from unknown source and \u003cem\u003eC. jejuni\u003c/em\u003e strain C57 and strain C220 (with accession CP059970.1 and CP059972.1 respectively), derived from the cloacal swab of a chicken host. Overall, this cluster exhibits some host structuring, with samples from chicken showing genetic relatedness as measured in the relative distance (Mash-like distances\u0026thinsp;\u0026lt;\u0026thinsp;0.1) with each other. In addition, the C2 cluster also showed a spatial signature with four of these sample being from the same kebele (Biftu Geda) and three of the four also being from the same household. The three samples that shared household were from mother and chicken source, indicating potential cross transmission of \u003cem\u003eC. jejuni\u003c/em\u003e between hosts. In contrast, \u003cb\u003eC3\u003c/b\u003e showed host structuring with all \u003cem\u003eC. jejuni\u003c/em\u003e in the group coming from human stool including two infant sources.\u003c/p\u003e \u003cp\u003eFor \u003cem\u003eC.infans\u003c/em\u003e, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB, the data reveals a cluster encompassing stool samples from infants, mothers, and siblings. This cluster is divided into two main groups, C1 and C2, with further subdivisions into subclusters C1.1 and C1.2, where household level concordance was observed for some of the samples within these clusters. In Cluster C1, samples from Households 3, 45, 84, and 111 demonstrate genetic clustering of \u003cem\u003eC.\u003c/em\u003e infans. Specifically, in Household 3, the samples represent an infant and a mother; in Household 45 (located in Amuna Kebele, within C1.2), infant and sibling stool samples show genetic similarity. Additionally, Household 84 in Kuro Kebele contains clustered stool samples from a mother and a sibling, and Household 111 includes samples from a mother and a sibling as well. These instances of genetic and spatial clustering suggest potential household-level cross transmission of \u003cem\u003eC.\u003c/em\u003e infans. Furthermore, in Household 135, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB, stool samples from an infant and a sibling appear genetically divergent despite the shared household environment. This finding indicates the presence of wider genetic diversity of \u003cem\u003eC.\u003c/em\u003e infans, with strains circulating in the same spatial setting.\u003c/p\u003e \u003cp\u003eFor \u003cem\u003eC. concisus\u003c/em\u003e two clusters, C1 and C2, were identified (\u003cb\u003eFig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA\u003c/b\u003e), with notable host structuring within each. Cluster C1 comprised \u003cem\u003eC. concisus\u003c/em\u003e isolates predominantly from small ruminants (sheep and goats), whereas cluster C2 grouping, was primarily associated with sibling and maternal stool samples. In contrast, \u003cem\u003eC. vicugnae\u003c/em\u003e was identified only in goat and sheep sources, and genetic variation was observed within small ruminant hosts (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eComparative analysis of stool microbiome diversity and shared bacterial communities across seven sample types\u003c/h2\u003e \u003cp\u003eFollowing the application of metagenomic filters, our analysis revealed the detection of more than 3,844 genera. Overall, the microbial composition showed a consistent bacterial dominance in the microbiota composition, with variations in other kingdoms depending on sample type. Bacterial communities constituted over 85% of sequences across all groups, with sheep and goat feces having the highest bacterial detection (92.49% and 92.91%, respectively). Eukaryote presence varied, ranging from 4.56% in goat feces to 10.38% in chicken feces. Archaeal communities were most abundant in cattle feces (2.97%), while infant stool exhibited the lowest archaeal presence (0.26%). Viral representation (mostly phages), though minimal in most samples, was notably higher in infant stool (5.52%) and sibling stool (3.06%) compared to other groups, where viruses ranged from 0.29\u0026ndash;2.35%.\u003c/p\u003e \u003cp\u003eComparisons of means for alpha diversity indices (Shannon and Simpson diversity) revealed significant differences in microbial diversity and evenness across the seven sample types (ANOVA, Shannon \u0026amp; Simpson, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Further pairwise comparisons using Tukey\u0026rsquo;s HSD indicated that infant stool samples exhibited significantly lower microbial diversity and richness compared to all other sample types (Tukey HSD for all pairs, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). For b-diversity, we used, principal component analysis (PCA) that differentiated the microbial communities into two main components explaining 62.5% of the total diversity (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) being as a result of sample types (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). Component 1 explained approximately 50% of the diversity in the total samples. This component consisted of 92% from the microbial community in sheep, 88% in goats, and 66% in cattle, with a minority of 0.03% in chicken. This demonstrates the substantial contribution of ruminant samples. In contrast, component 2, which explained about 13% of variability within the samples, was loaded from human stool (53% in siblings, 51% in mothers, and 11% in infants) and 44% from chicken feces. Furthermore, Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB, explains these differences by showing relationship of individual sample based on variation in microbial community structure across the seven sample types as denoted by clustering of similar colors.