high-fat diet-induced complement activation mediates...

Metabolism

High-Fat Diet-Induced Complement ActivationMediates Intestinal Inflammation and Neoplasia,Independent of ObesityStephanie K. Doerner1, Edimara S. Reis2, Elaine S. Leung3, Justine S. Ko1,Jason D. Heaney4,5, Nathan A. Berger1,6, John D. Lambris2, and Joseph H. Nadeau1,3

Abstract

Obesity and related metabolic disturbances are closely asso-ciated with pathologies that represent a significant burden toglobal health. Epidemiological and molecular evidence linksobesity and metabolic status with inflammation and increasedrisk of cancer. Here, using a mouse model of intestinal neo-plasia and strains that are susceptible or resistant to diet-induced obesity, it is demonstrated that high-fat diet-inducedinflammation, rather than obesity or metabolic status, is asso-ciated with increased intestinal neoplasia. The complementfragment C5a acts as the trigger for inflammation and intestinaltumorigenesis. High-fat diet induces complement activation

and generation of C5a, which in turn induces the production ofproinflammatory cytokines and expression of proto-oncogenes.Pharmacological and genetic targeting of the C5a receptorreduced both inflammation and intestinal polyposis, suggest-ing the use of complement inhibitors for preventing diet-induced neoplasia.

Implications: This study characterizes the relations between dietand metabolic conditions on risk for a common cancer andidentifies complement activation as a novel target for cancerprevention. Mol Cancer Res; 14(10); 953–65. �2016 AACR.

IntroductionObesity is an increasingly important risk factor for many

cancers (1, 2). Although the mechanisms underlying the associ-ation between obesity and cancer are not fully understood,hormonal changes and chronic inflammation support a favorableenvironment for tumor progression (1, 2). Alterations in adiposetissue that occurs in response to excess nutrient intake lead to animbalance between adipokines that are associated with insulinresistance and a proinflammatory environment that characterizesmetabolic syndrome. Whereas leptin potentiates tumor growththrough induction of cell proliferation, angiogenesis and inflam-mation, adiponectin has anti-inflammatory properties and isnegatively correlated with body mass index (BMI; ref. 3). Incolorectal cancer, imbalances in insulin and adipokine metabo-lism are associated with oxidative stress and increased levels ofproinflammatory cytokines (3, 4).

Supporting a role for diet-induced obesity (DIO) in promotinginflammation and cancer, levels of IL6 and tumor burden are bothreduced following azoxymethane treatment of obese Lepob/obmicefed a bean-based diet versus a standard diet (5). Furthermore,medications used to treat hyperlipidemia and hypertension pre-vent colorectal carcinogenesis by attenuating chronic inflamma-tion (6). Similarly, preclinical and clinical studies indicate thatnon-steroidal antiinflammatory drugs have antitumorigenicactivity by inhibiting cyclooxygenases (7). The mechanisms andpathways underlying these effects remain to be identified.

Although a link between inflammation and cancer is appre-ciated, it is unclear whether increased tumorigenesis results fromDIO or instead from independent effects of diet on metabolismand tumorigenesis. Increasing evidencepoints towarda crucial roleof diet in determining tumor growth. Mice fed a high-fat diet(HFD)with reduced vitaminDandcalciumshow increasedweightgain and tumorigenesis compared with counterparts fed a low-fatdiet (LFD; ref. 8). Apart from vitamin and nutrient deficiencies,both high-fat and sucrose-rich diets are associated with increasedtumorigenesis (9, 10). Mediterranean diets are associated withreduced cancer risk (11). Further, dietary derivatives such asvitamins, b-carotene, resveratrol, and the omega-3/6 fatty acidratio are directly implicated in maintaining immune homeostasisin the intestineandwithprotection fromcolorectal cancer (12, 13).Given the complex relations between diet and obesity as well asbetween obesity and cancer, identifying the individual contribu-tions of diet and obesity-related metabolic disturbances toincreased cancer risk is challenging. Such elucidation could con-tribute to development of effective ways to manage cancer.

To explore themechanistic relationships among diet, DIO, andtumorigenesis, we focused on the Min allele of the Apc tumorsuppressor gene, an established mouse model of intestinal neo-plasia (14). These mice spontaneously develop intestinal polyps

1Department ofGenetics,CaseWesternReserveUniversity,Cleveland,Ohio. 2Department of Pathology and Laboratory Medicine, Universityof Pennsylvania, Philadelphia, Pennsylvania. 3Pacific NorthwestResearch Institute, Seattle, Washington. 4Department of Molecularand Human Genetics, Baylor College of Medicine, Houston, Texas.5Dan L Duncan Cancer Center, Baylor College of Medicine, Houston,Texas. 6Case Comprehensive Cancer Center, Case Western ReserveUniversity, Cleveland, Ohio.

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

N.A. Berger, J.D. Lambris, and J.H. Nadeau equally supervised this project.

Corresponding Author: J. Nadeau, Pacific Northwest Research Institute, 720Broadway, Seattle, WA 98122. Phone: 206-860-6752; Fax: 206-860-6753;E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-16-0153

�2016 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 953

on August 10, 2019. © 2016 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst August 17, 2016; DOI: 10.1158/1541-7786.MCR-16-0153

http://mcr.aacrjournals.org/

in amanner resembling human familial adenomatous polyposis,a condition that carries a 100% risk of developing colorectalcancer (15). Mutations in Apc are observed in 80% of sporadiccolorectal cancer cases (15). In parallel, the previously character-ized C57BL/6J-ChrA/J chromosome substitution strains (CSS)were used to investigate the influence of DIO (16). CSSs areinbred strains that carry a single chromosome from a donor strain(A/J) replacing the corresponding chromosome of the host strain(C57BL/6J [B6]; ref. 16). When exposed to a HFD, strains such asCSS-Chr2A/J (A2) and CSS-Chr9A/J (A9) are susceptible to DIO,while others such asCSS-Chr7A/J (A7) andCSS-Chr17A/J (A17) areresistant (17). Using a series of crosses, we combined the ApcMin

mutation with each of these four CSSs to create CSS.ApcMin/þ andCSS.Apcþ/þ test and control strains that all share a common B6inbred genetic background. We hypothesized that if DIO isimportant for increased intestinal neoplasia, then the DIO-sus-ceptible (CSSs-2 and -9) but not DIO-resistant (CSSs-7 and -17)strains would show increased polyp numbers upon HFD expo-sure. By contrast, if increased neoplasia results directly from HFDexposure, independent of obesity, increased polyp numberswould be observed in all four CSSs fed a HFD. Thus, with B6.ApcMin/þ or wild-type B6.Apcþ/þ mice, alone or in combinationwith the selected CSSs with contrasting genetic susceptibility toDIO (17), we were able to test the independent effects of diet andDIO on intestinal tumor development.

Here, we show that the chemical composition of dietary fatdifferentially affects obesity and insulin metabolism. Notably,HFD-induced inflammation from specific dietary fats, rather thanobesity or metabolic status, is associated with increased intestinalpolyposis. Mechanistically, specific dietary fats activate comple-ment signaling and generate complement fragment C5a, whichacts as a key trigger for inflammation and intestinal tumorigenesis.Importantly, C5a-induced tumorigenesis occurs independently ofobesity or metabolic status. Pharmacological and genetic target-ing of C5a receptor (C5aR) prevented diet-induced local andsystemic inflammation and significantly reduced polyp burden.

