Perturbations of the Gut Microbiome and Metabolome in Children with Calcium Oxalate Kidney Stone Disease

Home , Oxalobacter, Oxalobacteraceae, Oxalobacter formigenes, Treponema

CLINICAL RESEARCH www.jasn.org

Michelle R. Denburg,1,2,3 Kristen Koepsell,4 Jung-Jin Lee ,5 Jeffrey Gerber,3,6 Kyle Bittinger,5 and Gregory E. Tasian2,3,4

1Division of Nephrology, Department of Pediatrics, The Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania 2Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania 3Center for Pediatric Clinical Effectiveness, The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 4Division of Pediatric Urology, Department of Surgery, The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 5Division of Gastroenterology, Department of Pediatrics, Hepatology, and Nutrition, The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 6Division of Infectious Diseases, Department of Pediatrics, The Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania

ABSTRACT Background The relationship between the composition and function of gut microbial communities and early-onset calcium oxalate kidney stone disease is unknown. Methods We conducted a case-control study of 88 individuals aged 4–18 years, which included 44 individuals with kidney stones containing $50% calcium oxalate and 44 controls matched for age, sex, and race. Shotgun metagenomic sequencing and untargeted metabolomics were performed on stool samples. Results Participants who were kidney stone formers had a significantly less diverse gut microbiome compared with controls. Among bacterial taxa with a prevalence .0.1%, 31 taxa were less abundant among individuals with nephrolithiasis. These included seven taxa that produce butyrate and three taxa that degrade oxalate. The lower abundance of these bacteria was reflected in decreased abundance of the gene encoding butyryl-coA dehydrogenase (P50.02). The relative abundance of these bacteria was correlated with the levels of 18 fecal metabolites, and levels of these metabolites differed in individuals with kidney stones compared with controls. The oxalate-degrading bacterial taxa identified as decreased in those who were kidney stone formers were components of a larger abundance correlation network that included Eggerthella lenta and several Lactobacillus species. The microbial (a) diversity was associated with age of stone onset, first decreasing and then increasing with age. For the individuals who were stone formers, we found the lowest a diversity among individuals who first formed stones at age 9–14 years, whereas controls displayed no age-related differences in diversity. Conclusions Loss of gut bacteria, particularly loss of those that produce butyrate and degrade oxalate, associates with perturbations of the metabolome that may be upstream determinants of early-onset calcium oxalate kidney stone disease.

JASN 31: 1358–1369, 2020. doi: https://doi.org/10.1681/ASN.2019101131

Correspondence: Dr. Gregory E. Tasian, Division of Urology, Received October 31, 2019. Accepted March 22, 2020. Children’s Hospital of Philadelphia, Wood Center, 3rd Floor, 34th Street and Civic Center Boulevard, Philadelphia, PA 19104. Published online ahead of print. Publication date available at Email: [email protected] www.jasn.org. Copyright © 2020 by the American Society of Nephrology

1358 ISSN : 1046-6673/3106-1358 JASN 31: 1358–1369, 2020 www.jasn.org CLINICAL RESEARCH

Kidney stone disease (nephrolithiasis) is highly prevalent, in- Significance Statement creasingly common, and is characterized by painful stone events that cause considerable morbidity. In addition, as a Although antibiotics have been associated with an increased risk of disorder of mineral metabolism, nephrolithiasis is associated kidney stones, particularly early in life, perturbations of the gut with increased risks of kidney function loss,1,2 decreased microbiome and metabolome in early-onset nephrolithiasis have not been investigated. Using shotgun metagenomic sequencing 325 bone2pt mineral density and fracture, and cardiovascular and untargeted metabolomics of stool samples in a study of 44 2 disease.6 8 Kidney stones affect one in 11 people in the United children with kidney stones and 44 controls matched for age, sex, States9 and result in annual healthcare costs .$10 billion.10 and race, the authors found that 31 bacterial taxa—including seven The prevalence of nephrolithiasis has increased by 70% over butyrate-producing taxa and three that degrade oxalate—were less 20 years.9,11 Our group and others have discovered dispropor- abundant among children with calcium oxalate stones. Levels of 18 metabolites differed between cases and controls and correlated tionate increases in the incidence of nephrolithiasis among with the fecal bacteria that were less abundant among children with 2 children, adolescents, and women.12 14 The shift in neph- nephrolithiasis. Such disruptions in the gut microbiome and me- rolithiasis to a younger age of onset has caused increasing tabolome may thus be determinants of early-onset disease and may hospitalizations, surgeries, and healthcare expenditures.15 explain the association between antibiotics and nephrolithiasis. In addition, the morbidity associated with nephrolithiasis appears to be more pronounced in younger individuals, and 50% was calcium oxalate) that spontaneously passed or were stone recurrence rates may be higher in children than removed surgically within the prior 3 years (stone analysis 4,16 adults. The reasons for this shift in the epidemiology of done by mass spectroscopy at various clinical laboratories). nephrolithiasis are unclear. However, the rapidity of the change We used a 3-year time window since the last stone event to suggests that the driving forces are external exposures such as capture individuals most likely to have active stone disease.16 diet and antibiotics. Many of these exposures may disrupt the Cases were recruited during outpatient visits to the CHOP gut-kidney axis, which is the complex interplay between the Kidney Stone Center. 17 intestinal and urinary tracts in human health and disease. Controls were healthy volunteers matched to cases on age Prior investigations have demonstrated perturbations of (62 years), sex, race, and ethnicity. The study was nested the gut microbiome among adults with nephrolithiasis. These within the CHOP healthcare system so that cases and controls studies found that the gut microbiome of those who formed arose from the same source population, and equal application 18,19 kidney stones is less diverse than controls and that bacteria of eligibility criteria for both cases and controls ensured that that degrade oxalate were less abundant in the stool of adults both groups were healthy, with the exception that cases were 20 with kidney stones. Currently, the composition and function individuals with kidney stones. To increase study efficiency, of the gut microbiome and metabolome among individuals participants with kidney stones were first matched to healthy with early-onset calcium oxalate kidney stone disease, the participants (n517) who had enrolled as controls in an in- 21 most common form of nephrolithiasis, is unknown. Discov- dependent study that used the same stool collection and ex- ery of the identity and function of microbial communities posure questionnaire as our study. These control participants and downstream metabolites perturbed in early-onset calcium were recruited from the CHOP emergency and dermatology oxalate kidney stone disease could reveal targets for novel ther- departments, oral surgery clinic, and urgent care centers. The apeutics for kidney stone prevention across the life span and remaining 27 control participants were recruited from seven help determine causes of the rapid shift in the epidemiology of suburban and urban practices in the Pediatric Research Con- nephrolithiasis. sortium, which includes the network of CHOP primary care practices in Pennsylvania and New Jersey.

METHODS Study Procedures Participants completed a baseline questionnaire, including Study Design and Population past medical history, early lifetime exposures (e.g., vaccina- We conducted a matched case-control study of 88 individuals tions), tobacco use, recent hospitalizations, and probiotic aged 4–18 years who received care in The Children’s Hospital use. In addition, research nutritionists administered 24-hour of Philadelphia (CHOP) healthcare system. We excluded in- dietary recalls over the telephone on 3 days (2 weekdays, dividuals who took antibiotics in the last 3 months and those 1 weekend day) to estimate participants’ daily nutrient and with inﬂammatory bowel disease, prior bariatric surgery, mineral intake. The 24-hour dietary recalls were collected using monogenic causes of kidney stones, cancer, immobility, cystic the Nutrition Data System for Research developed at the University ﬁbrosis, celiac disease, diabetes, congenital anomalies of the of Minnesota.22 Dietary intake data were gathered by a multiple- kidney and urinary tract, urinary tract obstruction, and renal pass interview approach,23 and values for 165 nutrients, nutrient tubular acidosis. The Institutional Review Board at CHOP ratios, and other food components were generated from a database approved this study. that includes .18,000 foods.24,25 Antibiotics, including route Cases included individuals with incident and recurrent and class, and medications taken over the last 1 year were kidney stones consisting of 100% calcium (of which at least ascertained by (1) self-report or interview of caregivers,