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eNotably, this analysis highlighted similarities between the microbial profiles of infant stool and chicken samples compared to other livestock samples, with maternal stool showing the closest relationship to the infant (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). Cluster analysis from the Bray Curtis dissimilarity metrics indicated that while infant stools shared a core microbiome with maternal and sibling samples, there was also a more pronounced compositional divergence. The infant microbiome is distinctly characterized by a high presence of \u003cem\u003eBifidobacterium, Collinsella, Streptococcus\u003c/em\u003e, and \u003cem\u003eOlsenella\u003c/em\u003e species (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eA\u003cb\u003e).\u003c/b\u003e These four genera are considerably less prevalent in the stools of mothers and siblings, and are absent in livestock samples, except for a single detection of \u003cem\u003eStreptococcus\u003c/em\u003e spp.in sheep feces which underscores the differences in microbial composition between infants and mothers and siblings.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAnalysis of the top 50 most abundant microbial taxa in infant stool revealed a distinct cluster uniquely present in human stool samples and absent in all livestock samples. This cluster includes the genera \u003cem\u003eBifidobacterium, Collinsella, Veillonella, Megasphaera, Bruquia, Stenotrophomonas, Prevotella, Haemophilus, Ureaplasma, Dialister\u003c/em\u003e, and \u003cem\u003eFaecalibacterium\u003c/em\u003e. Chicken feces exhibited microbial similarities with human stool samples (infant, mother, and sibling), sharing genera such as \u003cem\u003eLigilactobacillus, Rothia, Leclercia, Sutterella, Coriobacterium, Yokenella, Allisonella, Blautia, Dictyocoela\u003c/em\u003e, and \u003cem\u003eAnaerostipes\u003c/em\u003e. Additionally, \u003cem\u003eEuzebya\u003c/em\u003e was shared exclusively with infant stool, while \u003cem\u003eSanguibacter\u003c/em\u003e was shared with infant and sibling stools. In contrast, genera such as \u003cem\u003eShumwellia, Alcanivorax, Lonsdalea\u003c/em\u003e, and \u003cem\u003eAlkaliphilus\u003c/em\u003e appeared variably across most sample types \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eB\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003eSub-group analysis of enteric microbiome based on findings from the MAL-ED and GEMS studies.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThis sub-group analysis investigated the presence and co-occurrence of selected enteric pathogens across seven sample types, informed by pathogen selection at the genus level based on findings from the MAL-ED and GEMS studies and WHO FERG estimates, which highlight pathogens of concern for infant health in developing regions(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). The analysis focused on common enteric pathogens, \u003cem\u003eShigella, Vibrio, Salmonella, Entamoeba, Cryptosporidium\u003c/em\u003e, and \u003cem\u003eGiardia\u003c/em\u003e. Hierarchical clustering revealed distinct co-occurrence patterns among these pathogens, including \u003cem\u003eC. jejuni\u003c/em\u003e. These bacterial co-occurrence patterns involving eight species, \u003cem\u003eSalmonella enterica, Salmonella bongori, Shigella flexneri, Shigella sonnei, Shigella sp. genomosp. SF-2015, and C. jejuni\u003c/em\u003e showed a high tendency to co-occur across various sample types, particularly in human stool and chicken feces \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eA\u003cb\u003e).\u003c/b\u003e Based on reported cases of infant diarrhea (as reported by the mother) within the clustered samples, co-detection of three \u003cem\u003eShigella\u003c/em\u003e species alongside \u003cem\u003eC. jejuni\u003c/em\u003e in one diarrheal sample was obaserved. In contrast, four additional infant samples within the same cluster had no reported diarrheal history \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eB\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn total, \u003cem\u003eSalmonella\u003c/em\u003e spp. (\u003cem\u003eS. enterica and S. bongori\u003c/em\u003e) was detected in 30 samples, with a detection rate of 11.5%. Both species co-occurred in 13 samples, and their presence was most notable in chicken feces, as well as in stool samples from infants and siblings. \u003cem\u003eShigella\u003c/em\u003e spp. were identified in 20.7% of samples (n\u0026thinsp;=\u0026thinsp;54), with the four major species: \u003cem\u003eS. dysenteriae, S. flexneri, S. sonnei\u003c/em\u003e, and \u003cem\u003eS. boydii\u003c/em\u003e comprising 71% of detections. \u003cem\u003eS. flexneri\u003c/em\u003e was the most prevalent species, with \u003cem\u003eShigella\u003c/em\u003e spp. detected across all sample types except for cattle, and the highest burden observed in infant stool (n\u0026thinsp;=\u0026thinsp;31), sibling stool (n\u0026thinsp;=\u0026thinsp;30), and chicken feces (n\u0026thinsp;=\u0026thinsp;28) (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eA \u003cb\u003e\u0026amp;B\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eVibrio\u003c/em\u003e spp. was detected in 117 samples across all sample types, representing 56 species, with \u003cem\u003eVibrio cholerae\u003c/em\u003e found in a limited number of samples (n\u0026thinsp;=\u0026thinsp;7), including sheep feces (n\u0026thinsp;=\u0026thinsp;3), and stool samples from mothers (n\u0026thinsp;=\u0026thinsp;2) and infants (n\u0026thinsp;=\u0026thinsp;2). \u003cem\u003eVibrio parahaemolyticus\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;25) was frequently detected across a broader range of sample types, though there was no co-detection of \u003cem\u003eV. cholerae\u003c/em\u003e and V. \u003cem\u003eparahaemolyticus\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eA \u003cb\u003e\u0026amp;B\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eAmong parasitic agents, \u003cem\u003eEntamoeba\u003c/em\u003e spp. was detected in 20 samples, with species including \u003cem\u003eE. histolytica, E. dispar, E. invadens, and E. nuttalli\u003c/em\u003e. \u003cem\u003eCryptosporidium muris\u003c/em\u003e was found in a mother stool, while \u003cem\u003eGiardia intestinalis\u003c/em\u003e was detected in an infant stool sample \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eA \u003cb\u003e\u0026amp;B).