Materials and MethodsMice

C57BL/6J (B6, Jackson Laboratory Repository numberJR000664) and C57BL/6J-ApcMin/þ/J (B6.ApcMin/þ, JR002020)mice and the CSSs C57BL/6J-Chr2A/J/NaJ (A2, JR004380),C57BL/6J-Chr7A/J/NaJ (A7, JR004385), C57BL/6J-Chr9A/J/NaJ(A9, JR004387), and C57BL/6J-Chr17A/J/NaJ (A17, JR004395)were purchased from The Jackson Laboratory. CSSs were chosenaccording to their resistance (A7 and A17) or susceptibility (A2and A9) to significant diet-induced weight gain (17). Impor-tantly, at the time of selection, these CSSs were not known tocarry genetic modifiers that affected development of intestinalpolyps in B6.ApcMin/þ mice. CSSs were backcrossed onto a B6.ApcMin/þ background, and the resulting CSS.ApcMin/þ strains(A2.ApcMin/þ, A7.ApcMin/þ, A9.ApcMin/þ, and A17.ApcMin/þ) orthe corresponding wild-type controls (A2.Apcþ/þ, A7.Apcþ/þ,A9.Apcþ/þ, and A17.Apcþ/þ) were maintained on a 12-hourlight/dark cycle at the Wolstein Research Facility (CWRU).Complement-deficient mice, C57BL/6J-C3�/� (B6.C3�/�) andC57BL/6J-C5aR�/� (B6.C5aR�/�), have been previouslydescribed (18, 19). These strains were backcrossed for 10generations onto a C57BL/6J background at the University ofPennsylvania. All experiments used true littermates to ensure

exposure to similar microbiota composition among test andcontrol mice. Procedures were conducted in accordance withapproved Institutional Animal Care and Use Committee(IACUC) standards at Case Western Reserve University.

DietsDiets were obtained from Research Diets. Their composition is

detailed in Supplementary Table S1. These diets differed in thesource (coconut, corn, or olive oil) and amounts (high or low) offat. HFDs contained 58% kcal/g from hydrogenated coconut,corn, or olive oil (HFDCoco, HFDCorn, or HFDOlive, respectively),while LFDs contained 10.5% kcal/g from the same oils (LFDCoco,LFDCorn, or LFDOlive, respectively). The HFDCoco diet is rich(99.1%) in saturated fatty acids. HFDCorn is rich in omega-6polyunsaturated fatty acids (61.5%), similar to Western diets thatcontain a 25:1 ratio ofv-6:v-3 fatty acids. HFOlive diet is rich in themonounsaturated fatty acids (71.9%; ref. 20).

Study designFrom birth to 30 days of age, mice were fed autoclaved Purina

5010 LabDiet, and autoclaved water ad libitum. At 30 days, maleswere randomized to the HF or LF groups and fed ad libitum untilsacrifice after 3 (33days of age), 30 (60days of age), or 60 (90daysof age) days on the diet (Fig. 1A). Mice were fasted for 12 to 14hours and anesthetized with isoflurane prior to blood collection.Whole blood, plasma, and serum sampleswere collected from theretro-orbital sinus into tubes with or without EDTA. Body weightwas measured every other day, and body length was measured atthe final time point to calculate body mass index (BMI). At thefinal timepoint,micewere euthanizedby cervical dislocation, andthe epididymal fat pad mass (EFPM) was used as a measure ofadiposity. The small and large intestines were immediatelyremoved, flushed with cold PBS, and cut longitudinally. Polypswere counted, and cross-sectional diameter was measured in thesmall intestine and colon using a Leica MZ10F Modular Stereo-microscope. Polyp size and number were used to calculate totalpolyp mass for each mouse. Intestinal samples were immediatelycollected for RNA and protein analysis, frozen in liquid nitrogen,and stored at �80�C. Numbers of mice used in each experimentare also indicated individually with data points in each graph.Numbers for molecular analysis are 3 to 5 samples.

Metabolic and cytokine analysisFasting insulin levels were measured with Mercodia Ultrasen-

sitiveMouse Insulin ELISA. Fasting glucose wasmeasuredwith anOneTouchUltra glucometer (Life Scan, Inc.). Homeostatic modelassessment of insulin resistance (HOMA-IR; refs. 21, 22) wascalculated using fasting insulin and glucose levels [HOMA-IR ¼fasting insulin (pmol/L)� fasting blood glucose (mmol/L)/22.5].Leptin, adiponectin, and complement C5a were measured inplasma with ELISA kits from EMD Millipore (leptin and adipo-nectin) and R&D Systems (C5a).

Quantitative RT-PCR and protein analysisSize-matched polyp and normal tissues were collected from the

small intestine of B6.ApcMin/þ mice and B6.Apcþ/þ mice, respec-tively. Tissueswere frozen inQiagenRLTbuffer andRNAextractedwith Qiagen RNeasy Mini Kits. Quantitative RT-PCR was donewith SYBRGreen (Quanta BioSciences) tomeasure the expressionlevels of F4/80, IL6, IL1b, TNF, and Myc in adipose or intestinaltissues (see Supplementary Table S2 for primer sequences).

Doerner et al.

Mol Cancer Res; 14(10) October 2016 Molecular Cancer Research954




Figure 1.

Diet and genetic background determine obesityand insulin resistance. A, study design for dietstudies. Male mice were maintained on a stock5010 (gray) diet until 30 days of age (d.o.a.)before being randomly assigned to a HFD(purple) or LFD (blue). Mice remained on a dietstudy for 3 days (33 d.o.a.), 30 days (60 d.o.a.),or 60 days (90 d.o.a.). Wild-type B6.Apcþ/þ, A2.Apcþ/þ, A9.Apcþ/þ, A7.Apcþ/þ, and A17.Apcþ/þ

strains and the equivalentApcMin/þ (B6.ApcMin/þ,A2.ApcMin/þ, A9.ApcMin/þ, A7.ApcMin/þ, and A17.ApcMin/þ) strains were fed the LFDCoco

or HFDCoco for 60 days. B, body weight curvesfrom B6.Apcþ/þ and B6.ApcMin/þ. Final bodyweight (C) and HOMA-IR (D) were measured atday 60 of treatment. E and F, the total number ofpolyps (E) and polypmass (F) were measured inthe intestine of each mouse at day 60 oftreatment. Values represent means � SEM(n¼ 5–54 males/group). � , P < 0.05; �� , P < 0.01;�� , P < 0.001 in relation to the correspondingLFD-fed group. #####, P < 0.001, in relation to B6.ApcMin/þ fed HFD.

Complement Activation and Intestinal Tumorigenesis

www.aacrjournals.org Mol Cancer Res; 14(10) October 2016 955




Tissues for Western blot analysis were immediately submerged inRIPA buffer supplemented with phosphatase and protease inhi-bitors, frozen in liquid nitrogen, and stored at �80�C. All anti-bodies (AKT, pAKT, IKK, pIKK, NFkB, p65 NFkB, STAT3, andpSTAT3) were purchased from Cell Signaling Technology andused according to the manufacturer's instructions. Homogenatesfrom intestinal samples were used to measure intestinal inflam-mation using R&DSystems and IL1b, IL10, TNFa, VEGF, and IL23ELISAs and theMilliplexMapMouse Cytokine/Chemokine Panelfrom EMD Millipore.

Immunohistochemistry of intestinal tissueTissues from the small intestine (ileum)were fixed overnight at

4�C in 4% paraformaldehyde, pH 7.2, and embedded in paraffin.Cross-sections (5 mm) were de-paraffinized in xylene and rehy-drated in serial dilutions of ethanol and 1xPBS. Tissues wereblocked in 5% goat serum, 3% BSA, 0.3% Triton X100 in 1x PBSwithout antigen retrieval. Sections were incubated overnight at4�Cwith a 1:100dilution of a rabbit polyclonal anti-C3c antibody(ab15980, Abcam) in blocking solution. Sections were washed in1xPBS and, for secondary detection, incubated with a 1:500dilution of goat antirabbit AlexaFluor 555 (A-21429, Life Tech-nologies) in blocking solution at room temperature for 1.5 hours.After washes in 1xPBS, coverslips were mounted and nucleicounterstained in Vectashield hardset mounting medium withDAPI (H-1500, Vector Laboratories). Images were taken with aZeiss Axioplan2 and an AxioCam digital camera. Images wereprocessed and layered with Photoshop CS5.

