The Gut Microbiome Modulates Colon ...

[ Pobierz całość w formacie PDF ]
//-->Downloaded frommbio.asm.orgon January 13, 2014 - Published bymbio.asm.orgDownloaded frommbio.asm.orgon January 13, 2014 - Published bymbio.asm.orgThe Gut Microbiome Modulates ColonTumorigenesisJoseph P. Zackular, Nielson T. Baxter, Kathryn D. Iverson,et al.2013. The Gut Microbiome Modulates Colon Tumorigenesis. mBio 4(6): .doi:10.1128/mBio.00692-13.Updated information and services can be found at:SUPPLEMENTALMATERIALREFERENCESThis article cites 49 articles, 22 of which can be accessed free at:Receive: RSS Feeds, eTOCs, free email alerts (when new articles cite thisarticle),more>>CONTENT ALERTSInformation about commercial reprint orders:Information about Print on Demand and other content delivery options:To subscribe to another ASM Journal go to:Downloaded frommbio.asm.orgon January 13, 2014 - Published bymbio.asm.orgRESEARCH ARTICLEThe Gut Microbiome Modulates Colon TumorigenesisJoseph P. Zackular,aNielson T. Baxter,aKathryn D. Iverson,aWilliam D. Sadler,bJoseph F. Petrosino,cGrace Y. Chen,bPatrick D. SchlossaDepartment of Microbiology and Immunology, University of Michigan, Ann Arbor, Michigan, USAa; Department of Internal Medicine, Division of Hematology andOncology, University of Michigan, Ann Arbor, Michigan, USAb; Human Genome Sequencing Center and Department of Molecular Virology and Microbiology, BaylorCollege of Medicine, Houston, Texas, USAcG.Y.C. and P.D.S. contributed equally to this article.ABSTRACTRecent studies have shown that individuals with colorectal cancer have an altered gut microbiome compared tohealthy controls. It remains unclear whether these differences are a response to tumorigenesis or actively drive tumorigenesis.To determine the role of the gut microbiome in the development of colorectal cancer, we characterized the gut microbiome in amurine model of inflammation-associated colorectal cancer that mirrors what is seen in humans. We followed the developmentof an abnormal microbial community structure associated with inflammation and tumorigenesis in the colon. Tumor-bearingmice showed enrichment in operational taxonomic units (OTUs) affiliated with members of theBacteroides, Odoribacter,andAkkermansiagenera and decreases in OTUs affiliated with members of thePrevotellaceaeandPorphyromonadaceaefamilies.Conventionalization of germfree mice with microbiota from tumor-bearing mice significantly increased tumorigenesis in thecolon compared to that for animals colonized with a healthy gut microbiome from untreated mice. Furthermore, at the end ofthe model, germfree mice colonized with microbiota from tumor-bearing mice harbored a higher relative abundance of popula-tions associated with tumor formation in conventional animals. Manipulation of the gut microbiome with antibiotics resulted ina dramatic decrease in both the number and size of tumors. Our results demonstrate that changes in the gut microbiome associ-ated with inflammation and tumorigenesis directly contribute to tumorigenesis and suggest that interventions affecting thecomposition of the microbiome may be a strategy to prevent the development of colon cancer.IMPORTANCEThe trillions of bacteria that live in the gut, known collectively as the gut microbiome, are important for normalfunctioning of the intestine. There is now growing evidence that disruptive changes in the gut microbiome are strongly associ-ated with the development colorectal cancer. However, how the gut microbiome changes with time during tumorigenesis andwhether these changes directly contribute to disease have not been determined. We demonstrate using a mouse model ofinflammation-driven colon cancer that there are dramatic, continual alterations in the microbiome during the development oftumors, which are directly responsible for tumor development. Our results suggest that interventions that target these changesin the microbiome may be an effective strategy for preventing the development of colorectal cancer.Received19 August 2013Accepted2 October 2013Published5 November 2013CitationZackular JP, Baxter NT, Iverson KD, Sadler WD, Petrosino JF, Chen GY, Schloss PD. 2013. The gut microbiome modulates colon tumorigenesis. mBio 4(6):e00692-13.doi:10.1128/mBio.00692-13.EditorMartin Blaser, New York UniversityCopyright© 2013 Zackular et al. This is