key: cord-0272058-348xszwy authors: Delwart, Eric; Tisza, Michael J.; Altan, Eda; Li, Yanpeng; Deng, Xutao; Hartigan-O’Connor, J. Dennis; Ardeshir, Amir title: Idiopathic chronic diarrhea in rhesus macaques is not associated with enteric viral infections date: 2021-07-02 journal: bioRxiv DOI: 10.1101/2021.07.01.450785 sha: cd0eeb6d50e16164f144617bc3d59b46123ce288 doc_id: 272058 cord_uid: 348xszwy While recent changes in treatment have reduced the lethality of idiopathic chronic diarrhea (ICD), this condition remains one of the most common causes of rhesus macaque deaths in non-human primate research centers. We compared the eukaryotic viromes in fecal swabs from 52 animals with ICD and 41 healthy animals. Viral metagenomics targeting virus-like particles was used to identify viruses shed by each animal. Five viruses belonging to the Picornaviridae, one to the Caliciviridae, one to the Parvoviridae, and one to the Adenoviridae families were identified. The fraction of reads matching each viral species was then used to estimate and compare viral loads in ICD cases versus healthy controls. None of the eukaryotic viruses detected in fecal swabs were strongly associated with ICD. Other potential causes of ICD are discussed. Idiopapthic chronic diarrhea (ICD) is the leading cause of morbidity in captive colonies of rhesus macaques and a common cause of non-medical research mortality (1, 2) and is defined by chronic or recurring non-bloody diarrhea and microscopic colonic ulcers (2) (3) (4) (5) . ICD cases are typically diagnosed clinically before one year of age. The histopathological evaluations of the ICD cases show surface epithelium attenuation, goblet cell depletion and crypt branching and microscopic ulcers (6) . The lamina propria is also filled with infrates including lymphocytes and plasma cells (5, 6) . In addition, Rhesus macaques' enterocolitis is a T H 1 biased immune response with concurrent gut microbial dysbiosis (7) (8) (9) (10) . Bacterial infections with Shigella, Campylobacter, Yersinia, Salmonella, and/or Clostridium difficile as well as parasitic infections, including Cryptosporidium, are common known causes of diarrhea in macaque colonies (2, 11) . Cases of persistent and recurrent diarrhea in which no pathogenic bacteria, parasites or other etiological agents are found are defined as ICD (8) . Rhesus macaques with ICD respond poorly to treatment, including corticosteroids or antibiotics, resulting in frequent hospitalization from dehydration and weight loss and non-responsive animals are euthanized for ethical considerations (6) . A study comparing viral etiology of diarrheal diseases in rhesus macaques found that rhesus and pigtail macaques housed in the monkey farm of the Institute of Laboratory Animal Sciences, Chinese Academy of Medical Sciences shed enteroviruses, enteric adenoviruses, coronaviruses, and rotaviruses, while others at the Yerkes National Primate Research Center (Yerkes, GA) contained mainly enteroviruses and eneteric adenoviruses (12) . Adenoviruses and rotaviruses were also associated with diarrhea in captive rhesus macaques (12) (13) (14) . Inoculation with primate caliciviruses induced colitis in juvenile rhesus macaques (15) . SIV-induced enteropathy was associated with a greater diversity of enteric viruses and parvovirus viremia when compared to SIV-negative rhesus macaques (16). We previously described the virome in feces of healthy macaques as well as in those acute diarrhea or ICD and reported that shedding of several picornavirus genotypes were weakly associated with ICD while parvoviruses were weakly associated with healthy animals (17) . Here we extend these studies by analyzing fecal samples from 52 animals with ICD and 41 healthy controls. Study design: This study is a case-control to search for the possible differences in the stool virome of ICD and healthy animals. The study was not blinded. Animals: All animals in this study were born, and raised at the CNPRC and samples collected while they are housed in oudoor colony (healthy controls) or from necropsy (ICD). The healthy control animals (n=41) were selected from outdoor housed animals without any gastrointestinal issues (including diarrheal diseases) or antibiotic treatment within the last 6 months prior to sampling. ICD cases (n=52) were selected from the cases that were hospitalized for the third time for non-pathogneic diarhea within the past 365 days and then euthanized. ICD cases that were euthanized with other accompanying diseases were excluded from the study to control for confounding factors. Sample collection: Rectal swabs were collection from rhesus macaques while sedated for the bi-annual physical exam (controls) or during necropsy (ICDs) and immediately frozen at −70°C. (23). Abundance was calculated with the same methods as described (21) . Finally, sample-by-sample swarm plots were generated using DaBest python package (24) . Fecal swabs were collected from 52 macaques suffering from ICD and 41 healthy controls. Fecal material was filtered to enrich for virus-like particles. Host and bacterial nucleic acids were then depleted using nuclease enzymes. Total nucleic acids were extracted from digested and filtered materials and amplified using random RT-PCR. DNA was then converted to Illumina-compatible DNA and sequenced (see materials and methods). All raw data are available in GenBank under Bioproject PRJNA608547. of reads matching these different simian viruses was then plotted in order to compare the eukaryotic virome composition of ICD cases to that of healthy animals (Fig 1) . This analysis indicated that for the viruses detected in most animals (five picornaviruses), the median and mean read percentages were higher in healthy than sick animals, with the single exception of enterovirus A92 whose mean value in ICD animals was above that of the healthy control group. When the data for these five most common infections was plotted using whisker plots, no difference between the ICD and healthy cases could be seen, except again Viral reads of from rectal swabs from healthy and sick macaques for mean enterovirus A92 representation being lower in healthy animals (Fig 2) . (20, 21) . Such hallmark genes include structural genes such as capsid proteins, genome-packaging genes such as terminases, and genome-replication genes such as RNA-dependent RNA polymerase and replicative helicase (20, 21) . The abundance of these viral contigs was compared between ICD and healthy animals by measuring reads per kilobase per millions reads (RPKM) (Figure 4 ). Only contigs that Note that most picornaviruses contigs retrieved from de novo assembly represented sub-genomic fragments with >90% average nucleotide identity to previously discovered macaque viruses. We tested the hypothesis that a specific eukaryotic virus was associated with ICD. We used sequence homology of NGS reads derived from viral metagenomics to identify viruses in fecal swabs. The frequencies of reads to different viruses were then used as surrogates for viral loads. By comparing the viruses in rhesus macaques with ICD versus those in healthy controls, we were unable to associate any eukaryotic virus with ICD. These results differ from the conclusions of a previous analysis, in which we showed that some enterovirus genotypes were weakly associated with ICD (17) . During the interval between these studies the treatment for ICD changed and the frequency of ICD dropped significantly. Specifically, antibiotic therapy was no longer done on animals testing negative for enteropathogens; these animals instead received fluid therapy, and synbiotic or probiotic treatments. The ICD fecal swabs samples that were analyzed in this study were collected during necropsy from severe cases that required medical cull while feces analyzed in prior studies were feces collected from cage pans while the animals were still under treatment. Therefore, it is possible that during the later stage of ICD studied here, shedding of enteric viruses in watery diarrhea is reduced due to more extensive damage to the lining of the gut and consequent reduction in viral target cells. It is possible that earlier sampling of ICDassociated enteric virome may be necessary to identify viruses associated with this condition. We observed that the median level of viral shedding was generally higher in healthy than in ICD cases (Fig 1 and 2) , a result consistent with a reduction in viral target cells in diarrheic animals. In the analysis of virus contigs predicted by Cenote-Taker 2 (Figs. 4 and 5) , it was also observed that fewer reads to specific phages were generally found in swabs from ICD cases versus healthy animals. Such a reduction in phage shedding may reflect a gut content with fewer bacterial hosts as its content is flushed out due to chronic diarrhea. Risk factor analysis may provide clues to diarrhea prevention in outdoor-housed rhesus macaques (Macaca mulatta) Infectious agent and immune response characteristics of chronic enterocolitis in captive rhesus macaques Microscopic colitis: Common cause of unexplained nonbloody diarrhea Diarrhea rates and risk factors for developing chronic diarrhea in infant and juvenile rhesus monkeys Effects of the macrolide drug tylosin on chronic diarrhea in rhesus macaques (Macaca mulatta) Idiopathic microscopic colitis of rhesus macaques: quantitative assessment of colonic mucosa Inulin treatment leads to changes in intestinal microbiota and resolution of idiopathic chronic diarrhea in rhesus macaques Therapeutic helminth infection of macaques with idiopathic chronic diarrhea alters the inflammatory signature and mucosal microbiota of the colon Dysbiosis in inflammatory bowel disease The gut microbiota shapes intestinal immune responses during health and disease Epidemiology and etiology of diarrhea in colony-born Macaca nemestrina Detection of viral agents in fecal specimens of monkeys with diarrhea Virus detection in monkeys with diarrhea: the association of adenoviruses with diarrhea and the possible role of rotaviruses Rotavirus is associated with decompensated diarrhea among young rhesus macaques (Macaca mulatta) Experimental inoculation of juvenile rhesus macaques with primate enteric caliciviruses Pathogenic simian immunodeficiency virus infection is associated with expansion of the enteric virome Case-Control Comparison of Enteric Viromes in Captive Rhesus Macaques with Acute or Idiopathic Chronic Diarrhea BLAST: improvements for better sequence analysis An ensemble strategy that significantly improves de novo assembly of microbial genomes from metagenomic next-generation sequencing data Discovery of several thousand highly diverse circular DNA viruses Cenote-Taker 2 democratizes virus discovery and sequence annotation Using MUMmer to identify similar regions in large sequence sets Using RepeatMasker to identify repetitive elements in genomic sequences Moving beyond P values: data analysis with estimation graphics Acknowledgments: We thank E. Fahsbender for technical help and data generation and R. Bruhn for assistance for statistical analyses. We thank CNPRC for assistance with sample collection.