Integrated multi-omics highlights alterations of gut microbiome functions in prodromal and idiopathic Parkinson’s disease

Integrated multi-omics highlights alterations of gut microbiome functions in prodromal and idiopathic Parkinson’s disease | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (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];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Integrated multi-omics highlights alterations of gut microbiome functions in prodromal and idiopathic Parkinson’s disease View ORCID Profile Remy Villette , View ORCID Profile Júlia Ortís Sunyer , View ORCID Profile Polina V. Novikova , View ORCID Profile Velma T. E. Aho , View ORCID Profile Viacheslav A. Petrov , View ORCID Profile Oskar Hickl , Susheel Bhanu Busi , View ORCID Profile Charlotte De Rudder , View ORCID Profile Benoit Kunath , View ORCID Profile Anna Heintz-Buschart , View ORCID Profile Jean-Pierre Trezzi , View ORCID Profile Rashi Halder , Christian Jäger , Laura A. Lebrun , Annegrät Daujeumont , View ORCID Profile Sebastian Schade , View ORCID Profile Annette Janzen , View ORCID Profile Nico Jehmlich , Martin von Bergen , View ORCID Profile Cédric C. Laczny , View ORCID Profile Patrick May , Claudia Trenkwalder , View ORCID Profile Wolfgang Oertel , Brit Mollenhauer , View ORCID Profile Paul Wilmes doi: https://doi.org/10.1101/2024.12.13.628341 Remy Villette 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Remy Villette Júlia Ortís Sunyer 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Júlia Ortís Sunyer Polina V. Novikova 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Polina V. Novikova Velma T. E. Aho 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Velma T. E. Aho Viacheslav A. Petrov 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Viacheslav A. Petrov Oskar Hickl 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Oskar Hickl Susheel Bhanu Busi 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg 2 UK Centre for Ecology and Hydrology , Wallingford, Oxfordshire, United Kingdom Find this author on Google Scholar Find this author on PubMed Search for this author on this site Charlotte De Rudder 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Charlotte De Rudder Benoit Kunath 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Benoit Kunath Anna Heintz-Buschart 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg 3 Biosystems Data Analysis, Swammerdam Institute for Life Sciences, University of Amsterdam , The Netherlands Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Anna Heintz-Buschart Jean-Pierre Trezzi 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jean-Pierre Trezzi Rashi Halder 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Rashi Halder Christian Jäger 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site Laura A. Lebrun 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site Annegrät Daujeumont 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sebastian Schade 5 Paracelsus-Elena-Klinik , Kassel, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Sebastian Schade Annette Janzen 6 Department of Neurology, Philipps-University Marburg , Marburg, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Annette Janzen Nico Jehmlich 7 Helmholtz-Centre for Environmental Research GmbH – UFZ, Department of Molecular Toxicology , Leipzig, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Nico Jehmlich Martin von Bergen 7 Helmholtz-Centre for Environmental Research GmbH – UFZ, Department of Molecular Toxicology , Leipzig, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site Cédric C. Laczny 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Cédric C. Laczny Patrick May 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Patrick May Claudia Trenkwalder 8 Department of Neurosurgery, University Medical Center Göttingen , Göttingen, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site Wolfgang Oertel 6 Department of Neurology, Philipps-University Marburg , Marburg, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Wolfgang Oertel Brit Mollenhauer 4 Department of Neurology, University Medical Center Göttingen , Göttingen, Germany 5 Paracelsus-Elena-Klinik , Kassel, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site Paul Wilmes 1 Luxembourg Centre for Systems Biomedicine, University of Luxembourg , Esch-sur-Alzette, Luxembourg 9 Department of Life Sciences and Medicine, Faculty of Science, Technology and Medicine, University of Luxembourg , Belvaux, Luxembourg 10 Departments of Genetics and Medicine (Hematology, Blood and Marrow Transplantation), Stanford University , Stanford, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Paul Wilmes For correspondence: paul.wilmes{at}uni.lu Abstract Full Text Info/History Metrics Supplementary material Preview PDF Summary Parkinson’s disease (PD) is associated with gut microbiome shifts, but the functional consequences remain unclear. Here, we use an integrated multi-omics approach to compare the gut microbiomes of individuals with PD and prodromal PD as well as healthy individuals. After analyzing each omics, meta-metabolomic was selected to inform the analysis as it represents the most discriminatory and robust ome. We identified 11 metabolites that were differentially abundant between the groups, amongst which β-glutamate was increased in PD and prodromal PD, and correlated with the transcriptional activities of Methanobrevibacter smithii and Clostridium spp. We identified decreases in transcripts, but not in gene abundances, related to glutamate metabolism, bile acids, chemotaxis and flagellar assembly in PD, particularly in keystone genera such as Roseburia, Agathobacter and Blautia . Our findings, integrated into the Expobiome map, reveal multifactorial microbiome alterations which converge with PD pathways. Our study highlights the importance of investigating the gut microbiome’s functional dimensions to better resolve microbiome-host interactions in health and disease. Introduction Parkinson’s disease (PD), a neurodegenerative disease impacting movement due to dopaminergic neuron loss, is the second most prevalent neurodegenerative disease worldwide 1 . Individuals with PD are often characterized by an increase in gut permeability, inflammation and constipation which, together, suggest a link between the gut microbiome and PD etiology 2 – 4 . This potential link is supported by numerous studies reporting differences in the gut microbiome structure of individuals with PD compared to healthy individuals 5 – 11 . These findings have been further confirmed by recent meta-analyses 12 , 13 . Together, the studies highlight a decreased abundance for the genera Roseburia , Blautia, Butyricoccus and Faecalibacterium in PD while Methanobrevibacter , Akkermansia , Lactobacillus , Bifidobacterium and Hungatella , are typically enriched 5 – 13 . Similar changes in idiopathic REM sleep behavior disorder (iRBD), a prodromal stage of PD 14 , have been reported 7 , 10 . Moreover, the taxa decreased in PD are known producers of short-chain fatty acids (SCFAs), which correspondingly have also been found to be decreased in concentration in fecal samples of PD individuals 6 , 15 , 16 . In addition to SCFAs, several microbiome-derived metabolites such as bile acids (BAs), glycine and glutamate have been associated with PD, either in plasma, serum 17 , 18 or stool 19 – 21 . BAs are produced by the host and metabolized by the gut microbiome into secondary bile acids with different cytotoxic capacities but also immunomodulatory capacities 22 , 23 . Glutamate is the major excitatory neurotransmitter and exerts toxic activity on neuronal cells 24 . Its levels in serum and cerebrospinal fluid have been reported as either increased 25 , 26 , not different 27 , or decreased 28 , but decreased in the gut in PD compared to healthy controls (HC) 29 . Altogether, alterations of the gut microbiome are linked to PD, but less is known about iRBD or other prodromal stages of the disease. Moreover, most of the associations between PD and the gut microbiome are based on taxonomic and metabolomic analyses. The resulting data, although insightful, lacks functional and systemic information that could better capture the complex crosstalk between the gut microbiome and the host in the context of PD. To obtain such information, we performed an integrated multi-omics study on a cross-sectional cohort comprised of individuals with iRBD and PD alongside HC. Metagenomics (MG), metatranscriptomics (MT), metaproteomics (MP) and meta-metabolomics (MM) were used to characterize taxonomic (taxMG, taxMT, taxMP) and functional (funMG, funMT, funMP, MM) differences between HC, iRBD and PD gut microbiomes. We identified substantial differences in gut microbiome functions and metabolites between the groups, including an increase in β-glutamate levels in PD that were related to a dysregulation of glutamate-related gene expression. Alterations in glutamate-related genes were linked with chemotaxis and flagellar assembly pathways, for which we identified strong and distinct taxonomic differences in transcription between PD and HC. Collectively, our data highlight the importance of multi-omics approaches for the identification of microbiome-mediated effects on neurological, and more broadly, complex human diseases involving host-microbiome interactions. Results Study cohort Our initial set of subjects consisted of 50 individuals with PD, diagnosed according to United Kingdom Parkinson’s Disease Society Brain Bank (UKPDSBB) clinical diagnostic criteria 30 , and 30 people with polysomnography-confirmed iRBD as well as 50 healthy control (HC) subjects. The data from 4 PD and 3 iRBD as well as 1 HC were subsequently excluded (see Methods), leading to a final data set of 46 PD, 27 iRBD and 49 HCs. The subjects in the three groups were of similar age but had slightly different gender distributions, with males overrepresented in the iRBD and PD groups (Table 1, Fisher test, p = 0.004), as is typical for these conditions 31 , 32 . Constipation, a prevalent non-motor symptom of PD 33 , was also more common in the iRBD and PD groups compared to HC (Fisher test, p < 0.001). Microbiome function is altered in PD Alpha diversity comparisons revealed no statistically significant differences between the three groups when considering taxMG, taxMT, funMG, taxMP and MM ( Fig. 2A , Mann-Whitney test, p > 0.05). However, funMT and funMP showed a statistically significant increase in alpha diversity in PD compared to HC and iRBD compared to PD, respectively ( Figure 2A , Mann-Whitney test, p < 0.05). We then analyzed beta diversity for all omics layers. TaxMG, funMG and taxMP revealed no statistically significant differences between the three groups (Extended fig. 1 A, B and D, PERMANOVA, p= 0.2, 0.8 and 0.35, respectively), while taxMT, funMT, funMP and MM showed a statistically significant separation of the groups, especially for funMT ( Figure 2B and C , Extended fig. 1 C, PERMANOVA p = 0.001, 0.001, 0.005 and p = 0.008, respectively). Permutation analysis revealed that funMT, funMP and MM resulted in the best separation of the three groups while taxMG, funMG and taxMP exhibited