an open-access article distributed under the terms of theCreative Commons Attribution-Noncommercial-ShareAlike 3.0 Unportedlicense,which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original author and source are credited.Address correspondence to Patrick D. Schloss, pschloss@umich.edu, or Grace Y. Chen, gchenry@umich.edu.olorectal cancer (CRC) is one of the most commonly diag-nosed malignancies worldwide, resulting in over a half-million deaths annually (1). Significant risk factors for CRC in-clude diets rich in red and processed meat, alcohol consumption,and chronic inflammation of the gastrointestinal tract (2–5). Eachof these factors is closely associated with changes in compositionand function of the complex community of microorganisms thatinhabits our gastrointestinal tract. This community, known as thegut microbiome, promotes various physiological functions thatare associated with cancer, including cell proliferation, angiogen-esis, and apoptosis (6–9). Therefore, we hypothesized that thecomposition, structure, and functional capacity of the gut micro-biome all directly affect tumor development in the colon.Several recent studies have addressed this hypothesis by char-acterizing the composition of the gut microbiome associated withCpatients with CRC (10–16). Using culture-independent ap-proaches, each of these studies observed a significant shift in thecomposition of the gut microbiome in patients with CRC com-pared to that in healthy controls. This phenomenon, referred to asdysbiosis, can be observed in both the luminal microbiome fromfeces and the mucosa-associated microbiome from tumor biopsyspecimens. Interestingly, each of these studies obtained conflict-ing results regarding the composition and structure of the CRC-associated microbial community. Furthermore, there are no bac-terial populations that have consistently been identified acrosseach study that can be attributed to the development or presenceof CRC. These data clearly show an association between abnor-malities in the gut microbiome and CRC; however, the conflictingresults point out the need for a mechanistic understanding of therole of the gut microbiome in this process.November/December 2013 Volume 4 Issue 6 e00692-13®mbio.asm.org1Downloaded frommbio.asm.orgon January 13, 2014 - Published bymbio.asm.orgZackular et al.FIG 1Inflammation-induced tumorigenesis is commensal dependent. (A) Mice were injected with azoxymethane (AOM) on day 1, followed by 3 subsequentrounds of water-administered 2% DSS. Colons were harvested 73 days after AOM, and tumors were grossly counted. Black wedges indicate fecal samples usedfor gut microbiome analysis (n 12). (B) Representative mice were euthanized following each round of DSS to identify macroscopic tumors (n 5 for each timepoint). An antibiotic cocktail of metronidazole, streptomycin, and vancomycin was administered in the drinking water of a separate cohort of mice for theduration of the model (n 9). Statistical analysis was performed using a two-tailed Student’sttest. *,P0.01. (C) Representative images of tumors in the distalcolon of conventional mice treated with AOM/DSS (n 12) and mice treated with an antibiotic cocktail and AOM/DSS (n 9). Error bars represent SEM.The combination of factors that could lead to dysbiosis is com-plex and not well understood. In addition, the effect of the devel-opment of this abnormal community on colon tumorigenesis re-mains unclear. Recent evidence suggests that certain strains ofBacteroides fragilisandEscherichia colican directly affect tumordevelopment in the colon through the production of virulencefactors (e.g., toxins and gene products) (17, 18). Furthermore,bacterial populations that produce the short-chain fatty acid bu-tyrate have antitumor effects in the colon by promoting apoptosisof colonic cancer cells (19, 20). We reason that dysbiosis of the gutmicrobiome leads to both enrichment of cancer-promoting bac-terial populations and loss of protective populations. Thus, un-derstanding the dynamics changes in the gut microbiome on acommunity-wide scale will be essential for understanding colontumor development.The gut microbiome is also likely to contribute to CRCthrough the initiation of inflammation. The link between inflam-mation and cancer is well established, and patients with inflam-matory bowel