Pharmacological inhibition of C5aRC5aR was inhibited with the Ac-Phe-[Orn-Pro-(D-Cha)-Trp-

Arg] peptide (PMX-53; ref. 23). An inactive peptide containing thesame amino acids in a scrambled order served as control. Thetreatment with PMX-53 or control peptide was initiated concom-itantly with the specific LFD or HFD (i.e., 30 days of age). Drugswere diluted in PBS, and subcutaneous injections were adminis-tered every other day at 2.0 mg/kg.

Fluorescence-activated cell sorting (FACS) of intestinalimmune cells

Lamina propria lymphocytes were isolated using a Percollgradient. Antibodies for CD11b, CD11c, CD45, and GR1 werepurchased from Biolegend and used according to the manufac-turer's instructions. Alternatively, cells were stained with therespective isotypes. All cell preparations showed at least 95%viability as measured with DAPI staining (Life Technologies).Samples were analyzed using the BD LSR II flow cytometer (BDBiosciences) and FCS Express software (v4.0; De Novo Software).

Statistical analysisStudent t tests and one-way analysis of variance (ANOVA) were

used to determine statistical significance, which was accepted atthe P < 0.05 level after Bonferroni correction for multiple hypoth-esis testing.

ResultsDiet and genetic background determine obesity and insulinresistance

To investigate the mechanisms of DIO-induced inflammationand cancer, we first assessed the effect of diet and genetic back-

ground on obesity and glucose homeostasis. Previously char-acterized CSSs were used to distinguish between diet- versusobesity-induced effects (16, 17). Wild-type B6.Apcþ/þ and CSS.Apcþ/þmice were fed a calorically equivalent diet high or low inhydrogenated coconut oil, HFDCoco, or LFDCoco (Supplemen-tary Table S1). After 60 days on HFD versus LFD, B6.Apcþ/þ, A2.Apcþ/þ, and A9.Apcþ/þ wild-type mice showed increased finalbody weight (FBW), BMI, and EFPM as well as elevated levels offasting glucose and insulin with a correspondingly higherinsulin resistance index [HOMA-IR] (Fig. 1A–D; SupplementaryTable S3). Conversely, A7.Apcþ/þ and A17.Apcþ/þ mice wereresistant to DIO and maintained baseline metabolic parameters(Fig. 1C and D; Supplementary Table S3). These results dem-onstrate the importance of diet and genetic background ondevelopment of obesity as well as glucose homeostasis, andvalidate the use of CSSs that differ in DIO susceptibility as amodel for distinguishing the independent contributions of dietand obesity to intestinal tumorigenesis.

Diet, but not obesity or metabolic status, promotes intestinaltumorigenesis

To explore the association betweenDIO and tumorigenesis, weused mice heterozygous for a mutant form of the Apc tumorsuppressor gene (B6.ApcMin/þ) that were fed HFDCoco or LFDCoco.In contrast to B6.Apcþ/þ mice on HFDCoco, which were �10%heavier than animals fed the equivalent LFD after 60 days, B6.ApcMin/þ mice on HFDCoco gained weight during the initial 40days, but then lost weight progressively andweremoribund at the60-day time-point (90 days of age; Fig. 1B and C), consistent withthe development of multiple intestinal polyps and cachexia (24).

Total intestinal polyp number and mass were significantlyincreased in all strains fed the HFDCoco but not the LFDCoco after60 days (Fig. 1E and F; Supplementary Table S3), regardless ofobesity or metabolic status (Fig. 1C and D; Supplementary TableS3). While B6.ApcMin/þmice fed LFDCoco developed an average of14 polyps, counts increased 6.6-fold in response to HFDCoco.Polypnumbers ranged from6.6 to 27.4 inCSS.ApcMin/þ strains fedLFDCoco and were 3.3- to 6.6-fold higher in HFDCoco-fed mice,independent of weight gain or glucose and insulin levels (Fig. 1E;Supplementary Table S3). Similarly, polyp mass increased 8.4- to25.4-fold in strains fed HFDCoco compared with LFDCoco (Fig. 1F;Supplementary Table S3), indicating that dietary fat, rather thanobesity and associated metabolic changes, promotes intestinalneoplasia in genetically susceptible B6.ApcMin/þ mice.

Although increased polyp number and mass were observedin obesity-susceptible A2.ApcMin/þ mice fed HFDCoco comparedwith LFDCoco, these parameters were significantly reducedcompared with other strains (Fig. 1E and F). These data suggestpresence of a factor on chromosome 2 that modifies intestinaltumorigenesis. Furthermore, an enhanced inflammatory statewas observed in the ApcMin/þ mice fed the HFD when comparedwith their LFD counterparts, as evidenced by increased neutro-phil counts but were significantly lower in A2.ApcMin/þ (Sup-plementary Fig. S1).

We next investigated the impact of dietary energy content andchemical composition on intestinal neoplasia with diets contain-ing distinct sources of fat (coconut, corn, or olive oil; Supple-mentary Table S1). As observed in B6.Apcþ/þ mice exposed toHFDCoco, B6.Apc

þ/þmice fed HFDCorn were insulin resistant (Fig.2B; Supplementary Table S4). However, only HFDCoco- but notHFDCorn-fed mice displayed increases in FBW, EFPM, and BMI

Doerner et al.





Figure 2.

Diet, but not obesity or metabolic status,promotes intestinal tumorigenesis. B6.ApcMin/þ

mice were fed the LFD or HFD with coconut(purple), corn (green), or olive (blue) oil as asource of fat for 60 days. The final body weight(A), HOMA-IR (B), total number of polyps (C) andpolypmass (D)were assessed after day60of diettreatment. Values represent means � SEM(n ¼ 7–54 males/group). � , P < 0.05; �� , P < 0.01;�� , P < 0.001 in relation to the equivalentLFD-fed group. ###, P < 0.001 in relation toHFDCoco- or HFDCorn-fed groups.






compared with LFD (Fig. 2A; Supplementary Table S4), suggest-ing that a diet rich in corn oil induces insulin resistance in theabsence of obesity. Mice fed HFDOlive showed significant weightgain without corresponding metabolic changes compared withLFDOlive, as indicated by increased FBW, BMI, and EFPM, butnormal levels of fasting glucose and insulin (Fig. 2A and B;Supplementary Table S4), in line with previous findings (20).Thus, chemical composition of dietary fat differentially affectsobesity and insulin resistance.

Importantly, the various diets also differentially affected intes-tinal tumorigenesis. Total intestinal polyp number andmass weresignificantly increased in mice fed HFDCoco and HFDCorn com-pared with corresponding LFD-fed animals (Fig. 2C and D). Incontrast, no significant differences in total polyp number or masswere observed inB6.ApcMin/þ fedHFDOlive versus LFDOlive (Fig. 2Cand D), demonstrating that although a diet rich in olive oilresulted in DIO, it prevented insulin resistance and intestinalneoplasia, suggesting that tumorigenesis may be associated withspecific dietary fats rather than obesity.

Similar degrees of tumorigenesis were observed after shorterexposure to aHFD (30vs. 60days), independent of fat source (Fig.2C and D; Supplementary Fig. S2). This dietary effect occurredafter a 30-day exposure to dietary fat, prior to significant increasesin FBW, BMI, or EFPM (Supplementary Fig. S2; SupplementaryTable S4), indicating that conditions that favor polyp develop-ment are established prior to onset of obesity or insulin resistance.