JASN 31: 1358–1369, 2020 Gut Microbiome and Metabolome 1359 CLINICAL RESEARCH www.jasn.org

(2) prescriptions from CHOP,and (3) antibiotics reconciled kidney stones, a linear model was used to test the association in the electronic health record. Age at nephrolithiasis diag- between the log-transformed relative abundance of taxa nosis was determined through chart review. Stool speci- with linear and quadratic terms of the age at stone disease mens were collected at the participants’ homes, kept chilled, diagnosis for cases and their corresponding matched controls. and shipped in an insulated container on ice packs. Stool Taxa were included in the latter analysis only if their relative stabilizer was not used. All specimens were shipped by abundance was at least 0.1% in any individual who formed same-daydeliveryandarrivedattheCHOPMicrobiome kidney stones. Center within 12–24 hours. At the time of receipt, specimen temperature was checked. Any specimen .20°C was rejec- Untargeted Metabolomics ted. Of the 88 original specimens received, only one had to Metabolomic profiling (Metabolon, Inc., Durham, NC) was be recollected due to a temperature upon arrival of .20°C. performed using a Waters Acquity Ultra-Performance Liquid All study procedures were completed within 3 months of Chromatography and a Thermo Scientific Q-Exactive high- enrollment. resolution/accurate mass spectrometer interfaced with a heated electrospray ionization source and Orbitrap mass an- Shotgun Metagenomics alyzer operated at 35,000 mass resolution. Metabolites were Upon arrival, stool samples were maintained at 280°C until identified by comparison to library entries of purified stan- being processed. Genomic DNA was extracted from stool and dards of .3300 commercially available purified standard prepared for shotgun metagenomic sequencing using the Nex- compounds, and concentrations were quantified using the tera XT DNA Library Preparation Kit (Illumina, San Diego, area under the curve. Missing values were imputed with the CA). DNA extraction blanks and DNA-free water were in- minimum value, and metabolite values were normalized by cluded as negative control samples to assess environmental registering the run-day medians to equal one to correct for and reagent contamination. Laboratory-generated mock variation resulting from instruments.34 communities consisting of DNA from Vibrio campbellii, Cryp- Principal components analysis (PCA) and linear discrimi- tococcus diffluens,andl phage were included as positive con- nant analysis (LDA) were carried out to identify fecal metab- trols. DNA sequencing was carried out on an Illumina HiSeq olomic characteristics of the study groups. The PCA and LDA 2500 instrument, generating 125 bp paired-end sequence included 841 out of 867 measured, named biochemicals after reads. excluding noninformative (largely xenobiotic drug) metabo- Reads were quality filtered and trimmed to remove adapter lites. For PCA and LDA, metabolite values were normalized sequences using Trimmomatic version 0.36.26 Reads aligning such that the mean equaled zero and the SD equaled one. The to the human host genome (version hg38) were removed using paired-sample t test was also performed to test the association BWA version 0.7.17-r1188.27 Taxonomic annotations were between metabolite level and study group. We correlated me- generated by Kraken version 1.0 using the standard database tabolites and species abundances that differed between groups with all complete bacterial, archaeal, and viral genomes in the and validated the correlations by permutation testing to eval- National Center for Biotechnology Information’sRefSeq.28 To uate the possibility that spurious correlations may have been assess gene function, reads were aligned to the KEGG database detected by coincidence, particularly when a metabolite and of gene ortholog sequences29 using Diamond version 0.9.24.30 taxon were both identified as higher in cases and lower in Oxalate-degrading organisms were identified using the lists controls. Specifically, we generated the null distribution of compiled in Ticinesi et al.20 and Miller and Dearing.31 correlation values by randomly shuffling the taxon abun- Butyrate-producing organisms were identified by searching dances within cases and controls. Accordingly, we assessed bacterial species data in Bergey’sManualofSystematicsof the significance of the correlation between metabolite level Archaea and Bacteria.32 and abundance of bacterial species using P values obtained The Shannon index and richness for a diversity analysis from permutation tests and identified as statistically signifi- was calculated using the vegan package version 2.5-6 (https:// cant only metabolite-taxon pairs that were more correlated CRAN.R-project.org/package5vegan) in R (version 3.6.0). than expected while controlling for the study group. The association between a diversity and study group was eval- uated using a paired t test. To test for the association of taxon abundance with study group, we log-transformed the taxon RESULTS proportions and applied a paired t test. Taxa were tested if the abundance in any sample exceeded 0.1%. Oxalobacter formi- Characteristics of Study Population genes and Oxalobacteraceae were also tested, despite lower The study consisted of 44 individuals with incident or recur- abundance, due to their role in oxalate degradation. To correct rent calcium oxalate kidney stones that occurred at #18 years for multiple comparisons, we adjusted P values using the of age and 44 healthy controls (Table 1). Those who were stone method of Benjamini–Hochberg to control the false discovery formers were representative of the pediatric kidney stone pop- rate (FDR).33 To identify bacteria increasing or decreasing in ulation in the United States,35 with a median age of 15.6 years abundance with age of first stone among participants with at time of study, 13 years at stone diagnosis, 52% female and

1360 JASN JASN 31: 1358–1369, 2020 www.jasn.org CLINICAL RESEARCH

Table 1. Population characteristics To compare the taxonomic composi- Participants with Kidney tion of all participants, we tested 91 bac- Characteristics Controls (n544) Stones (n544) terial taxa that had .0.1% abundance in Median age at enrollment, yr (IQR) 15.6 (11.8, 17.1) 15.6 (12.4, 16.5) at least one sample. We found that 31 bac- Median age at first stone, yr (IQR) 13 (8.5, 15) N/A terial taxa differed in abundance between Sex, no. (%) participants with kidney stones and con- Male 21(47.7) 21(47.7) trols at a pre-defined threshold of FDR Female 23(52.3) 23(52.3) adjusted P,0.05 (Figure 1A). Of the taxa Ethnicity, no. (%) identified, all were less abundant among Hispanic or Latino 3 (6.8) 2 (4.5) participants who were stone formers than Not Hispanic or Latino 40 (90.9) 42 (95.5) controls. These included seven taxa that Missing 1 (2.3) 0 produce the short-chain fatty acid butyrate, Race, no. (%) including several Roseburia and Clostridium Black 1 (2.3) 2 (4.5) White 43 (97.7) 40 (90.9) species. Correspondingly, the gene abun- Multiple 0 2 (4.5) dance of butyryl-coA dehydrogenase, a key Median body mass index, percentile (IQR) 67.6 (28.5, 91.8) 61.4 (40.3, 80.1)a bacterial enzyme in the butyrate produc- Stone activity, no. (%)b tion pathway, was lower among those who Quiescent 21 (47.7) N/A were stone formers (P50.02, Figure 1B). Active 23 (52.3) N/A Sequencing also revealed a lower Recurrent, no. (%)c abundance of three oxalate-degrading No 18 (40.9) N/A bacterial taxa in participants with kidney Yes 26 (59.1) N/A stones: Enterococcus faecalis, Enterococcus Family history, no. (%) faecium,andBifidobacterium animalis. No 14 (31.8) N/A We detected the oxalate-degrading spe- Yes 30 (68.2) N/A cies O. formigenes at low relative abun- IQR, interquartile range; N/A, not applicable. aThree missing, n541. dance in two subjects from the control bQuiescent is the absence of growth of an existing stone and no formation of a new stone in the year group (Supplemental Figure 1). Neither before specimen collection. Active is the growth of an existing stone or formation of a new stone in the the presence/absence nor relative abun- year before specimen collection. cRecurrent stone formers are participants with a new symptomatic stone or new stone formation on dance of O. formigenes were different be- imaging between their first stone and the date of specimen collection. tween participants with kidney stones and controls due to the small number of sam- 98%whiterace.Oftheparticipantswithkidneystones, ples where it was detected (P50.4 for relative abundance, 59% had recurrent kidney stones and 68% had a family P50.5 for presence/absence). We detected a small proportion history of nephrolithiasis. Antibiotic exposures are sum- of reads assigned to the Oxalobacteraceae family in all controls marized in Table 2, and dietary intakes and current medi- and in all but one participant with kidney stones. There was no cations of the study population are presented in difference in Oxalobacteraceae abundance (P50.4). We tested Supplemental Tables 1 and 2. Dietary oxalate (P50.05), for an association between oxalate intake and bacterial species fructose (P50.03), and protein (P50.04) intake was higher abundance and found no correlations after correction for mul- among control participants, and antibiotic exposure within tiple comparisons. None of the oxalate-degrading species were 3–12 months of enrollment was higher among those with associated with oxalate intake, even before accounting for mul- kidney stones (P,0.001). For the 36 participants with kid- tiple comparisons. We aligned reads to a database of bacterial ney stones and 31 controls for whom data were available on gene sequences and tested for a difference in the abundance of antibiotic exposure within the first 3 years of life, this ex- oxalyl-CoA decarboxylase or formyl-CoA transferase, bacterial posure was also higher among those who formed kidney genes involved in oxalate degradation (Figure 1C). We found stones (P50.03). that the abundance of these genes was not different between groups (P$0.29). Shotgun Metagenomic Sequencing Reveals Taxonomic and Gene Alterations in Children with Kidney Stones Fecal Metabolome of Those with Kidney Stones Is The overall taxonomic profile of the gut microbiome among Correlated with Bacterial Abundance participants with kidney stones and controls was similar to We carried out untargeted metabolomics of fecal samples to that observed in previous studies,36 where the Bacteroidetes identify metabolite products that may be associated with dif- and Clostridia accounted for most of the bacterial population ferences in the microbiome. As with the taxonomic analysis, (Supplemental Figure 1). Among the Bacteroidetes, the species the overall profile of metabolites was similar between partic- Bacteroides vulgatus was especially prominent in a subset of ipants with kidney stones and controls (Figure 2A). We carried samples, with abundance values .40%. out an LDA to determine if a subset of the metabolites could be

