Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
1 Organ level protein networks as a reference for the host effects of the microbiome
2 3 Robert H. Millsa,b,c,d, Jacob M. Wozniaka,b, Alison Vrbanacc, Anaamika Campeaua,b, Benoit 4 Chassainge,f,g,h, Andrew Gewirtze, Rob Knightc,d, and David J. Gonzaleza,b,d,# 5 6 a Department of Pharmacology, University of California, San Diego, California, USA 7 b Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego, 8 California, USA 9 c Department of Pediatrics, and Department of Computer Science and Engineering, University of 10 California, San Diego California, USA 11 d Center for Microbiome Innovation, University of California, San Diego, California, USA 12 e Center for Inflammation, Immunity and Infection, Institute for Biomedical Sciences, Georgia State 13 University, Atlanta, GA, USA 14 f Neuroscience Institute, Georgia State University, Atlanta, GA, USA 15 g INSERM, U1016, Paris, France. 16 h Université de Paris, Paris, France. 17 18 Key words: Microbiota, Tandem Mass Tags, Organ Proteomics, Gnotobiotic Mice, Germ-free Mice, 19 Protein Networks, Proteomics 20 21 # Address Correspondence to: 22 David J. Gonzalez, PhD 23 Department of Pharmacology and Pharmacy 24 University of California, San Diego 25 La Jolla, CA 92093 26 E-mail: [email protected] 27 Phone: 858-822-1218 28
1 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
29 Abstract
30 Connections between the microbiome and health are rapidly emerging in a wide range of
31 diseases. However, a detailed mechanistic understanding of how different microbial communities are
32 influencing their hosts is often lacking. One method researchers have used to understand these effects are
33 germ-free mouse models. Differences found within the organ systems within these model organisms may
34 highlight generalizable mechanisms that microbiome dysbioses have throughout the host. Here, we
35 applied multiplexed, quantitative proteomics on the brains, spleens, hearts, small intestines and colons of
36 conventionally raised and germ-free mice, identifying associations to colonization state in over 7,000
37 proteins. Highly ranked associations were constructed into protein-protein interaction networks and
38 visualized onto an interactive 3D mouse model for user-guided exploration. These results act as a
39 resource for microbiome researchers hoping to identify host effects of microbiome colonization on a
40 given organ of interest. Our results include validation of previously reported effects in xenobiotic
41 metabolism, the innate immune system and glutamate-associated proteins while simultaneously providing
42 organism-wide context. We highlight organism-wide differences in mitochondrial proteins including
43 consistent increases in NNT, a mitochondrial protein with essential roles in influencing levels of NADH
44 and NADPH, in all analyzed organs of conventional mice. Our networks also reveal new associations for
45 further exploration including protease responses in the spleen, high-density lipoproteins in the heart, and
46 glutamatergic signaling in the brain. In total, our study provides a resource for microbiome researchers
47 through detailed tables and visualization of the protein-level effects of microbial colonization on several
48 organ systems.
49
50 Introduction
51 The gut microbiome is emerging as a critical component of human health. It has been shown that
52 the microbial communities colonizing our bodies play important roles in the immune development of
53 infants(Milani et al. 2017) and the regulation of the innate immune system(Thaiss et al. 2016). Further, a
2 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
54 dysbiosis of the gut microbiome has been correlated with many diseases including Inflammatory Bowel
55 Disease (IBD)(Sartor and Wu 2017), diabetes(Tilg and Moschen 2014), obesity(Bouter et al. 2017),
56 cardiovascular disease(Ahmadmehrabi and Tang 2017) and mental health disorders(Nguyen et al. 2018).
57 Microbial production or modification of metabolites such as bile acids, choline derivatives, vitamins and
58 lipids provide some insight into the underlying host-microbe interactions in these diseases(Nicholson et
59 al. 2012). However, many mechanisms mediating these disease states remain unknown.
60 Germ-free (GF) mouse models, wherein a mouse is raised without any exposure to microbes,
61 have been an invaluable tool for assessing causal effects in microbiome research(Bhattarai and Kashyap
62 2016). GF models also provide an opportunity to understand the fundamental effects of microbial
63 colonization at an organismal scale. Systems level analyses of the tissues of GF mice have been
64 performed, but these studies have generally highlighted a select few organ tissues. Protein-level studies
65 have shown varying responses to colonization along different regions of the gastrointestinal (GI) tract
66 (Lichtman et al. 2016), changes in drug metabolizing proteins in livers and kidneys(Kuno et al. 2016), and
67 differences in circulating fatty acids from an analysis of serum and livers(Kindt et al. 2018). They have
68 also shown that microbial colonization alters post-translational modifications including histone
69 acetylation and methylation in liver, colon, and adipose tissue(Krautkramer et al. 2016), as well as lysine
70 acetylation in the gut and liver(Simon et al. 2012). An important transcriptomic study revealed a strong
71 connection between colonization and increased Nnt, a mitochondrial protein that has functions in redox
72 homeostasis and biosynthetic pathways through the generation of NADH and NADP+(Mardinoglu et al.
73 2015). The authors found Nnt transcripts increased in several conventional mouse tissues including
74 sections of the small intestine, colon and liver, which correlated with significant alterations in host amino
75 acid levels and glutathione metabolism(Mardinoglu et al. 2015). Other related studies found transcript
76 differences in the brain(Diaz Heijtz et al. 2011), and further highlighted the GI-dependent transcript
77 effects of microbial colonization in Myd88 deficient mice(Larsson et al. 2012).
3 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
78 Here, we sought to further detail the protein effects of microbiota colonization occurring both
79 inside and outside of the GI tract as a reference for microbiome researchers interested in a given host
80 protein, organ system or protein-network. Associations of highly ranked proteins were constructed into
81 protein-protein interaction networks for the brain, spleen, heart, small intestine and colon, as well as a
82 global network. While the brains and gastrointestinal tract of GF mice have been characterized in several
83 studies, other organs such as the heart and spleen may be of interest given the emerging roles of the
84 microbiota in atherosclerosis(Karlsson et al. 2012) and immune development(Chung et al. 2012). We
85 hypothesized that applying methods for improved accuracy in quantitative proteomics (Ting et al. 2011)
86 would further define the influence of the microbiota in each of these organs. With this, we hope to reveal
87 microbiota-induced changes that could underlie disease states. Our results validate a body of literature in
88 the field, provide visualization tools for contextualizing the organism-wide effects of microbial
89 colonization, and identify several new host-microbiota associations for further investigation.
90
91 Results
92 Construction of protein-protein interaction networks
93 To determine the protein-level consequences of microbial colonization, three biological replicates
94 of five different tissues (brain, small intestine, colon, spleen, heart) were analyzed from conventionally-
95 raised or GF mice. Multiplexed proteomic analysis of tissue homogenates resulted in the quantification of
96 7752 proteins overall, of which 4663 were quantified across all samples. These 4663 proteins were used
97 for downstream analysis. A separate pilot study of the brain tissue resulted in the quantification of 6203
98 proteins.
99 On a per-organ basis, we contrasted the protein abundances of tissue collected from conventional
100 mice against tissue from GF mice. We ranked the association of each protein to colonization state by
101 accounting for both significance level and fold-change (Supplemental Table S1). Interaction networks
102 were built in order to identify groups of proteins whose expression was modulated by microbial
4 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
103 colonization. We first constructed organ-specific protein interaction networks containing all the proteins
104 with a highly ranked (|π| > 1) association to the colonization state of a given organ (Fig. 1). Functional
105 enrichments within these organ networks were then assessed by gene-set enrichment analysis
106 (Supplemental Table S2, Fig. 2). To identify overlapping mechanisms occurring throughout all organs,
107 we compiled all highly ranked proteins within each organ (|π| > 1) into a single protein network (Fig. 3).
108 Colonization state of the mouse appeared to have larger impacts on organs in direct contact with
109 the gut microbiota, namely the small intestine and colon. The GI organs analyzed had an average of 210
110 proteins associated with colonization state while the three organs outside of the GI tract averaged 52. The
111 brain yielded the lowest number of associated proteins with only 22. GI tract organs also displayed a
112 higher percentage of interconnectivity, with an average of 71% of associated proteins within GI organs
113 having a moderate-confidence connection to another associated protein within the organ, while organs
114 outside the GI tract averaged 39% (Fig. 1). We hypothesize that the interconnectivity and number of
115 associations to colonization is related to the direct contact of GI organs with microbiota. However, it is
116 possible that including a higher portion of intestinal tissue may have influenced these results.
117 Validation of protein-networks through previously reported associations
118 Our networks provided support for previously reported broad-scale effects on GI organs, as well
119 as abundance shifts from specific transcripts or proteins. As shown in a previous proteomic
120 study(Lichtman et al. 2016), the small intestine and colon displayed distinct changes as a result of
121 microbial colonization. One example of this was the larger portion of proteins associated with GF status
122 within the small intestine (66%) than in the colon (42%, Fig. 1).
123 Other studies identified similar results at a pathway and individual protein level. After a literature
124 search for protein or transcript differences within GF models, we identified seven publications with
125 related findings (Supplemental Table S3). In brief, changes in xenobiotic metabolism were reported in the
126 GI, liver and kidney, both at the transcriptional level(Fu et al. 2017) and the protein level(Kuno et al.
127 2016). Our networks also highlighted changes in glutamate related proteins which were previously
5 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
128 reported at the transcriptional level(El Aidy et al. 2013). We also report the regulation of innate immune
129 proteins including the antimicrobial peptide REG3G(Larsson et al. 2012), and the regulation of
130 NNT(Mardinoglu et al. 2015), which will be discussed further below.
131
132 Organ-specific network results
133 Functional enrichment analysis of GI tract organs resulted in stronger and more diverse
134 associations to colonization status than organs outside the GI tract. Several enrichments emerged with
135 potential links to redox shifts in the small intestine. These included disulfide bonds, oxidoreductase
136 activity, NADP, and xenobiotic metabolism through cytochrome P450s (Fig. 2, Fig. 3). Functional
137 differences in the colon highlighted pancreatic secretion, immunoglobulins, proteins of the heat shock
138 protein 70 family, digestion, and stress as the functions increased in conventional mice (Fig. 2, Fig. 3).