diseases, such as ulcerative colitis, are at a greaterrisk of developing CRC in their lifetime. In the case of ulcerativecolitis, the risk for cancer is related to both the duration and se-verity of inflammation, with an increasing rate of 0.5 to 1% peryear after the first decade (2, 21, 22). Chronic inflammation of thecolon leads to the production of various inflammatory cytokinesand reactive oxygen species that work in concert to generate atumor microenvironment that promotes carcinogenesis (21, 23,24). It has been suggested that this process is microbe driven, butit is unclear how the normally beneficial gut microbiome becomesinflammatory.To determine the role of the gut microbiome in inflammationand colon tumorigenesis, we used a well-established model ofcolitis-associated CRC that recapitulates the progression fromchronic inflammation to dysplasia and adenocarcinoma in hu-mans (25). We characterized the dynamics of the gut microbiomein this model and demonstrated that community-wide changespromote tumorigenesis in the colon. Our data support a model inwhich epithelial cell mutation and inflammatory perturbations ofthe gut microbiome lead to the development of an abnormal mi-crobial community with enhanced tumor-promoting activity.RESULTSInflammation-associated colon tumorigenesis.We were able toreplicate an inflammation-based murine model of tumorigenesisin specific-pathogen-free (SPF) C57BL/6 mice (n 12) using anintraperitoneal injection of the chemical carcinogen azoxymeth-ane (AOM) followed by three subsequent rounds of water-administered 2% dextran sodium sulfate (DSS) treatment (26, 27)(Fig. 1A). The mice showed a consistent pattern of weight lossfollowing each round of DSS treatment, with the most pro-nounced change occurring after the first round of DSS (seeFig. S1A in the supplemental material). We did not observe mac-2®mbio.asm.orgNovember/December 2013 Volume 4 Issue 6 e00692-13Downloaded frommbio.asm.orgon January 13, 2014 - Published bymbio.asm.orgGut Microbiome Modulates Colon Tumorigenesisroscopic tumors following the first round of DSS administration;however, we did observe increased infiltration by immune cells,lytter effsignificant epithelial damage, and submucosal edema (seeFig. S1B). In addition, we observed a significant increase in theproinflammatory mediators macrophage inflammatory protein 2(MIP-2), gamma interferon (IFN- ), tumor necrosis factor alpha(TNF- ), interleukin 6 (IL-6), and IL-1 (see Fig. S1A). Macro-scopic tumors and epithelial hyperplasia were apparent followingthe second round of DSS (Fig. 1B; see also Fig. S1B). At the end ofthe model, the cohort had a median of 14.5 tumors per mouse (n12), the majority of which were greater than 1 mm in diameterand located in the distal colon and rectum (see Fig. S1B). Theseresults demonstrate that our cohort of AOM/DSS-treated micedeveloped a significant number of colonic tumors with completepenetrance that could be detected as early as 7 weeks after AOMinjection.To determine whether tumor incidence and penetrance weredependent on the gut microbiome, we treated mice (n 9) withan antibiotic cocktail of metronidazole, vancomycin, and strepto-mycinad libitumfor 2 weeks prior to AOM and then throughoutthe model, including the days of AOM injection and throughoutthe DSS treatment and recovery periods. Antibiotic-treated micehad significantly fewer tumors in the colon than untreated mice(Fisher exact test,P0.001) (Fig. 1B). Tumors that were presentin antibiotic-treated mice were also significantly smaller thanthose observed in untreated mice (Student’sttest,P0.002)(Fig. 1C; see also Fig. S5A in the supplemental material). Theseresults suggest that specific populations within the microbiomewere essential for tumorigenesis. To determine whether the rela-tive change in bacterial density following antibiotic treatment wasdue to a change in the bacterial load, we performed quantitativePCR (qPCR) with the 16S rRNA gene from stool samples ofantibiotic-treated mice. The number of 16S rRNA gene copies permg of feces was not significantly different from that for untreatedstool samples (P 0.21) (see Fig. S2). Combined, these resultsindicate that changes to the structure of the community ratherthan total bacterial numbers affected tumorigenesis.Significant shifts in the microbiome are associated with co-lon