the lowest separation capacity (Extended fig. 1D, PERMANOVA R² of 0.043, 0.041, 0.038, 0.018, 0.014 and 0.017, respectively). Pairwise comparisons using PERMANOVA demonstrated that most differences were found between either HC and PD or HC and iRBD, with only funMP and funMT showing statistically significant differences between PD and iRBD ( Figure 2E ). Based on these differences, we next assessed how the confounding factors sex, age and constipation may impact beta diversity. Sex and constipation were found to have a significant association with beta diversity for taxMT and funMT, while age was associated with taxMG ( Figure 2F ). Importantly, MM and funMP were not found to be associated with any confounders ( Figure 2F ). Download figure Open in new tab Figure 1. Schematic representation of the analytical workflow. Metagenomic (MG), metatranscriptomic (MT), metaproteomic (MP) and meta-metabolomic (MM) data were generated for each sample. Pre-processed MG and MT reads were sample-wise assembled using the iterative hybrid assembly pipeline of the Integrated Meta-omics Pipeline (IMP). After assembly, taxonomic annotation was performed at the read and contig levels, followed by gene prediction and functional annotation on the assembled contigs. Expressed proteins (MP) were identified using the predicted genes from the MG/MT hybrid assembly. For these three omics levels, we generated taxonomic and functional profiles that are referred to as taxMG, taxMT and taxMP for the taxonomic level, and funMG, funMT and funMP for the functional level, respectively. Community-based networks were reconstructed from gene annotations. Finally, the meta-metabolome (MM) was integrated with the other omics data at the network level. The integrated multi-omics analysis was performed with the available clinical metadata. Download figure Open in new tab Figure 2. Microbiome structure is altered in PD and iRBD vs HC. A. Shannon index for different omics between Healthy Controls (HC), idiopathic REM sleep behaviour disorder (iRBD) and Parkinson’s Disease (PD). P-values are based on pairwise Mann-Whitney tests. Non-metric Multidimensional Scaling (NMDS) based on Bray-Curtis dissimilarity for taxonomic annotation of metatranscriptomic (taxMT) ( B ) and functional metatranscriptomic (funMT) data ( C ). D . Principal Component Analysis (PCA) for meta-metabolomic (MM) data based on untargeted, targeted SCFA and targeted bile acids abundance. All three quantifications have been sum-normalised before any merging. PCA was then computed on the merged matrix. All tests are based on PERMANOVA with 1000 permutations. E . Pairwise PERMANOVA between groups for each omics. F . PERMANOVA analysis for “Age”, “Sex” and “Constipation” for each omics. Size of rectangle is based on – log10(p-value) and colour on R² value. G . Differential abundance analysis using SIAMCAT for taxMG and taxMT. Values are pseudo fold changes for pairwise comparison between groups and size is based on –log10(p-value). Shape is based on significance before and after FDR correction. We next looked at differential abundances on taxonomic layers to highlight the compositional differences between the groups. No significant differences were found in taxMG at the genus and species levels between any of the three groups ( Figure 2G and Extended fig. 2C, q > 0.05, SIAMCAT). For taxMT, SIAMCAT highlighted an increase in Alistipes obesi and Ruthenibacterium lactatiformans in iRBD vs HC, Roseburia incertae sedis , Blautia massiliensis, B. obeum and Clostridium sp. were decreased in PD vs HC and no significant differences were found between PD and iRBD ( Figure 2G , q < 0.05, q 0.05, respectively). We identified the genus Eubacterium as being decreased in PD compared to HC in taxMT (Extended fig. 2C, q < 0.05, SIAMCAT). The ALDEx2 algorithm highlighted only Roseburia incertae sedis to be depleted in PD after FDR correction (Extended fig. 2B). Subsequently, we investigated overall differences in gene abundances and expressions of the microbial genes linked to the observed differences in metabolites using the KEGG database. We observed no statistically significant differences in gene abundances between HC vs PD or HC vs iRBD (funMG, Extended fig. 4A and B, q > 0.05). Gene expression highlighted one transcript upregulated in HC vs iRBD, but 145 transcripts upregulated in HC vs PD (funMT, Extended fig. 4C and D, q 0.05, respectively). No genes or transcripts were significantly higher in iRBD or PD compared to HC (Extended fig. 4A-D, q > 0.05). Finally, we found no statistically significant differences in gene or transcript abundances between PD and iRBD (data not shown). Altered metabolome of PD patients is linked to microbial abundance and activity Considering that MM is associated with PD and iRBD but not with confounders, we chose MM as a robust guide for further statistical comparisons. In addition, MM can be considered as one of the final outputs of microbial activity and an important driver of microbial effects on the host. For this purpose, we removed unidentified compounds from the statistical testing, because we cannot associate microbial genes with them. Our analyses revealed 11 statistically significantly different compounds between the groups, including alanine, β-glutamate, serine, and glycerol ( Figure 3A , q < 0.05). We found a significant increase in isovalerate, isobutyrate and valerate in PD patients ( Figure 3A , q < 0.05) but no differences for butyrate, acetate, formate, propionate or total SCFAs (data not shown). Primary bile acids glycocholic and chenodeoxycholic acids were decreased in PD and in iRBD and PD, respectively ( Figure 3A , q < 0.05). Glutamate was not differentially abundant between the three groups (data not shown). To unravel the effect of confounding factors, we measured the variance explained by each factor on the compounds’ abundances. Diagnosis explained more variance than constipation or sex ( Fig. 3B , diagnosis: mean= 4.18%; median = 3.33%, sex: mean = 2.08%, median = 1.3%; constipation: mean = 3.7%, median = 2.68%). We found only malic acid as being significantly associated with sex and no compound to be associated with constipation (Extended fig. 3A and B, q 0.05). Based on the PERMANOVA results and variance analysis, the differences observed in MM were most strongly associated with the disease status and, importantly, not with the confounding factors. Download figure Open in new tab Figure 3. Altered metabolome is linked with microbial activity and transcripts. A. Metabolite relative abundances for significant compounds. Metabolomic data from untargeted meta-metabolomics, targeted SCFA and targeted bile acids were combined after normalization by sum for each. Dunn test, FDR corrected. B . Spearman correlation between taxMT at the species level with statistically significant metabolites. All p-values are FDR corrected. Genera are selected based on differential abundance or relevance in the literature. C . Absolute log2 fold change between HC and PD for funMG and funMT associated to significant compounds. Dots are scaled by the –log10(p-value), colorized and shaped according to p-value significance before (triangle shape) and after FDR correction (round shape). D . Chemotaxis and flagellin assembly pathway genes expression and Shannon index fold change between HC and PD group. Dots are colorized and shaped according to p-value significance before (triangle shape) and after FDR correction (round shape). Wilcoxon test, FDR corrected. E . Functional diversity comparison for funMT KO found as significantly different in pairwise differential analysis. Shannon index was calculated for each KO for funMG and funMT. Only genes significantly different after FDR correction on a Kruskal-Wallis test are plotted. P-values are calculated using Dunn post hoc test. We next checked correlations between metabolite abundances and microbial abundances alongside activities. Metabolite abundances were linked to microbial abundances for taxMG and taxMT. TaxMG exhibited fewer significant correlations with metabolite abundances compared to taxMT, except for Methanobrevibacter smithii and Clostridium sp. CAG:273, which were positively correlated with β-glutamate levels (Extended fig. 3C, Spearman test, q < 0.01). Conversely, Prevotella copri and Faecalibacterium prausnitzii were negatively correlated with β-glutamate levels, while Roseburia hominis was positively correlated with glycerol levels (Extended fig. 3C, Spearman test, q < 0.001). Additionally, for taxMT we noted positive correlations with β-glutamate, isovalerate and isobutyrate with A. muciniphila and Clostridium sp . CAG:448 alongside Methanobrevibacter smithii ( Figure 3C , Spearman test, q < 0.01, q < 0.001 and q < 0.05, respectively), while F. prausnitzii, Blautia massilliensis, Blautia obeum/wexlerae, Blautia obeum and Clostridim spp. were negatively correlated with β-glutamate. In addition , R. hominis, Roseburia spp. and Clostridium spp. were positively correlated with glycerol ( Figure 3C , Spearman test, q < 0.05). Of note, while performing the correlations at the genus level, we noted negative correlations between Eubacterium and isovalerate, isobutyrate and valerate (taxMT, q < 0.05, Spearman test, data not shown). Strikingly, A. muciniphila , some Clostridium spp. and M. smithii were positively correlated with compounds increased in PD while R. hominis , F. prausnitzii, B. species and P. copri were negatively correlated with these same compounds, highlighting groups of bacteria being linked with either individuals with PD and iRBD or HC ( Figure 3C ). Expression of genes linked to glutamate, bile acids and flagella is dysregulated in the iRBD and PD gut microbiome To reinforce the results of the differential analysis, we acquired all the orthologs related to the metabolites which were statistically significant between the three groups. More specifically, we used regular expression matching to retrieve all orthologous genes that are linked to the above-mentioned metabolites in KEGG. Because the KEGG database only has one metabolite entry annotated as β-glutamate, we selected all glutamate-related genes instead (linked to both L-and D-glutamate). We found no statistically significant differences in gene abundances after correction between any of the three groups (funMG, Figure 3D and Extended fig. 3D, q > 0.05, Mann-Whitney test). However, we did find a decrease in transcripts in PD for the three known glutamate synthase genes (funMT, GLT1 :K00264, GLU :K00284 and gltB :K00265, Figure 3D , Mann-Whitney test, q = 0.004, q = 0.004 and p = 0.006, respectively) alongside a decrease in cheB in PD (K03412, protein-glutamate methylesterase/glutaminase, Figure 3D , q < 0.01, Mann-Whitney test). Furthermore, we found an increase in cofE, mainly found in Archaea and involved in methanogenesis (K12234, coenzyme F420-0:L-glutamate ligase, q < 0.05, Mann-Whitney test). In addition, we found a decrease in BA-related transcripts in PD while transcripts related to serine and isovalerate were increased in PD ( Figure 3D , q < 0.05). HC vs iRBD comparisons revealed significant differences in alanine-related transcripts but not for