139 GF colons had enrichments for transporter activity, Calycin, fatty-acid binding, metalloproteases and
140 peroxisome proliferator-activated receptor (PPAR) signaling (Fig. 2). Together, these results may indicate
141 stress response as an important factor mediating host-microbiota interactions in the colon and redox states
142 influencing interactions of the small intestine.
143 Organs outside of the GI tract have been less studied in regards to their regulatory responses to
144 the microbiome. Within the spleen, proteins increased in conventional mice were found to be part of a
145 highly connected functional protein network consisting primarily of pancreatic digestive enzymes
146 (CELA2A, CELA3B, CELA1, CPA1, CPA2, CTRB1, CTRL, CTSE, TRY5, etc.), iron-binding proteins
147 (FST1, TFRC, STEAP3, FECH, LTF and HP) and innate immune mediators (LCN2, S100A9, NGP,
148 ITGAM and CHIL3) (Fig. 1, Fig. 3). Many of the digestive enzymes have roles in the GI tract and were
149 similarly associated to colonization state within the small intestine and colon (Fig. 3). Iron-binding and
150 immune proteins were similarly regulated in other organs (Fig. 3), but the differential regulation of LTF
151 (Lactotransferrin) and FSLT1 (Follistatin-related protein 1) were primarily restricted to the spleen. Given
6 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
152 the biological roles of the spleen, the influence of these proteins in mediating immune processes may be
153 of significant interest.
154 The networks associated with the brain and the heart displayed fewer interconnected proteins.
155 Only 34% and 27% of proteins had a connection within the heart network and brain network respectively.
156 However, a group of primarily GF-associated proteins that included APOA1 and APOE was found among
157 the heart proteins (Fig. 1). These results may suggest changes in lipid profiles in the heart; increased high-
158 density lipoproteins (HDL) and chylomicrons in GF compared to conventional mice. The brain showed
159 limited functional enrichments (Supplemental Table S2). However, both RASGRF1 and RASGRF2 were
160 increased among conventional mice. These proteins may be of interest given implications in
161 glutamatergic excitatory synaptic signaling, and in facilitating long-term potentiation leading to enhanced
162 memory, learning, and synaptic plasticity(Drake et al. 2011; Schwechter et al. 2013).
163 Organism-wide network results
164 One prominent finding from our generalized network was a significant increase in N
165 NT in all conventional tissues. This protein is a key regulator of generalized biosynthetic processes, and is
166 related to glutamate synthesis(Mardinoglu et al. 2015). Statistically, NNT was among the strongest
167 relationships found within all organs analyzed (π = 3.8, 4, 2.3, 7, and 3.7 for small intestine, colon,
168 spleen, heart and brain respectively). We also identified sub-networks related to mitochondrial glutamate
169 metabolism and mitochondrial respiratory chain NADH dehydrogenase, with most of the related proteins
170 downregulated in the small intestine of conventional mice (Fig. 3). In addition, proteins relating to the
171 mitochondrial reduction of glutathione, GLUD1 and GLS had confirmed associations to previous
172 transcriptomic analysis (Supplemental Table S3). While this sub-network was largely derived from small
173 intestine proteins, there was evidence that this system may also be affected in the brain through AMT,
174 which is involved in the mitochondrial metabolism of glycine.
175 The protein networks identified generalized differences among all organs related to the innate and
176 adaptive immune systems. Proteins related to innate immunity tended to be increased in conventional
7 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
177 mice, while proteins associated with neutrophil degranulation were associated with germ-free mice (Fig.
178 3). REG3G, a protein associated with Toll-like receptor (TLR) signaling subsequent to pathogen-
179 associated molecular pattern (PAMP) activation in Paneth cells, was increased among conventional mice.
180 Additionally, proteins related to antigen processing were moderately enriched in GF mice.
181 We next evaluated protein relationships to GF status at an organismal-level by applying a
182 compositionally aware multinomial regression technique. This technique accounts for organ type to assess
183 organism-wide protein associations through ranks. Our top ranked protein associated with conventional
184 status was NNT, while the protein with the strongest association to GF status was IGKV5-39, a protein
185 involved in immune response and immunoglobin production (Fig. 4A). As observed from the traditional
186 statistical approach, NNT was significantly higher in conventional mice within all organs included in the
187 regression (Fig. 4B). IGKV5-39 was more strongly associated with the spleen, colon and heart than the
188 small intestine (Fig. 4B).
189 Next we assessed the 150 top and bottom ranked proteins associated with GF status from the
190 multinomial regression and created a protein-protein interaction network (Fig. 4C). Of interest was a
191 cluster of proteins related to redox states in mitochondria. Several proteins including ECSIT, NDUFS4,
192 NUBPL, NDUFA11, NDUFB6, NDUFA8, ATPAF1 are all related to mitochondrial complex 1 of the
193 electron transport chain. Other evidence of the organism-wide impact of microbial colonization on
194 mitochondria included shifts in mitochondrial ribosomal proteins (MRPS23, MRPL50, MRPS17,
195 MRPL49 and MRPL17) and proteins related to mitochondrial heat shock response (HSPD1, HSPE1,
196 SOD2, LONP1). The data presented here suggests that microbial colonization may have organismal-level
197 impacts on mitochondrial function.
198 Interactive 3D Visualization of Associations to Colonization Status
199 To encourage user interaction with our data, we created a web-based display of our results
200 projected onto a 3D model of a mouse. This interactive display allows for user-guided exploration of the
201 protein and pathway-level associations to colonization status. To access the model, users should access
8 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
202 the ‘ili webserver (https://ili.embl.de/), then drag-and-drop the provided texturization file (Supplemental
203 File S1) and a supplemental table for either the proteins (Supplemental Table S3) or pathways
204 (Supplemental Table S4) identified in this study. With this tool users can search for proteins and
205 pathways of interest and project the association scores onto all the organs analyzed in this study.
206 For the protein-level visualization, each protein is listed by the protein name with any Gene
207 Ontology (GO) molecular function terms associated with the protein. With this, users are able to search
208 for a protein of interest or identify proteins of interest by molecular functions. The ‘ili-compatible table of
209 association scores (Supplemental Table S4) uses the π-statistic for conventional/GF status. These scores
210 range from 6.98 (significantly increased in conventional mice) to -5.11 (significantly increased in GF
211 mice). An example use case is displayed in Figure 5A, which depicts the strong association to
212 conventional status for NNT within all organs.
213 The pathway-level visualizations help to summarize and explore the gene set enrichment
214 analyses. On a per-organ basis, an association to conventional or GF status for each functional category
215 was calculated by comparing the statistical strength of gene set enrichments (Supplemental Table S2).
216 These association scores have been summarized in an ‘ili-compatible file in Supplemental Table S5.
217 Pathway association scores ranged from 12.90 (highly associated to conventional status) to -7.25 (highly
218 associated with GF status). We demonstrate the use of this tool to visualize the associations to
219 “Oxidoreductase” in Figure 5B.
220
221 Discussion
222 Our multi-organ analysis was utilized to generate protein interaction maps, and 3D visualization
223 tools that researchers can use to understand organ-specific and organism-wide changes that may underlie
224 host-microbiome interactions. From our networks we can identify common themes found from previous
225 analyses of GF tissue. For example, we found changes in innate and adaptive immune responses to
226 microbial colonization(Larsson et al. 2012), changes in the glutamine and glutamate pathway(El Aidy et
9 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
227 al. 2013; Mardinoglu et al. 2015), and changes in xenobiotic degradation pathways (Kuno et al. 2016; Fu
228 et al. 2017; Kindt et al. 2018). This study contributes to the field by moving toward an organism-wide
229 understanding of host-microbiome interactions. Here, we incorporate new statistical and visualization
230 tools for multi-organ analysis and include understudied organs from outside of the GI tract. As roles for
231 the microbiome are expanding into immunity (Thaiss et al. 2016), cardiovascular disease(Ahmadmehrabi
232 and Tang 2017), and mental health disorders(Nguyen et al. 2018), defining the influence of the
233 microbiome on these tissues may be of importance. Indeed, our networks provided several putative
234 organism-level roles for the microbiome.