JASN 31: 1358–1369, 2020 Gut Microbiome and Metabolome 1361 CLINICAL RESEARCH www.jasn.org

Table 2. Antibiotic exposure within 3–12 months more abundant among those with kidney fi Count (%) stones were classi ed under the super- P Exposure pathway of amino acids and derivatives, Participants with Kidney Stones Controls Valuea (n539) (n537) whereas metabolites less abundant in those with kidney stones were classified Any antibiotic 28 (72) 9 (24) ,0.001 1 7 (18) 5 (14) as lipids and lipid-like molecules. 2 11 (28) 4 (11) We hypothesized that the levels of 3 6 (15) 0 (0) these metabolites could be associated 4 1 (3) 0 (0) with the relative abundance of bacterial 6 1 (3) 0 (0) taxa identified in our study. We computed 7 2 (5) 0 (0) a correlation matrix between metabolites Sulfa 7 (18) 0 (0) 0.06 and species abundances (Figure 2D), 1 6 (15) 0 (0) which revealed strong correlations for 2 1 (3) 0 (0) many of the taxon-metabolite pairs. , Cephalosporin 18 (46) 1 (3) 0.001 E. faecalis was positively correlated 1 8 (21) 1 (3) with three amino acid derivatives as 2 8 (21) 0 (0) 3 1 (3) 0 (0) well as other metabolites. Other bacterial 4 1 (3) 0 (0) species were correlated with several me- Fluoroquinolone 8 (21) 0 (0) 0.02 tabolites, in both a positive and negative 1 4 (10) 0 (0) direction. Thus, we found a set of metab- 2 3 (8) 0 (0) olites that distinguished those with kidney 3 1 (3) 0 (0) stones and controls, and which were Nitrofurantoin 0 (0) 0 (0) — highly correlated with a select group of Penicillin 7 (18) 3 (8) 0.29 bacterial taxa. 1 6 (15) 3 (8) 2 1 (3) 0 (0) Oxalate-Degrading Bacteria in Broad spectrum penicillin 0 (0) 0 (0) — Those with Kidney Stones Form a Metronidazole 0 (0) 0 (0) — Macrolide 3 (8) 5 (14) 0.73 Modular Network 1 2 (5) 3 (8) Following our untargeted analysis, we 2 1 (3) 2 (5) focused on oxalate-degrading bacteria, Helicobacter pylori 0(0) 0(0) — which have received considerable atten- treatment tion in the literature on nephrolithiasis Tetracycline 2 (5) 0 (0) 0.50 given the high prevalence of calcium ox- 1 1 (3) 0 (0) alate kidney stones.37239 We constructed 2 1 (3) 0 (0) an abundance correlation network for — Antimycobacterial 0 (0) 0 (0) oxalate-degrading bacteria detected in . Lincosamide 1 (3) 0 (0) 0.99 this study (Figure 3A), and found that 2 1 (3) 0 (0) it consisted of positive correlations Aminoglycoside 2 (5) 0 (0) 0.50 Unknown class 0 (0) 1 (3) .0.99 among oxalate-degrading taxa and 2 0 (0) 1 (3) lacked any strong negative correlations. aExact McNemar significance probability. The network consisted of two modules: one populated by relatively high- abundance, oxalate-degrading taxa and used to distinguish the two groups, and found that a linear the other by low-abundance Lactobacilli (Figure 3B). In the discriminant separated those with kidney stones from controls high-abundance module, we observed that E. feacium was cor- (Figure 2B) with 77% accuracy. Thus, we found a subset of related with both Eggerthella lenta and E. faecalis. B. animalis fecal metabolites that differed in participants with kidney was also connected to E. faecium, but was not otherwise at- stones. tached to the module. We compared metabolite abundances and identified 18 Three of the species attached to the high-abundance module metabolitesthatweresignificantly different between partici- were identified as lower in those who were stone formers versus pants with kidney stones and controlsat a pre-specified nom- controls: E. faecalis, E. faecium,andB. animalis. Moreover, inal P,0.01 (Figure 2C). Ten metabolites were more abundant these species were associated with metabolites that distin- in those who were stone formers versus controls, whereas guished the groups. Thus, we observed that a module of eight metabolites were less abundant in those with kidney oxalate-degrading bacteria varied in a coordinated manner stones than controls. Most of the metabolites that were that tracked with metabolite differences.

1362 JASN JASN 31: 1358–1369, 2020 JASN bacterial of range. abundance interquartile Relative IQR, plot. the degradation. of oxalate right (C) the and to production marked butyrate are (B) species for oxalate-degrading genes identi and Taxa Butyrate-producing (A) disease. controls. stone with kidney with participants among lower 1. Figure 31: C A Lachnoclostridium phytofermentans Relative abundance Ruminococcus champanellensis 1358 1e−06 1e−05 1e−04 h eaieaudneo 2bceiltx n h bnac ftebceilgn uyy-o eyrgns were dehydrogenase butyryl-CoA gene bacterial the of abundance the and taxa bacterial 32 of abundance relative The Ethanoligenens harbinense Desulfovibrio desulfuricans Butyrivibrio proteoclasticus Treponema succinifaciens Oscillibacter valericigenes – Slackia heliotrinireducens Streptococcus pyogenes Faecalitalea cylindroides 39 2020 1369, Bifidobacterium animalis Mageeibacillus indolicus Clostridium sp.SY8519 Clostridium perfringens Lachnospiraceae uncl. Enterococcus faecium Enterococcus faecalis Eubacterium limosum Clostridium botulinum Desulfovibrio vulgaris Campylobacter jejuni Clostridioides difficile Ruminococcus albus Proteobacteria uncl. Campylobacter coli Streptococcus suis Roseburia hominis Clostridiales uncl. Clostridium uncl. oxalyl−CoA decarboxylase Firmicutes uncl. Bacteria uncl. Oxalate degrader Butyrate producer P=0.99 10 −5 Relative abundance(medianandIQR) 10 −4 1e−05 3e−05 1e−04 Case Control 10 −3 formyl−CoA transferase Case Control 10 −2 P=0.29 ﬁ da ifrnilyaudn ntoewt inysoe compared stones kidney with those in abundant differentially as ed B Relative abundance Relative abundanceRelative abundance Relative abundance 3e−04 5e−04 1e−03 1e−04 3e−04 1e−03 1e−06 3e−06 1e−05 3e−05 1e−06 1e−05 1e−04 u irboeadMetabolome and Microbiome Gut www.jasn.org butyryl−CoA dehydrogenase acetate CoA−transferase acetate CoA−transferase butyrate kinase alpha subunit beta subunit P=0.78 P=0.02 P=0.61 P=0.76 LNCLRESEARCH CLINICAL 1363 CLINICAL RESEARCH www.jasn.org