Intestinal neoplasia is associated with diet-inducedinflammation

To identify mechanisms linking dietary fat with intestinaltumorigenesis, we focused on HFDCoco effects over a 30-dayperiod, before the onset of DIO, metabolic changes, or tumor-induced cachexia. Circulating levels of the proinflammatory cyto-kines IL6, IL1b, and monocyte chemotactic protein (MCP-1) andthe adipose tissue-derived hormone, leptin, were elevated inmicefed HFDCoco compared with LFDCoco (Fig. 3A–C and E, cf. ref. 1).Levels of antiinflammatory cytokine IL10were not affected by dietor strain (Fig. 3D). In contrast, the level of adiponectin, anadipokine that is involved in fatty acid oxidation and has antiin-flammatory properties, was reduced inHFDCoco-fedmice (Fig. 3F).mRNA expression levels of IL6 and the macrophage marker F4/80were significantly elevated in adipose tissue from B6.ApcMin/þ inresponse to HFDCoco (Fig. 3G and H). Expression of the proin-flammatory cytokines IL1b, IL6, and TNFa was also increased inintestinal tissue of both B6.Apcþ/þ and B6.ApcMin/þ mice fed HFD,while c-Mycwas significantly elevated in response toHFDCoco onlyin B6.ApcMin/þ (Fig. 3I–L), supporting the concept that HFDCoco

triggers a proinflammatory environment that promotes polypdevelopment.

To examine the influence of diet on early tumor environ-ment, mice were exposed to HFD or LFD for 3 days. Prior to anychanges in FBW or HOMA-IR (Fig. 4A and B) after HFDCoco

exposure, increased intestinal polyp numbers were observed(Fig. 4C and D). Circulating levels of adiponectin weredecreased in the HFD-fed mice when compared with the LFDcounterparts (Fig. 4F). Remarkably, the short HFD exposuresignificantly increased IL6, IL23, TNFa, and VEGF but not IL10in the intestines of mice fed HFDCoco compared with LFDCoco

(Fig. 4H–L), suggesting that HFDCoco rapidly triggers a proin-flammatory and proangiogenic environment that promotespolyp development.

Source of dietary fat determines local proinflammatoryenvironment and intestinal polyposis

In light of our observation that HFDOlive did not promotepolyposis in genetically susceptible mice despite DIO (Fig. 2),we evaluated intestinal inflammation in mice fed HFDCoco,HFDCorn, or HFDOlive. After a 30-day exposure of wild-typeB6.Apcþ/þ mice to HFD or the corresponding LFD, levels ofVEGF, TNFa, IL23, and IL1b were significantly elevated inmice fed HFDCoco and HFDCorn compared with the LFD (Fig.5A–D). Although no increase in these inflammatory factorswas observed in mice fed HFDOlive, the antiinflammatorycytokine, IL10, was significantly elevated in HFDOlive-fed mice(Fig. 5E).

CSS A2.ApcMin/þ, which showed reduced tumorigenesis regard-less of dietary exposure or obesity status, was reassessed to explorethe link between diet and inflammation. Notably, chromosome 2in the A/J strain as well as in CSS A2.ApcMin/þ carries a dysfunc-tional complement component C5 gene (16). C5 is cleaved inresponse to complement activation, with the subsequent gener-ation of the C5a fragment, an established proinflammatorymedi-ator with a detrimental role in cancer (25–29). This cleavageproduct signals through theG-coupled receptor C5aR (30). To testwhether dietary fat modulates complement activation, plasmaC5a levels were measured in B6.Apcþ/þ, CSS.Apcþ/þ, and thecorresponding ApcMin/þ strains fed HFDCoco or LFDCoco for 60days. All strains, except A2.Apcþ/þ and A2.ApcMin/þ, showed asignificant increase in plasma C5a in response to HFDCoco,regardless of obesity status or obesity-associated metabolicchanges (Supplementary Fig. S3).

Next, circulating levels of plasma C5a levels were measuredin response to the different sources of dietary fat. After 30 daysof diet exposure, C5a levels were significantly elevated in bothB6.Apcþ/þ and B6.ApcMin/þ strains fed HFDCoco or HFDCorn butnot in HFDOlive when compared with the corresponding LFD-fed mice (Fig. 6A). After a short 3-day exposure to HFDCoco, C5alevels were significantly increased in plasma and the intestine(Fig. 4E and G).

Additionally, using an antibody that recognizes comple-ment C3 fragment C3c, and the C3c epitope of native C3 andC3b, increased deposition of complement C3 in the normalsmall intestine was observed in response to HFDCoco whencompared with LFDCoco (Supplementary Fig. S4A–S4D). Spe-cifically, antibody staining accumulated within the laminapropria and along the epithelial basement membrane in theunderlying connective tissue. These results are consistent withprevious observations that activated complement C3 is depos-ited along the basement membrane in the inflamed intestinalmucosa of ulcerative colitis (UC) and Crohn disease (CD)patients and that complement C3 is expressed by subepithelialmyofibroblasts (SEMF), which are positioned subadjacent toIECs along basement membrane, in response to inflammation(31, 32). Curiously, complement C3 also accumulated withinpolyps of B6.ApcMin/þ mice fed HFDCoco for 30 days (Sup-plementary Fig. S4E and S4F). Complement C3 localizedprimarily to the stroma of polyp lesions but was expressed atgreatly diminished levels or was absent in the stroma sur-rounding normal epithelial cells in deep crypts under polyplesions and positioned the greatest distance from the intestinallumen. These results indicate that interactions with specificdietary fats trigger C3 activation and C5a generation shortlyafter HFD exposure.

Doerner et al.





Complement C5a mediates diet-induced inflammation andintestinal neoplasia

The link between HFD-induced complement activation andintestinal tumorigenesis was confirmed with a genetically engi-neered knockout mutation in the C3 complement component(C3) gene that was backcrossed onto the B6.ApcMin/þ background.Absent (homozygotes) or reduced (heterozygotes) C3 levels,which is normally essential for complement activation (33),resulted in reduced polyp number and mass in the intestine ofHFDCoco-fedmice (Supplementary Fig. S5). Similarly, diminishedtumorigenesis was observed in C5aR�/� mice on a B6.ApcMin/þ

background fedHFDCoco (Fig. 6B andC), supporting a role for the

C3–C5a–C5aR axis in diet-induced intestinal neoplasia. Inaddition, pharmacological targeting of C5aR with the antago-nist peptide PMX-53 (23), but not an inactive control peptide,resulted in a significant reduction in polyp number and massresulting from HFD exposure (Fig. 6B and C; SupplementaryTable S5). Complement deficiency did not influence bodyweight or glucose and insulin levels in mice fed the HFDCoco

for 30 days (Supplementary Table S5). Further, genetic orpharmacological inhibition of C5aR did not significantlyreduce polyp number or mass in males fed LFDCoco (Supple-mentary Table S5), supporting the notion that C5a specificallymediates HFD-induced inflammation.