235 Our protein networks highlight the potential of organism-wide redox state changes being linked
236 to the microbiome. This is perhaps best highlighted in the distal increases of Nnt associated with
237 microbial colonization. NNT is a mitochondrial protein with well-described roles in glutathione redox
238 reactions through the conversion of NADPH to NADP+(Ronchi et al. 2013). Glutathione is
239 interconverted with glutathione disulfide, collectively representing the most abundant redox pair in the
240 body(Circu and Aw 2011). Abundances of this redox pair are often used as an indication of the general
241 redox state(Circu and Aw 2011), which is involved in a large variety of biological processes(Circu and
242 Aw 2011; Birben et al. 2012; Ray et al. 2012). Evidence of intestinal redox differences in GF animals has
243 been known since the 1970s(Koopman et al. 1975; Celesk et al. 1976) and evidence of microbiota-
244 dependent regulation of Nnt in the GI tract has been previously described(Mardinoglu et al. 2015). Here
245 we find evidence of increased NNT throughout the entire conventional mouse. Additionally, both the
246 traditional and multivariate statistical methods used for the identification of organism-wide effects found
247 networks of proteins related to mitochondria, including mitochondrial complex I proteins. Mitochondrial
248 complex I activity might also be linked to redox states as complex I activity was shown to be dependent
249 on glutathione transport into the mitochondria(Kamga et al. 2010). In summary, many of our protein
250 networks may have been influenced by redox states, including shifts in mitochondrial proteins, the innate
10 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
251 immune processes such as neutrophil degranulation, and degradation of xenobiotics through cytochrome
252 P450s (Kramer and Darley-Usmar 2015).
253 Though our strongest relationships to microbial colonization were found within the GI tract,
254 which was unsurprisingly the focus of most previous GF organ analyses(Larsson et al. 2012; El Aidy et
255 al. 2013; Mardinoglu et al. 2015; Lichtman et al. 2016), the analyses of the brain, spleen and heart did
256 yield several interesting results. Within the heart, our analyses indicated a relationship to HDL and the
257 germ-free state. Our results within the heart may be of importance given the increasing literature
258 regarding microbiota regulation of lipids and lipoproteins including HDL(Nakaya and Ikewaki 2018), as
259 well as the microbial links to metabolites leading to atherosclerosis(Wang et al. 2011b). Our findings may
260 indicate a mechanism in which microbiota influence the makeup of the heart through changes in
261 lipoproteins.
262 The data of the spleen proteomes revealed networks of proteases with similar increases within the
263 small intestine and colon in the presence of the microbiota. These networks also suggested potential links
264 between the gut microbiota and pancreatic secretion. Pancreatic secretion of enzymes is thought to be
265 largely regulated through circulating hormones(Singh and Webster 1978). The gut microbiome has
266 several potential links to pancreatitis and pancreatic cancer, which may be mediated through sensing of
267 microbial compounds such as lipopolysaccharide through TLR4(Leal-Lopes et al. 2015). Here, we have
268 observed a link between microbial colonization and increased pancreatic secretion of digestive proteins,
269 which may be of interest for further investigation.
270 The organism-wide effects of microbial colonization illustrated in our networks may have
271 implications in several disease states. Dysregulation of Complex I has been implicated in microbiome
272 related diseases including Ulcerative Colitis(Haberman et al. 2019) and there is accumulating evidence of
273 mitochondrial dysfunction in Crohn’s disease(Mottawea et al. 2016). This association between the
274 microbiome and mitochondria are thought to be mediated through three key microbiome metabolites:
275 short-chain fatty acids, the urolithins and lactate(Franco-Obregon and Gilbert 2017). While speculative, it
11 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
276 is possible that the IBD microbiota may influence these interactions. Glutamatergic signaling has been
277 suggested as a potential target for treating mood disorders(Zarate et al. 2010). Our identification of
278 proteins in the brain related to glutamatergic signaling and glutamate (i.e. RASGRF proteins and NNT)
279 may be of relevance to the discussions surrounding utilization of the microbiome to treat mood
280 disorders(Mangiola et al. 2016).
281 There are likely many unknown mechanisms mediating host-microbiota interactions. Our detailed
282 maps and visualization tools for understanding the organ-level impacts of microbial colonization give
283 insight into the unique and common protein-level changes occurring throughout mice. These networks
284 suggest changes throughout the mice related to mitochondrial dysfunction and redox states. Though fewer
285 changes were found in organs apart from the GI tract, our protein networks of the spleen, brain and heart
286 may provide insight for researchers establishing connections between the microbiota and diseases related
287 to these organ systems. We hope these networks and visualization tools may be useful in the microbiome
288 research community to help dissect the specific effects that microbial communities have in a given organ-
289 system, protein or protein network of interest. In total, we view our study as a step toward better
290 understanding the role of the microbiota in health and disease.
291
12 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
292 Materials and Methods
293 Gnotobiotic Mice
294 Three male germ-free C57/BL6 mice were kept under germ-free conditions in a Park Bioservices isolator
295 in GSU’s germ-free facility, and three male conventional C57/BL6 were kept in regular housing at GSU’s
296 animal facility. At 5 months of age, mice were euthanized and organs were collected followed by
297 immediate snap-freezing. All mice were bred and housed at Georgia State University, Atlanta, Georgia,
298 USA. under institutionally-approved protocols (IACUC # A14033).
299
300 Protein digestion and TMT labeling
301 All organ proteome methods were preformed as previously described(Lapek et al. 2018). Snap
302 frozen organs (stored at -80 ºC beforehand) were suspended in PBS and homogenized using a Mini
303 BeadBeater (Biospec). Our final analysis contained 3 biological replicates of each colonization condition
304 per organ, and no technical replicates were collected given strong correlation observed between biological
305 replicates in our past work (Lapek et al. 2018). Organ homogenates were lysed in 1 mL of buffer
306 composed of 75 mM NaCl (Sigma-Aldrich), 3% sodium dodecyl sulfate (SDS, Thermo Fisher Scientific),
307 1 mM NaF (Sigma-Aldrich), 1 mM beta-glycerophosphate (Sigma-Aldrich), 1 mM sodium orthovanadate
308 (Sigma-Aldrich), 10 mM sodium pyrophosphate (Sigma-Aldrich), 1 mM phenylmethylsulfonyl fluoride
309 (PMSF, Sigma-Aldrich), and a Complete Mini EDTA free protease inhibitors (1 tablet per 10 mL, Roche)
310 in 50 mM HEPES (Sigma-Aldrich), pH 8.5(Villen and Gygi 2008). To ensure full lysis, homogenates
311 were passed through a 21-gauge syringe 20 times. Insoluble debris was then pelleted by centrifugation for
312 5 minutes at 14,000 rpm. Supernatants were transferred to new tubes and an equal volume of 8 M urea in
313 50 mM HEPES, pH 8.5 was added to each sample. Samples were then vortexed and further lysed through
314 two 10-second intervals of probe sonication at 25% amplitude.
315 Proteins were reduced with dithiothreitol (DTT, Sigma-Aldrich) and alkylated with
316 iodoacetamide (IAA, Sigma-Aldrich)(Haas et al. 2006). Proteins were next precipitated via methanol-
13 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
317 chloroform precipitation(Wessel and Flugge 1984). Precipitated proteins were re-solubilized in 300 μL of
318 1 M urea (Thermo Fisher Scientific) in 50 mM HEPES, pH 8.5. Proteins were digested in a two-step
319 digestion process. First, 3 μg of LysC (Wako) was added to each sample, and samples were digested
320 overnight at room temperature. Next, 3 μg of trypsin was added, and samples were digested for six hours
321 at 37 ºC. Digests were acidified with trifluoroacetic acid (TFA, Pierce) to quench the digestion reaction.
322 Peptides were desalted with C18 Sep-Paks (Waters) as previously described(Tolonen and Haas 2014).
323 Concentration of desalted peptides was determined using a Pierce Quantitative Colorimetric Peptide
324 Assay (Thermo Fisher Scientific), and peptides were aliquoted into 50 μg portions. Aliquots were dried
325 under vacuum and stored at -80 ºC until they were labeled with TMT reagents.
326 Peptides were labeled with 10-plex TMT reagents (Thermo Fisher Scientific)(Thompson et al.
327 2003; McAlister et al. 2014) as previously described(Wang et al. 2011a). TMT reagents were
328 reconstituted in dry acetonitrile (Sigma-Aldrich) at 20 μg/μL. Dried peptides were re-suspended in 30%
329 dry acetonitrile in 200 mM HEPES, pH 8.5, and 7 μL of the appropriate TMT reagent was added to
330 peptides. Reagents 126 and 131 (Thermo Fisher Scientific) were used to label peptide aliquots composed
331 of an equal concentration of every sample within all mass spec runs. These composite samples acted as a
332 reference within all mass spec runs for data normalization purposes described later. Remaining reagents
333 were used to label samples in random order with no bias regarding animal of origin, organ or colonization
334 status. This randomization was performed to prevent known batch-effects in mass-spectrometry
335 experiments(Brenes et al. 2019). The brain samples were analyzed as a pilot experiment before the other
336 organs. For brains, all procedures were performed as above, though the TMT experiment was separate
337 from other organs. Within the brain TMT experiment, one channel, 129C, consisted of a 50 μg average of
338 all peptides from brain samples. Labeling was carried out for 1 hour at room temperature, and was
339 quenched by adding 8 μL of 5% hydroxylamine (Sigma-Aldrich). TMT-labeled peptides from each of the
340 organ samples were acidified by adding 50 μL of 1% TFA and subsequently combined into a composite
341 sample per TMT 10-plex experiment. During the pilot experiment with the brains, the pooled samples
14 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
342 were desalted and fractionated using a High pH Reversed-Phase Peptide Fractionation Kit (Pierce) per
343 manufacturer instructions. For the larger studied, samples were pooled per 10-plex experiment, desalted
344 with C18 Sep-Paks, and further fractionated as described below.
345
346 Collection of LC-MS2/MS3 spectra for protein identification and quantification
347 Data acquisition methods were performed as previously described(Lapek et al. 2018). Sample
348 fractionation for excluding the brains, was performed by basic pH reverse-phase liquid chromatography
349 with concatenated fractions as previously described(Wang et al. 2011a). Briefly, samples were re-
350 suspended in 5% formic acid/5% acetonitrile and separated over a 4.6 mm × 250 mm C18 column
351 (Thermo Fisher Scientific) on an Ultimate 3000 HPLC fitted with a fraction collector, degasser and
352 variable wavelength detector. The separation was performed over a 22% to 35%, 60-minute linear
353 gradient of acetonitrile in 10 mM ammonium bicarbonate (Thermo Fisher Scientific) at 0.5 mL/min. The
354 resulting 96 fractions were combined as previously described(Wang et al. 2011a). All fractions were dried
355 under vacuum and re-suspended in 5% formic acid/5% acetonitrile and analyzed by liquid
356 chromatography (LC)-MS2/MS3 for identification and quantitation.