the other metabolites (Extended fig. 4A, q < 0.05). β-glutamate abundance was negatively correlated with the transcripts of the three glutamate synthases and carbamoyl-phosphate synthases, but positively correlated with methyl aspartate mutase, methylamine-glutamate N-methyltransferase and glutaminase (Extended fig. 4B, Spearman correlation test, q < 0.01). Since cheB is part of the chemotaxis and flagellar assembly (FA) KEGG pathways, we further inspected these pathways to assess the microbial capacity for motility. We found no statistically significant differences in gene abundances (funMG) for either flagellar assembly or chemotaxis pathways between the three different groups (data not shown, q > 0.05). In contrast however, based on funMT, the chemotaxis pathway (14/26 transcripts decreased in PD, 0/26 transcripts increased, Figure 3E , q < 0.05) and the FA pathway (25/46 transcripts decreased in PD, 1/46 transcripts increased in PD, Mann-Whitney, Figure 3E , q < 0.05) were strongly downregulated in PD. In addition, we found transcripts decreased in both PD and iRBD for five transcripts belonging to chemotaxis and eight belonging to flagella assembly (q < 0.05, Mann-Whitney, Figure 3E ). Moreover, eight transcripts showed a decrease in alpha diversity for chemotaxis and flagellin assembly pathways (p < 0.01, Mann-Whitney, Figure 3E ). Finally, we observed a decrease in alpha diversity for GLU , flagellar assembly transcripts ( fliJ , fliQ and fliC ) and cheD in PD compared to the other two groups ( Figure 3F , q < 0.05, Dunn test) while pilD and aaaT were elevated in iRBD ( Figure 3F , Dunn test, p < 0.05). Flagellin and chemotaxis are differentially expressed depending on taxonomy and disease status We next assessed the taxa expressing secondary bile acids biosynthesis (SBAS), FA and chemotaxis pathway genes. While resolving taxonomically SBAS at the genus level we found an interesting divergence between the differential analysis at the transcript the taxonomic level. Indeed, we previously found that baiCD , baiE , baiB , baiH and baiF were all significantly decreased in PD, while baiA and baiN were not ( Figure 2C ). We found 17 statistically significant differences in transcripts for baiA and baiN after FDR correction and 31 before correction when resolving their expression at the genus levels, but not for the other transcripts previously found as differentially expressed (q < 0.05 and p < 0.05, respectively, Extended fig. 5A). We found a significant decrease of expression for baiN in PD for Blautia , Fusicatenibacter , Choladocola , Oliverpabstia and Ruminococcus genera and an increase in expression for baiA encoded by Lawsonibacter genus in PD (q < 0.05, Mann-Whitney test, Extended fig. 5A). When resolving the taxonomic expression of FA and chemotaxis genes we noted a differential expression according to disease and taxonomy. More specifically, we identified different clusters of taxa expressing FA and chemotaxis genes. The first cluster was composed of microbes expressing these genes principally in PD, including Ruminiclostridium , Enterocloster , Dysosmobacter and Butyvibrio . A second cluster was composed of microbes expressing these genes principally in HC, including Roseburia , Agathobacter and Eubacterium ( Figure 4 ). Strikingly, we found that the third cluster is composed of taxa expressing flagellin or chemotaxis genes only in PD, including Escherichia , Cellulosilyticum , Citrobacter or Eisenbergiella . A fourth cluster was composed only of taxa expressing in HC including Flavonifractor , Succinivibrio , Eisenbergiella or CAG-603 ( Figure 4 ). We subsequently investigated the expression levels of genes of the extracellular parts of flagella in the cluster wherein Roseburia and most Lachnospiraceae were located (Cluster 2). Overall, we found a decrease in flagellin ( fliC ), filament cap ( fliD , fliS ) and hook-filament junction genes ( flgK , flgL ) in PD or iRBD for Roseburia , CAG-115 and Agathobacter genera (Extended fig. 5B, p < 0.05, Mann-Whitney test). Download figure Open in new tab Figure 4. Flagellar assembly and chemotaxis gene expressions according to taxonomy. Top 50 genera expressing flagellar assembly and/or chemotaxis related genes are depicted by the mean of transcripts per million per disease status. Values are depicted for each genus and genes on the left panel and the sum of all genes per genus on the right panel with a square root transformation. Genera are clustered based on the log2FC(HC/PD) for each gene using a Canberra distance and hclust() using “ward.D2” method. Metabolites and metabolites-related genes associated with PD are central to the microbial ecosystem metabolism To quantify the importance of glutamate derivatives and related genes in microbial metabolism, we next reconstructed microbiome-wide metabolic networks as previously described 34 . We mapped genes related to the compounds identified earlier as significantly different between the different groups. The metabolic network highlighted glutamate-related genes as compact and placed in the middle of the network while glycerol was more scattered across the network ( Figure 5A ). Strikingly, glutamate-related genes formed a subnetwork central to the overall network with a betweenness centrality measure (BC) of 93.0 compared to 0.0009 for the whole community ( Figure 5B ). Crucially, L-glutamate was the most central non-cofactor metabolite in the network and 2-oxoglutarate (L-glutamate derivative) was the second most central metabolite ( Figure 5C ). Considering this, it is apparent that glutamate and glutamate-related genes are central to microbiome metabolism and that modifications in the levels of these metabolites or transcripts reflect profound modifications of microbial metabolism. Download figure Open in new tab Figure 5. Metabolic network of whole community interactions. A. Metabolic network of whole community interactions, with KEGG KOs represented as nodes and associated metabolites as edges. Node sizes reflect MT/MG ratio of normalized read counts for each KO. Highlighted is the overlap between glutamate-, thymine-, glycerol-, serine-, alanine-and glucuronate-associated subnetworks mapped on the whole community network. B. Betweenness Centrality calculated for the key metabolites highlighted in the whole-community network based on genes as nodes. C. The network was inverted to calculate Betweenness centrality for metabolites, here metabolites are nodes and genes are edges. Multi-Omics Factor Analysis validates β-glutamate and flagella links with PD To validate our findings, we used an unsupervised method with the Multi-Omics Factor Analysis (MOFA). The resulting MOFA model included 10 factors (F1-10 hereafter) whereby F1 showed a strong association with disease status (p < 0.05, ANOVA, Figure 6A ). A complete description of the MOFA model is provided in the Extended information. We found that F1 mainly explained the variance of funMT, taxMT, and MM (17.6%, 9.6%, and 5.7%, respectively, Extended fig. 6A-B) and showed separation of HC and PD, but not iRBD versus other groups (p < 0.05, ANOVA, Figure 6B ). Specifically, the microbiome of PD was characterized by the joint increase in abundance of M. smithii , archaeal proteins and genes (based on MP and MG data), A. muciniphila (based on taxMT data), β-glutamate, isovalerate, isobutyrate, hexadecanoic, and hyocholic acids, whereas the abundance of R. hominis (taxMG data), flagellin (funMT), GCA and glycerol were decreased in PD ( Figure 6C ). Overall, the MOFA results are strongly consistent with the per omic layer analysis and validate these findings. Download figure Open in new tab Figure 6 MOFA analysis validates the findings of per omics layer analyses. A. Associations of MOFA factors with the diagnosis and confounders, the color of rectangles represents partial R² values, significant associations (FDR-adjusted p-value < 0.05) marked with an asterisk. B. Abundance of the Factor 1 in studied groups. C . Min-max scaled weights of top 10 features per omics layer contributing to Factor 1. The sign of the weight indicates the direction of the effect, the abundance of features with positive weights is positively associated with the Factor 1 level, and the abundance of features with negative weights is negatively associated with the Factor 1 level. From the microbiome to the host: expanding the Expobiome map To contextualize our newly gained insights about the links between microbial taxa, their molecules, and mechanistic evidence from the literature on host physiology, we integrated our results into the Expobiome map 35 ( https://expobiome.lcsb.uni.lu ). Specifically, we added R. intestinalis and hominis , B. wexlerae, coccoides and producta , Clostridium butyricum and F. plautii as new taxonomic elements linking flagellin, lipoteichoic acid (LTA) and SCFAs signaling to immune-relevant human proteins ( Figure 7 ). We also added mechanistic links on F. prausnitzii and its effects mediated by the short-chain fatty acid butyrate. Thereby, we integrated new mechanistic details linking the combinatorial functions from these taxa and pro-inflammatory components effectors such as TLR5 and its downstream activation of NLRP3 inflammasome and NF-κB, IL-6 and TNF-α secretion. Thereby these new additions to the Expobiome map highlight the convergent immunomodulatory functions of commensal taxa, perturbed in the context of PD. Download figure Open in new tab Figure 7. Schematic representation of the Expobiome map new taxonomic and molecular additions. Schematic representation of the new additions to the Expobiome map. In short, we collected literature highlighting mechanistic evidences linking microbes to the host physiology. We integrated Roseburia , Blautia , Faecalibacterium , Flavonifractor and Clostridium species as new taxonomic elements and linked their molecules to the host physiology. Finally, the Expobiome map is now connected to Parkinson’s disease map, providing direct overlay of the shared proteins between the maps. To facilitate the investigation of potential relationships between microbial taxa, microbial compounds, and human proteins that have been identified as crucial players in PD, we have furthermore linked the Expobiome map to the Parkinson’s disease map ( https://pdmap.uni.lu ). Notable overlaps exist between host proteins that have been established to interact with microbial components in both the immune system and the central nervous system. For example, dendritic cells and macrophages have well characterized interactions with flagellin through TLR5 binding. TLR5 and other antigen sensing receptors are present on microglia, a cell type of critical importance in neuroinflammation and PD. This integration of information from both thus-far disconnected maps allows for a more comprehensive understanding of these interactions and helps to draw new hypotheses on the intricate links between the gut microbiome, the immune system and PD. Discussion Here we investigate the links between the gut microbiome and PD using an integrated multi-omics approach based on standardized sample collection and extraction. We include individuals with iRBD as a prodrome of PD to compare