0

0 0

5

10

15

20

25

500

1,000

1,500

2,000

2,500

0 50 0

5

10

15

20

75

100

125

10

20

30

40

50

LF+/+ Min/+

HF HFLF

LF+/+ Min/+

HF HFLF LF+/+ Min/+

HF HFLF LF+/+ Min/+

HF HFLF LF+/+ Min/+

HF HFLF

LF+/+ Min/+

HF HFLFLF+/+ Min/+

HF HFLFLF+/+ Min/+

HF HFLFLF+/+ Min/+

HF HFLF

LF+/+ Min/+

HF HFLF LF+/+ Min/+

HF HFLF LF+/+ Min/+

HF HFLF

5

10

15

20

25A B C D

E F G H

I J K L

Blood

Blood

IntestineIntestine Intestine Intestine

Blood

0 0

00

1

2

3

4

5

5

10

15

1

2

3

4

5

10

15

20Adipose Adipose

Blood Blood Blood

IL6

(pg

/mL

)

Ad

ipo

nec

tin

(mg

/mL

)

Lep

tin

(p

g/m

L)

IL6

(Rel

ativ

e ex

pre

ssio

n)

F4/

80 (

Rel

ativ

e ex

pre

ssio

n)

Myc

(R

elat

ive

exp

ress

ion

)

TN

Fa

(Rel

ativ

e ex

pre

ssio

n)

00

5

10

1520

40

60

80

100

1

2

3

4

5

IL1b

(R

elat

ive

exp

ress

ion

)

IL6

(Rel

ativ

e ex

pre

ssio

n)

IL1b

(p

g/m

L)

MC

P-1

(p

g/m

L)

IL10

(p

g/m

L)

Figure 3.

Intestinal neoplasia is associated with diet-induced inflammation. B6.Apcþ/þ and B6.ApcMin/þ mice were fed the LFDCoco or HFDCoco for 30 days. Levels ofanti- and proinflammatory mediators implicated in metabolic syndrome and cancer were measured in the blood (A–F), adipose tissue (G and H), and intestinaltissue (I–L) at day 30 of treatment. Values represent means � SEM (n ¼ 5–7 males/group). � , P < 0.05; �� , P < 0.01; �� , P < 0.001 in relation to the samestrain fed the LFDCoco. ##, P < 0.01 in relation to B6.Apcþ/þ fedHFDCoco.






Mechanistically, genetic and pharmacologic inhibition of C5aRsignaling reduced HFD-induced inflammation as indicated byreduced levels of intestinal VEGF, TNFa, and IL23 (Fig. 6D–I). Inaddition, abrogation of C5aR signaling with PMX-53 inhibited

AKT-NFkB signaling, which has been implicated in inflammationand cancer (34).Whereas increased levels of phosphorylated AKT,IKK, and NFkB were observed in polyp tissue from B6.ApcMin/þ

mice fed HFDCoco as compared with normal B6.Apcþ/þ tissue,activation was attenuated in mice treated with PMX-53 but not inthe control peptide (Fig. 6M). Furthermore, pharmacologicaltargeting of C5aR resulted in reduced mRNA expression levelsof IL1b and c-Myc in polyp tissue (Fig. 6J and K), and preventedinfiltration of inflammatory neutrophils to the lamina propria(Fig. 6L), which has been implicated as a key inflammatory triggerin DIO response (35).

DiscussionHere, we show that complement C5a-induced inflammation

promotes intestinal tumorigenesis. Complement signaling mod-ulates various pathophysiological processes and has been previ-ously implicated in tumorigenesis (33). Recent findings indicatethat complement activation in the tumor microenvironmentleads to tumor growth andmetastasis. In vivomodels using breast,lung and colon cancer cell lines showed that C5aR-mediatedsignaling creates a microenvironment that promotes growththrough recruitment ofmyeloid-derived suppressor cells (MDSC)and inhibition of antitumor immune responses mediated byCD8þ andCD4þT cells (25, 27–29, 36). In addition,C5a-inducedinflammation is associated with promotion of angiogenesis andtumor progression. In the absence of C3 or C5aR-mediatedsignals, mice genetically susceptible to ovarian cancer developsmall, poorly vascularized tumors. While MDSCs do not seem toplay a role in this model, complement activation is associatedwith increased tumorigenesis as indicated by elevated regulatoryT-cell numbers and increased VEGF-induced angiogenesis (37).Complement activation also promotes colitis-associated carcino-genesis via induction of proinflammatory IL1b and IL17 by bonemarrow-derived neutrophils (38). In addition to creating a proan-giogenic and protumorigenic microenvironment, C3a and C5aderived from tumor cells have an autocrine effect on tumorproliferation via activation of the PI3K/AKT signaling pathway(26). Such findings agree with our observations that the C5aRantagonist PMX-53 prevents activation of AKT/IKK/NFkB follow-ing HFDCoco exposure (Fig. 6M). Additionally, deficiency of PTX3has been associated with increased susceptibility tomesenchymaland epithelial carcinogenesis due to poor regulation of comple-ment activation and consequent inflammation (39). Further,development of hepatocellular carcinoma mediated by HFD-induced inflammation involved c-Myc and NFkB signaling path-ways in genetically C5-sufficient (C57BL/6) but not in C5-defi-cient (A/J)mice (40), pointing to complement activation as a linkbetween DIO and cancer. Finally, recent reports suggest thatC3aR- and C5aR-mediated signaling modulates neutrophil andCD8þ T-cell function with consequent increased tumorigenesis(41–43).

Obesity has become a major public health concern because ofits association with medical conditions such as hypertension,diabetes, cardiovascular disease, and certain cancers (44).Although mutations in genes such as leptin, leptin receptor, andmelanocortin-4 receptor affect satiety control, glucose homeosta-sis and energy metabolism, common obesity is considered amultifactorial disorder controlled by multiple genes combinedwith environment factors, such as diet, physical activity, andintestinal microbiota (45, 46). Given the complexity involved in

14

0

0 0

5

10

15

20

25

30

500C5a

(ng/

mL)

Adi

pone

ctin

(µg/

mL)

1,000

1,500

0

0

0

50

100

150

0

50

100

150

0

1,000

2,000

3,000

500

1,000

1,500

0

50

100

150

500C5a

(pg/

mL)

VE

GF

(pg/

mL)

TNFa

(pg/

mL)

IL10

(pg/

mL)

IL23

(pg/

mL)

IL6

(pg/

mL)

1,000

1,500

2,000

Blood

Intestine Intestine

Intestine Intestine

Intestine Intestine

Blood

0.0

0.5

1.0

1.5

2.0

10

20

30

40

50

0

5

10

HO

MA

-IR

15

20

LF HF+/+ Min/+

LF HFLF HF

+/+ Min/+LF HF

LF HF+/+ Min/+

LF HFLF HF+/+ Min/+

LF HF

LF HF+/+ Min/+

LF HF LF HF+/+ Min/+

LF HF

LF HF+/+ Min/+


LF HF

LF HF+/+ Min/+


LF HF

LF HF+/+ Min/+


LF HF

16

18

Fina

l bod

y w

eigh

t (g)

Tota

l pol

yp n

umbe

r

Tota

l pol

yp m

ass

(mm

2 )

20

22A B

C D

E F

G H

I J

K L

Figure 4.

Inflammation and intestinal neoplasia are triggered soon after exposure toHFD- the 3 days study. B6.Apcþ/þ and B6.ApcMin/þ mice were fed theLFDCoco or HFDCoco for 3 days. Final body weight (A) and HOMA-IR (B) weremeasured at day 3 of treatment. Intestinal polyp number (C) and mass(D) were analyzed. Circulating levels of C5a (E) and adiponectin (F) weremeasured in plasma. Levels of anti- and proinflammatory mediators implicatedin inflammation and cancer: C5a (G), IL6 (H), VEGF (I), IL23 (J), TNFa(K), and IL10 (L) were measured in the intestine of B6.Apcþ/þ or B6.ApcMin/þ atday 3 of treatment. Values represent means� SEM (A–F, n¼ 6–9 males/group;G–L, n ¼ 3–5). � , P < 0.05; �� , P < 0.01; �� , P < 0.001 in relation to thecorresponding LFD-fed group.

Doerner et al.





Figure 5.