\u003c/b\u003e Viral detection was minimal overall, with \u003cem\u003eHuman mastadenovirus\u003c/em\u003e found in both infant and sibling stool samples. Phage-related viral detections were common, with notable exceptions were \u003cem\u003eParapoxvirus\u003c/em\u003e, Orf virus, in ruminant samples and \u003cem\u003eGallid herpesvirus 1\u003c/em\u003e in chicken samples. The Orf virus had the highest overall detection (n\u0026thinsp;=\u0026thinsp;19), particularly in goat feces (n\u0026thinsp;=\u0026thinsp;12).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003e \u003cem\u003eCampylobacter\u003c/em\u003e infection is a significant cause of diarrheal illness among children under five in LMIC including Ethiopia (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). Systematic reviews and meta-analysis indicate that around 10% of children in this age group are affected by \u003cem\u003eCampylobacter\u003c/em\u003e, with prevalence in animals reaching 14.6% compared to 9% in humans, suggesting its zoonotic nature (\u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). Prior research primarily focused on thermophilic \u003cem\u003eCampylobacter\u003c/em\u003e species like \u003cem\u003eC. jejuni, C. coli, C. lari\u003c/em\u003e, and \u003cem\u003eC. upsaliensis\u003c/em\u003e found in humans and domestic animals, while the prevalence of non-thermophilic species, such as \u003cem\u003eC. hyointestinalis, Candidatus C. infans, C. fetus, C. showae\u003c/em\u003e, and \u003cem\u003eC. concisus\u003c/em\u003e, remains largely unexamined.\u003c/p\u003e \u003cp\u003eThis study leveraged samples from a cohort of 106 infants in rural eastern Ethiopia, representing households across 10 of Haramaya\u0026rsquo;s 36 rural kebeles, for shotgun metagenomic sequencing. A recent report from our group indicated detection by qPCR of \u003cem\u003eCampylobacter\u003c/em\u003e at the genus level in 64% of infant stool samples, with prevalence increasing with age as reported previously (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). By analyzing stool samples from infants, their associated environment (mothers, siblings, and livestock feces \u0026lsquo;[from cattle, goats, sheep, and chickens), we captured a detailed view of \u003cem\u003eCampylobacter\u003c/em\u003e infection dynamics and other enteric pathogens across multiple hosts, shedding light on potential pathways of early life colonization.\u003c/p\u003e \u003cp\u003eMetagenomic sequencing identified 21 major \u003cem\u003eCampylobacter\u003c/em\u003e species across different sample types. Among these, five species, \u003cem\u003eC. jejuni, C. concisus, C. coli, C. infans\u003c/em\u003e, and \u003cem\u003eC. upsaliensis\u003c/em\u003e were found to colonize infants. The result highlighted early exposure of infants to diverse \u003cem\u003eCampylobacter\u003c/em\u003e strains from both human (mother and sibling) and livestock sources, highlighting the effectiveness of metagenomic approaches in characterizing \u003cem\u003eCampylobacter\u003c/em\u003e diversity and colonization patterns often missed by traditional methods (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). Notably, the detection of species like \u003cem\u003eC. concisus\u003c/em\u003e and \u003cem\u003eC. infans\u003c/em\u003e, especially in infants, is significant; \u003cem\u003eC. infans\u003c/em\u003e is a relatively new proposed species, while \u003cem\u003eC. concisus\u003c/em\u003e has been linked with chronic gastrointestinal conditions such as inflammatory bowel disease and has been observed in cases of gastroenteritis particularly in young children and immunocompromised patients (\u003cspan additionalcitationids=\"CR34\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). Likewise, C. \u003cem\u003einfans\u003c/em\u003e was identified as the second most dominant species in breastfed infants in a global enteric multicenter study and has been associated with diarrhea in humans and non-human primates (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Additionally, C. \u003cem\u003eupsaliensis\u003c/em\u003e is relatively rare in infants, with its isolation often associated with dogs and cats (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). However, these findings originate from developed countries, where both healthy and sick dogs and cats have been identified as sources of \u003cem\u003eC. upsaliensis\u003c/em\u003e infection in humans (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e) In African contexts, research on \u003cem\u003eC. upsaliensis\u003c/em\u003e is limited. A study in a rural South African setting did report the presence of \u003cem\u003eC. upsaliensis\u003c/em\u003e in dogs, indicating that pets could similarly act as reservoirs in African regions (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). However, overall, data supporting \u003cem\u003eC. upsaliensis\u003c/em\u003e reservoir hosts in rural Africa setting remains sparse.\u003c/p\u003e \u003cp\u003eThe study also found host specific and mixed host infection patterns among \u003cem\u003eCampylobacter\u003c/em\u003e species. The detection of \u003cem\u003eC. jejuni\u003c/em\u003e across all sample types underscores its ecological versatility and adaptive potential, aligning with previous findings on the zoonotic nature of this species(\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e).