357 All LC-MS2/MS3 experiments were carried out on an Orbitrap Fusion (Thermo Fisher Scientific)
358 with an in-line Easy-nLC 1000 (Thermo Fisher Scientific) and chilled autosampler. Home-pulled, home-
359 packed columns (100 μm ID × 30 cm, 360 μm OD) were used for analysis. Analytical columns were
360 triple-packed with 5 μm C4 resin, 3 μm C18 resin, and 1.8 μm C18 resin (Sepax) to lengths of 0.5 cm,
361 0.5 cm, and 30 cm respectively. Peptides were loaded at 500 bar and eluted with a linear gradient of 11%
362 to 30% acetonitrile in 0.125% formic acid over 165 minutes at a flow rate of 300 nL/minute, with the
363 column heated to 60 ºC. Nano-electrospray ionization was performed by applying 2000 V through a
364 stainless-steel T-junction at the inlet of the microcapillary column.
365 The mass spectrometer was run in data-dependent mode, where a survey scan was performed
366 over 500-1200 m/z at a resolution of 60,000 in the Orbitrap. Automatic gain control (AGC) was set to 15 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
367 2×105 for MS1 with a maximum ion injection time of 100 ms. The S-lens RF was set to 60 and centroided
368 data was collected. Top-N mode was used to select the most abundant ions in the MS1 scan for MS2 and
369 MS3 with N set to 10.
370 The decision tree option was used for MS2 analysis, using charge state and m/z range as
371 qualifiers. Ions carrying 2 charges were analyzed from the m/z range of 600-1200, and ions carrying 3
372 and 4 charges were selected from the m/z range of 500-1200. An ion intensity threshold of 5×104 was
373 used. MS2 spectra were obtained using quadrupole isolation at a 0.5 Th window and fragmented using
374 Collision Induced Dissociation with a normalized collision energy of 30%. Fragment ions were detected
375 and centroided data collected in the linear ion trap using rapid scan rate with an AGC target of 1×104 and
376 maximum ion injection time of 35 ms.
377 MS3 analysis was performed using synchronous precursor selection (SPS) enabled to maximize
378 TMT quantitation sensitivity(McAlister et al. 2014). A maximum of 10 MS2 precursors was specified for
379 the SPS setting, which were simultaneously isolated and fragmented for MS3 analysis. Higher-Energy
380 Collisional Dissociation fragmentation was used for MS3 analysis with a normalized collision energy of
381 55%. Resultant fragment ions (MS3) were detected in the Orbitrap at a resolution of 60,000 with a low
382 mass cut-off of 110 m/z. AGC for MS3 spectra was set to 1×105 with a maximum ion injection time of
383 100 ms. Centroided data were collected, and MS2 ions between the range of 40 m/z below and 15 m/z
384 above the precursor m/z were excluded by SPS.
385
386 Data processing and normalization
387 Data were processed using Proteome Discover 2.1 (Thermo Fisher Scientific). MS2 data were
388 searched against UniProt mouse databases (downloaded 7/2/2018 and 5/11/2017 for the brain analysis
389 and other organs respectively) using the Sequest algorithm(Eng et al. 1994). A decoy search was also
390 conducted with sequences in reverse order(Peng et al. 2003; Elias et al. 2005; Elias and Gygi 2007). A
391 precursor mass tolerance of 50 ppm(Beausoleil et al. 2006; Huttlin et al. 2010) was specified and 0.6 Da
16 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
392 tolerance for MS2 fragments. Static modification of TMT 10-plex tags on lysine and peptide n-termini
393 (+229.162932 Da) and carbamidomethylation of cysteines (+57.02146 Da) were specified. Variable
394 oxidation of methionine (+15.99492 Da) was also included in search parameters. Data were filtered to 1%
395 peptide and protein level false discovery rates with the target-decoy strategy through Percolator(Kall et al.
396 2007; Spivak et al. 2009).
397 TMT reporter ion intensities were extracted from MS3 spectra for quantitative analysis, and
398 signal-to-noise values were used for quantitation. Spectra were filtered and summed as previously
399 described(Lapek et al. 2017b). Data were normalized in a multi-step process, whereby they were first
400 normalized to the pooled standards (TMT-126 and -131) for each protein, and then to the median signal
401 across the pooled standards from all experiments(Lapek et al. 2017a). An average of these normalizations
402 was used for the next step. As the brain samples did not contain a composite bridge channel for
403 normalization, raw signal/noise ratios were normalized by the average signal of the given protein divided
404 by the median of all protein averages. To account for slight differences in amounts of protein labeled,
405 these values were then normalized to the median of the entire dataset and reported as final normalized
406 summed signal-to-noise ratios per protein per sample.
407
408 Statistical analysis
409 Bioinformatic analysis was performed in Python (version 3.5.1) and records are available online
410 in Jupyter Notebooks (https://github.com/rhmills/Germ-free-organ-proteomics). To prevent statistical
411 artifacts generated due to the various methods of dealing with missing values, only proteins with
412 quantification in all samples were used. A Student’s t-test with unequal variance was performed through
413 the package, SciPy (https://www.scipy.org). For ranking purposes, we evaluated associations to
414 colonization state through π–score, which accounts for both fold change and p-value(Xiao et al. 2014). A
415 statistical cutoff of |π| > 1 was chosen based on previous work (Tran et al. 2019). This statistical measure
17 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
416 corresponds to a significance level of α ~ 0.05, and allowed for moderate stringency while including an
417 adequate number of proteins for protein network construction and functional enrichment analysis.
418 Protein-protein interaction networks were created through STRING-db(Szklarczyk et al. 2015).
419 Associations between proteins were determined through default settings, accounting for text mining,
420 experiments, databases, co-expression, neighborhood, gene fusion and co-occurrence. Connections were
421 restricted to interactions between proteins within the query list only. Networks were subsequently
422 visualized through Cytoscape (version 3.5.1)(Shannon et al. 2003). Edges within protein networks were
423 based on the combined evidence scores, with thicker edges indicating higher confidence. Per-organ
424 networks were performed with a medium minimum confidence (0.4) to visualize connectivity through
425 maximizing potential connections, while combined organ networks utilized a high minimum confidence
426 (0.8) to identify putative functional groupings.
427 Functional enrichment analysis was performed through the DAVID server(Huang da et al. 2009b;
428 Huang da et al. 2009a) to identify significant groups of proteins per organ, split between Germ-free and
429 colonized states. Parameters were set as previously described(Nicolay et al. 2015). Benjamini-hochberg
430 corrected p-values were reported for the most significant groupings. Bar plots were visualized through
431 GraphPad Prism (version 7.0b).
432 Songbird(Morton et al. 2019) was used to implement the multinomial regression analysis with
433 organ, mouse, and colonization status used in the regression formula. Model parameters were as follows:
434 10,000 epochs, batch size of 5, differential prior of 1.0, learning rate of 0.001, gradient clipping size of
435 10, and proteins with >5 counts were included. As the brain tissue was processed as an independent pilot
436 study, we could not include this data as part of the multinomial regression analysis given the independent
437 normalization used in each experiment.
438
439 3D Mouse Model
18 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
440 The 3D Mouse model was generated as described previously(Quinn et al. 2019). In brief, MRI
441 images were acquired at the UCSD Center for Functional MRI from a euthanized 8-week-old female
442 C57BL/6 mouse with a Bruker 7T/20 MRI scanner using a quadrature birdcage transceiver. MRI
443 Parameters were as follows: 3D FLASH protocol with TE/TR=6 ms/15 ms and matrix size 128×64×156,
444 field of view prescribed to match the body size. InVesalius
445 (https://link.springer.com/chapter/10.1007/978-3-319-27857-5_5) was used to trace individual organs in
446 each MRI slice to generate the 3D model. The model was then processed with Blender
447 (https://www.blender.org/) for smoothing.
448 Interactive display of associations can be done through the following steps: 1) Accessing the ‘ili-
449 web browser (https://ili.embl.de) (Protsyuk et al. 2018). 2) Uploading the 3D mouse model visualization
450 file (Supplemental File S1) 3) Uploading either the protein-based associations table (Supplemental Table
451 S4) or the pathway-based associations table (Supplemental Table S5).
452 Pathway-level association scores were determined by comparing the –Log10(Benjamini-Hochberg
453 corrected p-values) for each functional enrichment in Supplemental Table S2. The statistical strength of
454 the functional enrichment in conventional tissue was subtracted from the statistical strength of the
455 functional enrichment in the GF tissue. The following formula summarizes the calculation:
456 Association Score = (-Log10(Adj. p-value for conventional status) – -Log10(Adj. p-value for GF Status))
457
458 Data Access
459 The proteomic data generated in this study have been submitted to the Mass Spectrometry Interactive
460 Virtual Environment (MassIVE) Repository (https://massive.ucsd.edu) under the study ID
461 MSV000083874. Code is available through GitHub (https://github.com/rhmills/Germ-free-organ-
462 proteomics), and as Supplemental Code.
463
464 Acknowledgements 19 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
465 D.J.G was supported by the Ray Thomas Edwards Foundation and the UC Office of the President.
466 R.H.M. was supported by UCSD’s Gastroenterology T32 Predoctoral fellowship. A.C. was supported by
467 the UCSD Microbial Sciences Initiative Graduate Research Fellowship and by the UCSD Graduate
468 Training Program in Cellular and Molecular Pharmacology through an institutional training grant from
469 the National Institute of General Medical Sciences, T32 GM007752.
470
471 Authors’ contributions
472
473 RHM, AV, RK, AG, DJG conceived and designed the experiments. RHM, BC and AV performed the
474 experiments. RHM, JW, AV and AC analyzed the data. RHM, JW, AC, AV and DJG wrote the
475 manuscript. All authors read and approved the final manuscript.