A 20 B 6

10 4 2 Control 0 Control 0 Case 6 −10 4 PC2 axis (5.47%) Number of samples 2 Case −20 0 −2 0 2 −40 −20 0 Linear discriminant axis for fecal metabolites PC1 axis (13%) C D Slackia heliotrinireducens ******** ***** Ruminococcus champanellensis ******** ***** Oscillibacter valericigenes ******** ***** Desulfovibrio desulfuricans ******** ****** 2−piperidinone Enterococcus faecalis **** ***** N−acetyl−isoputreanine* Bacteria uncl. ************ dehydroepiandrosterone sulfate (DHEA−S) Mageeibacillus indolicus ***** * *** diacetylspermidine* Ruminococcus albus ***** ****** homocitrulline Treponema succinifaciens ***** * *** Firmicutes uncl. imidazole propionate **** * *** Proteobacteria uncl. N−acetylhistamine ******* ***** Ethanoligenens harbinense 4−acetamidobenzoate ****** ***** Streptococcus suis homoserine **** * *** Roseburia hominis N6,N6−dimethyllysine ***** * *** Eubacterium limosum suberate (C8−DC) *** ** * * * * Faecalitalea cylindroides *** *** sebacate (C10−DC) Clostridioides difficile octadecanedioate (C18−DC) *** * * Clostridium sp. SY8519 **** * *** azelate (C9−DC) Desulfovibrio vulgaris triethanolamine ******* ***** Lachnoclostridium phytofermentans *** * * linoleate (18:2n6) Enterococcus faecium *** oleate/vaccenate (18:1) Clostridium uncl. *** ****** 5−hydroxyhexanoate Clostridium botulinum **** * **** −2 0 2 4 Clostridiales uncl. *** * *** Butyrivibrio proteoclasticus Estimated difference in level **** * * Lachnospiraceae uncl. *** *** (95% confidence interval) Clostridium perfringens **** ****** Streptococcus pyogenes ******* * * Bifidobacterium animalis ****** Campylobacter coli Campylobacter jejuni

Correlation 0.50 homoserine

0.25 homocitulliner 2−piperidinone triethanolamine azelate (C9−DC) 0.00 linoleate (18:2n6) suberate (C8−DC) N−acetylhistamine diacetylspermidine* sebacate (C10−DC) imidazole propionate 5−hydroxyhexanoate N6,N6−dimethyllysine

−0.25 4−acetamidobenzoate oleate/vaccenate (18:1) −0.50 N−acetyl−isoputreanine* octadecanedioate (C18−DC) dehydroepiandrosterone sulfate (DHEA−S)

Figure 2. The fecal metabolome is distinct in participants with kidney stone disease. (A) PCA of untargeted fecal metabolite survey. (B) LDA distinguishes those with kidney stones from controls with 77% accuracy. (C) Metabolites increased (right) and decreased (left) in those who form kidney stones. (D) Correlation of metabolite concentration and bacterial abundance. Stars indicate signiﬁcant correlations, after controlling for study group. PC, principal component.

Children with Kidney Stones Exhibit a Unique those who were kidney stone formers (linear term P50.030, Age-Dependent Microbiota Profile quadratic term P50.029 for richness, linear term P50.008, As expected, a diversity of the gut microbiome was lower in quadratic term P50.009 for Shannon index). The lowest a participants with kidney stones compared with controls, as- diversity was found among individuals who first formed kid- sessed by richness (P,0.001) and Shannon index (P50.01; ney stones between 9 and 14 years of age. In contrast, the a Figure 4A). In addition to an overall lower value in cases, the a diversity of the microbiome of control participants was similar diversity exhibited an age-dependent association in those with across the age spectrum, and there were no significant associ- kidney stones but not in controls (Figure 4B). The association ations found with age. was not described by a simple linear relationship: bacterial In an exploratory analysis, we performed a regression anal- diversity first decreased and then increased with age among ysis of bacterial taxon abundance to determine which bacteria

1364 JASN JASN 31: 1358–1369, 2020 www.jasn.org CLINICAL RESEARCH

1e−02 Lactobacillus gasseri Control Case

1e−03

r 1.0 Lactobacillus acidophilus 1e−04 Lactobacillus casei 0.5 0.0 Eggerthella lenta

Enterococcus faecalis -0.5 Relative abundance 1e−05 Lactobacillus salivarius -1.0

Lactobacillus johnsonii Abundance Lactobacillus rhamnosus 1e−06 1e−04 Enterococcus faecium 2e−04 3e−04 Lactobacillus plantarum 4e−04 5e−04 Eggerthella lenta Lactobacillus casei Lactobacillus gasseri Enterococcus faecalis Enterococcus faecium Lactobacillus johnsonii Lactobacillus salivarius Lactobacillus plantarum Oxalobacter formigenes Bifidobacterium animalis Lactobacillus rhamnosus Lactobacillus acidophilus

Leuconostoc mesenteroides Bifidobacterium animalis Leuconostoc mesenteroides taxa

Figure 3. Oxalate-degrading bacterial taxa identified as decreased among kidney stone formers were components of a larger correlation network. (A) Correlation network of oxalate-degrading bacteria observed in the study. Taxa are connected if the absolute correlation was .0.5. Node size corresponds to mean taxon abundance. (B) Abundance of oxalate-degrading species. might be associated with the age-dependent pattern of a di- the function of the gut-kidney axis in early-onset calcium versity in those with kidney stones. We tested 77 taxa with a oxalate kidney stone disease and suggest that loss of bacteria, relative abundance of at least 0.1% in any participant who was particularly those that produce butyrate and degrade oxalate, a kidney stone former, and identified 13 taxa that exhibited a may act synergistically to influence gut metabolism and favor similar pattern as the overall a diversity (Figure 4C). For these kidney stone formation. taxa, the uncorrected P values for their associations with linear Calcium oxalate kidney stone disease, which is the most or quadratic terms for age at first stone were ,0.05, but none common form of nephrolithiasis,21 occurs when calcium were significantly associated after FDR correction. Two of the and oxalate become supersaturated in the urine and crystal- species, E. lenta and E. faecium, are oxalate degraders that were lization occurs. To date, evaluation and treatment of neph- identified in our correlation network analysis. We observed rolithiasis has focused on the kidney rather than upstream partial overlap between age-dependent species and species perturbations that may affect urine chemistries and other me- identified as less abundant in participants with kidney stones. diators of stone formation. The gut-kidney axis represents a potential causal pathway between the gut microbiome, intestinal metabolites, urine metabolites, and urine chemistries.17 In DISCUSSION the context of kidney stone disease, this axis provides a frame- work for understanding how exposures that perturb the com- We found that the gut microbiome of children and adolescents position of the gut microbiome might have downstream effects with calcium oxalate kidney stone disease is less diverse than on the intestinal and urinary tracts. Our group recently dem- that of controls, consistent with prior studies in adults.18,19 In onstrated that exposure to certain oral antibiotics was associ- particular, butyrate-producing bacteria and a network of ated with a 1.3- to 2.3-fold increased odds of developing kidney oxalate-degrading species were less abundant in children stones.40 The risk was greatest for those exposed at younger with calcium oxalate kidney stones compared with controls. ages, which is consistent with reports that exposure to antibi- In addition, levels of 18 metabolites, largely in the amino acid otics early in life produce greater alterations of host metabolism and lipid superfamilies, were different in the stool of youth than those later in life.41 This study, in which prior antibiotic with early-onset kidney stone disease compared with healthy exposure was also higher among individuals with nephrolithia- children. These metabolites were correlated with the fecal bac- sis, provides a potential link between antibiotic exposure, gut teria that were less abundant in those with kidney stones versus dysbiosis,18,19 and nephrolithiasis. controls. We also identified an age dependence of microbial In addition to an overall lower diversity of the gut micro- diversity among those who were kidney stone formers, and biome, shotgun metagenomics revealed two themes of al- we identified candidate bacteria that may underlie the age- tered bacterial metabolism (decreased butyrate production dependent pattern observed. Collectively, our results highlight and decreased oxalate degradation) among individuals with