\u003cem\u003eC. jejuni\u003c/em\u003e is well recognized for its host diversity and environmental resilience, which contribute to its status as one of the most prevalent causes of bacterial gastroenteritis worldwide (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). The mixed host clustering of \u003cem\u003eC. coli and C. concisus\u003c/em\u003e also highlights their potential for cross-host transmission. Although \u003cem\u003eC. coli\u003c/em\u003e is primarily associated with livestock, particularly pigs and poultry, it has been increasingly detected in humans and other animal hosts, indicating its adaptability and potential zoonotic threat (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e). In this study, \u003cem\u003eC. coli\u003c/em\u003e was detected in both infant stool and goat feces, reflecting its potential for mixed-host infection where direct or indirect contact between humans and livestock may have facilitated transmission. On the other hand, \u003cem\u003eC. concisus\u003c/em\u003e was detected in all sample types except chicken feces, pointing to its relatively broad host range with a preference for mammalian hosts.\u003c/p\u003e \u003cp\u003eA major finding in this study is also the co-infection patterns among \u003cem\u003eCampylobacter\u003c/em\u003e species revealed significant associations, particularly for \u003cem\u003eC. jejuni\u003c/em\u003e, which demonstrated a high probability (P\u0026thinsp;\u0026gt;\u0026thinsp;0.5) of co-occurrence with \u003cem\u003eC. coli, C. concisus\u003c/em\u003e, and \u003cem\u003eC. showae\u003c/em\u003e. These observations indicate the potential for interactions within the host environment, where multiple \u003cem\u003eCampylobacter\u003c/em\u003e species may simultaneously colonize, potentially leading to complex infection dynamics influencing disease conditions in infants. These co-infections are likely due to synergistic effects or shared virulence mechanisms that facilitate persistence and survival within host microbiomes (\u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). High co-occurrence probability of these pairs highlights the need for further investigation into the implications of such mixed \u003cem\u003eCampylobacter\u003c/em\u003e infections in infant health.\u003c/p\u003e \u003cp\u003eIn addition to the \u003cem\u003eCampylobacter\u003c/em\u003e species, our study showed that \u003cem\u003eC. jejuni\u003c/em\u003e exhibited strong clustering with other enteric pathogens, including \u003cem\u003eSalmonella enterica, Salmonella bongori\u003c/em\u003e, and multiple \u003cem\u003eShigella\u003c/em\u003e species (\u003cem\u003eS. flexneri, S. sonnei, and Shigella sp. genomosp. SF-2015\u003c/em\u003e). This clustering was particularly pronounced in human stool and chicken feces samples, suggesting shared environmental or dietary transmission routes. The frequent co-detection of \u003cem\u003eC. jejuni\u003c/em\u003e with these pathogens aligns with other studies that have observed overlapping transmission pathways and reservoirs, particularly in resource-limited settings where sanitation and food safety practices are less stringent (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eEven though the metagenomic investigation in this study was centered on \u003cem\u003eCampylobacter\u003c/em\u003e and key enteric pathogens, it was also complemented by a global microbiome analysis that revealed two distinct microbiota clusters. The first cluster (principal component) associated with ruminants and another consisting of human and chicken samples. Using PCA, we demonstrated that the chicken microbiome shares organisms including enteric pathogens with the infant microbiome while the ruminant microbiome is more distinct. Using multivariate analysis, we found that \u003cem\u003eCampylobacter\u003c/em\u003e species from chicken samples show a strong correlation with human stool samples (r\u0026thinsp;\u0026gt;\u0026thinsp;0.54) but only a moderate correlation with cattle (r\u0026thinsp;=\u0026thinsp;0.46). Further analysis of \u003cem\u003eC. jejuni\u003c/em\u003e sub-clusters revealed that human and chicken sources frequently grouped together, providing evidence of cross-host transmission between chickens and infants.\u003c/p\u003e \u003cp\u003eIn conclusion, the study provides evidence for cross-host zoonotic transmission potential of some \u003cem\u003eCampylobacter\u003c/em\u003e species while other species appeared to be host-restricted. There was significant correlation among the \u003cem\u003eCampylobacter\u003c/em\u003e species detected in human stools, suggesting a potential pathway for infant \u003cem\u003eCampylobacter\u003c/em\u003e colonization through exposure via sibling and mother. Also, strong correlation with chickens indicates they may be a direct source of \u003cem\u003eCampylobacter\u003c/em\u003e infection for humans, whereas cattle may also contribute to the environmental pool of \u003cem\u003eCampylobacter\u003c/em\u003e in multi-host environments.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEthical approval was obtained from the University of Florida Internal Review Board (IRB201903141); the Haramaya University Institutional Health Research Ethics Committee (COHMS/1010/3796/20), and the Ethiopia National Research Ethics Review Committee (SM/14.1/1059/20). Written informed consent was obtained from all participating households (husband and wife) using a form in the local language (Afan Oromo).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConsent for publication\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAvailability of data\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe metagenomics data is submitted to NCBI under Bioproject number PRJNA1127034 and will be made available once the manuscript is accepted.