early and later stages of the disease but do not find statistically significant differences between iRBD and PD, especially in comparison to the more pronounced differences found between HC and PD. Previous studies have shown differences in the different early stages of PD, but these studies were performed using 16S rRNA gene amplicon data 7 , 10 . In contrast to the amplicon-based results, no differentiation between iRBD and PD are found using metagenomic data 5 . More specifically, and contrary to previous findings 5 including our own work 7 , we do not find significant differences between HC and PD individuals in the metagenome with respect to alpha or beta diversity. However, we show differences in transcriptional activity for R. hominis , B. obeum, B. wexlerae and P. copri which are known to be found at lower abundance in PD 5 , 6 , 8 – 13 . Moreover, we find several species, including A. muciniphila , M. smithii and Clostridium spp, to be positively correlated with the abundance of compounds increased in PD such as β-glutamate, isovalerate or isobutyrate, while other species are inversely correlated with these compounds (species belonging to Faecalibacterium , Blautia , Eubacterium and Roseburia ). Collectively, these findings are in line with findings associating those genera with PD 5 , 6 , 8 – 13 . We identify metabolites that are associated with PD, amongst which BAs, alanine, serine and β-glutamate showed differences in linked gene expression. The BA chenodeoxycholic acid (CDCA) is decreased in PD and iRBD whereas glycocholic acid (GCA) is decreased only in iRBD. We also detect a decrease in transcripts related to BAs in PD but not iRBD. BAs have a wide spectrum of effects on the immune system 22 , 23 , metabolism 35 , hormones 36 and on the CNS 37 . GCA has been found to be increased in the CNS of PD and associated with disease duration 37 , stressing the importance of analyzing BAs in PD and prodromal stages of PD. Interestingly, when we taxonomically resolve the differentially expressed SBAS genes, we find different signals than when we compare the genes by themselves. Indeed, for the differential expression analysis we found a significant decrease in PD for baiB , baiCD , baiE , baiF and baiH but not for baiA and baiN , when we resolved the taxonomic expression of the SBAS genes we find differences only for baiA and baiN . This finding underscores the importance of considering the microbiome at the taxonomic and functional level at once. In contrast to previous studies 6 , 15 , we do not find a decrease in butyrate, propionate, acetate or overall levels of SCFAs. However, we find an increase in isovalerate, isobutyrate and valerate as described previously 15 . Previous studies have highlighted the correlation of fecal concentrations of isobutyrate and isovalerate with PD severity 38 and concentrations of valerate with disease duration 39 . There currently is an apparent lack of knowledge regarding isobutyrate or isovalerate with respect to host-microbiome interactions and very few related genes annotated in the KEGG database, thereby hindering interpretations concerning potential links between these metabolites and PD. β-glutamate and glutamate-related genes are of particular interest in the context of PD because of the reported toxic effects of L-glutamate on neurons 24 and because of its association with microbial activity. In addition, glutamate levels have been reported to be increased in PD individuals’ blood sera 17 , 18 . β-glutamate has strikingly only one reaction described in the KEGG database, in contrast, L-glutamate has 213 reactions and 54 pathways, while D- glutamate has 12 reactions and 2 pathways. We find that glutamate and glutamate-related genes are central to microbial metabolism, which underpins the notion that the highlighted differences reflect a pronounced impact on gut microbiome function. Of note, glutamate likely has local effects on enteric neurons with subsequent influences on the CNS 40 . Although we show significant differences in β-glutamate and glutamate-related genes in the stool of PD patients, we detect no significant differences in the glutamate levels (both enantiomers) between the three groups. We highlight significant differences in β-glutamate for which we currently cannot evaluate the effects on the ENS and CNS. Overall, based on our results, microbiome-driven glutamate metabolism and its impact on the glutamatergic system must be comprehensively studied in the future to disentangle its link with PD. Glutamate-related genes are further involved in chemotaxis and flagellar assembly pathways, highlighting a modification of the latter genes’ expression with the majority being decreased in PD individuals and some of these genes in iRBD too. A decrease in flagellar assembly gene abundances has been previously reported in a metagenomic analyses of PD 9 . We do not see significant differences in the linked gene abundances, but their expression levels are significantly different, highlighting altered regulation of transcription in PD/iRBD compared to HC. Flagellar assembly and chemotaxis genes are also differentially expressed by specific microbes; the genera Escherichia and Cellulosilyticum for instance are expressing flagellar assembly genes only in the context of PD without being statistically differentially abundant. Flagellin is a known immunogenic molecule, a potent pro-inflammatory compound in pathogens 41 – 43 and is thus targeted by secretory IgA 44 . However, flagellin in commensals, especially in the Lachnospiraceae family, has been shown to be either ‘silently recognised’ 45 or elicit anti-inflammatory effects 46 – 48 . Amongst the Lachnospiraceae family, the genus Roseburia shows a decrease in the transcription of flagellin in the gut microbiome of PD. This in turn may be linked to immune system dysregulation and exert indirect effect on the CNS, particularly in microglia as shown in a previous study using a murine model of PD 49 . Previous studies have shown an increased inflammatory state in PD, with increased pro-inflammatory circulating immune cells 50 , cytokines 51 , and activated microglia 52 . In addition, microglia can be activated by α-synuclein via NLRP3 following TLR2 (lipopolysaccharide (LPS) sensing) and TLR5 (flagellin sensing) activation 53 . Therefore, the activation or inhibition of TLRs by distinct flagellins may modulate microglia activation by competing with α-synuclein and affect PD progression. It is however not known if flagellin can reach the CNS directly or if this potential effect would be mediated by an indirect effect through the immune system. Evidence exists that a local, microbiome-mediated effect on the enteric immune system can impact distant sites such as the brain 54 or tumors 55 . This provides solid grounds for the hypothesis that flagellin can regulate the CNS and modulate the neuroinflammation occurring in PD as resolved in the Expobiome map ( Figure 7 ). Knowledge on microbial antigens is based on studies of pathogens, mainly Salmonella typhimurium , pathogenic Escherichia coli , Staphylococcus aureus or Bacillus subtilis . The results of these studies have led the scientific community to consider bacterial antigens as pro-inflammatory and danger signals for the host 56 . Our findings associating flagellin expression with health, along with recent reports showing the anti-inflammatory effects of flagellin from R. hominis and intestinalis 45 , 46 , LTA from F. plautii 57 and LPS from Bacteroides dorei 58 , call for a re-evaluation of the impact of common antigens on host physiology. Finally, the diverse compounds produced by the gut microbiome demand detailed investigation, with regard to the taxonomy and commensal/pathogen phenotype of the producing microbe, to fully understand their modulating effects on the immune and nervous systems alongside metabolism in health and disease. In conclusion, our work clearly highlights the importance of studying microbiome functions rather than restricting microbiome analyses to taxonomic structure. Specifically, the combination of MT and MM provides clear insights into the activity of specific microbial taxa in relation to disease. In our present work, MT reveals that disease association is not solely determined by gene expression levels; the diversity of microbes capable of expressing a specific function and the specific taxa expressing those functions are also of immediate interest and relevance. While microbiome interventions are mainly designed to deal with dysbiosis at the taxonomic level, we propose here to focus on restoring the lost functions from dysbiotic gut microbiomes. The future of microbiome research might lie in understanding how we can modulate, re-activate or shut down specific microbial functions in vivo in order to improve knowledge and later improve or functionally tailor microbiome interventions. STAR Methods Key resources table View this table: View inline View popup EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS All subjects from both cohorts provided informed written consent, and the sample analysis was approved by the Comité National d’Ethique de Recherche of Luxembourg (reference no.: 140174_ND). Kassel Cohort The DeNoPa cohort represents a prospective, biannual follow-up study of (initially de novo) Parkinson’s disease (PD) patients at the Paracelsus-Elena Klinik, Kassel, Germany. Fecal samples from PD patients (46) and healthy controls (29) were collected during the 4-year follow-up visit for the cohort. Details on inclusion and exclusion criteria and ancillary investigations have been published previously 82 , 83 . Subjects with idiopathic rapid-eye-movement sleep behaviour disorder (iRBD, 13) were recruited at the same clinic, diagnosed according to consensus criteria of the International RBD study group 84 using video-assisted polysomnography, and were included only if they showed no signs of a neurodegenerative disorder. DeNoPa subjects were required to have a 4-week antibiotic free interval before fecal sample collection. As additional control subjects, we collected fecal samples from (20) neurologically healthy subjects living in the same household as the DeNoPa participants. Samples of de novo PD patients from a cross-sectional cohort at the same clinic were included if subjects were recently diagnosed, drug-naïve and met United Kingdom Parkinson’s Disease Society Brain Bank (UKPDSBB) clinical diagnostic criteria 30 . All subjects except household HC were interviewed and examined by an expert in movement disorders. The study conformed to the Declaration of Helsinki and was approved by the ethics committee of the Physician’s Board Hessen, Germany (FF 89/2008). The DeNoPa trial is registered at the German Register for Clinical trials (DRKS00000540). Marburg Cohort We also added samples from 14 patients with polysomnography-confirmed iRBD which were recruited from the outpatient clinic of the Department of Neurology, Philipps-University, Marburg, Germany, between November 2015 and November 2016. iRBD was diagnosed according to the guidelines of the American Academy of Sleep Medicine (AASM ICSD-3) 85 . A detailed medical history was recorded, and