tumorigenesis.To further test the hypothesis that specificchanges in the microbial community structure were associatedwith inflammation and tumorigenesis, we examined the dynamicsof the gut microbiome throughout the model using stool samplesfrom a subset of the original cohort of conventional mice treatedwith AOM/DSS for Fig. 1 (n 10). We used the fecal samplestaken prior to AOM injection as a baseline control for each mouseand then took samples following each subsequent round of DSSadministration (Fig. 1A). Mice showed a significant decrease inmicrobial diversity in the gut microbiome following the firstround of DSS administration through tumor development (P0.001) (Fig. 2A and B). Ordination of the distances between fecalsamples showed that at the time of euthanization, tumor-bearingmice developed a significantly altered microbiome that clusteredseparately from that in baseline samples taken prior to the firstround of DSS (Fig. 2C). Further examination of fecal samplescollected at various time points during the AOM/DSS tumor in-duction protocol revealed that significant alterations in the micro-biome could be observed as early as the first round of DSS admin-istration in 7 of the 10 mice. Each round of DSS treatment resultedin a significant change in the structure of the microbiome(Fig. 2D). Fecal samples taken from tumor-bearing mice after thethird round of DSS until the time of euthanization also clusteredseparately from earlier samples. The distances between clusterswere significantly higher than the distances within clusters(Fig. 2D). These clusters were observed using operation taxo-nomic unit (OTU) and phylogenetics-based metrics of -diversity(i.e.,YCand unweighted or weighted UniFrac) and could be dis-tinguished from one another using the Random Forest machinelearning algorithm (accuracy for each group: baseline, 100%; DSSround 1, 72.4%; DSS round 2, 71.9%; DSS round 3, 80.6%). Theseresults highlighted the association between a dramatically alteredmicrobiome structure and the presence of tumors.To determine the effect of inflammation on the microbialcommunity independent of tumorigenesis, we treated mice withthree rounds of DSS without the AOM injection (n 5). Therewas an initial community shift following the first round of DSS,but the subsequent stepwise shifts that occurred in AOM/DSS-treated mice were not observed in mice treated with DSS only (seeFig. S3 in the supplemental material). Furthermore, we did notobserve the sustained drop in microbial diversity that was ob-served in AOM/DSS-treated animals (see Fig. S3A). These resultssuggest that inflammation alone is not sufficient to cause micro-bial community changes. Rather, the synergistic effects of theAOM/DSS model are necessary for the development of the alteredmicrobiome structure and tumorigenesis.We next identified which OTUs were responsible for the dra-matic shifts in the microbial community structure during inflam-mation and tumorigenesis (Fig. 3). Consistent with our commu-nitywide -diversity analyses, we observed changes in 37 bacterialpopulations (after excluding OTUs representing 0.5% of thecommunity) during the time course of the model relative to thosein baseline samples prior to treatment. Fecal samples taken afterthe first round of DSS were enriched in the relative abundance ofOTUs affiliated with members of the genusBacteroides(OTUs 1and 13). We also observed a significant decrease in the relativeabundances of OTUs associated with members of the genusPre-votellaand unclassified genera within the familyPorphyromon-adaceae.Following the second round of DSS, we observed a fur-ther loss of the samePrevotella(OTUs 4 and 5) andPorphyromonadaceae(OTUs 7, 12, 15, 22, 31, and 48) and thecontinued enrichment ofBacteroides(OTUs 1 and 13). Samplestaken from mice following the third round of DSS showed signif-icant differences compared to those taken following the firstround of DSS and from healthy baseline mice (Fig. 3) (allPvalueswere 0.001 as determined by analysis of molecular variance[AMOVA]). Tumor-bearing mice showed enrichment in OTUsaffiliated withBacteroides(OTU 1),Odoribacter(OTU 3), andTuricibacter(OTU 20). Additionally, we detected a marked bloomof a member of theErysipelotrichaceaefamily (OTU 26), whichwas undetectable in all of the mice prior to the second round ofDSS, when tumors are not