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCompeting interests\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Bill and Melinda Gates Foundation funded the CAGED project to address food insecurity issues in Ethiopia and Burkina Faso through the project Equip\u0026mdash;Strengthening Smallholder Livestock Systems for the Future (grant number OPP11755487). These funds are administered by the Feed the Future Innovation Lab for Livestock Systems, established with funding from the U.S. Agency for International Development (USAID) and co-led by the University of Florida\u0026rsquo;s Institute of Food and Agricultural Sciences and the International Livestock Research Institute. Support for the Feed the Future Innovation Lab for Livestock Systems is made possible by the generous support of the American people through USAID. The contents are the authors\u0026apos; responsibility and do not necessarily reflect the views of USAID or the U.S. Government. REDCap is hosted at the University of Florida Clinical and Translational Science Institute (CTSI) and supported by NIH National Center for Advancing Translational Sciences grant UL1TR000064. This project is funded by the U.S. Agency for International Development Bureau for Food Security under agreement number AID-OAA-L-15-00003 as part of Feed the Future Innovation Lab for Livestock Systems. Any opinions, findings, conclusions, or recommendations expressed here are those of the authors alone. Research reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, partly funded by the NIH National Center for Advancing Translational Sciences under award number UL1TR001427.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAuthor Contributions\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: AHH, GR, ZM; Data curation: ZM, AO, LD, Formal Analysis: ZM, AO, DC, ; Funding acquisition: AHH, GR; Investigation: \u0026nbsp;ZM, AO, BM, LD, NS; Methodology: ZM, LD, AO, NS, GR, AHH; Project Administration: ZM, GR, AHH, WG; Resources: WG, GR, AHH, Supervision: WG, GR, AHH; Writing \u0026ndash; original: ZM, AHH, GR; Writing \u0026ndash; review \u0026amp; editing: ZM, AO, LD, BM, NS, WG, AHH, GR. All authors approved the final version for submission.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAcknowledgments\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work is a result of the CAGED Research Team, whose members include:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAbadir Jemal Seran, Abdulmuen Mohammed Ibrahim, Bahar Mummed Hassen, Belisa Usmael Ahmedo, Cyrus Saleem, Dehao Chen, Efrah Ali Yusuf, Getnet Yimer, Ibsa Abdusemed Ahmed, Ibsa Aliyi Usmane, Jafer Kedir Amin, Jemal Yusuf Hassen, Kedir Abdi Hassen, Kunuza Adem Umer, Karah Mechlowitz, Kedir Teji Roba, Loic Deblais, Mussie Bhrane, Mark J. Manary, Mawardi M. Dawid, Mahammad Mahammad Usmail, Nigel P. French, Nur Shaikh, Sarah L. McKune, Xiaolong Li, Yenenesh Demisie Weldesenbet, Yang Yang. This study would not have been possible without cooperation of study communities and local administration of the study kebeles. We want to express our appreciation to the study households, the Community Advisory Board and all who supported the study directly or otherwise.\u003c/p\u003e\n\u003cp\u003eResearch reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, which was supported in part by the NIH National Center for Advancing Translational Sciences under award number UL1TR001427. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTroeger CE, Khalil IA, Blacker BF, Biehl MH, Albertson SB, Zimsen SRM, et al. Quantifying risks and interventions that have affected the burden of diarrhoea among children younger than 5 years: an analysis of the Global Burden of Disease Study 2017. The Lancet Infectious Diseases. 2020;20(1):37\u0026ndash;59.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO 2015. Estimates of the global burden of foodborne diseases: foodborne disease burden epidemiology reference group 2007\u0026ndash;2015 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.foodbornediseaseburden.org/ferg/estimates\u003c/span\u003e\u003cspan address=\"https://www.foodbornediseaseburden.org/ferg/estimates\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e: World Health Organization; [\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHavelaar AH, Kirk MD, Torgerson PR, Gibb HJ, Hald T, Lake RJ, et al. World Health Organization global estimates and regional comparisons of the burden of foodborne disease in 2010. PLoS medicine. 2015;12(12):e1001923.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirk MD, Pires SM, Black RE, Caipo M, Crump JA, Devleesschauwer B, et al. World Health Organization Estimates of the Global and Regional Disease Burden of 22 Foodborne Bacterial, Protozoal, and Viral Diseases, 2010: A Data Synthesis. PLOS Medicine. 2015;12(12):e1001921.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe MAL-ED Network Investigators. The MAL-ED Study: A Multinational and Multidisciplinary Approach to Understand the Relationship Between Enteric Pathogens, Malnutrition, Gut Physiology, Physical Growth, Cognitive Development, and Immune Responses in Infants and Children Up to 2 Years of Age in Resource-Poor Environments. Clinical Infectious Diseases2014. p. S193-S206.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Bank 2017. Drug-Resistant Infections: A Threat to Our Economic Future. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJaffee S, Henson S, Unnevehr L, Grace D, Cassou E. The safe food imperative: Accelerating progress in low-and middle-income countries: World Bank Publications; 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFenta SM, Nigussie TZ. Factors associated with childhood diarrheal in Ethiopia; a multilevel analysis. Archives of Public Health. 2021;79(1):123.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlemayehu K, Oljira L, Demena M, Birhanu A, Workineh D. Prevalence and Determinants of Diarrheal Diseases among Under-Five Children in Horo Guduru Wollega Zone, Oromia Region, Western Ethiopia: A Community-Based Cross-Sectional Study. Canadian Journal of Infectious Diseases and Medical Microbiology. 2021;2021:5547742.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO. WHO estimates of the global burden of foodborne diseases: foodborne disease burden epidemiology reference group 2007\u0026ndash;2015: World Health Organization; 2015.