a complete neurological examination performed to verify the subjects’ suitability. Inclusion criteria were age above 18 years, no dopamimetic therapy, and no diagnosis of PD, MSA, DLB or PSP. Exclusion criteria were smoking, antibiotic therapy in the last 24 months, history of other neurological diseases or disorders of the gastrointestinal tract. Non-motor and autonomic symptoms were evaluated with the SCOPA-AUT 86 and PD-NMS 87 questionnaires. Motor function was evaluated with the UPDRS 88 . Additionally, patients were asked to complete the RBD-Sleep questionnaire 89 . The study conformed to the Declaration of Helsinki and was approved by the ethics committee of the Medical Faculty of the Philipps-University, Marburg, Germany (46/14). METHOD DETAILS Fecal sample collection Fecal samples were collected at the clinics via a Faeces Collection Device (MedAuxil) and collection tubes (Sarstedt), as previously described 7 . Samples were immediately flash-frozen on dry ice after collection. Samples were subsequently stored at –80 °C and shipped on dry ice. Sample exclusions The initial set of samples consisted of 50 PD, 30 iRBD and 50 healthy control subjects (HC). Three PD and two iRBD cases were subsequently excluded for clinical reasons (adjusted diagnosis), one iRBD and one PD subject for logistical reasons, and one control due to a combination of microbiome-altering medications (metformin, antidepressants, statins, and proton pump inhibitors). Additional samples were excluded due to missing values (metabolomics) or a low amount of identified analytes (metaproteomics), leading to the final numbers of samples summarized below: ‒ Metagenomics (MG) & metatranscriptomics (MT): 49 HC, 27 iRBD, 46 PD ‒ Metaproteomics (MP): 42 HC, 22 iRBD, 40 PD ‒ Meta-metabolomics: 49 HC, 27 iRBD, 41 PD Metagenomic and metatranscriptomic sequencing Extractions from fecal samples were performed according to a previously published protocol 90 , conducted on a customized robotic system (Tecan Freedom EVO 200). After extraction, DNA and RNA were purified prior the sequencing analysis by using the following commercial kits respectively: Zymo DNA Clean&Concentrator-5 (D4014) and Zymo RNA Clean&Concentrator-5 (R1014). RNA quality was assessed and quantified with an Agilent 2100 Bioanalyser (Agilent Technologies) and the Agilent RNA 6000 Nano kit, and genomic DNA and RNA fractions with a NanoDrop Spectrophotometer 1000 (Thermo Scientific) as well as commercial kits from Qubit (Qubit ds DNA BR Assay kit, Q32850; Qubit RNA BR Assay kit, Q10210). All DNA samples were subjected to random shotgun sequencing. Following DNA isolation, 200-300 ng of DNA was sheared using a Bioruptor NGS (Diagenode) with 30s ON and 30s OFF for 20 cycles. Sequencing libraries were prepared using the TruSeq Nano DNA library preparation kit (Illumina) following the manufacturer’s protocol, with 350 bp average insert size. For MT, 1 µg of isolated RNA was rRNA-depleted using the RiboZero kit (Illumina, MRZB12424). Library preparation was performed using the TruSeq Stranded mRNA library preparation kit (Illumina) following the manufacturer’s protocol, apart from omitting the initial steps for mRNA pull down. MG and MT analyses, the qualities of the libraries were checked using a Bioanalyzer (Agilent) and quantified using Qubit (Invitrogen). Libraries were sequenced on an Illumina NextSeq500 instrument with 2×150 bp read length. Metaproteomics 20 µL protein extract were processed using the paramagnetic bead approach with SP3 carboxylate coated beads 91 , 92 . Briefly, the protein samples were reduced with 2µL 25 mM DTT in 20 mM ammonium bicarbonate (Sigma-Aldrich) for 1 h at 60°C. Subsequently, 4 μL 100 mM iodoacetamide (Merck) in 20 mM ammonium bicarbonate was added and incubated for 30 min at 37°C in the dark. Next, 5 μL of 10% formic acid was added as well as 70 μL 100% acetonitrile (ACN) to reach a final organic content higher than 50% (v/v). 2 μL SP3 beads per sample were washed with water three times with subsequent addition of the sample. After protein binding to the beads, the supernatant was discarded. The beads were washed twice with 200 μL 70% (v/v) ethanol, and once with 200 µL ACN. The protein lysates were proteolytically cleaved using trypsin (1:50) over night at 37 °C. Since trypsin is added in aqueous solution to the samples, the proteins are not bound to the beads during enzymatic cleavage. ACN was added to each sample to reach a final organic content higher than 95% (v/v). After peptide binding to the beads, the samples were washed with pure ACN on the magnetic rack. Finally, the peptides were eluted in two steps. First, with 200 μL 87% ACN (v/v) containing 10 mM ammonium formate (pH 10), and next with two times adding 50 μL water containing 2 % (v/v) DMSO and combination of the two aqueous supernatants. Thus, two fractions of peptides were generated, which were evaporated and re-dissolved in water containing 0.1 % formic acid (20 µL) and analyzed on a Q Exactive HF instrument (Thermo Fisher Scientific) equipped with a TriVersa NanoMate source (Advion) in LC chip coupling mode. Peptide lysates were injected on a trapping column (Acclaim PepMap 100 C18, 3 μm, nanoViper, 75 μm x 2 cm, Thermo Fisher Scientific) with 5 μL/min by using 98% water/2% ACN 0.5% trifluoroacetic acid, and separated on an analytical column (Acclaim PepMap 100 C18, 3 μm, nanoViper, 75 μm x 25 cm, Thermo Fisher Scientific) with a flow rate of 300 nL/min. Mobile phase was 0.1% formic acid in water (A) and 80 % ACN/0.08 % formic acid in water (B). Full MS spectra (350–1,550 m/z ) were acquired in the Orbitrap at a resolution of 120,000 with automatic gain control (AGC) target value of 3×10 6 ions. Meta-metabolomics Untargeted GC-MS as well as targeted measurements (SCFA GC-MS/MS and bile acids LC- MS/MS) from fecal samples were performed according to a previously published protocol 93 . All GC-MS chromatograms were processed using MetaboliteDetector, v3.220190704 94 while LC-MS chromatogram were acquired with Thermo Xcalibur software (version 4.1.31.9) and analyzed with TaceFinder (Version 4.1). Compounds were initially annotated by retention time and mass spectrum using an in-house mass spectral library. Internal standards were added at the same concentration to every medium sample to correct for uncontrolled sample losses and analyte degradation during metabolite extraction. The data was normalized by using the response ratio of the integrated peak area of the analyte and the integrated peak area of the internal standard. Sequencing data processing and analysis For all samples, MG and MT sequencing data were processed and hybrid-assembled using the Integrated Meta-omic Pipeline (IMP) 95 ( https://git-r3lab.uni.lu/IMP/imp3 , commit 8c1bd6fa443d064511909c9eede20703f45e6c69). It includes steps for the trimming and quality filtering of the reads, the filtering of rRNA from the MT data, and the removal of human reads after mapping against the human genome (hg38). Pre-processed MG and MT reads were assembled using the IMP-based iterative hybrid-assembly pipeline using MEGAHIT 96 1.0.3. After assembly, the prediction and annotation of structural features such as open-reading frames (ORFs) was performed using a modified version of Prokka 66 and followed by functional annotation of those using Mantis 68 . Structural features were quantified on MG and MT level using featureCounts 97 . Taxonomic annotation of reads and contigs was performed using Kraken2 98 with a GTDB release207 database ( http://ftp.tue.mpg.de/ebio/projects/struo2/GTDB_release207/kraken2 ) and a 0.5 confidence threshold. Additionally, taxon abundances were estimated using mOTUs 2.5.1 99 . The mOTU abundances were used to generate abundance matrices for each taxonomic rank (phylum, class, order, family, genus and species) by summing up taxon marker read counts at the respective levels. Metaproteomics prediction and annotation For each sample, the predicted proteins were concatenated with a cRAP database of contaminants and the human UniProtKB Reference Proteome prior to the MP search. In addition, reversed sequences of all protein entries were added to the databases for the estimation of false discovery rates. The search was performed using Sipros v1.1 100 as search engine with the following parameters: trypsin was used as the digestion enzyme and a maximum of two missed cleavages was allowed. The tolerance levels for matching to the database were 1 Da for MS1 and 0.01 Da for MS2. Peptides with large errors for parent ions were later filtered out by setting the Filter Mass Tolerance Parent Ion parameter to 0.05 Da. Carbamidomethylation of cysteine residues was set as a fixed modification and oxidation of methionines was allowed as a variable modification. Peptides with length between 7 and 60 amino acids, with a charge state composed between +2 and +4 and a maximum missed cleavages of 3 were considered for identification. The results from all identifications were filtered by Sipros using at least one unique peptide per protein and peptide false discovery rate (FDR) was dynamically set to achieve a 1% of protein FDR. Data analysis was performed on all samples with at least 2000 proteins identified. A summary matrix of all selected samples consisting of the KO annotations from the integrated MG and MT analysis and the spectral count from the MP identification was then generated and used for statistical analysis. QUANTIFICATION AND STATISTICAL ANALYSIS Dimensionality reduction and ordination Beta diversity for MG and MT was assessed using the Bray-Curtis dissimilarity and subjected to a Non-Metric MultiDimensionnal Scaling (NMDS) for both the taxonomic and functional levels, using the metaMDS() function from the vegan package (2.6.2). Principal Component Analysis (PCA) was performed for MP and MM using the rda() function in the vegan package (2.6.4). PERMANOVA was used to assess statistical differences between groups using the Bray-Curtis dissimilarity and conducted in the vegan R package with the adonis2() function. Differential abundance analysis and correlations Differential abundance was done in two different approaches. The first approach consisted in using SIAMCAT 72 and ALDEx2 80 algorithm to find all the taxa and genes differentially expressed between groups without prior assumption. We used two different algorithms to have a sensitive algorithm (SIAMCAT, less prone to have false negative) and a more conservative one (ALDEx2, less prone to have false positive). The second approach consisted in using the MM significant compounds to drive the analysis on the functional level for MG and MT. Therefore, differential abundance tests and multiple correlation tests were conducted with a classical approach. We used Mann-Whitney or Kruskal-Wallis followed by a Dunn test (depending on the number of groups) and Spearman correlation tests. We applied FDR correction using the Benjamini & Hochberg method 101 . We depicted both FDR corrected as q-values and non-FDR corrected p-values to represent most of the differences found in our datasets. All statistical tests were done using the rstatix package (0.7.2). Variance analysis Variance analysis was used to