476
477
478
479
20 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
480 Figure Legends
481 Figure 1. Organ-specific protein networks modulated by microbial colonization. Proteins
482 significantly increased in either germ-free or conventional animals were analyzed for interactions through
483 String-db. Edges in each node represent the combined score accounting for all interaction sources. The
484 edges are sized by the combined score with the minimum threshold being 0.4 (of a maximum confidence
485 1). Nodes represent gene names of significant proteins with a minimum statistical cutoff of |π| >1. Red
486 indicates a significantly higher presence in conventional mice and grey indicates the opposite. The nodes
487 are sized by the level of significance as assessed by π-score.
488
489 Figure 2. Functional enrichments associated with microbial colonization within each organ system.
490 Proteins significantly increased in either germ-free or conventional animals within each organ were
491 analyzed for functional enrichments using DAVID. All proteins identified within the experiment were
492 used as a background. Displayed are bar plots showing the -Log10(adj. p-values) associated with selected
493 functional groupings. Benjamini-hochberg correction was applied to account for multiple hypothesis
494 testing. The bars are plotted in red if they are associated with the proteins enriched among conventional
495 mice and grey if they are associated with germ-free mice.
496
497 Figure 3. Combined organ protein networks modulated by microbial colonization. Proteins
498 significantly increased in either GF or conventional animals were analyzed for interactions through
499 String-db. Edges in each node represent the combined score accounting for all interaction sources. The
500 edges are sized by the combined score with the minimum threshold being 0.8 (of a maximum confidence
501 1). Nodes represent gene names of proteins with a highly ranked association (a minimum statistical
502 cutoff of |π| >1) within at least one organ. Nodes are sized by the number of organs the protein had a
503 strong association with. The level of association of each node to a particular organ is colored according to
504 the fraction of the total |π|-score each organ provides. Within each node is a bar plot of the π-scores for
21 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
505 each organ within the node. Putative functional groupings within the network are highlighted. Select
506 sections are highlighted in colored boxes and shown in 2× zoom.
507
508 Figure 4. Organism level protein networks modulated by the microbiome.
509 A multinomial regression controlling for organ and microbial colonization state of the mice was used to
510 assess proteins associated with colonization status. (A) Proteins ranked by regression coefficient; proteins
511 with coefficients of the greatest magnitude are most associated with colonization status. Proteins with
512 positive coefficients are more abundant in conventional mice, while proteins with negative coefficients
513 are more abundant in GF mice. (B) Log abundance of NNT or IGKV5-39 over the entire proteome in
514 each organ. (C) Protein-protein interaction networks from the top ranked proteins from the multinomial
515 regression associated with both conventional and GF status when controlling for organ and mouse. The
516 top 150 proteins associated with both GF and conventional status were analyzed (300 proteins total), and
517 proteins with high confidence interactions (0.8) are shown. Nodes are sized by the absolute value of the
518 regression coefficient and colored by association to GF (grey) or conventional (red) status. Putative
519 functional groupings are indicated.
520
521 Figure 5. Interactive 3D Visualization of Associations to Colonization Status. A 3D mouse model was
522 generated for use on the web-based ili platform (https://ili.embl.de/). (A) An example use case for the
523 protein-level association visualizations are shown through plotting the π-score enrichment for
524 conventional colonization status. (B) An example use case for the pathway level association visualizations
525 are shown through highlighting the enrichment scores for “Oxidoreductase”. Pathway association scores
526 were generated through –Log10(Benjamini-corrected p-values) of the conventional organs minus the GF
527 organs (from Supplemental Table S2).
528
22 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
529 References
530 Ahmadmehrabi S, Tang WHW. 2017. Gut microbiome and its role in cardiovascular 531 diseases. Curr Opin Cardiol 32: 761-766. 532 Beausoleil SA, Villen J, Gerber SA, Rush J, Gygi SP. 2006. A probability-based approach for 533 high-throughput protein phosphorylation analysis and site localization. Nat 534 Biotechnol 24: 1285-1292. 535 Bhattarai Y, Kashyap PC. 2016. Germ-Free Mice Model for Studying Host-Microbial 536 Interactions. Methods Mol Biol 1438: 123-135. 537 Birben E, Sahiner UM, Sackesen C, Erzurum S, Kalayci O. 2012. Oxidative stress and 538 antioxidant defense. World Allergy Organ J 5: 9-19. 539 Bouter KE, van Raalte DH, Groen AK, Nieuwdorp M. 2017. Role of the Gut Microbiome in the 540 Pathogenesis of Obesity and Obesity-Related Metabolic Dysfunction. 541 Gastroenterology 152: 1671-1678. 542 Brenes A, Hukelmann J, Bensaddek D, Lamond AI. 2019. Multibatch TMT Reveals False 543 Positives, Batch Effects and Missing Values. Mol Cell Proteomics 18: 1967-1980. 544 Celesk RA, Asano T, Wagner M. 1976. The size pH, and redox potential of the cecum in mice 545 associated with various microbial floras. Proc Soc Exp Biol Med 151: 260-263. 546 Chung H, Pamp SJ, Hill JA, Surana NK, Edelman SM, Troy EB, Reading NC, Villablanca EJ, 547 Wang S, Mora JR et al. 2012. Gut immune maturation depends on colonization with a 548 host-specific microbiota. Cell 149: 1578-1593. 549 Circu ML, Aw TY. 2011. Redox biology of the intestine. Free Radic Res 45: 1245-1266. 550 Diaz Heijtz R, Wang S, Anuar F, Qian Y, Bjorkholm B, Samuelsson A, Hibberd ML, Forssberg 551 H, Pettersson S. 2011. Normal gut microbiota modulates brain development and 552 behavior. Proc Natl Acad Sci U S A 108: 3047-3052. 553 Drake NM, DeVito LM, Cleland TA, Soloway PD. 2011. Imprinted Rasgrf1 expression in 554 neonatal mice affects olfactory learning and memory. Genes Brain Behav 10: 392- 555 403. 556 El Aidy S, Derrien M, Merrifield CA, Levenez F, Dore J, Boekschoten MV, Dekker J, Holmes E, 557 Zoetendal EG, van Baarlen P et al. 2013. Gut bacteria-host metabolic interplay 558 during conventionalisation of the mouse germfree colon. ISME J 7: 743-755. 559 Elias JE, Gygi SP. 2007. Target-decoy search strategy for increased confidence in large-scale 560 protein identifications by mass spectrometry. Nature Methods 4: 207-214. 561 Elias JE, Haas W, Faherty BK, Gygi SP. 2005. Comparative evaluation of mass spectrometry 562 platforms used in large-scale proteomics investigations. Nat Methods 2: 667-675. 563 Eng JK, McCormack AL, Yates JR. 1994. An approach to correlate tandem mass spectral data 564 of peptides with amino acid sequences in a protein database. J Am Soc Mass 565 Spectrom 5: 976-989. 566 Franco-Obregon A, Gilbert JA. 2017. The Microbiome-Mitochondrion Connection: Common 567 Ancestries, Common Mechanisms, Common Goals. mSystems 2. 568 Fu ZD, Selwyn FP, Cui JY, Klaassen CD. 2017. RNA-seq Profiling of Intestinal Expression of 569 Xenobiotic Processing Genes in Germ-Free Mice. Drug Metab Dispos 45: 1225-1238. 570 Haas W, Faherty BK, Gerber SA, Elias JE, Beausoleil SA, Bakalarski CE, Li X, Villen J, Gygi SP. 571 2006. Optimization and use of peptide mass measurement accuracy in shotgun 572 proteomics. Mol Cell Proteomics 5: 1326-1337. 23 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
573 Haberman Y, Karns R, Dexheimer PJ, Schirmer M, Somekh J, Jurickova I, Braun T, Novak E, 574 Bauman L, Collins MH et al. 2019. Ulcerative colitis mucosal transcriptomes reveal 575 mitochondriopathy and personalized mechanisms underlying disease severity and 576 treatment response. Nat Commun 10: 38. 577 Huang da W, Sherman BT, Lempicki RA. 2009a. Bioinformatics enrichment tools: paths 578 toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 579 37: 1-13. 580 Huang da W, Sherman BT, Lempicki RA. 2009b. Systematic and integrative analysis of large 581 gene lists using DAVID bioinformatics resources. Nat Protoc 4: 44-57. 582 Huttlin EL, Jedrychowski MP, Elias JE, Goswami T, Rad R, Beausoleil SA, Villen J, Haas W, 583 Sowa ME, Gygi SP. 2010. A tissue-specific atlas of mouse protein phosphorylation 584 and expression. Cell 143: 1174-1189. 585 Kall L, Canterbury JD, Weston J, Noble WS, MacCoss MJ. 2007. Semi-supervised learning for 586 peptide identification from shotgun proteomics datasets. Nat Methods 4: 923-925. 587 Kamga CK, Zhang SX, Wang Y. 2010. Dicarboxylate carrier-mediated glutathione transport 588 is essential for reactive oxygen species homeostasis and normal respiration in rat 589 brain mitochondria. Am J Physiol Cell Physiol 299: C497-505. 590 Karlsson FH, Fak F, Nookaew I, Tremaroli V, Fagerberg B, Petranovic D, Backhed F, Nielsen 591 J. 2012. Symptomatic atherosclerosis is associated with an altered gut metagenome. 592 Nat Commun 3: 1245. 593 Kindt A, Liebisch G, Clavel T, Haller D, Hormannsperger G, Yoon H, Kolmeder D, Sigruener 594 A, Krautbauer S, Seeliger C et al. 2018. The gut microbiota promotes hepatic fatty 595 acid desaturation and elongation in mice. Nat Commun 9: 3760. 596 Koopman JP, Janssen FG, van Druten JA. 1975. Oxidation-reduction potentials in the cecal 597 contents of rats and mice. Proc Soc Exp Biol Med 149: 995-999. 598 Kramer PA, Darley-Usmar VM. 2015. The emerging theme of redox bioenergetics in health 599 and disease. Biomed J 38: 294-300. 600 Krautkramer KA, Kreznar JH, Romano KA, Vivas EI, Barrett-Wilt GA, Rabaglia ME, Keller 601 MP, Attie AD, Rey FE, Denu JM. 2016. Diet-Microbiota Interactions Mediate Global 602 Epigenetic Programming in Multiple Host Tissues. Mol Cell 64: 982-992. 603 Kuno T, Hirayama-Kurogi M, Ito S, Ohtsuki S. 2016. Effect of Intestinal Flora on Protein 604 Expression of Drug-Metabolizing Enzymes and Transporters in the Liver and Kidney 605 of Germ-Free and Antibiotics-Treated Mice. Mol Pharm 13: 2691-2701. 606 Lapek JD, Greninger P, Morris R, Amzallag A, Pruteanu-Malinici I, Benes CH, Haas W. 2017a. 607 Detection of Dysregulated Protein Assoiciation Networks by High-Throughput 608 Proteomics Predicts Cancer Vulnerabilities. Nat Biotechnol In Press. 609 Lapek JD, Jr., Lewinski MK, Wozniak JM, Guatelli J, Gonzalez DJ. 2017b. Quantitative 610 Temporal Viromics of an Inducible HIV-1 Model Yields Insight to Global Host 611 Targets and Phospho-Dynamics Associated with Vpr. Mol Cell Proteomics 612 doi:10.1074/mcp.M116.066019. 613 Lapek JD, Jr., Mills RH, Wozniak JM, Campeau A, Fang RH, Wei X, van de Groep K, Perez- 614 Lopez A, van Sorge NM, Raffatellu M et al. 2018. Defining Host Responses during 615 Systemic Bacterial Infection through Construction of a Murine Organ Proteome 616 Atlas. Cell Syst doi:10.1016/j.cels.2018.04.010.