JASN 31: 1358–1369, 2020 Gut Microbiome and Metabolome 1365 CLINICAL RESEARCH www.jasn.org

A C Bifidobacteium Alistipes Alistipes finegoldii Alistipes shahii adolescentis p=0.0003 1e−01

400 1e−02 1e−03

200 1e−04 Richness 1e−05 (10k reads/sample)

0 Clostridioides difficile Eggerthella lenta Enterococcus faecium Eubacterium limosum Control 4 p=0.011 1e−01 Case 1e−02 3 1e−03 2 1e−04

1 1e−05 Shannon diversity Gordonibacter Oscillibacter Porphyromonas Porphyromonas 0 pamelaeae valericigenes asaccharolytica gingivalis 1e−01

Relative abundance 1e−02 B Control Case

500 (10k reads/sample) 1e−03 400

Richness 1e−04 300 1e−05 200 Streptococcus 6 9 12 15 6 9 12 15 6 9 12 15 thermophilus 100 Age p = 0.214 Age p = 0.030 1e−01 0 Age2 p = 0.236 Age2 p = 0.029 1e−02 Shannon diversity 3 1e−03

2 1e−04

1e−05 1 Age p = 0.164 Age p = 0.008 2 2 6 9 12 15 0 Age p = 0.196 Age p = 0.009 Age of first stone 6 9 12 15 6 9 12 15 Age of first stone

Figure 4. The lowest diversity of the gut microbiome was found among individuals who first formed kidney stones between 9 and 14 years of age. (A) Species richness and Shannon diversity of bacterial communities in participants with kidney stones and controls. (B) Quadratic fits of diversity versus age of stone formation. (C) Candidate taxa associated with age-dependent diversity profile in those who form kidney stones. early-onset calcium oxalate kidney stone disease. Butyrate- Ruminococcus, Bifidobacterium,andOscillospira.Incontrast producing organisms that were lower in abundance among to the untargeted approach we took, they focused their anal- those who formed kidney stones included Roseburia, which ysis on microbial networks involved in oxalate homeostasis accounts for 1% of all bacteria present in the gut microbiome (taxa associated with Oxalobacter species or those stimulated by and is particularly sensitive to dietary intake.42 Butyrate is a oxalate in a rodent model). The absence of Oxalobacter in our short-chain fatty acid that is a mediator of inflammation43,44 samples could explain the differences in the composition of the and, importantly, helps maintain the gut mucosal barrier and oxalate-degrading network between studies, as could differ- regulates expression of SLC26 oxalate transporters in the in- ences in the gut microbiome between children and adults. A testine.45247 These functions of butyrate suggest that loss of recent study found that bacterial genes involved in oxalate butyrate production would increase absorption of oxalate in degradation were less abundant in the stool of adults with the gut and lead to increased urinary oxalate excretion. kidney stones and showed that these genes were represented We also found lower abundance of oxalate-degrading in several bacterial species (but not Oxalobacter), whose bacteria, namely E. faecalis, E. faecium,andB. animalis. cumulative abundance inversely correlated with urine oxa- These three taxa were part of an oxalate-degrading network, late excretion.20 Although our study did not identify over- which included lower abundance taxa in the Lactobacillus all differences in gene abundance, we identified several genus. However, we found no differences in the abundance oxalate-degrading species that were reduced in abundance, of O. formigenes, which is an anaerobic gram-negative bac- indicating a species-dependent difference in oxalate degra- teria that has been the focus of much attention37,38,48 due to dation function for the microbiome of children with kidney its use of oxalate as its carbon and energy source. Miller stones. Cumulatively, these results suggest that the commu- et al.49 have also reported a microbial network associated nity of oxalate-degrading bacteria might influence the de- with oxalate degradation among adults, which included velopment of calcium oxalate stone disease across the life

1366 JASN JASN 31: 1358–1369, 2020 www.jasn.org CLINICAL RESEARCH span and is one that includes far more bacteria than O. the highest rates of administration for children ,10 years formigenes. old and women.50 Children receive more antibiotics than We found strong evidence of perturbations in the gut me- any other age group, and 30% of antibiotics prescribed during tabolome among children with kidney stone disease. In par- ambulatory care visits are inappropriate.51253 Future studies ticular, 18 metabolites differed between cases and controls, should determine if the differences found in the gut microbiome with ten metabolites more abundant and eight metabolites and metabolome in early-onset kidney stone disease are due to less abundant among individuals with nephrolithiasis. Most antibiotic exposure. If confirmed in other studies, the unique of the metabolites that were more abundant among individ- microbial communities and their associated functions we iden- uals with nephrolithiasis were in the amino acid or derivatives tified could lead to new therapies for kidney stone prevention. superfamily, whereas most of the metabolites that were less For example, recent murine studies have demonstrated that abundant were in the lipid superfamily. These differences in downregulating the NLRP1-mediated inflammasome pathway fecal metabolites were largely explained by the lower abun- can expand the community of organisms that produce butyrate, dance of the 31 specific bacterial taxa among individuals with which may help prevent inflammatory bowel disease.54 early-onset nephrolithiasis, with oleate/vaccinate being the We acknowledge there are several limitations to this study. only metabolite not associated with any bacterial taxa. These As in all observational studies, unmeasured confounding and metabolites could help focus future investigations that seek to bias are possible. We reduced selection bias by matching on identify the causal pathway between external exposures that age, sex, and race, and nesting controls in the same health perturb the gut microbiome and nephrolithiasis. system from which participants with kidney stones were de- Our results provide the first evidence of an altered gut mi- rived. We also extensively phenotyped participants and col- crobiome in early-onset kidney stone disease and the first ev- lected information on antibiotics and other exposures that idence of an associated altered gut metabolome in individuals affect the microbiome (e.g., diet and other medications). How- of any age with calcium oxalate nephrolithiasis. This study is ever, the reliance on self-report/interview for some of the also the first to use shotgun metagenomics to investigate the medication exposures is a limitation, particularly for more gut microbiome in nephrolithiasis, which improves resolution remote prescriptions for which missingness was greater. Al- of species-level identification of bacteria and their biologic though the current sample size is too small to determine how functions compared with 16S ribosomal RNA sequencing. diet and antibiotics perturb the gut microbiome and metab- This resolution allowed us to investigate overall and species- olome and affect kidney stone disease, we are currently ex- level differences in gut microbiome diversity by the age of panding the population to perform these mediation analyses. disease onset. In particular, the lowest microbial diversity It is also possible that those who form kidney stones, particu- was found among children who first formed kidney stones larly children with a strong family history of kidney stones, between 9 and 14 years of age, with 13 taxa nominally associated could have genetic differences in intestinal physiology that with the age of nephrolithiasis onset. We found no association affect the microbiome and metabolome as an alternate expla- between age and a diversity in matched control samples. These nation for the differences observed. Second, this study could results are consistent with our prior study, which demonstrated not assess the structure and function of the microbiome before that the incidence rates of nephrolithiasis begin to increase at stone formation. Third, microbiome data are high dimen- 10 years of age.14 We hypothesize that this nadir of microbial sional and thus require special considerations to reduce spu- diversity may increase susceptibility to other well described rious results. In addition, many species and biochemicals exposures that increase the risk of incident nephrolithiasis, appear rarely in microbiome and metabolomic data, respec- such as increased dietary sodium or low fluid intake, which tively. Therefore, when conducting multiple tests, we controlled were similar between participants with kidney stones and con- for a prespecified FDR, did not test rare species, and confirmed trols in this study. Although more work is needed to follow up the significance of the taxon-metabolite associations with per- on age-dependent associations in children who form kidney mutation testing. Fourth, butyrate, which is volatile, could not stones, we have identified a set of bacterial candidates worthy of be directly measured in the untargeted metabolomics assay. We further investigation. also did not analyze the urine microbiome or measure urine Our findings provide insight into understanding the in- metabolites and chemistries. Future studies should extend crease in the prevalence of nephrolithiasis and the shift toward analyses to urine to examine the effect of perturbations of an earlier age of onset.14 The rapidity of this change suggests the gut microbiome on the kidney. Finally, imaging was not that changes in external, potentially modifiable, factors are obtained for controls. Thus, some may have had asymptomatic driving this change. It seems unlikely that traditional risk fac- stones, reported to be found among 4% of adults, but would tors for nephrolithiasis, such as low fluid intake, have changed likely be markedly less common among children.55 The pres- that dramatically to cause both a 70% increase in the preva- ence of stones in controls would bias results to the null, and lence of nephrolithiasis over the past two decades and rising would cause us to potentially underestimate associations with incidence during childhood. However, antibiotic use has in- kidney stone formation. creased over this time period. Over 250 million antibiotic Loss of gut bacteria, particularly those that produce buty- courses were prescribed in the United States in 2011, with rate and degrade oxalate, is associated with perturbations of