assess the importance of each clinical factor on MM. To verify the covariance of factors and to assess which factors explained the most variance in our datasets, we computed the total variance for each clinical factor (removing the NAs for each factor) and the variance explained by each group within a clinical factor. Explained variance was calculated as follows: Microbiome-wide metabolic network analysis The microbiome-wide metabolic network analysis was conducted by establishing an association between KEGG KOs and corresponding ChEBI IDs. The networks were visualized utilizing the NetworkX package (release 3.3) 73 . In this network, the nodes were represented by KEGG KOs, while the edges were denoted by the corresponding metabolites (either products or reactants) 34 . Such compounds as water, energy transporters and cofactors were removed, to only consider main compounds of a given reaction. The analysis was restricted to genes that were present in a minimum of 50% of the samples. To construct metabolite-specific networks, we used KEGG KOs which have either a reactant or product in KEGG. Glutamate-, thymine-, glycerol-, serine-, alanine-and glucuronate-specific subnetworks were composed of 146, 9, 66, 43, 70, 18 genes, respectively. The network topology metric ‘Betweenness centrality’ was used to underscore the importance of a metabolite in microbiome-wide metabolism 34 . Integrated multi-omics analysis using MOFA2 Integrative analysis for the seven omics layers was conducted with the Multi-Omics Factor Analysis (MOFA) 2 R package (version 1.10.0) 102 . Before the analysis, data were preprocessed as follows: (I) funMG, funMT, funMP, taxMG, taxMT, and taxMP data were filtered based on the number of non-zero features, a feature was kept if it was present in at least in 25% of samples in each group (PD, HC, iRBD) or at least in 75% in any of the groups; (II) funMG, funMT, taxMG, and taxMT count data were separately residualized in a linear model to remove variance explained by differences in sequencing depth; (III) funMP and taxMP data were residualized by the sum of protein counts per sample and information on the number of high-quality proteins recovered per sample; (IV) regression residuals were cubic-root transformed to account for heteroscedasticity; (V) MM data were transformed using a centered log-ratio transformation. Each dataset was then additionally filtered to retain the features with the largest variance for the subsequent analysis. For funMG and funMT, we included features with variances equal to or larger than 90% feature variance for a dataset, for the other datasets we included features with variance equal to or larger than the median feature variance for a given dataset. In the results, the feature size for omics layers was as follows: 759 for funMG, 657 for funMT, 410 for funMP, 109 for taxMG, 71 for taxMT, 115 for taxMP, and 34 for MM. MOFA analysis was run on scaled omics data with fifteen initial factors. All factors that explained less than 2% of the variance were excluded from the model. The remaining factors were tested for differential abundance between the groups studied using the linear regression followed by ANOVA type II controlling the participants’ sex, age, and recruitment cohort. Extended figure 1. NMDS analysis of A. metagenomic taxonomic composition, B. metagenomic functions and C. meta-proteomic taxonomic composition, using a Bray-Curtis dissimilarity matrix. D. PCA analysis of metaproteomic functions. E. PERMANOVA analysis for the three groups and all omics. Colour represents R² values and size is –log10(p-value). All PERMANOVA analysis were run using 1000 permutations using a Bray-Curtis dissimilarity matrix. Extended figure 2. A. Differential abundance analysis at the genus level using SIAMCAT algorithm. B. Differential abundance analysis using ALDEx2 algorithm at the species level. Values are pseudo fold changes for HC/PD and size is based on –log10(p-value). Shape is referring to level of significance, triangular shape for p-value significance before and round shape for p- values < 0.05 after FDR correction. Extended figure 3. A. Percentage of variance explained for each metabolite for “Sex” and “Diagnosis” (left panel) and Percentage of variance explained for each metabolite for “Constipation” and “Diagnosis” (right panel). Metabolites are including untargeted meta-metabolomics, targeted SCFA and targeted bile acids, normalized by sum before merging and variance quantification. B. Variance for each metabolite associated to the clinical factors “Diagnosis”, “Sex” and “Constipation”. Metabolites are including untargeted meta-metabolomics, targeted SCFA and targeted bile acids, normalized by sum before merging and variance quantification. C. Spearman correlation between metabolites and taxMG species. P-values are FDR corrected. D. Absolute log2 fold change between HC and iRBD for funMG and funMT associated to significant compounds. Dots are scaled by the –log10(p-value), colorized and shaped according to p-value significance before (triangle shape) and after FDR correction (round shape). Extended figure 4. A to D. ALDEx2 differential abundance analysis on funMG for HC vs PD ( A. ) and HC vs iRBD ( B. ); funMT for HC vs PD ( C. ) and HC vs iRBD ( D. ). All genes and transcripts are colorized and shaped according to p-value significance before (triangle shape) and after FDR correction (round shape) E. Spearman correlation between beta-glutamate relative abundance and funMG-funMT KEGG orthologs related to glutamate species. Only genes with at least one significant correlation are plotted. All p-values are FDR corrected. Extended figure 5. A. Bile acids transcripts found significantly different between the groups. P-values are corrected with FDR. B. Flagellar assembly transcripts encoding for extracellular component of the flagella for the genus present in Cluster 2. All tests are Wilcoxon tests. Extended figure 6. Multiomics variance explained by MOFA factors. A . Variance explained by the MOFA factors across different omics layers, total. B . Variance explained by the MOFA factors across different omics layers, splitted by factors. Data availability Due to privacy restrictions, the datasets generated by this study are available upon reasonable request by contacting Paul Wilmes. Code availability The IMP pipeline, which was used for analysis of metagenomic and metatranscriptomic data, is available at https://gitlab.lcsb.uni.lu/IMP/imp3 . The R and python code used for statistical analyses and visualizations is available at https://gitlab.lcsb.uni.lu/ESB/[TBA ]. Contributions Conceptualization: J.-P.T., A.H.-B., W.O., B.M., P.W. Patient recruitment, clinical coordination, and sampling: S.S., A.J., C.T., W.O., B.M. Multi-omic data generation: C.D.R., J.-P.T., C.J., L.A.L, A.D, N.J, M.v.B. Bioinformatics, statistics and data visualization: R.V, J.O.S., P.N., V.T.E.A, V.P, O.H, B.K, P.M. Initial manuscript draft: R.V, J.O.S, P.N, V.T.E.A, V.P. Review and editing: C.C.L., S.B.B., P.M. and P.W. Funding acquisition: C.C.L, W.O, B.M, P.W. All authors read and approved of the submitted version. Rights Retention Statement This research was funded in whole, or in part, by the Luxembourg National Research Fund (FNR), grant references CORE/16/BM/11333923 (MiBiPa), CORE/15/BM/10404093 (microCancer/MUST), PRIDE/11823097 (MICROH DTU), and CORE/19/BM/13684739 (metaPUF). For the purpose of open access, and in fulfilment of the obligations arising from the grant agreement, the author has applied a Creative Commons Attribution 4.0 International (CC BY 4.0) license to any Author Accepted Manuscript version arising from this submission. Extended information Multi-omics data overview Using our previously developed methodological framework 107 , 108 , we performed a systematic multi-omic analysis of DNA, RNA, protein, and metabolite fractions isolated from flash-frozen fecal samples. We used MG, MT, MP and MM data to find biomarkers associated with the PD phenotype ( Fig. 1 ). We generated a mean of 7.5 (std 1.7) Gbps and 7.5 (std 1.4) Gbps of sequencing data for MG and MT, respectively. After trimming and filtering, we retained a mean of 6.8 (std 1.7) Gbps and 3.2 (std 1.3) Gbps for MG and MT, respectively. The mean assembly size was 0.4 (std 0.1) Gbps, with on average 5.9×10 5 (std 1.7×10 5 ) genes predicted. Finally, protein databases contained a mean of 7.2×10 5 (std 1.8×10 5 ) proteins, an average of 4.1×10 4 (std 0.6×10 4 ) MS spectra per sample were acquired, and a mean of 3.4×10 3 (std 1.7×10 3 ) proteins were identified. MOFA model description MOFA is an unsupervised machine learning approach for the integration of multi-omics data sets 102 . It allows for the identification of highly informative features across multiple omics. It has previously been used in the study of the gut microbiome in several diseases, giving critical insights into the link between the gut microbiome, health, and disease 104 – 106 . The biggest proportion of variance was explained by funMG and funMT, followed by the taxMT and funMP datasets (Extended fig. Aa). F1-2 incorporated most of the variance related to the funMT and taxMT, whereas funMG and taxMG variance was predominantly covered by F3- F5 (Extended fig. 8B). The funMP variance was explained mostly by F6, and MM variance was explained by F1 and F6. MOFA factors were tested in a linear model followed by ANOVA with disease status, as well as confounders including patients’ sex, age, and recruitment cohort. Among the MOFA factors, F1 showed an association with the disease status, whereas F4 and F9 were associated with patients’ sex and recruitment cohort, respectively ( Fig. 4A ). Acknowledgments We thank the staff of the Luxembourg Centre for Systems Biomedicine (LCSB), particularly the sequencing platform and the metabolomics platform, for running the sequencing and metabolomic analyses. The bioinformatics presented in this paper were carried out using the HPC facilities of the University of Luxembourg 103 . This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 863664), and was further supported by the Luxembourg National Research Fund (FNR) CORE/16/BM/11333923 (MiBiPa), CORE/15/BM/10404093 (microCancer/MUST), PRIDE/11823097 (MICROH DTU), and CORE/19/BM/13684739 (metaPUF), the Michael J. Fox Foundation under grant IDs 14701 (MiBiPa-PLUS) and MJFF-019228 (PARKdiet), the Parkinson’s Foundation (MiBiPa Saliva), the Institute for Advanced Studies of the University of Luxembourg through an AUDACITY grant (ref. no. MCI-BIOME_2019), as well as an “Espoîr en tête” grant from the Rotary Club Luxembourg to P.W. This work was also supported by a Fulbright Research Scholarship from the Commission for Educational Exchange between the United States, Belgium and Luxembourg to P.W. Additional funding was provided by the FNR under INTERMOBILITY/23/17856242. C.d.R. received funding from the European Union’s Horizon 2020 Widening Fellowships (N° 101026381). References 1. ↵ Elbaz , A. , Carcaillon , L. , Kab , S. & Moisan , F . Epidemiology of Parkinson’s disease . Revue Neurologique 172 , 14 – 26 ( 2016 ). OpenUrl CrossRef PubMed 2. ↵ Schwiertz , A. et al. Fecal markers of intestinal inflammation and intestinal permeability are elevated in Parkinson’s disease . Parkinsonism & Related Disorders 50 , 104 – 107 ( 2018 ). OpenUrl CrossRef PubMed 3. Clairembault , T. et al. Structural alterations of the intestinal epithelial barrier in Parkinson’s disease . Acta Neuropathologica Communications 3 , 12 ( 2015 ). 