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePlatts-Mills JA, Kosek M. Update on the burden of Campylobacter in developing countries. Current opinion in infectious diseases. 2014;27(5):444.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZenebe T, Zegeye N, Eguale T. Prevalence of Campylobacter species in human, animal and food of animal origin and their antimicrobial susceptibility in Ethiopia: a systematic review and meta-analysis. Annals of clinical microbiology and antimicrobials. 2020;19(1):1\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTerefe Y, Deblais L, Ghanem M, Helmy YA, Mummed B, Chen D, et al. Co-occurrence of Campylobacter Species in Children From Eastern Ethiopia, and Their Association With Environmental Enteric Dysfunction, Diarrhea, and Host Microbiome. Frontiers in Public Health. 2020;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHavelaar AH, Brhane M, Ahmed IA, Kedir J, Chen D, Deblais L, et al. Unravelling the reservoirs for colonisation of infants with Campylobacter spp. in rural Ethiopia: protocol for a longitudinal study during a global pandemic and political tensions. BMJ Open. 2022;12(10):e061311.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeblais L, Ojeda A, Brhane M, Mummed B, Hassen KA, Ahmedo BU, et al. Prevalence and Load of the Campylobacter Genus in Infants and Associated Household Contacts in Rural Eastern Ethiopia: a Longitudinal Study from the Campylobacter Genomics and Environmental Enteric Dysfunction (CAGED) Project. Appl Environ Microbiol. 2023;89(7):e0042423.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParker CT, Schiaffino F, Huynh S, Paredes Olortegui M, Pe\u0026ntilde;ataro Yori P, Garcia Bardales PF, et al. Shotgun metagenomics of fecal samples from children in Peru reveals frequent complex co-infections with multiple Campylobacter species. PLoS Negl Trop Dis. 2022;16(10):e0010815.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndersen SC, Kiil K, Harder CB, Josefsen MH, Persson S, Nielsen EM, et al. Towards diagnostic metagenomics of Campylobacter in fecal samples. BMC Microbiol. 2017;17(1):133.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaverkamp TH, Spilsberg B, Johannessen GS, Torp M, Sekse C. Detection of Campylobacter in air samples from poultry houses using shot-gun metagenomics\u0026ndash;a pilot study. bioRxiv. 2021:2021.05. 17.444449.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePlatts-Mills JA, Liu J, Gratz J, Mduma E, Amour C, Swai N, et al. Detection of Campylobacter in stool and determination of significance by culture, enzyme immunoassay, and PCR in developing countries. J Clin Microbiol. 2014;52(4):1074\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDerelle R, von Wachsmann J, M\u0026auml;klin T, Hellewell J, Russell T, Lalvani A, et al. Seamless, rapid and accurate analyses of outbreak genomic data using Split K-mer Analysis (SKA). bioRxiv. 2024:2024.03.25.586631.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris SR. SKA: Split Kmer Analysis Toolkit for Bacterial Genomic Epidemiology. bioRxiv. 2018:453142.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller WG, Chapman MH, Williams TG, Wood DF, Bono JL, Kelly DJ. Campylobacter californiensis sp. nov., isolated from cattle and feral swine. International Journal of Systematic and Evolutionary Microbiology. 2024;74(10).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller WG, Lopes BS, Ramjee M, Jay-Russell MT, Chapman MH, Williams TG, et al. Campylobacter devanensis sp. nov., Campylobacter porcelli sp. nov., and Campylobacter vicugnae sp. nov., three novel Campylobacter lanienae-like species recovered from swine, small ruminants, and camelids. International Journal of Systematic and Evolutionary Microbiology. 2024;74(6).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark CG, Chen C-y, Berry C, Walker M, McCorrister SJ, Chong PM, et al. Comparison of genomes and proteomes of four whole genome-sequenced Campylobacter jejuni from different phylogenetic backgrounds. PLOS ONE. 2018;13(1):e0190836.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLengerh A, Moges F, Unakal C, Anagaw B. Prevalence, associated risk factors and antimicrobial susceptibility pattern of Campylobacter species among under five diarrheic children at Gondar University Hospital, Northwest Ethiopia. BMC pediatrics. 2013;13:1\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbamecha A, Assebe G, Tafa B, Wondafrash B. Prevalence of thermophilic Campylobacter and their antimicrobial resistance profile in food animals in Lare District, Nuer Zone, Gambella, Ethiopia. J Drug Res Dev. 2015;1(2):2470\u0026ndash;1009.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiriba K, Awulachew E, Anja A. Prevalence and associated factor of Campylobacter species among less than 5-year-old children in Ethiopia: a systematic review and meta-analysis. European Journal of Medical Research. 2021;26(1):1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiriba K, Awulachew E, Anja A. Prevalence and associated factor of Campylobacter species among less than 5-year-old children in Ethiopia: a systematic review and meta-analysis. European Journal of Medical Research. 2021;26(1):2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZenebe T, Zegeye N, Eguale T. Prevalence of Campylobacter species in human, animal and food of animal origin and their antimicrobial susceptibility in Ethiopia: a systematic review and meta-analysis. Annals of Clinical Microbiology and Antimicrobials. 2020;19(1):61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBelina D, Gobena T, Kebede A, Chimdessa M, Mummed B, Thystrup CAN, et al. Occurrence and diversity of Campylobacter species in diarrheic children and their exposure environments in Ethiopia. PLOS Global Public Health. 2024;4(10):e0003885.