JASN 31: 1358–1369, 2020 Gut Microbiome and Metabolome 1367 CLINICAL RESEARCH www.jasn.org the metabolome that may be upstream determinants of early- using the health improvement network. Clin J Am Soc Nephrol 9: onset kidney stone disease. 2133–2140, 2014 5. Taylor EN, Feskanich D, Paik JM, Curhan GC: Nephrolithiasis and risk of incident bone fracture. JUrol195: 1482–1486, 2016 6. Alexander RT, Hemmelgarn BR, Wiebe N, Bello A, Samuel S, ACKNOWLEDGMENTS Klarenbach SW, et al.; Alberta Kidney Disease Network: Kidney stones and cardiovascular events: A cohort study. Clin J Am Soc Nephrol 9: 506–512, 2014 The views expressed in this article are those of the authors and do 7. Ferraro PM, Taylor EN, Eisner BH, Gambaro G, Rimm EB, Mukamal KJ, necessarily represent the ofﬁcial view of the National Institute of et al.: History of kidney stones and the risk of coronary heart disease. Diabetes and Digestive and Kidney Diseases (NIDDK). JAMA 310: 408–415, 2013 8. Rule AD, Roger VL, Melton LJ 3rd, Bergstralh EJ, Li X, Peyser PA, et al.: Kidney stones associate with increased risk for myocardial infarction. J Am Soc Nephrol 21: 1641–1644, 2010 DISCLOSURES 9. Scales CD Jr, Smith AC, Hanley JM, Saigal CS; Urologic Diseases in America Project: Prevalence of kidney stones in the United States. Eur Dr. Denburg reports grants from CHOP, National Center for Complemen- Urol 62: 160–165, 2012 tary and Integrative Health, NIDDK, Mallinckrodt Pharmaceuticals, Patient- 10. Pearle MS, Calhoun EA, Curhan GC; Urologic Diseases of America Centered Outcomes Research Institute (PCORI), outside the submitted work. Project: Urologic diseases in America project: Urolithiasis. J Urol 173: Dr. Tasian reports grants from CHOP (Foerderer grant), National Institutes of 848–857, 2005 Health (NIH)/NIDDK, PCORI, Pennsylvania Department of Health, during 11. Stamatelou KK, Francis ME, Jones CA, Nyberg LM, Curhan GC: Time the conduct of the study; and personal fees from Allena Pharmaceuticals and trends in reported prevalence of kidney stones in the United States: Lumenis Inc., outside the submitted work. All remaining authors have nothing 1976-1994. Kidney Int 63: 1817–1823, 2003 to disclose. 12. Dwyer ME, Krambeck AE, Bergstralh EJ, Milliner DS, Lieske JC, Rule AD: Temporal trends in incidence of kidney stones among children: A 25-year population based study. JUrol188: 247–252, 2012 FUNDING 13. Scales CD Jr, Curtis LH, Norris RD, Springhart WP, Sur RL, Schulman KA, et al.: Changing gender prevalence of stone disease. JUrol177: 979–982, 2007 This study was supported by funding from the CHOP Foerderer grant, the 14. Tasian GE, Ross ME, Song L, Sas DJ, Keren R, Denburg MR, et al.: Commonwealth Universal Research Enhancement (CURE) program’sTo- Annual incidence of nephrolithiasis among children and adults in bacco Formula grant SAP #4100068710, and NIH/NIDDK grant South Carolina from 1997 to 2012. Clin J Am Soc Nephrol 11: K23DK106428 (to Dr. Tasian). 488–496, 2016 15. Wang HH, Wiener JS, Lipkin ME, Scales CD Jr, Ross SS, Routh JC: Es- timating the nationwide, hospital based economic impact of pediatric SUPPLEMENTAL MATERIAL urolithiasis. JUrol193[Suppl 5]: 1855–1859, 2015 16. Tasian GE, Kabarriti AE, Kalmus A, Furth SL: Kidney stone recurrence among children and adolescents. J Urol 197: 246–252, 2017 This article contains the following supplemental material online at 17. Robijn S, Hoppe B, Vervaet BA, D’Haese PC, Verhulst A: Hyperoxaluria: http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2019101131/-/ Agut-kidneyaxis?Kidney Int 80: 1146–1158, 2011 DCSupplemental. 18. Stern JM, Moazami S, Qiu Y, Kurland I, Chen Z, Agalliu I, et al.: Evidence Supplemental Table 1. Daily dietary intake in kidney stone formers for a distinct gut microbiome in kidney stone formers compared to non- stone formers. Urolithiasis 44: 399–407, 2016 versus controls. 19. Tang R, Jiang Y, Tan A, Ye J, Xian X, Xie Y, et al.: 16S rRNA gene se- Supplemental Table 2. Current/recent medication use in kidney quencing reveals altered composition of gut microbiota in individuals stone formers versus controls. with kidney stones. Urolithiasis 46: 503–514, 2018 Supplemental Figure 1. Heatmap of bacterial taxa in children with 20. Ticinesi A, Milani C, Guerra A, Allegri F, Lauretani F, Nouvenne A, et al.: kidney stone disease and matched healthy controls. Understanding the gut-kidney axis in nephrolithiasis: An analysis of the gut microbiota composition and functionality of stone formers. Gut 67: 2097–2106, 2018 21. Lieske JC, Rule AD, Krambeck AE, Williams JC, Bergstralh EJ, Mehta REFERENCES RA, et al.: Stone composition as a function of age and sex. Clin J Am Soc Nephrol 9: 2141–2146, 2014 1. Alexander RT, Hemmelgarn BR, Wiebe N, Bello A, Morgan C, Samuel S, 22. Feskanich D, Sielaff BH, Chong K, Buzzard IM: Computerized collection et al.: Alberta Kidney Disease Network: Kidney stones and kidney and analysis of dietary intake information. Comput Methods Programs function loss: A cohort study. BMJ 345: e5287, 2012 Biomed 30: 47–57, 1989 2. Denburg MR, Jemielita TO, Tasian GE, Haynes K, Mucksavage P, Shults 23. Johnson RK, Driscoll P, Goran MI: Comparison of multiple-pass 24-hour J, et al.: Assessing the risk of incident hypertension and chronic kidney recall estimates of energy intake with total energy expenditure de- disease after exposure to shock wave lithotripsy and ureteroscopy. termined by the doubly labeled water method in young children. JAm Kidney Int 89: 185–192, 2016 Diet Assoc 96: 1140–1144, 1996 3. Carbone LD, Hovey KM, Andrews CA, Thomas F, Sorensen MD, 24. Sievert YA, Schakel SF, Buzzard IM: Maintenance of a nutrient database Crandall CJ, et al.: Urinary tract stones and osteoporosis: Findings from for clinical trials. Control Clin Trials 10: 416–425, 1989 the women’s health initiative. JBoneMinerRes30: 2096–2102, 2015 25. Schakel SF, Sievert YA, Buzzard IM: Sources of data for developing and 4. Denburg MR, Leonard MB, Haynes K, Tuchman S, Tasian G, Shults J, maintaining a nutrient database. J Am Diet Assoc 88: 1268–1271, et al.: Risk of fracture in urolithiasis: A population-based cohort study 1988