4. ↵ Scd , van Ij . & P, D. The Intestinal Barrier in Parkinson’s Disease: Current State of Knowledge . Journal of Parkinson’s disease 9 , ( 2019 ). 5. ↵ Palacios , N. et al. Metagenomics of the Gut Microbiome in Parkinson’s Disease: Prodromal Changes . Annals of Neurology 94 , 486 – 501 ( 2023 ). OpenUrl CrossRef PubMed 6. ↵ Aho , V. T. E. et al. Relationships of gut microbiota, short-chain fatty acids, inflammation, and the gut barrier in Parkinson’s disease . Mol Neurodegeneration 16 , 6 ( 2021 ). 7. ↵ Heintz-Buschart , A. et al. The nasal and gut microbiome in Parkinson’s disease and idiopathic rapid eye movement sleep behavior disorder . Movement Disorders 33 , 88 – 98 ( 2018 ). OpenUrl CrossRef PubMed 8. ↵ Cirstea , M. S. et al. Microbiota Composition and Metabolism Are Associated With Gut Function in Parkinson’s Disease . Mov Disord 35 , 1208 – 1217 ( 2020 ). OpenUrl CrossRef PubMed 9. ↵ Boktor , J. C. et al. Integrated Multi-Cohort Analysis of the Parkinson’s Disease Gut Metagenome . Movement Disorders 38 , 399 – 409 ( 2023 ). OpenUrl CrossRef PubMed 10. ↵ Huang , B. et al. Gut microbiome dysbiosis across early Parkinson’s disease, REM sleep behavior disorder and their first-degree relatives . Nat Commun 14 , 2501 ( 2023 ). 11. ↵ Wallen , Z. D. et al. Metagenomics of Parkinson’s disease implicates the gut microbiome in multiple disease mechanisms . Nat Commun 13 , 6958 ( 2022 ). 12. ↵ Romano , S. et al. Meta-analysis of the Parkinson’s disease gut microbiome suggests alterations linked to intestinal inflammation . npj Parkinsons Dis . 7 , 1 – 13 ( 2021 ). OpenUrl CrossRef PubMed 13. ↵ Toh , T. S. et al. Gut microbiome in Parkinson’s disease: New insights from meta-analysis . Parkinsonism & Related Disorders 94 , 1 – 9 ( 2022 ). OpenUrl CrossRef PubMed 14. ↵ Howell , M. J. & Schenck , C. H . Rapid Eye Movement Sleep Behavior Disorder and Neurodegenerative Disease . JAMA Neurology 72 , 707 – 712 ( 2015 ). OpenUrl CrossRef PubMed 15. ↵ Unger , M. M. et al. Short chain fatty acids and gut microbiota differ between patients with Parkinson’s disease and age-matched controls . Parkinsonism & Related Disorders 32 , 66 – 72 ( 2016 ). OpenUrl CrossRef PubMed 16. ↵ Chen , S.-J. & Lin , C.-H . Gut microenvironmental changes as a potential trigger in Parkinson’s disease through the gut–brain axis . J Biomed Sci 29 , 54 ( 2022 ). 17. ↵ Hirayama , M. , Tsunoda , M. , Yamamoto , M. , Tsuda , T. & Ohno , K . Serum Tyrosine-to-Phenylalanine Ratio is Low in Parkinson’s Disease . Journal of Parkinson’s Disease 6 , 423 – 431 ( 2016 ). OpenUrl CrossRef 18. ↵ Figura , M. et al. Serum amino acid profile in patients with Parkinson’s disease . PLOS ONE 13 , e0191670 ( 2018 ). OpenUrl CrossRef PubMed 19. ↵ Vascellari , S. et al. Gut Microbiota and Metabolome Alterations Associated with Parkinson’s Disease . mSystems 5 , doi: 10.1128/msystems.00561-20 ( 2020 ). OpenUrl CrossRef 20. Li , P. et al. Gut Microbiota Dysbiosis Is Associated with Elevated Bile Acids in Parkinson’s Disease . Metabolites 11 , 29 ( 2021 ). 21. ↵ Hertel , J. et al. Integrated Analyses of Microbiome and Longitudinal Metabolome Data Reveal Microbial-Host Interactions on Sulfur Metabolism in Parkinson’s Disease . Cell Reports 29 , 1767 – 1777 .e8 ( 2019 ). OpenUrl CrossRef PubMed 22. ↵ Hang , S. et al. Bile acid metabolites control T H 17 and T reg cell differentiation . Nature 576 , 143 – 148 ( 2019 ). OpenUrl CrossRef PubMed 23. ↵ Song , X. et al. Microbial bile acid metabolites modulate gut RORγ + regulatory T cell homeostasis . Nature 577 , 410 – 415 ( 2020 ). OpenUrl CrossRef PubMed 24. ↵ Iovino , L. , Tremblay , M. E. & Civiero , L . Glutamate-induced excitotoxicity in Parkinson’s disease: The role of glial cells . Journal of Pharmacological Sciences 144 , 151 – 164 ( 2020 ). OpenUrl CrossRef PubMed 25. ↵ Iwasaki , Y. , Ikeda , K. , Shiojima , T. & Kinoshita , M . Increased plasma concentrations of aspartate, glutamate and glycine in Parkinson’s disease . Neurosci Lett 145 , 175 – 177 ( 1992 ). OpenUrl CrossRef PubMed 26. ↵ Jiménez-Jiménez , F. J. et al. Neurotransmitter amino acids in cerebrospinal fluid of patients with Parkinson’s disease . J Neurol Sci 141 , 39 – 44 ( 1996 ). OpenUrl CrossRef PubMed Web of Science 27. ↵ Mally , J. , Szalai , G. & Stone , T. W . Changes in the concentration of amino acids in serum and cerebrospinal fluid of patients with Parkinson’s disease . J Neurol Sci 151 , 159 – 162 ( 1997 ). OpenUrl CrossRef PubMed 28. ↵ Yuan , Y.-S. et al. Change in Plasma Levels of Amino Acid Neurotransmitters and its Correlation with Clinical Heterogeneity in Early Parkinson’s Disease Patients . CNS Neuroscience & Therapeutics 19 , 889 – 896 ( 2013 ). OpenUrl CrossRef PubMed 29. ↵ Wang , W. et al. Interactions between gut microbiota and Parkinson’s disease: The role of microbiota-derived amino acid metabolism . Front. Aging Neurosci . 14 , 976316 ( 2022 ). 30. ↵ Hughes , A. J. , Daniel , S. E. , Kilford , L. & Lees , A. J . Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. Journal of Neurology , Neurosurgery & Psychiatry 55 , 181 – 184 ( 1992 ). OpenUrl CrossRef 31. ↵ Beheshti , I. , Booth , S. & Ko , J. H . Differences in brain aging between sexes in Parkinson’s disease . npj Parkinsons Dis . 10 , 1 – 16 ( 2024 ). OpenUrl CrossRef PubMed 32. ↵ Li , X. et al. Sex differences in rapid eye movement sleep behavior disorder: A systematic review and meta-analysis . Sleep Med Rev 71 , 101810 ( 2023 ). 33. ↵ Mertsalmi , T. H. et al. More than constipation – bowel symptoms in Parkinson’s disease and their connection to gut microbiota . European Journal of Neurology 24 , 1375 – 1383 ( 2017 ). OpenUrl CrossRef PubMed 34. ↵ Roume , H. et al. Comparative integrated omics: identification of key functionalities in microbial community-wide metabolic networks . npj Biofilms Microbiomes 1, 1 – 11 ( 2015 ). 35. ↵ Villette , R. et al. Unraveling Host-Gut Microbiota Dialogue and Its Impact on Cholesterol Levels . Front. Pharmacol . 11 , ( 2020 ). 36. ↵ Hurley , M. J. , Bates , R. , Macnaughtan , J. & Schapira , A. H. V. Bile acids and neurological disease . Pharmacology & Therapeutics 240 , 108311 ( 2022 ). OpenUrl CrossRef PubMed 37. ↵ Kalecký , K. & Bottiglieri , T . Targeted metabolomic analysis in Parkinson’s disease brain frontal cortex and putamen with relation to cognitive impairment . npj Parkinsons Dis . 9 , 1 – 16 ( 2023 ). OpenUrl CrossRef PubMed 38. ↵ Chen , S.-J. et al. Association of Fecal and Plasma Levels of Short-Chain Fatty Acids With Gut Microbiota and Clinical Severity in Patients With Parkinson Disease . Neurology 98 , e848 – e858 ( 2022 ). OpenUrl CrossRef PubMed 39. ↵ Baert , F. et al. Parkinson’s disease patients’ short chain fatty acids production capacity after in vitro fecal fiber fermentation . npj Parkinsons Dis . 7 , 1 – 14 ( 2021 ). OpenUrl CrossRef PubMed 40. ↵ Tan , S. et al. Interaction between the gut microbiota and colonic enteroendocrine cells regulates host metabolism . Nat Metab 6 , 1076 – 1091 ( 2024 ). OpenUrl CrossRef PubMed 41. ↵ Qian , F. , et al. Salmonella flagellin is a potent carrier–adjuvant for peptide conjugate to induce peptide-specific antibody response in mice . Vaccine 33 , 2038 – 2044 ( 2015 ). OpenUrl CrossRef PubMed 42. Gram , A. M. et al. Salmonella Flagellin Activates NAIP/NLRC4 and Canonical NLRP3 Inflammasomes in Human Macrophages . J Immunol 206 , 631 – 640 ( 2021 ). OpenUrl Abstract / FREE Full Text 43. ↵ Tran , H. Q. , Ley , R. E. , Gewirtz , A. T. & Chassaing , B . Flagellin-elicited adaptive immunity suppresses flagellated microbiota and vaccinates against chronic inflammatory diseases . Nat Commun 10 , 5650 ( 2019 ). 44. ↵ Cullender , T. C. et al. Innate and Adaptive Immunity Interact to Quench Microbiome Flagellar Motility in the Gut . Cell Host & Microbe 14 , 571 – 581 ( 2013 ). OpenUrl CrossRef PubMed 45. ↵ Clasen , S. J. et al. Silent recognition of flagellins from human gut commensal bacteria by Toll-like receptor 5 . Science Immunology 8 , eabq7001 ( 2023 ). 46. ↵ Quan , Y. , et al. Roseburia intestinalis -derived flagellin is a negative regulator of intestinal inflammation . Biochemical and Biophysical Research Communications 501 , 791 – 799 ( 2018 ). OpenUrl CrossRef PubMed 47. Shen , Z. et al. Roseburia intestinalis stimulates TLR5-dependent intestinal immunity against Crohn’s disease . eBioMedicine 85 , ( 2022 ). 48. ↵ Wu , X. et al. Roseburia intestinalis-derived flagellin ameliorates colitis by targeting miR-223-3p-mediated activation of NLRP3 inflammasome and pyroptosis . Molecular Medicine Reports 22 , 2695 – 2704 ( 2020 ). OpenUrl PubMed 49. ↵ Liu , J. et al. Microbiota-microglia crosstalk between Blautia producta and neuroinflammation of Parkinson’s disease: A bench-to-bedside translational approach . Brain, Behavior, and Immunity 117 , 270 – 282 ( 2024 ). OpenUrl CrossRef PubMed 50. ↵ Tian , J. et al. Specific immune status in Parkinson’s disease at different ages of onset . NPJ Parkinsons Dis 8 , 5 ( 2022 ). 51. ↵ Tansey , M. G. et al. Inflammation and immune dysfunction in Parkinson disease . Nat Rev Immunol 22 , 657 – 673 ( 2022 ). OpenUrl CrossRef PubMed 52. ↵ Ho , M. S . Microglia in Parkinson’s Disease . Adv Exp Med Biol 1175, 335 – 353 ( 2019 ). 53. ↵ Scheiblich , H. et al. Microglial NLRP3 Inflammasome Activation upon TLR2 and TLR5 Ligation by Distinct α-Synuclein Assemblies . The Journal of Immunology 207 , 2143 – 2154 ( 2021 ). OpenUrl CrossRef PubMed 54. ↵ Bostick , J. W. , Schonhoff , A. M. & Mazmanian , S. K . Gut microbiome-mediated regulation of neuroinflammation . Current Opinion in Immunology 76 , 102177 ( 2022 ). 55. ↵ Gopalakrishnan , V. et al. Gut microbiome modulates response to anti–PD-1 immunotherapy in melanoma patients . Science 359 , 97 – 103 ( 2018 ). OpenUrl Abstract / FREE Full Text 56. ↵ Gallucci , S. & Matzinger , P . Danger signals: SOS to the immune system . Current Opinion in Immunology 13 , 114 – 119 ( 2001 ). OpenUrl CrossRef PubMed Web of Science 57. ↵ Mikami , A. et al. Oral Administration of Flavonifractor plautii, a Bacteria Increased With Green Tea Consumption, Promotes Recovery From Acute Colitis in Mice via Suppression of IL-17 . Front Nutr 7 , 610946 ( 2020 ). OpenUrl PubMed 58. ↵ Vatanen , T. et al. Variation in Microbiome LPS Immunogenicity Contributes to Autoimmunity in Humans . Cell 165 , 842 – 853 ( 2016 ). OpenUrl CrossRef PubMed 59. Parks , D. H. et al. GTDB: an ongoing census of bacterial and archaeal diversity through a phylogenetically consistent, rank normalized and complete genome-based taxonomy . Nucleic Acids Research 50 , D785 – D794 ( 2022 ). OpenUrl CrossRef PubMed 60. Quast , C. et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools . Nucleic Acids Research 41 , D590 – D596 ( 2013 ). OpenUrl CrossRef PubMed Web of Science 61. Kanehisa , M. & Goto , S . KEGG: kyoto encyclopedia of genes and genomes . Nucleic Acids Res 28 , 27 – 30 ( 2000 ). OpenUrl CrossRef PubMed Web of Science 62. Schneider , M. , Bairoch , A. , Wu , C. H. & Apweiler , R . Plant protein annotation in the UniProt Knowledgebase . Plant Physiol 138 , 59 – 66 ( 2005 ). OpenUrl Abstract / FREE Full Text 63. Narayanasamy , S. et al. IMP: a pipeline for reproducible reference-independent integrated metagenomic and metatranscriptomic analyses . Genome Biology 17 , 260 ( 2016 ). 