24 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
617 Larsson E, Tremaroli V, Lee YS, Koren O, Nookaew I, Fricker A, Nielsen J, Ley RE, Backhed F. 618 2012. Analysis of gut microbial regulation of host gene expression along the length 619 of the gut and regulation of gut microbial ecology through MyD88. Gut 61: 1124- 620 1131. 621 Leal-Lopes C, Velloso FJ, Campopiano JC, Sogayar MC, Correa RG. 2015. Roles of Commensal 622 Microbiota in Pancreas Homeostasis and Pancreatic Pathologies. J Diabetes Res 623 2015: 284680. 624 Lichtman JS, Alsentzer E, Jaffe M, Sprockett D, Masutani E, Ikwa E, Fragiadakis GK, Clifford 625 D, Huang BE, Sonnenburg JL et al. 2016. The effect of microbial colonization on the 626 host proteome varies by gastrointestinal location. ISME J 10: 1170-1181. 627 Mangiola F, Ianiro G, Franceschi F, Fagiuoli S, Gasbarrini G, Gasbarrini A. 2016. Gut 628 microbiota in autism and mood disorders. World J Gastroenterol 22: 361-368. 629 Mardinoglu A, Shoaie S, Bergentall M, Ghaffari P, Zhang C, Larsson E, Backhed F, Nielsen J. 630 2015. The gut microbiota modulates host amino acid and glutathione metabolism in 631 mice. Mol Syst Biol 11: 834. 632 McAlister GC, Nusinow DP, Jedrychowski MP, Wuhr M, Huttlin EL, Erickson BK, Rad R, Haas 633 W, Gygi SP. 2014. MultiNotch MS3 enables accurate, sensitive, and multiplexed 634 detection of differential expression across cancer cell line proteomes. Anal Chem 86: 635 7150-7158. 636 Milani C, Duranti S, Bottacini F, Casey E, Turroni F, Mahony J, Belzer C, Delgado Palacio S, 637 Arboleya Montes S, Mancabelli L et al. 2017. The First Microbial Colonizers of the 638 Human Gut: Composition, Activities, and Health Implications of the Infant Gut 639 Microbiota. Microbiol Mol Biol Rev 81. 640 Morton JT, Marotz C, Washburne A, Silverman J, Zaramela LS, Edlund A, Zengler K, Knight R. 641 2019. Establishing microbial composition measurement standards with reference 642 frames. Nat Commun 10: 2719. 643 Mottawea W, Chiang CK, Muhlbauer M, Starr AE, Butcher J, Abujamel T, Deeke SA, Brandel 644 A, Zhou H, Shokralla S et al. 2016. Altered intestinal microbiota-host mitochondria 645 crosstalk in new onset Crohn's disease. Nat Commun 7: 13419. 646 Nakaya K, Ikewaki K. 2018. Microbiota and HDL metabolism. Curr Opin Lipidol 29: 18-23. 647 Nguyen TT, Kosciolek T, Eyler LT, Knight R, Jeste DV. 2018. Overview and systematic 648 review of studies of microbiome in schizophrenia and bipolar disorder. J Psychiatr 649 Res 99: 50-61. 650 Nicholson JK, Holmes E, Kinross J, Burcelin R, Gibson G, Jia W, Pettersson S. 2012. Host-gut 651 microbiota metabolic interactions. Science 336: 1262-1267. 652 Nicolay BN, Danielian PS, Kottakis F, Lapek JD, Jr., Sanidas I, Miles WO, Dehnad M, Tschop K, 653 Gierut JJ, Manning AL et al. 2015. Proteomic analysis of pRb loss highlights a 654 signature of decreased mitochondrial oxidative phosphorylation. Genes Dev 29: 655 1875-1889. 656 Peng J, Elias JE, Thoreen CC, Licklider LJ, Gygi SP. 2003. Evaluation of multidimensional 657 chromatography coupled with tandem mass spectrometry (LC/LC-MS/MS) for 658 large-scale protein analysis: the yeast proteome. J Proteome Res 2: 43-50. 659 Protsyuk I, Melnik AV, Nothias LF, Rappez L, Phapale P, Aksenov AA, Bouslimani A, 660 Ryazanov S, Dorrestein PC, Alexandrov T. 2018. 3D molecular cartography using LC- 661 MS facilitated by Optimus and 'ili software. Nat Protoc 13: 134-154. 25 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
662 Quinn RA, Vrbanac A, Melnik AV, Patras KA, Christy M, Nelson AT, Aksenov A, Tripathi A, 663 Humphrey G, da Silva R et al. 2019. Chemical Impacts of the Microbiome Across 664 Scales Reveal Novel Conjugated Bile Acids. bioRxiv doi:10.1101/654756: 654756. 665 Ray PD, Huang BW, Tsuji Y. 2012. Reactive oxygen species (ROS) homeostasis and redox 666 regulation in cellular signaling. Cell Signal 24: 981-990. 667 Ronchi JA, Figueira TR, Ravagnani FG, Oliveira HC, Vercesi AE, Castilho RF. 2013. A 668 spontaneous mutation in the nicotinamide nucleotide transhydrogenase gene of 669 C57BL/6J mice results in mitochondrial redox abnormalities. Free Radic Biol Med 670 63: 446-456. 671 Sartor RB, Wu GD. 2017. Roles for Intestinal Bacteria, Viruses, and Fungi in Pathogenesis of 672 Inflammatory Bowel Diseases and Therapeutic Approaches. Gastroenterology 152: 673 327-339 e324. 674 Schwechter B, Rosenmund C, Tolias KF. 2013. RasGRF2 Rac-GEF activity couples NMDA 675 receptor calcium flux to enhanced synaptic transmission. Proc Natl Acad Sci U S A 676 110: 14462-14467. 677 Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, 678 Ideker T. 2003. Cytoscape: a software environment for integrated models of 679 biomolecular interaction networks. Genome Res 13: 2498-2504. 680 Simon GM, Cheng J, Gordon JI. 2012. Quantitative assessment of the impact of the gut 681 microbiota on lysine epsilon-acetylation of host proteins using gnotobiotic mice. 682 Proc Natl Acad Sci U S A 109: 11133-11138. 683 Singh M, Webster PD. 1978. Neurohormonal control of pancreatic secretion. A review. 684 Gastroenterology 74: 294-309. 685 Spivak M, Weston J, Bottou L, Kall L, Noble WS. 2009. Improvements to the percolator 686 algorithm for Peptide identification from shotgun proteomics data sets. J Proteome 687 Res 8: 3737-3745. 688 Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, Simonovic M, 689 Roth A, Santos A, Tsafou KP et al. 2015. STRING v10: protein-protein interaction 690 networks, integrated over the tree of life. Nucleic Acids Res 43: D447-452. 691 Thaiss CA, Zmora N, Levy M, Elinav E. 2016. The microbiome and innate immunity. Nature 692 535: 65-74. 693 Thompson A, Schafer J, Kuhn K, Kienle S, Schwarz J, Schmidt G, Neumann T, Johnstone R, 694 Mohammed AK, Hamon C. 2003. Tandem mass tags: a novel quantification strategy 695 for comparative analysis of complex protein mixtures by MS/MS. Anal Chem 75: 696 1895-1904. 697 Tilg H, Moschen AR. 2014. Microbiota and diabetes: an evolving relationship. Gut 63: 1513- 698 1521. 699 Ting L, Rad R, Gygi SP, Haas W. 2011. MS3 eliminates ratio distortion in isobaric 700 multiplexed quantitative proteomics. Nat Methods 8: 937-940. 701 Tolonen AC, Haas W. 2014. Quantitative proteomics using reductive dimethylation for 702 stable isotope labeling. J Vis Exp doi:10.3791/51416. 703 Tran HQ, Mills RH, Peters NV, Holder MK, de Vries GJ, Knight R, Chassaing B, Gonzalez DJ, 704 Gewirtz AT. 2019. Associations of the fecal microbial proteome composition and 705 proneness to diet-induced obesity. Mol Cell Proteomics 706 doi:10.1074/mcp.RA119.001623. 26 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