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaakoush NO, Casta\u0026ntilde;o-Rodr\u0026iacute;guez N, Mitchell HM, Man SM. Global Epidemiology of Campylobacter Infection. Clin Microbiol Rev. 2015;28(3):687\u0026ndash;720.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuince C, Walker AW, Simpson JT, Loman NJ, Segata N. Shotgun metagenomics, from sampling to analysis. Nature Biotechnology. 2017;35(9):833\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBian X, Garber Jolene M, Cooper Kerry K, Huynh S, Jones J, Mills Michael K, et al. Campylobacter Abundance in Breastfed Infants and Identification of a New Species in the Global Enterics Multicenter Study. mSphere. 2020;5(1):\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1128/msphere.00735\u0026thinsp;\u0026ndash;\u0026thinsp;19\u003c/span\u003e\u003cspan address=\"10.1128/msphere.00735\u0026thinsp;\u0026ndash;\u0026thinsp;19\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaakoush NO, Mitchell HM. Campylobacter concisus\u0026ndash;a new player in intestinal disease. Frontiers in cellular and infection microbiology. 2012;2:4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIstivan T, Huq M. Biofilms of Campylobacter concisus: a potential survival mechanism in the oral cavity. Microbiology Australia. 2023;44(2):100\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNapit R, Manandhar P, Poudel A, Rajbhandari PG, Watson S, Shakya S, et al. Novel strains of Campylobacter cause diarrheal outbreak in Rhesus macaques (Macaca mulatta) of Kathmandu Valley. PLOS ONE. 2023;18(3):e0270778.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarbonero A, Torralbo A, Borge C, Garc\u0026iacute;a-Bocanegra I, Arenas A, Perea A. Campylobacter spp., C. jejuni and C. upsaliensis infection-associated factors in healthy and ill dogs from clinics in Cordoba, Spain. Screening tests for antimicrobial susceptibility. Comparative immunology, microbiology and infectious diseases. 2012;35(6):505\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParsons B, Porter C, Stavisky J, Williams N, Birtles R, Miller W, et al. Multilocus sequence typing of human and canine C. upsaliensis isolates. Veterinary microbiology. 2012;157(3\u0026ndash;4):391\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJaime AL, Joan S, Lee B, Nancy S, Sydney MH, Eleanor L, et al. Campylobacter upsaliensis: Another Pathogen for Consideration in the United States. Clinical Infectious Diseases. 2002;34(11):e59-e60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarama M, Etter E, McCrindle C, El-Ashram S, Prosperi A, Ombui JN, et al. Prevalence and risk factors associated with Campylobacter spp. occurrence in healthy dogs visiting four rural community veterinary clinics in South Africa. Onderstepoort Journal of Veterinary Research. 2019;86(1):1\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuong T, Konkel ME. Comparative studies of Campylobacter jejuni genomic diversity reveal the importance of core and dispensable genes in the biology of this enigmatic food-borne pathogen. Curr Opin Biotechnol. 2009;20(2):158\u0026ndash;65.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHermans D, Pasmans F, Messens W, Martel A, Van Immerseel F, Rasschaert G, et al. Poultry as a host for the zoonotic pathogen Campylobacter jejuni. Vector-Borne and Zoonotic Diseases. 2012;12(2):89\u0026ndash;98.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGripp E, Hlahla D, Didelot X, Kops F, Maurischat S, Tedin K, et al. Closely related Campylobacter jejuni strains from different sources reveal a generalist rather than a specialist lifestyle. BMC genomics. 2011;12:1\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHorrocks S, Anderson R, Nisbet D, Ricke S. Incidence and ecology of Campylobacter jejuni and coli in animals. Anaerobe. 2009;15(1\u0026ndash;2):18\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHumphrey T, O'Brien S, Madsen M. Campylobacters as zoonotic pathogens: a food production perspective. International journal of food microbiology. 2007;117(3):237\u0026ndash;57.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGodschalk PC, Gilbert M, Jacobs BC, Kramers T, Tio-Gillen AP, Ang CW, et al. Co-infection with two different Campylobacter jejuni strains in a patient with the Guillain-Barr\u0026eacute; syndrome. Microbes Infect. 2006;8(1):248\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCasta\u0026ntilde;o-Rodr\u0026iacute;guez N, Kaakoush NO, Lee WS, Mitchell HM. Dual role of Helicobacter and Campylobacter species in IBD: a systematic review and meta-analysis. Gut. 2017;66(2):235\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang G, He Y, Jin X, Zhou Y, Chen X, Zhao J, et al. The effect of co-infection of food-borne pathogenic bacteria on the progression of Campylobacter jejuni infection in mice. Frontiers in microbiology. 2018;9:1977.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNataro James P, Kaper James B. Diarrheagenic Escherichia coli. Clinical Microbiology Reviews. 1998;11(1):142\u0026ndash;201.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBhattarai V, Sharma S, Rijal KR, Banjara MR. Co-infection with Campylobacter and rotavirus in less than 5 year old children with acute gastroenteritis in Nepal during 2017\u0026ndash;2018. BMC pediatrics. 2020;20:1\u0026ndash;8.\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":"microbiome","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mbio","sideBox":"Learn more about [Microbiome](http://microbiomejournal.biomedcentral.com/)","snPcode":"40168","submissionUrl":"https://submission.nature.com/new-submission/40168/3","title":"Microbiome","twitterHandle":"@MicrobiomeJ","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Campylobacter, Enteric Pathogens, Shotgun metagenomics, Eastern Ethiopia, Longitudinal Cohort, Human, Livestock","lastPublishedDoi":"10.21203/rs.3.rs-5736322/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5736322/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eLivestock are recognized as major reservoirs for \u003cem\u003eCampylobacter\u003c/em\u003e species and other enteric pathogens, posing substantial infection risks to humans. High prevalence of \u003cem\u003eCampylobacter\u003c/em\u003e during early childhood has been linked to environmental enteric dysfunction and stunting, particularly in low-resource settings.