64. Hiller , K. et al. MetaboliteDetector: comprehensive analysis tool for targeted and nontargeted GC/MS based metabolome analysis . Anal Chem 81 , 3429 – 3439 ( 2009 ). OpenUrl CrossRef PubMed 65. Li , D. et al. MEGAHIT v1.0: A fast and scalable metagenome assembler driven by advanced methodologies and community practices . Methods 102 , 3 – 11 ( 2016 ). OpenUrl CrossRef PubMed 66. ↵ Seemann , T . Prokka: rapid prokaryotic genome annotation . Bioinformatics 30 , 2068 – 2069 ( 2014 ). OpenUrl CrossRef PubMed Web of Science 67. Liao , Y. , Smyth , G. K. & Shi , W . featureCounts: an efficient general purpose program for assigning sequence reads to genomic features . Bioinformatics 30 , 923 – 930 ( 2014 ). OpenUrl CrossRef PubMed Web of Science 68. ↵ Queirós , P. , Delogu , F. , Hickl , O. , May , P. & Wilmes , P . Mantis: flexible and consensus-driven genome annotation . GigaScience 10 , giab042 ( 2021 ). 69. Milanese , A. et al. Microbial abundance, activity and population genomic profiling with mOTUs2 . Nat Commun 10 , 1014 ( 2019 ). 70. Davis , M. P. A. , van Dongen , S. , Abreu-Goodger , C. , Bartonicek , N. & Enright , A. J . Kraken: a set of tools for quality control and analysis of high-throughput sequence data . Methods 63 , 41 – 49 ( 2013 ). OpenUrl CrossRef PubMed Web of Science 71. Guo , X. et al. Sipros Ensemble improves database searching and filtering for complex metaproteomics . Bioinformatics 34 , 795 – 802 ( 2018 ). OpenUrl CrossRef PubMed 72. ↵ Wirbel , J. et al. Microbiome meta-analysis and cross-disease comparison enabled by the SIAMCAT machine learning toolbox . Genome Biol 22 , 1 – 27 ( 2021 ). OpenUrl CrossRef PubMed 73. ↵ Hagberg , A. A. , Schult , D. A. & Swart , P. J. Exploring Network Structure, Dynamics, and Function using NetworkX . ( 2008 ). 74. Wickham H ( 2016 ). ggplot2: Elegant Graphics for Data Analysis . Springer-Verlag New York . ISBN 978-3-319-24277-4, http://ggplot2.org . 75. Oksanen J , Blanchet FG , Kindt R , et al : vegan: Community Ecology Package . R package version 2 . 3 – 5 . 2016 . OpenUrl 76. Kassambara , A. Pipe-Friendly Framework for Basic Statistical Tests [R package rstatix version 0.6.0]. in ( 2020 ). 77. Gu , Z. Complex heatmap visualization . iMeta 1 , e43 ( 2022 ). OpenUrl CrossRef 78. Kassambara , A. Ggpubr: ‘ggplot2’ Based Publication Ready Plots . ( 2023 ). 79. Love , M. I. , Huber , W. & Anders , S . Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 . Genome Biology 15 , 550 ( 2014 ). 80. ↵ Fernandes , A. D. , Macklaim , J. M. , Linn , T. G. , Reid , G. & Gloor , G. B . ANOVA-Like Differential Expression (ALDEx) Analysis for Mixed Population RNA-Seq . PLOS ONE 8 , e67019 ( 2013 ). OpenUrl CrossRef PubMed 81. Argelaguet , R. et al. Multi-Omics Factor Analysis—a framework for unsupervised integration of multi-omics data sets . Molecular Systems Biology 14 , e8124 ( 2018 ). OpenUrl Abstract / FREE Full Text 82. ↵ Mollenhauer , B. et al. Nonmotor and diagnostic findings in subjects with de novo Parkinson disease of the DeNoPa cohort . Neurology 81 , 1226 – 1234 ( 2013 ). OpenUrl CrossRef PubMed 83. ↵ Mollenhauer , B. et al. Monitoring of 30 marker candidates in early Parkinson disease as progression markers . Neurology 87 , 168 – 177 ( 2016 ). OpenUrl CrossRef PubMed 84. ↵ Schenck , C. H. et al. Rapid eye movement sleep behavior disorder: devising controlled active treatment studies for symptomatic and neuroprotective therapy—a consensus statement from the International Rapid Eye Movement Sleep Behavior Disorder Study Group . Sleep Medicine 14 , 795 – 806 ( 2013 ). OpenUrl CrossRef PubMed 85. ↵ Sateia , M. J . International Classification of Sleep Disorders-Third Edition . Chest 146 , 1387 – 1394 ( 2014 ). OpenUrl CrossRef PubMed 86. ↵ Visser , M. , Marinus , J. , Stiggelbout , A. M. & Van Hilten , J. J . Assessment of autonomic dysfunction in Parkinson’s disease: The SCOPA-AUT . Movement Disorders 19 , 1306 – 1312 ( 2004 ). OpenUrl CrossRef PubMed Web of Science 87. ↵ Chaudhuri , K. R. et al. The metric properties of a novel non-motor symptoms scale for Parkinson’s disease: Results from an international pilot study . Movement Disorders 22 , 1901 – 1911 ( 2007 ). OpenUrl CrossRef PubMed 88. ↵ Fahn , S. , Elton , R. & Members of the UPDRS Development Committee. Unified Parkinson’s disease rating scale . Recent developments in Parkinson’s disease 0 , 153 – 163 ( 1987 ). OpenUrl 89. ↵ Stiasny-Kolster , K. , et al. The REM sleep behavior disorder screening questionnaire—A new diagnostic instrument . https://movementdisorders.onlinelibrary.wiley.com/doi/abs/10.1002/mds.21740 ( 2007 ). 90. ↵ Roume , H. et al. A biomolecular isolation framework for eco-systems biology . ISME J 7 , 110 – 121 ( 2013 ). OpenUrl CrossRef PubMed 91. ↵ Hughes , C. S. et al. Single-pot, solid-phase-enhanced sample preparation for proteomics experiments . Nat Protoc 14 , 68 – 85 ( 2019 ). OpenUrl CrossRef PubMed 92. ↵ Bannuscher , A. et al. A multi-omics approach reveals mechanisms of nanomaterial toxicity and structure–activity relationships in alveolar macrophages . Nanotoxicology 14 , 181 – 195 ( 2020 ). OpenUrl CrossRef PubMed 93. ↵ De Saedeleer , B. et al. Systematic characterization of human gut microbiome-secreted molecules by integrated multi-omics . ISME COMMUN . 1 , 1 – 6 ( 2021 ). OpenUrl CrossRef PubMed 94. ↵ Hiller , K. et al. MetaboliteDetector: Comprehensive Analysis Tool for Targeted and Nontargeted GC/MS Based Metabolome Analysis . Anal. Chem . 81 , 3429 – 3439 ( 2009 ). OpenUrl CrossRef PubMed 95. ↵ Narayanasamy , S. et al. IMP: a pipeline for reproducible reference-independent integrated metagenomic and metatranscriptomic analyses . Genome Biology 17 , 260 ( 2016 ). 96. ↵ Li , D. , Liu , C.-M. , Luo , R. , Sadakane , K. & Lam , T.-W . MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph . Bioinformatics 31 , 1674 – 1676 ( 2015 ). OpenUrl CrossRef PubMed 97. ↵ Liao , Y. , Smyth , G. K. & Shi , W . featureCounts: an efficient general purpose program for assigning sequence reads to genomic features . Bioinformatics 30 , 923 – 930 ( 2014 ). OpenUrl CrossRef PubMed Web of Science 98. ↵ Wood , D. E. , Lu , J. & Langmead , B . Improved metagenomic analysis with Kraken 2 . Genome Biology 20 , 257 ( 2019 ). 99. ↵ Milanese , A. et al. Microbial abundance, activity and population genomic profiling with mOTUs2 . Nat Commun 10 , 1014 ( 2019 ). 100. ↵ Guo , X. et al. Sipros Ensemble improves database searching and filtering for complex metaproteomics . Bioinformatics 34 , 795 – 802 ( 2018 ). OpenUrl CrossRef PubMed 101. ↵ Benjamini , Y. & Hochberg , Y . Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing . Journal of the Royal Statistical Society: Series B (Methodological ) 57 , 289 – 300 ( 1995 ). OpenUrl CrossRef PubMed Web of Science 102. ↵ Argelaguet , R. , et al. Multi-Omics Factor Analysis—a framework for unsupervised integration of multi-omics data sets | Molecular Systems Biology . EMBOpress https://www.embopress.org/doi/full/10.15252/msb.20178124 ( 2018 ). 103. ↵ Varrette , S. , Bouvry , P. , Cartiaux , H. & Georgatos , F . Management of an academic HPC cluster: The UL experience . in 2014 International Conference on High Performance Computing & Simulation (HPCS) 959–967 ( 2014 ). doi: 10.1109/HPCSim.2014.6903792 . OpenUrl CrossRef 104. ↵ Haak , B. W. , et al. Integrative Transkingdom Analysis of the Gut Microbiome in Antibiotic Perturbation and Critical Illness . mSystems 6 , doi: 10.1128/msystems.01148-20 ( 2021 ). OpenUrl CrossRef 105. Al Bataineh , M. T. et al. Uncovering the relationship between gut microbial dysbiosis, metabolomics, and dietary intake in type 2 diabetes mellitus and in healthy volunteers: a multi-omics analysis . Sci Rep 13 , 17943 ( 2023 ). 106. ↵ Arıkan , M. et al. Metaproteogenomic analysis of saliva samples from Parkinson’s disease patients with cognitive impairment . npj Biofilms Microbiomes 9 , 1 – 10 ( 2023 ). OpenUrl CrossRef PubMed 107. ↵ Wilmes , P. , Muller , E. , Heintz-Buschart , A. , Roume , H. & Lebrun , L. The sequential isolation of metabolites, RNA, DNA, and proteins from a single, undivided mixed microbial community sample . Preprint at doi: 10.1038/protex.2014.051 ( 2014 ). OpenUrl CrossRef 108. ↵ Heintz-Buschart , A. et al. Integrated multi-omics of the human gut microbiome in a case study of familial type 1 diabetes . Nat Microbiol 2 , 1 – 13 ( 2016 ). OpenUrl View the discussion thread. Back to top Previous Next Posted December 17, 2024. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Integrated multi-omics highlights alterations of gut microbiome functions in prodromal and idiopathic Parkinson’s disease Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Integrated multi-omics highlights alterations of gut microbiome functions in prodromal and idiopathic Parkinson’s disease Remy Villette , Júlia Ortís Sunyer , Polina V. Novikova , Velma T. E. Aho , Viacheslav A. Petrov , Oskar Hickl , Susheel Bhanu Busi , Charlotte De Rudder , Benoit Kunath , Anna Heintz-Buschart , Jean-Pierre Trezzi , Rashi Halder , Christian Jäger , Laura A. Lebrun , Annegrät Daujeumont , Sebastian Schade , Annette Janzen , Nico Jehmlich , Martin von Bergen , Cédric C. Laczny , Patrick May , Claudia Trenkwalder , Wolfgang Oertel , Brit Mollenhauer , Paul Wilmes bioRxiv 2024.12.13.628341; doi: https://doi.org/10.1101/2024.12.13.628341 Share This Article: Copy Citation Tools Integrated multi-omics highlights alterations of gut microbiome functions in prodromal and idiopathic Parkinson’s disease Remy Villette , Júlia Ortís Sunyer , Polina V. Novikova , Velma T. E. Aho , Viacheslav A. Petrov , Oskar Hickl , Susheel Bhanu Busi , Charlotte De Rudder , Benoit Kunath , Anna Heintz-Buschart , Jean-Pierre Trezzi , Rashi Halder , Christian Jäger , Laura A. Lebrun , Annegrät Daujeumont , Sebastian Schade , Annette Janzen , Nico Jehmlich , Martin von Bergen , Cédric C. Laczny , Patrick May , Claudia Trenkwalder , Wolfgang Oertel , Brit Mollenhauer , Paul Wilmes bioRxiv 2024.12.13.628341; doi: https://doi.org/10.1101/2024.12.13.628341 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Systems Biology Subject Areas All Articles Animal Behavior and Cognition (7653) Biochemistry (17763) Bioengineering (13944) Bioinformatics (42101) Biophysics (21509) Cancer Biology (18667) Cell Biology (25592) Clinical Trials (138) Developmental Biology (13413) Ecology (19969) Epidemiology (2067) Evolutionary Biology (24393) Genetics (15647) Genomics (22582) Immunology (17791) Microbiology (40524) Molecular Biology (17222) Neuroscience (88860) Paleontology (667) Pathology (2848) Pharmacology and Toxicology (4841) Physiology (7670) Plant Biology (15182) Scientific Communication and Education (2048) Synthetic Biology (4312) Systems Biology (9843) Zoology (2274)