1368 JASN JASN 31: 1358–1369, 2020 www.jasn.org CLINICAL RESEARCH

26. Bolger AM, Lohse M, Usadel B: Trimmomatic: A flexible trimmer for 42. Louis P, Flint HJ: Diversity, metabolism and microbial ecology of Illumina sequence data. Bioinformatics 30: 2114–2120, 2014 butyrate-producing bacteria from the human large intestine. FEMS 27. Ouyang Y, Ye F, Liang Y: A modified electronegativity equalization Microbiol Lett 294: 1–8, 2009 method for fast and accurate calculation of atomic charges in large 43. Puddu A, Sanguineti R, Montecucco F, Viviani GL: Evidence for the gut biological molecules. Phys Chem Chem Phys 11: 6082–6089, 2009 microbiota short-chain fatty acids as key pathophysiological molecules 28. Wood DE, Salzberg SL: Kraken: Ultrafast metagenomic sequence improving diabetes. Mediators Inflamm 2014: 162021, 2014 classification using exact alignments. Genome Biol 15: R46, 2014 44. Vieira EL, Leonel AJ, Sad AP, Beltrão NR, Costa TF, Ferreira TM, et al.: 29. Kanehisa M, Goto S: KEGG: Kyoto encyclopedia of genes and ge- Oral administration of sodium butyrate attenuates inflammation and nomes. Nucleic Acids Res 28: 27–30, 2000 mucosal lesion in experimental acute ulcerative colitis. J Nutr Biochem 30. Buchfink B, Xie C, Huson DH: Fast and sensitive protein alignment 23: 430–436, 2012 using DIAMOND. Nat Methods 12: 59–60, 2015 45. Canani RB, Terrin G, Elce A, Pezzella V, Heinz-Erian P, Pedrolli A, et al.: 31. Miller AW, Dearing D: The metabolic and ecological interactions of Genotype-dependency of butyrate efficacy in children with congenital oxalate-degrading bacteria in the mammalian gut. Pathogens 2: chloride diarrhea. Orphanet J Rare Dis 8: 194, 2013 636–652, 2013 46. Jiang Z, Asplin JR, Evan AP, Rajendran VM, Velazquez H, Nottoli TP, 32. Whitman WB, editor: Bergey’s Manual of Systematics of Archaea and et al.: Calcium oxalate urolithiasis in mice lacking anion transporter Bacteria, Hoboken, NJ, Wiley, 2015 Slc26a6. Nat Genet 38: 474–478, 2006 33. Benjamini Y, Hochberg Y: Controlling the false discovery rate: a practical and 47. Landry GM, Hirata T, Anderson JB, Cabrero P, Gallo CJ, Dow JA, et al.: powerful approach to multiple testing. JRStatSocB57: 289–300, 1995 Sulfate and thiosulfate inhibit oxalate transport via a dPrestin 34. Evans AM, Bridgewater BR, Liu Q, Mitchell MW, Robinson RJ, Dai H, (Slc26a6)-dependent mechanism in an insect model of calcium oxalate et al.: High resolution mass spectrometry improves data quantity and nephrolithiasis. Am J Physiol Renal Physiol 310: F152–F159, 2016 quality as compared to unit mass resolution mass spectrometry in high- 48. Azcarate-Peril MA, Bruno-Bárcena JM, Hassan HM, Klaenhammer TR: throughput profiling metabolomics. Metabolomics 4: 132, 2014 Transcriptional and functional analysis of oxalyl-coenzyme A (CoA) 35. Ward JB, Feinstein L, Pierce C, Lim J, Abbott KC, Bavendam T, et al.: decarboxylase and formyl-CoA transferase genes from Lactobacillus NIDDK Urologic Diseases in America Project: Pediatric urinary stone acidophilus. Appl Environ Microbiol 72: 1891–1899, 2006 disease in the United States: The urologic diseases in America project. 49. Miller AW, Choy D, Penniston KL, Lange D: Inhibition of urinary stone Urology 129: 180–187, 2019 disease by a multi-species bacterial network ensures healthy oxalate 36. Human Microbiome Project Consortium: Structure, function and di- homeostasis. Kidney Int 96: 180–188, 2019 versity of the healthy human microbiome. Nature 486: 207–214, 2012 50. Hicks LA, Bartoces MG, Roberts RM, Suda KJ, Hunkler RJ, Taylor TH Jr, 37. Kaufman DW, Kelly JP, Curhan GC, Anderson TE, Dretler SP, Preminger et al.: US outpatient antibiotic prescribing variation according to ge- GM, et al.: Oxalobacter formigenes may reduce the risk of calcium ography, patient population, and provider specialty in 2011. Clin Infect oxalate kidney stones. J Am Soc Nephrol 19: 1197–1203, 2008 Dis 60: 1308–1316, 2015 38. Kwak C, Kim HK, Kim EC, Choi MS, Kim HH: Urinary oxalate levels and 51. Pichichero ME: Dynamics of antibiotic prescribing for children. JAMA the enteric bacterium Oxalobacter formigenes in patients with calcium 287: 3133–3135, 2002 oxalate urolithiasis. Eur Urol 44: 475–481, 2003 52. Hicks LA, Taylor TH Jr, Hunkler RJ: U.S. outpatient antibiotic pre- 39. Siener R, Bangen U, Sidhu H, Hönow R, von Unruh G, Hesse A: The role scribing, 2010. N Engl J Med 368: 1461–1462, 2013 of Oxalobacter formigenes colonization in calcium oxalate stone dis- 53. Fleming-Dutra KE, Hersh AL, Shapiro DJ, Bartoces M, Enns EA, File TM ease. Kidney Int 83: 1144–1149, 2013 Jr, et al.: Prevalence of inappropriate antibiotic prescriptions among US 40. Tasian GE, Jemielita T, Goldfarb DS, Copelovitch L, Gerber JS, Wu Q, ambulatory care visits, 2010-2011. JAMA 315: 1864–1873, 2016 et al.: Oral antibiotic exposure and kidney stone disease. JAmSoc 54. Tye H, Yu CH, Simms LA, de Zoete MR, Kim ML, Zakrzewski M, et al.: Nephrol 29: 1731–1740, 2018 NLRP1 restricts butyrate producing commensals to exacerbate in- 41. Cox LM, Yamanishi S, Sohn J, Alekseyenko AV, Leung JM, Cho I, et al.: flammatory bowel disease. Nat Commun 9: 3728, 2018 Altering the intestinal microbiota during a critical developmental win- 55. Bansal AD, Hui J, Goldfarb DS: Asymptomatic nephrolithiasis detected dow has lasting metabolic consequences. Cell 158: 705–721, 2014 by ultrasound. Clin J Am Soc Nephrol 4: 680–684, 2009

JASN 31: 1358–1369, 2020 Gut Microbiome and Metabolome 1369 Table of Contents

Supplementary Table 1. Daily dietary intake in kidney stone formers versus controls.

Supplementary Table 2. Current/recent medication use in kidney stone formers versus controls.