707 Villen J, Gygi SP. 2008. The SCX/IMAC enrichment approach for global phosphorylation 708 analysis by mass spectrometry. Nat Protoc 3: 1630-1638. 709 Wang Y, Yang F, Gritsenko MA, Wang Y, Clauss T, Liu T, Shen Y, Monroe ME, Lopez-Ferrer D, 710 Reno T et al. 2011a. Reversed-phase chromatography with multiple fraction 711 concatenation strategy for proteome profiling of human MCF10A cells. Proteomics 712 11: 2019-2026. 713 Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS, Dugar B, Feldstein AE, Britt EB, Fu X, 714 Chung YM et al. 2011b. Gut flora metabolism of phosphatidylcholine promotes 715 cardiovascular disease. Nature 472: 57-63. 716 Wessel D, Flugge UI. 1984. A method for the quantitative recovery of protein in dilute 717 solution in the presence of detergents and lipids. Anal Biochem 138: 141-143. 718 Xiao Y, Hsiao TH, Suresh U, Chen HI, Wu X, Wolf SE, Chen Y. 2014. A novel significance score 719 for gene selection and ranking. Bioinformatics 30: 801-807. 720 Zarate C, Jr., Machado-Vieira R, Henter I, Ibrahim L, Diazgranados N, Salvadore G. 2010. 721 Glutamatergic modulators: the future of treating mood disorders? Harv Rev 722 Psychiatry 18: 293-303. 723
27 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
SPLEEN SMALL NNTMTHFD1 PGRMC1 AKR1C14PTP4A1ANO6 AGPAT2 INTESTINE HMBS RRM1 ABHD14B TDP2 CNDP2 GLS AKR1C13 TYMNUP15S 3 STEAP3 FECH PSAT1 KCNAB2 DHRS1 RDH11GLUD1 ANXA13 RRM2 NR3C1 ALDH1A3 TFRC CTSE CTH AKR1C12 MAOADHRS3 ASNS SERPINA6 DHRS11 LCT FTL1GM5771 RDH19 GCLC GALK1 CTSB ATP6V1F HP CHIL3 MAOB EPDGAX T1 HINT1 PRSS1 GSTO1GSTA4 TBXAS1 SORAKRD 1B7 MECR TRY5 ABCB1A ALDH1A7 PPA2 LTF LCN2 RPL17 ZFP622 ALDH1A1 DCXR CARS NDUFA8 MCG_15095 ALOX15 GSTA1 PCK1 TREH CPA2 CPB1 NGP COX7C CYP2C65ABCD3 CAT RPS19 AMY2A2 PNLIP PLB1 WARS S100A9 UGT2B34 ACE CPA1 CTRL CYP2B10 ACAD11 CELA2A CELA1 UGT2A3 RDH7 NDUFS8 CEL MRPS34 AMY1 ACP5 HCLS1 IVDECHDC2 SERPINA6 ACE2 ENPEP CELA3B ST6GAL1 SCP2 CES2G TNKS1BP1 STARD5 CTRB1SYCN SARS2 CYP2D26 DPYD PHB UBAP2L IGHM MAPK13 AIFM2 HSD3B3 UGDH FYB APOA1 YBX1 SULT1DHSD3B1 3 ARAF TOMM22 MAN2B2 CASP7 GM49368 HCFC1 CES1E SPARCL1
CCN3 IGKV5-39 CNKSR3 AMY2A5 EPHX2 XPNPEP2 SARNP MYL4 ACOT7 PDZK1 PVALB ABRAXAS2 IGHG1 CES2E VNN1 PRKCQ CDC42SE2 ECI1 CES2DUOG X2 FYB1 SART1 CYP4V3IYD CES1F CMPK2 2210010C04RIK MS4A1 MEP1A IGHA WDR18 NNT CDV3 NDRG1 PLD3 CMBL AACS WBSCR16 LY6D HEBP1 ORC2SRSF6
CDA PIGR LRGREG31 G WASF2 DQX1 GM5771 FCGRT ATG3 ERO1L WIPI2 SORBSSRGAP3 2 LYZ1 CELA2A HEART TRY5 IGJ PLS1FRK CELA1 ARHGEF10L CTTN CTRL PLIN4 MZB1 PRSS1 GPAM INA MAPRE3 TXNDC5ACTR10 CPA2CTRB1 GNPAT NEFM PLIN5 FKBP11 CWF19L1 CPB1 MCG_15095 SCIN CHMP1A PSMD9 HEXB SERPINA6 CTDSP1 HPGD GPLD1 PSMB5 IGKV17-127 APOA1 DHX8 CHMP4B FHL2 TMEM236 SPPL2A B3GALAMY2AT5 2
RCN1 IGKV1-117 NMT1 PRPF38A ENTPD5 GRB7 TRIM2 APOE FDPS HP DHX16 NNT SCAMP3 FSTL1 CENPV MAP2K3 TM9SF1 GM20521 SEC11A CD82 PBLD1 PRMT5 CST3 NTAN1FBP1 ACAT2 AKR1C19 ZNF622 ERO1A ANP32E DENND4B CA3 EMC3 TBCB ENPP7 BTNL1 MPP1 ABHD6 ACTR3 FBXL18 NGP IMPACT TDP2 MAST4 SIAE CHORDC1 IGLC1 ARHGDIB IGKV9-120 GK CYP4V2 SUCLA2 ESS2 CLYBL SULT1C2 NAALADL1 RCC1L CORO1A UBE2Z C530008M17RIK KDM5A FAM210A IGKV1-117 GM43738 IGKV6-15 TLK1 LSM14A NDUFB3 FNTA
RASSF4 HMBS SNRK IGKV1-135 PRRC2B SLC16A1 ATPAF1 SDCBP2 FCRL5 ORC2 TBCE IGHV1-22 MYH6 KRT86 PURA IGKV5-39 GATAD2B COLON ISYNA1 TERF2IP PRMT3 CTDSPL PAFAH1B1 GNAQ ACAT3 POGZ SELENBP1 SND1 LHPP MAOA FXR2 ARHGEF17 HEBP1 SERPINA6 ASPN UBAP2L PROS1 HYDIN CARD19 FXR1 BRAIN SLC26A3 MUP8 HEBP1IRAK1 ABI1 NCK1 MUP11 NNT WASF2NNT UBL4 EIF3J2 SERPINA3K MCCC2 IYD ACADL EIF3K LIFR S100A5 ABCFAR1 FGAP3 RPS26 IGHG1 PRPF31 EIF1 SRP72 CDADC1 MTG1 ASNS CD59A RNASE4 SF3A1 RBP2 CPT1B GOLGA2AUP1 SRP68 COPZ1 RBBP5 FABP1 TAF15 EEF1B2 WDFY1 MRTO4 OSTC HPX FRAS1 CELA3B ANG4 SNRPF HSPA1B AMY2A2 APOA4 RPLP0 SYMPKFABP3 DNAJC3 SSSCA1 GSTZ1 SYCN MTX1 RPLP2 HSPA2 HSPA5 GYS1 RASGRF1 HDLBP SCGN CTRB1 REG3G FABP5 MTCH1 HSPA1L HSPA1A CPA2 TIMM50 DNAJB11 TAGLN2 UBA5 AMT CEL PRSS1 GPS1 CTRL MCG_15095 UROD HSPB1 LMO7 CPA1 CKMT1GPLD1 P4HB DNAJC10 GM5771 HEBP2 GOLM1 KRT1ATP6AP2 TXNDC5 PNLIP NDUFA1LDH2 B MZB1 PNLIPRP2CPB1 KRT7 LAMB1 CES2A CELA2A PKP2 MDH2 ALPI CWF19L1 NDUFB4 FKBP11 AMY1 KRT18 SYP CYP2C55 KRT10 MIA3 CUTC RASGRF2 GSTM5 GSTO1 MEP1A CHL1 MAPKAPK3CDC42SE2 CYP3A13 ANO6 AGR2 ACE2 PITRM1 GPX2 SORD TGM3 XPNPEP1 Increased in Conventional CLCA1PPP1R1B MGAM GALM SIS NLN SLC15A1 ABAT SLC3A1 Increased in Germ-Free AIF1 CPQ DPP4RGS10 NCF2 FAHD1 CHIA1 CTDSP1 GLOD5 CD200 NAALADL1 OBSCN NIPSNAP2 TRIM14 NEBL MUP22 IGHG1 MYOM2 AW551984 IGKV5-39 Low High EMC7 DHRS7C THEM4 ADH6A
OCIAD2 Significance Significance HPGD CADM3 ARSB CALM1 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
SPLEEN SMALL
Disulfide bond INTESTINE Peptidase S1A Disulfide bond Trypsin-like Peptidase S1A,chymotrypsin-type peptidases GO:0005615~extracellular space Proteolysis GO:0007586~digestion Secreted Immunoglobulin-like domain Pancreatic secretion SM00406:IGv Protein digestion and Protein digestion and absorption absorption mmu04972:Pancreatic secretion Iron binding IPR013106:Immunoglobulin V-set 0 5 10 15 IPR007677:Gasdermin -Log (Adj. p-value) Necrosis Metabolism of xenobiotics by cytochrome P450 mmu00340:Histidine metabolism HEART Lipid metabolism beta-amyloid mmu00140:Steroid hormone biosynthesis binding PIRSF000097:aldo-keto reductase glycerol-3-P O-acyltransferase mmu00983:Drug metabolism - other enzymes HDL mmu00982:Drug metabolism - cytochrome P450 high-density lipoprotein particle NADP positive regulation of triglyceride GO:0016491~oxidoreductase activity biosynthetic process Signal 10 5 0 5 10 IG -Log (Adj. p-value) Perilipin Immunoglobulin V-set Intermediate filament head, DNA-binding domain Keratin, type I COLON Phospholipid/glycerol Pancreatic