Source of dietary fat determines localproinflammatory environment and intestinalpolyposis. B6.Apcþ/þ and B6.ApcMin/þmice werefed diets supplemented with coconut (purple),corn (green), or olive (blue) oil for 30days. Levelsof anti- and proinflammatory mediatorsimplicated in inflammation and cancer such asVEGF (A), TNFa (B), IL23 (C), IL1b (D), and IL10(E)weremeasured in intestinal tissue at day30oftreatment. Values represent means � SEM(n¼ 3–5 males/group). � , P < 0.05; �� , P < 0.01 inrelation to the same strain fed the LFDCoco. #,P <0.05;##,P<0.01 in relation to the same strainfed the corresponding LFDCoco or HFDCoco.






clinical studies to evaluate obesity-related diseases in humans,mouse strains are essential for studying thephysiologyofDIOandmetabolic syndrome (47, 48). CSSs in particular are a valuableparadigm for disentangling the complex relations among diet,obesity, and metabolic status (Fig. 1C and D; refs. 16, 17). TheseCSSs together with cancer promoting genetic models such asApcMin/þ (49) are a genetically defined, reproducible experimentalmodel system that can be used to exploremechanisms underlyinginteractions among diet, metabolism, and inflammation ontumorigenesis.

Extensive epidemiological research has recently establishedan association between obesity and increased risk of cancer

(40, 50, 51). Obesity is considered a leading environmental riskfor colorectal cancer and obesity-induced inflammation hasbeen correlated to various stages of cancer progression, includ-ing cellular transformation, promotion, survival, invasion,angiogenesis and metastasis (51). Obesity results in the accu-mulation of large adipocytes that upon necrosis recruit macro-phages that secrete proinflammatory cytokines and are impli-cated in the process of insulin resistance. In addition to theinfluence of diet on tumorigenesis in the GI tract, HFDspromote tumors with earlier appearance, greater frequency,accelerated growth, larger size, and in some cases more frequentmetastasis (50). For example, HFD has been shown to lead to

Figure 6.

Complement C5a mediates diet-induced inflammation and intestinal neoplasia. B6.Apcþ/þ and B6.ApcMin/þ mice were fed the LFD or HFD with coconut (purple),corn (green), or olive oil (blue) as a source of fat for 30 days. Circulating levels of the complement activation fragment C5a were measured in the plasmaat day 30 of treatment (A). B–F, B6.ApcMin/þ (þ/þ; Min/þ) and the complement-deficient C5aRþ/�;ApcMin/þ (C5aRþ/�; Min/þ), and C5aR�/�; ApcMin/þ (C5aR�/�;Min/þ) strains were fed HFDCoco or LFDCoco for 30 days. The total number of polyps (B) and polyp mass (C) were measured in the intestine of each mouseat day 30 of treatment (n¼ 5–20/group). Intestinal inflammationwasmeasured inmice genetically deficient in C5aR (D–F). Protein levels of VEGF (D), TNFa (E), andIL23 (F) were measured in the intestine (n ¼ 3–5/group) B6.Apcþ/þ and B6.ApcMin/þ mice were fed HFDCoco for 30 days, and concomitantly treated withthe C5aR antagonist peptide (PMX53) or a scrambled inactive peptide (control; 2 mg/kg s.c.) every other day (B and C). D and E, intestinal inflammation wasmeasured in males treated with PMX53 or control inhibitor (n ¼ 3–5). Total polyp number (B) and polyp mass (C) were measured. Protein levels of VEGF(G), TNFa (H), and IL23 (I) and expression of IL1b (J) andMyc (K)weremeasured in the intestine ofmales treatedwith PMX53or the control inhibitor (n¼ 3–6/group).L, the percentage of viable CD11bþ/GR-1þ neutrophils was measured with flow cytometry in the intestinal lamina propria of B6.Apcþ/þ andB6.ApcMin/þ malestreated with PMX53 or the control peptide (n ¼ 4/group). Levels of phosphorylated AKT (pS473), IKK (pS176), and NFkB (pS536) were assessed with Westernblotting (M) in polyp tissue of B6.ApcMin/þ males or normal intestinal tissue of B6.Apcþ/þ males treated with PMX53 or the control peptide. Levels of AKTand NFkBweremeasured as a control for the protein amount loaded. Results are representative of 3 independent experiments. � , P < 0.05; �� , P < 0.01; �� , P < 0.001compared with same strain fed corresponding LFD. #, P < 0.05; ##, P < 0.01; ###, P < 0.001 compared with B6.Apcþ/þ or B6.ApcMin/þ fed the same diet.$, P < 0.05; $$, P < 0.01; $$$, P < 0.001 compared with same strain treated with the control inhibitor.

Doerner et al.





hepatocellular cancer in C57BL/6J mice (40) and feeding aWestern type diet, enriched in fat and cholesterol, acceleratestumor progression in mice with an autochthonous model ofprostate cancer (52). Further demonstration of diet on tumor-igenesis, outside the intestine, is provided by demonstrationthat a high fat, lard-based diet, fed to obesity-resistant BALB/cmice promoted faster tumor growth and increased metastasis inmice orthotopically inoculated with syngeneic mammary car-cinoma cells (53).

In this study, specific HFDs induced upregulation of the proin-flammatory cytokines MCP-1 IL1b, IL6, IL23, TNFa, and VEGF inadipose tissue, intestine and blood, which correlated withincreased tumorigenesis (Figs. 3–6). The tumor microenviron-ment favors a collaboration betweenmalignant and immune cellsas well as proinflammatory mediators secreted by tumor-infil-trating lymphocytes that promote cell proliferation and angio-genesis (54, 55). Increased levels of proinflammatory cytokinesand angiogenic factors are found in serum and adipose tissue ofcolorectal cancer patients and are correlated with poor survival(54). Interestingly, leanmice fedHFDpresented increased inflam-mation and tumorigenesis, indicating that the HFD itself and notobesity or the related metabolic status determines the localenvironment that favors tumor growth. It has recently been shownthat HFD promotes tumorigenesis in the small intestine of genet-ically susceptible K-rasG12Dint mice, independent of obesity, as aconsequence of a shift in gut microbiota composition and adecrease in Paneth-cell–mediated antimicrobial host defense(10). Dietary interventions can significantly influence gut micro-flora, thereby affecting disease outcomes. In this study, HFDscomposed of specific dietary fats induced complement activationwith consequent release of the proinflammatory C5a fragment assoon as after 3 days of diet exposure, suggesting that complementactivation may occur prior to the microbiota shift and mayinfluence the composition of intestinal bacteria given the anti-microbial properties of the complement component system

(30, 33). Indeed, inhibition of C5aR-mediated signaling has beenpreviously shown to alter the composition and diversity of skinmicrobiota along with the proinflammatory profile (56).

Adipose tissue is a relevant source of complement proteins,including C3, Factor B, and properdin, that are required foractivation of the alternative pathway. Local complement activa-tion and consequent production of C5a regulate infiltration ofproinflammatory macrophages into adipose tissue and promoteinsulin resistance (57). This study showed that specific HFDsinduced complement activation, adding to the current under-standing on the local role of complement in adipose tissue. To ourknowledge, only one in vitrobut no in vivo study has demonstratedcomplement activation by a specific dietary component (gluten;ref. 58). Clearly, further investigation is needed to dissect theprecise fatty acids responsible for complement activation. Nev-ertheless, such observations open new avenues for research ondiet-induced inflammatory diseases and places complement as akey target for therapeutic intervention in such conditions.

Apart from complement, fatty acids are involved in immuno-logical processes via direct activation of immune cells and releaseof proinflammatory mediators (59). Whereas saturated fatty acidscontribute to promoting inflammation, unsaturated fatty acidsshow both pro- and antiinflammatory properties and the balancebetween omega-3 and omega-6 is considered critical for mainte-nance of healthy levels of fat (12). Reduced levels of complementactivation, inflammatory cytokines, and tumorigenesis wereobserved in mice fed HFDOlive despite similar levels of obesitycomparedwith those fedHFDCoco (Figs. 2, 5, 6A). Olive oil, whichis rich in theunsaturatedoleic acid, has antioxidant properties and,together with other components such as phenolic compounds,induces antiproliferative and antiinflammatory responses (20).