Supplementary Figure 1. Heatmap of bacterial taxa in children with kidney stone disease and matched healthy controls. Each column of the heatmap represents one fecal sample and each row represents one bacterial taxon. Taxa were included if the abundance in any sample exceeded 0.1%. (with the exception of Oxalobacteraceae and Oxalobacter formigenes which were included despite lower abdundance due to their role in oxalate degradation). White cells denote taxa that were not observed in the sample. Supplement Table 1. Median daily dietary intake for participants Cases Controls Dietary Intake (N = 44) (N = 44) P value* Calcium (mg) 838.30 1016.90 0.11 Fructose (mg) 18.18 22.38 0.03 Magnesium (mg) 223.20 264.50 0.13 Oxalic Acid (mg) 125.50 157.80 0.05 Phytic Acid (mg) 561.10 645.90 0.24 Potassium (mg) 1867.10 2174.80 0.08 Butyric Acid (mg) 0.50 0.51 0.62 Sodium (mg) 2793.50 2980.40 0.33 Sucrose (mg) 36.25 47.12 0.07 Total Protein (g) 62.71 74.02 0.04 Water (mL) 2177.60 1922.30 0.02 Zinc (mg) 8.63 10.65 0.08 *Wilcoxon signed-rank test

Supplement Table 2. Current/recent medication use among cases and controls* Medication Cases Controls Pairs P value** Antacids 31 0.25 No 33 (87%) 36 (100%) Yes 5 (13%) 0 (0%) NSAIDS 31 0.61 No 31 (82%) 26 (72%) Yes 7 (18%) 10 (28%) Proton pump inhibitors 40 1.00 No 42 (100%) 42 (100%) Yes 0 (0%) 0 (0%) Histamine receptor antagonists 38 0.13 No 40 (98%) 35 (85%) Yes 1 (2%) 6 (15%) Narcotics 38 0.50 No 39 (95%) 41 (100%) Yes 2 (5%) 0 (0%) Tricyclic antidepressants 39 0.50 No 38 (90%) 41 (100%) Yes 4 (10%) 0 (0%) Anticholinergics 39 1.00 No 42 (100%) 41 (100%) Yes 0 (%) 0 (%) Metoclopramide 39 1.00 No 42 (100%) 41 (100%) Yes 0 (0%) 0 (0%) Anti-diarrheal medications No 42 (98%) 41 (100%) 40 1.00 Yes 1 (2%) 0 (0%) Laxatives No 39 (98%) 40 (100%) 36 1.00 Yes 1 (3%) 0 (0%) Current medication No 23 (53%) 32 (74%) 42 0.08 Yes 20 (47%) 11 (26%) Number of meds 1 6 (14%) 9 (21%) 2 6 (14%) 1 (2%) 3 6 (14%) 0 (0%) 4 1 (2%) 0 (0%) 7 0 (0%) 1 (2%) 8 1 (2%) 0 (0%) *Antacids and NSAIDS within 2 weeks. Other listed medications within 4 weeks. **Exact McNemar significance probability. Study group 1 Study group k__Bacteria p__Actinobacteria g__Adlercreutzia s__equolifaciens Control p__Actinobacteria g__Bifidobacterium 0.8 p__Actinobacteria g__Bifidobacterium s__adolescentis Case p__Actinobacteria g__Bifidobacterium s__animalis p__Actinobacteria g__Bifidobacterium s__bifidum p__Actinobacteria g__Bifidobacterium s__breve 0.6 p__Actinobacteria g__Bifidobacterium s__dentium p__Actinobacteria g__Bifidobacterium s__longum p__Actinobacteria g__Bifidobacterium s__thermophilum p__Actinobacteria g__Eggerthella s__lenta 0.4 p__Actinobacteria g__Eggerthella s__sp. YY7918 p__Actinobacteria g__Gordonibacter s__pamelaeae p__Actinobacteria g__Slackia s__heliotrinireducens 0.2 p__Bacteroidetes g__Alistipes p__Bacteroidetes g__Alistipes s__finegoldii p__Bacteroidetes g__Alistipes s__shahii p__Bacteroidetes g__Bacteroides 0 p__Bacteroidetes g__Bacteroides s__fragilis p__Bacteroidetes g__Bacteroides s__helcogenes p__Bacteroidetes g__Bacteroides s__salanitronis p__Bacteroidetes g__Bacteroides s__thetaiotaomicron p__Bacteroidetes g__Bacteroides s__vulgatus p__Bacteroidetes g__Odoribacter s__splanchnicus p__Bacteroidetes g__Parabacteroides s__distasonis p__Bacteroidetes g__Porphyromonas s__asaccharolytica p__Bacteroidetes g__Porphyromonas s__gingivalis p__Bacteroidetes g__Prevotella p__Bacteroidetes g__Prevotella s__dentalis p__Bacteroidetes g__Prevotella s__denticola p__Bacteroidetes g__Prevotella s__intermedia p__Bacteroidetes g__Prevotella s__melaninogenica p__Bacteroidetes g__Prevotella s__ruminicola p__Bacteroidetes g__Prevotella s__sp. oral taxon 299 p__Bacteroidetes g__Riemerella s__anatipestifer p__Bacteroidetes g__Tannerella s__forsythia p__Bacteroidetes o__Bacteroidales p__Firmicutes p__Firmicutes f__Lachnospiraceae p__Firmicutes g__ s__[Eubacterium] rectale p__Firmicutes g__Acidaminococcus s__intestini p__Firmicutes g__Butyrivibrio s__proteoclasticus p__Firmicutes g__Clostridioides s__difficile p__Firmicutes g__Clostridium p__Firmicutes g__Clostridium s__botulinum p__Firmicutes g__Clostridium s__perfringens p__Firmicutes g__Clostridium s__sp. SY8519 p__Firmicutes g__Enterococcus s__faecalis p__Firmicutes g__Enterococcus s__faecium p__Firmicutes g__Ethanoligenens s__harbinense p__Firmicutes g__Eubacterium s__[Eubacterium] eligens p__Firmicutes g__Eubacterium s__limosum p__Firmicutes g__Faecalitalea s__cylindroides p__Firmicutes g__Lachnoclostridium s__[Clostridium] saccharolyticum p__Firmicutes g__Lachnoclostridium s__phytofermentans p__Firmicutes g__Lactobacillus s__fermentum p__Firmicutes g__Lactobacillus s__rhamnosus p__Firmicutes g__Lactococcus s__lactis p__Firmicutes g__Mageeibacillus s__indolicus p__Firmicutes g__Megasphaera s__elsdenii p__Firmicutes g__Oscillibacter s__valericigenes p__Firmicutes g__Roseburia s__hominis p__Firmicutes g__Ruminococcus s__albus p__Firmicutes g__Ruminococcus s__champanellensis p__Firmicutes g__Streptococcus p__Firmicutes g__Streptococcus s__lutetiensis p__Firmicutes g__Streptococcus s__parasanguinis p__Firmicutes g__Streptococcus s__pyogenes p__Firmicutes g__Streptococcus s__salivarius p__Firmicutes g__Streptococcus s__suis p__Firmicutes g__Streptococcus s__thermophilus p__Firmicutes g__Veillonella s__parvula p__Firmicutes o__Clostridiales p__Proteobacteria p__Proteobacteria f__Enterobacteriaceae p__Proteobacteria f__Oxalobacteraceae p__Proteobacteria g__Campylobacter p__Proteobacteria g__Campylobacter s__coli p__Proteobacteria g__Campylobacter s__hominis p__Proteobacteria g__Campylobacter s__jejuni p__Proteobacteria g__Desulfovibrio s__desulfuricans p__Proteobacteria g__Desulfovibrio s__vulgaris p__Proteobacteria g__Escherichia p__Proteobacteria g__Escherichia s__coli p__Proteobacteria g__Haemophilus s__parainfluenzae p__Proteobacteria g__Klebsiella s__pneumoniae p__Proteobacteria g__Methylobacterium s__radiotolerans p__Proteobacteria g__Oxalobacter s__formigenes p__Proteobacteria g__Raoultella s__ornithinolytica p__Proteobacteria o__Enterobacterales p__Spirochaetes g__Treponema s__succinifaciens p__Verrucomicrobia g__Akkermansia s__muciniphila