secretion acyltransferase SM00406:IGv 1.5 1.0 0.5 0.0 -Log (Adj. p-value) Heat shock protein 70 family Protein processing in ER digestion BRAIN Stress response Heat shock protein 70 Peptidase S1A, chymotrypsin-type ER chaperone complex endoplasmic reticulum lumen Regulation of Ras protein Secreted signal transduction IPR007110:Immunoglobulin-like domain mmu03320:PPAR signaling pathway Extracellular region IPR000463:Cytosolic fatty-acid binding Metalloprotease GO:0008237~metallopeptidase activity 0.0 0.2 0.4 0.6 0.8 Lipocalin/cytosolic fatty-acid binding protein domain -Log (Adj. p-value) IPR011038:Calycin-like IPR012674:Calycin GO:0005215~transporter activity Increased in Conventional 4 2 0 2 4 Increased in Germ-Free -Log (Adj. p-value) CASP7 PRPF31 DHX8 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press SART1 DNAJC10 MZB1 CASP7 FSTL1 DNAJB11 SNRPF B3GALTLCN5 2 SPARCL1 HSPA5 MIA3 YBX1 HSPA1B LTF RCN1 FABP1 DNAJC10 MZB1 SYMPK GOLM1 P4HB NUP153 DHX16 1 HPX HSPA2 PRPF38A CST3 APOE DNAJB11 REG3G HSPA1L LAMB1 APOA4 WDR1SPARCL8 1 HSPA5 SRSF6 HSPA1A ZFP622 SF3A1 GM5771 DNAJC3 HSPB1 LYZ1 NR3C1 9 MRTO4 ERO1L APOA1 HSPA1B HEBP2 EIF3K UBE2Z LMO7 FABP1 GPS1 TRY5 FRK P4HB CDA HP HDLBP RPS19 RPLP2 HSPA2 NUP153 EIF3J2 RPL17 FBXL18 FABP5 UBA5 RPLP0 SRP68 RPS26 LRG1 FTL1 HSPA1L WDR18 PSMD9 2 PRSS1 TXNDC5 APOA4 SRP72 OSTC HSPA1A HEXB HSPB1 PSME1 PSMB5 3 MAPKAPK3 AKR1C14 ACTR10 CELA1 AKR1C12 PKP2 PAFAH1B1 AKR1C13 10 RBP2 MAP2K3 DPYD APOA1 CELA3B CPA2 MAPK13 EIF3K CELA2A RRM2 SYCN AKR1C19 MTHFD1 TYMS StressB3galt Response5 NEFM HDLBP COPZ1 ARFGAP3 CMPK2 CTRL CPA1 ORC2 RRM1 CEL NCK1 CPB1 8 GOLM1 GOLGA2 HPX PNLIPRP2PNLIP CD59A CST3 PGRMC1 ABI1 MCG_15095 RETSAT REG3G ACTR3 KCNAB2 CTRB1 DGAT1 ALDH1A1 WASF2 DNAJC3 RDH11 TFRC KRT1 GM5771 RDH7 SIS ATP6AP2 CTTN KRT10 LYZ1 ANO6 DHRS3 ATP6V1F MAOA MGAM ALDH1A7 KRT18 ACE2 AMY1 AGPAT2 HCLS1 KRT7 11 AKR1B7 MAOB HEBP2 GPLD1 KRT86 XPNPEP2 DPP4 LCT TRY5 FRK GPAM CDA LY6D GALM HP MEP1A ALPI GNPAT TREH FECH HMBS 4 7 GALK1 UGT2B34 FABP5 VNN1 SORD DCXR CYP2D26 EPHX2 UGDH UROD PIGR SCP2 CYP2C65 CBR3 ABCD3 CAT CYP2B10 UGT2A3 GSTM5 GPX2 CYP2C55 LRG1 FTL1 CES1E CTH ABAT GSTA1 CYP3A13 TXNDC5 LDHB GCLC PRSS1 NNT NDUFA1NDU2 FB4 GSTO1 GSTA4 ALDH1A3 5 NDUFB3 AMT GLS CNDP2 ALOX15 NDUFS8 COX7CPSME1 HEXB NDUFA8 GLUD1 SUCLA2 HINT1 PTP4A1 MDH2 CHMP4B CHMP1A ACAT2 PTPN23 SARS2 PCK1 AACS Innate Immunity AMY2A2 CARS 6 MYL4 OBSCN WARS MYH6 TIMM50 TOMM22
PSMB5 ASNSPRMT3PRMT5 FNTA IVD MCCC2 EMC3EMC7 WIPI2 ATG3 FXR2FXR1 PSAT1 LCT MAPKAPK3 ARAF PHB FAHD1GSTZ1 MECRACADL LHPPPPA2 FDPS GALM CELA1 TREH PKP2 GALK1 UGT2B34 DCXR CYP2D26 CELA3B CPASORD2 CELA2A UGDH CYP2C65 CBR3 SYCN CYP2B10 UGT2A3 GSTM5 GPX2 CYP2C55 CTRL CPA1 CES1E CEL GSTA1 CYP3A13 CPB1 GSTO1 GSTA4
PNLIPRP2PNLIP ALOX15
MCG_15095 RETSAT Xenobiotic metabolism with Cytochrome P450s CTRB1 DGAT1
SIS ANO6 Other Functional Groups Digestive enzymes from pancrease 1. Spliceosome 9. Translation 2. Proteosome 10. Nucleotide Color Bar Graph Size 3. Antigen processing biosynhesis 4. Protein digestion 11. Keratin Brain Small COX7C Signi cant in 5. Mitochondrial respiratory Intestine Increased in 1 Organ chain NADH dehydrogenase Colon Conventional 6. Mitochondrial glutamate Spleen Increased in metabolism Heart Signi cant in Germ-Free 7. Neutrophil degranulation 5 Organs 8. Bile acid synthesis Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
A B NNT
NNT
NNT
IGKV5-39
IGKV5-39
IGKV5-39
ATP5F1 C WDR43 UTP6 Mitochondrial NOC3L NUBPLNDUFB3 Complex 1 HEATR1 UTP11LSURF6 NDUFS4 NDUFS8 NDUFA11 IPO4 NDUFAF7 PDHB NDUFA3 NDUFA8 NDUFA5 NDUFB6 PRKAA2 Splicing PDK1 PCX SSB MTHFD1Innate RRAGA SNRNP70 LRG1 Immunity SNRPF SOD2 PRSS1 UGT2A3 MRPL49 REG3G RPS20 SF3A1 DHX16 CDA RPL3LMRPL17 MRPS17 HSPE1 HNRNPA0 TRY5 SIS SRSF6 GLA UGDH MRPL50 NPC2 RPL23A-PS3 SF3A2 GALM RPS26 HSPD1 GALT DCP1A Mitochondrial MRPS23 RPS10 LONP1 heat shock PSMD1 NFKB2 LCT RPL37A Mitochondrial ARF5 PSMD14 SYCN TRAPPC6B OSTC Ribosome YKT6 COPZ1 TRAPPC10 BCL2L14 PSME4 CELA3B DDOST PLD1 TSPAN14 ARHGAP15 CTRB1 RANBP9 TRAPPC12 TERF2 NRAS CTRL KCNAB2 SUZ12 RHOF GOLM1 CKAP4 SYNE2 NBEAL2 TBL1XR1 DOK3 ARHGEF7 SPHK2 CALML4 NOS1 HIST2H2AA1 PTK2 CAMK4
PPAP2B CTTN GRB7 DMD Increased in DIAP2 NUP62 LPL NUP107 ITGA1 Conventional SCARB2 TPM1 MYH8 HSPA12B Increased in Germ-Free BAG3 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
A Brain Spleen Colon
Heart Small Intestine
Protein Level Associations GF Conventional Associated Associated
B Brain Spleen Colon
Heart Small Intestine
Pathway Level GF Conventional Associations Associated Associated Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press
Organ level protein networks as a reference for the host effects of the microbiome
Robert H Mills, Jacob M Wozniak, Alison Vrbanac, et al.
Genome Res. published online January 28, 2020 Access the most recent version at doi:10.1101/gr.256875.119
Supplemental http://genome.cshlp.org/content/suppl/2020/02/11/gr.256875.119.DC1 Material
P
Accepted Peer-reviewed and accepted for publication but not copyedited or typeset; accepted Manuscript manuscript is likely to differ from the final, published version.
Creative This article is distributed exclusively by Cold Spring Harbor Laboratory Press for the Commons first six months after the full-issue publication date (see License http://genome.cshlp.org/site/misc/terms.xhtml). After six months, it is available under a Creative Commons License (Attribution-NonCommercial 4.0 International), as described at http://creativecommons.org/licenses/by-nc/4.0/.
Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the Service top right corner of the article or click here.
Advance online articles have been peer reviewed and accepted for publication but have not yet appeared in the paper journal (edited, typeset versions may be posted when available prior to final publication). Advance online articles are citable and establish publication priority; they are indexed by PubMed from initial publication. Citations to Advance online articles must include the digital object identifier (DOIs) and date of initial publication.
To subscribe to Genome Research go to: https://genome.cshlp.org/subscriptions
Published by Cold Spring Harbor Laboratory Press