Together, we found that certain HFDs activate complement,regardless of DIO, and generate C5a, which in turn promotes aproinflammatory environment by triggering signaling pathwaysthat control expression of proto-oncogenes and release of

Figure 7.

Molecularmechanisms determining HFD-induced intestinal polyp progression. Diets high in specific dietary fats induce an inflammatory environment in the intestinethat promotes intestinal tumorigenesis in genetically susceptible APCmin/þmice, independent of obesity and related metabolic conditions. HFD containingcoconut or corn oil as a source of saturated fat induce complement activation with subsequent release of C5a. C5a in turn induces the infiltration ofinflammatory neutrophils to the intestinal lamina propria and activate the AKT/IKK/NFkB signaling pathway in intestinal cells with consequent production ofproinflammatory mediators such as IL1b, IL6, TNFa, and the proto-oncogene c-myc. Conversely, mice fed a HFD containing olive oil as a source of fatgained significant body weight, but did not promote a proinflammatory environment or development of polyps in the intestine.






inflammatory mediators, consequently favoring formation ofintestinal polyps in genetically susceptible mice (Fig. 7). Impor-tantly, three observations were made that indicate dissociationbetween DIO and high dietary fat in tumor development:increased intestinal tumorigenesis in both DIO-susceptible andDIO-resistant CSS mice (Fig. 1C–F), increased polyp numbers inleanmice after only 3 days of HFD exposure (Fig. 4A–D), and lackof tumor progression in obeseHFDOlive-fedmice (Fig. 2). As such,these results show that HFD-induced complement activation andC5a generation promote inflammation and intestinal tumorigen-esis prior to the onset of obesity. Given that HFD-induced com-plement activation is an upstream event and is crucial for theestablishment of inflammation, and given the beneficial effects ofantiinflammatory therapies in cancer (7), these data point tocomplement-targeted therapeutics (60) as a potentially powerfulmeans of preventing diet-induced neoplasia.

Disclosure of Potential Conflicts of InterestS.K. Doerner, N.A. Berger, J.D. Lambris, and J.H. Nadeau are co-inventors

in a patent application titled "A method to treat polyp formation."No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: S.K. Doerner, N.A. Berger, J.D. Lambris, J.H. NadeauDevelopment of methodology: S.K. Doerner, N.A. Berger, J.H. Nadeau

Acquisition of data (provided animals, acquired and managed pati-ents, provided facilities, etc.): S.K. Doerner, E.S. Leung, J.S. Ko, J.D. Heaney,N.A. Berger, J.H. NadeauAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S.K. Doerner, E.S. Reis, J.D. Heaney, N.A. Berger,J.H. NadeauWriting, review, and/or revision of the manuscript: S.K. Doerner, E.S. Reis,J.D. Heaney, N.A. Berger, J.D. Lambris, J.H. NadeauAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S.K. Doerner, N.A. Berger, J.H. NadeauStudy supervision: N.A. Berger, J.D. Lambris, J.H. NadeauOther (data collection, performed experiments): E.S. Leung, J.S. Ko

AcknowledgmentsWe thank David A. DeSantis and Carmen Fiuza-Luces for technical

assistance. This work was supported by the Transdisciplinary Research inEnergy Balance and Cancer Grant #U54 CA116867 to N.A. Berger andJ.H. Nadeau, AI030040 and AI068730 to J.D. Lambris and P40 RR012305to J.H. Nadeau and N.A. Berger.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received May 2, 2016; revised July 21, 2016; accepted July 24, 2016;published OnlineFirst August 17, 2016.

References1. Renehan AG, Zwahlen M, Egger M. Adiposity and cancer risk: new mech-

anistic insights from epidemiology. Nat Rev Cancer 2015;15:484–98.2. Iyengar NM, Hudis CA, Dannenberg AJ. Obesity and cancer: local and

systemic mechanisms. Annu Rev Med 2015;66:297–309.3. Despres JP, Lemieux I. Abdominal obesity and metabolic syndrome.

Nature 2006;444:881–7.4. Gregor MF, Hotamisligil GS. Inflammatory mechanisms in obesity. Annu

Rev Immunol 2011;29:415–45.5. Mentor-Marcel RA, Bobe G, Barrett KG, YoungMR, Albert PS, BenninkMR,

et al. Inflammation-associated serum and colon markers as indicators ofdietary attenuation of colon carcinogenesis in ob/obmice. Cancer Prev Res(Phila) 2009;2:60–9.

6. Kubota M, Shimizu M, Sakai H, Yasuda Y, Ohno T, Kochi T, et al. Renin-angiotensin system inhibitors suppress azoxymethane-induced colonicpreneoplastic lesions in C57BL/KsJ-db/db obese mice. Biochem BiophysRes Commun 2011;410:108–13.

7. Ulrich CM, Bigler J, Potter JD. Non-steroidal anti-inflammatory drugs forcancer prevention: promise, perils and pharmacogenetics. Nat Rev Cancer2006;6:130–40.

8. Newmark HL, Yang K, Kurihara N, Fan K, Augenlicht LH, Lipkin M.Western-style diet-induced colonic tumors and their modulation bycalcium and vitamin D in C57Bl/6 mice: a preclinical model for humansporadic colon cancer. Carcinogenesis 2009;30:88–92.

9. Jiang Y, Pan Y, Rhea PR, Tan L, GageaM, Cohen L, et al. A sucrose-enricheddiet promotes tumorigenesis in mammary gland in part through the 12-lipoxygenase pathway. Cancer Res 2016;76:24–9.

10. SchulzMD,AtayC,Heringer J, Romrig FK, Schwitalla S, AydinB, et al.High-fat-diet-mediated dysbiosis promotes intestinal carcinogenesis indepen-dently of obesity. Nature 2014;514:508–12.

11. Couto E, Boffetta P, Lagiou P, Ferrari P, Buckland G, Overvad K, et al.Mediterranean dietary pattern and cancer risk in the EPIC cohort. Br JCancer 2011;104:1493–9.

12. Veldhoen M, Brucklacher-Waldert V. Dietary influences on intestinalimmunity. Nat Rev Immunol 2012;12:696–708.

13. Song M, Garrett WS, Chan AT. Nutrients, foods, and colorectal cancerprevention. Gastroenterology 2015;148:1244–60e16.

14. Su LK, Kinzler KW, Vogelstein B, Preisinger AC, Moser AR, Luongo C, et al.Multiple intestinal neoplasia caused by amutation in themurine homologof the APC gene. Science 1992;256:668–70.

15. Joslyn G, CarlsonM, Thliveris A, Albertsen H, Gelbert L, Samowitz W, et al.Identification of deletion mutations and three new genes at the familialpolyposis locus. Cell 1991;66:601–13.

16. Singer JB, Hill AE, Burrage LC, Olszens KR, Song J, Justice M, et al. Geneticdissection of complex traits with chromosome substitution strains ofmice.Science 2004;304:445–8.

17. Burrage LC, Baskin-Hill AE, Sinasac DS, Singer JB, Croniger CM, Kirby A,et al. Genetic resistance to diet-induced obesity in chromosome substitu-tion strains of mice. Mamm Genome 2010;21:115–29.

18. Wessels MR, Butko P, Ma M, Warren HB, Lage AL, Carroll MC. Studies ofgroup B streptococcal infection in mice deficient in complement compo-nentC3 orC4demonstrate an essential role for complement in both innateand acquired immunity. Proc Natl Acad Sci U S A 1995;92:11490–4.