evident. Simultaneous with the bloom-ing of several bacterial populations, there was a significant de-crease in the relative abundance of OTUs associated with mem-bers of the genusPrevotella(OTUs 4 and 5) and the familyPorphyromonadaceae(OTUs 7, 12, 15, 22, 31, and 48). An OTUassociated with theBacteroidesgenus (OTU 13), which bloomedduring the onset of inflammation, decreased significantly follow-ing the third round of DSS. These results strongly suggest thatboth inflammation and tumorigenesis promote gut microbiomedysbiosis, as highlighted by major shifts in bacterial populationsfrom a wide range of taxonomic groups.November/December 2013 Volume 4 Issue 6 e00692-13®mbio.asm.org3Downloaded frommbio.asm.orgon January 13, 2014 - Published bymbio.asm.orgZackular et al.FIG 2Development of a dysbiotic gut microbiome during colon tumorigenesis. Microbiome analysis was performed with fecal samples from 10 representativemice; color coding is as indicated in Fig. 1A. (A) Inverse Simpson’s diversity index. (B) Observed community richness estimate. Statistical analysis was performedusing repeated-measures paired group analysis of variance. (C) Nonmetric multidimensional scaling (NMDS) ordination based onycdistances for all 10 miceduring the AOM/DSS model. (D) AverageYCdistance within (black) and between (gray) phases of the model. Error bars represent SEM.We hypothesized that the variability in tumor burden amongAOM/DSS-treated mice was associated with variability in the gutmicrobiome between mice (coefficient of variation for tumor bur-den 37.9) (Fig. 1C). We identified an OTU related to an unclas-sified genus within the familyPorphyromonadaceae(OTU 12) thatwas negatively correlated with tumor burden (Spearman correla-tion0.73;P0.05). The relative abundance of this bacterialpopulation decreased with each round of DSS, and this drop inabundance was more pronounced in mice with higher tumor bur-dens. These results suggest that alterations in the relative abun-dances of specific bacterial populations were associated not onlywith the incidence of tumors but also with their prevalence.Tumor-associated alterations in the microbiome increasetumorigenesis in germfree mice.To determine whether thecommunity-wide microbiome changes directly contributed to tu-mor incidence in the colon, we conventionalized germfree micewith either the healthy microbiome of untreated mice or the mi-crobiome of tumor-bearing mice analyzed in Fig. 1. To ensure thatmice were repeatedly inoculated and stably colonized, we trans-ferred fresh feces and bedding to two groups of germfree mice (n10/group). One group was housed with the bedding fromhealthy, untreated SPF mice, and a second group was housed withbedding from tumor-bearing AOM/DSS-treated mice. To mini-mize litter effects, each group was comprised of two cages of 5mice collected from separate litters that were randomly assignedto each of the cages. Following conventionalization, mice weretreated with AOM/DSS under germfree conditions, as describedabove (Fig. 1). All bacterial phyla and 90% (62 of 69) of genus-level taxa detected in donor samples were detected within the re-cipient germfree mice (see Table S3 in the supplemental material),which is higher than has been previously reported (28, 29). Fur-thermore, 81% of the sequences we obtained from the donor micebelonged to OTUs that were found in the recipient germfree mice.Mice conventionalized with the microbiome of tumor-bearingmice had a 2-fold increase in tumor burden (P 0.002) relative tothat of mice conventionalized with a healthy microbiome(Fig. 4A). Additionally, tumors from these mice were significantlylarger than those observed in recipients of a healthy microbiome(P 0.002) (see Fig. S5B). Similar to our results with SPF mice,germfree mice conventionalized with the community of tumor-bearing mice had a significantly less diverse gut microbiome (P0.001). Using community-wide -diversity analyses, we deter-mined that conventionalization with these two treatments of bed-ding resulted in two distinct microbial community structures(AMOVA.P0.001) (Fig. 4C). Germfree mice conventionalizedwith the microbiome of tumor-bearing mice showed significant4®mbio.asm.orgNovember/December 2013 Volume 4 Issue 6 e00692-13 [ Pobierz całość w formacie PDF ]

zanotowane.pl

doc.pisz.pl

pdf.pisz.pl

shinnobi.opx.pl