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA longitudinal study of 106 infants was conducted from December 2020 to June 2022. Monthly stool samples were collected from infants beginning in the first month after birth. Additional stool samples from mothers, siblings, and livestock (goats, cattle, sheep, and chickens) were collected biannually. A subset of 280 samples from \u003cem\u003eCampylobacter\u003c/em\u003e positive households with complete metadata were analyzed by shotgun metagenomic sequencing followed by bioinformatic analysis via the CZ-ID metagenomic pipeline (Illumina mNGS Pipeline v7.1). Further statistical analyses in JMP PRO 16 explored the microbiome, emphasizing \u003cem\u003eCampylobacter\u003c/em\u003e and other enteric pathogens. Two-way hierarchical clustering and split k-mer analysis examined host structuring, patterns of co-infections and genetic relationships. Principal component analysis was used to characterize microbiome composition across the seven sample types.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eMore than 3,844 genera were detected in the 263 samples. Twenty-one dominant \u003cem\u003eCampylobacter\u003c/em\u003e species were detected with distinct clustering patterns for humans, ruminants, and broad hosts. The generalist (broad-host) cluster included the most prevalent species, \u003cem\u003eC. jejuni, C. concisus\u003c/em\u003e, and \u003cem\u003eC. coli\u003c/em\u003e, present across sample types. Among \u003cem\u003eC. jejuni\u003c/em\u003e a major cluster involving humans, chickens, and ruminants isolates, was detected, indicating potential zoonotic transmission to infants and mothers. \u003cem\u003eCandidatus\u003c/em\u003e C. infans was only detected in human hosts. \u003cem\u003eCampylobacter\u003c/em\u003e species from chickens showed strong positive correlations with mothers (r\u0026thinsp;=\u0026thinsp;0.76), siblings (r\u0026thinsp;=\u0026thinsp;0.61) and infants (r\u0026thinsp;=\u0026thinsp;0.54), while no to weak correlation was observed between \u003cem\u003eCampylobacter\u003c/em\u003e species from chickens and small ruminants (sheep and goats) with (r\u0026thinsp;=\u0026thinsp;0.15, r\u0026thinsp;=\u0026thinsp;0.0, respectively). Co-occurrence analysis revealed a higher likelihood (p\u0026thinsp;\u0026gt;\u0026thinsp;0.5) of pairs such as \u003cem\u003eC. jejuni\u003c/em\u003e with \u003cem\u003eC. coli, C. concisus\u003c/em\u003e, and \u003cem\u003eC. showae\u003c/em\u003e. Overall microbiome composition was strongly host driven, with two principal components accounting for 62% of the total variation. Analysis of the top 50 most abundant microbial taxa in infant stool revealed a distinct cluster uniquely present in human stool samples and absent in all livestock samples. Hierarchical clustering revealed frequent co-occurrence of \u003cem\u003eC. jejuni\u003c/em\u003e with other enteric pathogens such as \u003cem\u003eSalmonella\u003c/em\u003e, and \u003cem\u003eShigella\u003c/em\u003e, particularly in human and chicken samples. Additionally, instances of \u003cem\u003eCandidatus C. infans\u003c/em\u003e were identified co-occurring with \u003cem\u003eSalmonella\u003c/em\u003e and \u003cem\u003eShigella\u003c/em\u003e species in stool samples from infants, mothers, and siblings.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eA comprehensive analysis of \u003cem\u003eCampylobacter\u003c/em\u003e diversity in humans and livestock in a low-resource setting, revealed that infants can be exposed to multiple \u003cem\u003eCampylobacter\u003c/em\u003e species early in life. \u003cem\u003eC. jejuni\u003c/em\u003e is the dominant species with a propensity for co-occurrence with other notable enteric bacterial pathogens, including \u003cem\u003eSalmonella\u003c/em\u003e, and \u003cem\u003eShigella\u003c/em\u003e, especially among infants.\u003c/p\u003e","manuscriptTitle":"Host Clustering of Campylobacter Species and Other Enteric Pathogens in a Longitudinal Cohort of Infants, Family Members and Livestock in Rural Eastern Ethiopia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-18 11:52:46","doi":"10.21203/rs.3.rs-5736322/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-01-13T23:26:33+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-01-13T23:25:36+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-01-03T04:51:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"Microbiome","date":"2024-12-30T14:33:46+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"microbiome","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mbio","sideBox":"Learn more about [Microbiome](http://microbiomejournal.biomedcentral.com/)","snPcode":"40168","submissionUrl":"https://submission.nature.com/new-submission/40168/3","title":"Microbiome","twitterHandle":"@MicrobiomeJ","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"576225dd-9b9e-4ceb-abf3-4d7572555d60","owner":[],"postedDate":"March 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-11-10T15:58:42+00:00","versionOfRecord":{"articleIdentity":"rs-5736322","link":"https://doi.org/10.1186/s40168-025-02203-w","journal":{"identity":"microbiome","isVorOnly":false,"title":"Microbiome"},"publishedOn":"2025-11-03 15:56:55","publishedOnDateReadable":"November 3rd, 2025"},"versionCreatedAt":"2025-03-18 11:52:46","video":"","vorDoi":"10.1186/s40168-025-02203-w","vorDoiUrl":"https://doi.org/10.1186/s40168-025-02203-w","workflowStages":[]},"version":"v1","identity":"rs-5736322","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5736322","identity":"rs-5736322","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}