19. Hopken UE, Lu B, Gerard NP, Gerard C. The C5a chemoattractant receptormediates mucosal defence to infection. Nature 1996;383:86–9.

20. Lucas L, Russell A, Keast R. Molecular mechanisms of inflammation. Anti-inflammatory benefits of virgin olive oil and the phenolic compoundoleocanthal. Curr Pharm Des 2011;17:754–68.

21. Berglund ED, Li CY, Ayala JE,McGuinnessOP,WassermanDH. Regulationof endogenous glucose production in glucose transporter 4 over-expressingmice. PLoS One 2012;7:e52355.

22. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC.Homeostasis model assessment: insulin resistance and beta-cell functionfrom fasting plasma glucose and insulin concentrations in man. Diabe-tologia 1985;28:412–9.

23. Paczkowski NJ, Finch AM, Whitmore JB, Short AJ, Wong AK, Monk PN,et al. Pharmacological characterization of antagonists of the C5a receptor.Br J Pharmacol 1999;128:1461–6.

24. Puppa MJ, White JP, Sato S, Cairns M, Baynes JW, Carson JA. Gut barrierdysfunction in the Apc(Min/þ) mouse model of colon cancer cachexia.Biochim Biophys Acta 2011;1812:1601–6.

25. Corrales L, Ajona D, Rafail S, Lasarte JJ, Riezu-Boj JI, Lambris JD, et al.Anaphylatoxin C5a creates a favorable microenvironment for lung cancerprogression. J Immunol 2012;189:4674–83.

26. Cho MS, Vasquez HG, Rupaimoole R, Pradeep S, Wu S, Zand B, et al.Autocrine effects of tumor-derived complement. Cell Rep2014;6:1085–95.

27. Vadrevu SK, Chintala NK, Sharma SK, Sharma P, Cleveland C, Riediger L,et al. Complement c5a receptor facilitates cancer metastasis by altering T-cell responses in the metastatic niche. Cancer Res 2014;74:3454–65.

Doerner et al.





28. Markiewski MM, DeAngelis RA, Benencia F, Ricklin-Lichtsteiner SK, Kou-toulaki A, Gerard C, et al. Modulation of the antitumor immune responseby complement. Nat Immunol 2008;9:1225–35.

29. Piao C, Cai L, Qiu S, Jia L, SongW, Du J. Complement 5a enhances hepaticmetastases of colon cancer via monocyte chemoattractant protein-1-medi-ated inflammatory cell infiltration. J Biol Chem 2015;290:10667–76.

30. Klos A,Wende E,WarehamKJ,Monk PN. International Union of Basic andClinical Pharmacology. [corrected]. LXXXVII. Complement peptide C5a,C4a, and C3a receptors. Pharmacol Rev 2013;65:500–43.

31. Ueki T, Mizuno M, Uesu T, Kiso T, Nasu J, Inaba T, et al. Distribution ofactivated complement, C3b, and its degraded fragments, iC3b/C3dg, in thecolonic mucosa of ulcerative colitis (UC). Clin Exp Immunol 1996;104:286–92.

32. Sugihara T, Kobori A, ImaedaH, Tsujikawa T, Amagase K, Takeuchi K, et al.The increased mucosal mRNA expressions of complement C3 and inter-leukin-17 in inflammatory bowel disease. Clin Exp Immunol 2010;160:386–93.

33. Ricklin D, Hajishengallis G, Yang K, Lambris JD. Complement: a keysystem for immune surveillance and homeostasis. Nat Immunol 2010;11:785–97.

34. DiDonato JA, Mercurio F, Karin M. NF-kappaB and the link betweeninflammation and cancer. Immunol Rev 2012;246:379–400.

35. Talukdar S, Oh da Y, Bandyopadhyay G, Li D, Xu J, McNelis J, et al.Neutrophils mediate insulin resistance in mice fed a high-fat diet throughsecreted elastase. Nat Med 2012;18:1407–12.

36. Downs-Canner S,MaggeD, Ravindranathan R, O'MalleyME, Francis L, LiuZ, et al. Complement inhibition: a novel formof immunotherapy for coloncancer. Ann Surg Oncol 2016;23:655–62.

37. Nunez-Cruz S, Gimotty PA, Guerra MW, Connolly DC, Wu YQ, DeAngelisRA, et al. Genetic and pharmacologic inhibition of complement impairsendothelial cell function and ablates ovarian cancer neovascularization.Neoplasia 2012;14:994–1004.

38. Ning C, Li YY, Wang Y, Han GC, Wang RX, Xiao H, et al. Complementactivation promotes colitis-associated carcinogenesis through activatingintestinal IL-1beta/IL-17A axis. Mucosal Immunol 2015;8:1275–84.

39. Bonavita E, Gentile S, Rubino M, Maina V, Papait R, Kunderfranco P, et al.PTX3 is an extrinsic oncosuppressor regulating complement-dependentinflammation in cancer. Cell 2015;160:700–14.

40. Hill-Baskin AE, Markiewski MM, Buchner DA, Shao H, DeSantis D, HsiaoG, et al. Diet-induced hepatocellular carcinoma in genetically predisposedmice. Hum Mol Genet 2009;18:2975–88.

41. Guglietta S, Chiavelli A, Zagato E, Krieg C, Gandini S, Ravenda PS, et al.Coagulation induced by C3aR-dependent NETosis drives protumorigenicneutrophils during small intestinal tumorigenesis. Nat Commun2016;7:11037.

42. Nabizadeh JA, Manthey HD, Steyn FJ, Chen W, Widiapradja A, Md AkhirFN, et al. The complement C3a receptor contributes to melanoma tumor-igenesis by inhibiting neutrophil and CD4þ T cell responses. J Immunol2016;196:4783–92.

43. Wang Y, Sun SN, LiuQ, Yu YY, Guo J,Wang K, et al. Autocrine complementinhibits IL10-dependent T-cell mediated antitumor immunity to promotetumor progression. Cancer Discov 2016 Jun 13. [Epub ahead of print].

44. Levi JS, L.M, Laurent RS, Lang A, Rayburn J. F as in Fat: how obesitythreatens America's future 2012. Trust for America's Health Reporthttp://healthyamericans.org/report/100/.2012.

45. van der Klaauw AA, Farooqi IS. The hunger genes: pathways to obesity. Cell2015;161:119–32.

46. ZhangC, YinA, LiH,WangR,WuG, Shen J, et al.Dietarymodulationof gutmicrobiota contributes to alleviation of both genetic and simple obesity inchildren. EBioMedicine 2015;2:966–82.

47. Sinasac DS, Riordan JD, Spiezio SH, Yandell BS, Croniger CM, Nadeau JH.Genetic control of obesity, glucose homeostasis, dyslipidemia and fattyliver in a mouse model of diet-induced metabolic syndrome. Int J Obes(Lond) 2016;40:346–55.

48. Spiezio SH, Amon LM, McMillen TS, Vick CM, Houston BA, Caldwell M,et al. Genetic determinants of atherosclerosis, obesity, and energy balancein consomic mice. Mamm Genome 2014;25:549–63.

49. Wang L, Zhang Q. Application of the Apc(Min/þ) mouse model forstudying inflammation-associated intestinal tumor. Biomed Pharmac-other 2015;71:216–21.

50. Berger NA. Obesity and cancer pathogenesis. Ann N Y Acad Sci 2014;1311:57–76.

51. Guffey CR, Fan D, Singh UP, Murphy EA. Linking obesity to colorectalcancer: recent insights into plausible biological mechanisms. Curr OpinClin Nutr Metab Care 2013;16:595–600.

52. Llaverias G, Danilo C,Wang Y,Witkiewicz AK, Daumer K, Lisanti MP, et al.A Western-type diet accelerates tumor progression in an autochthonousmouse model of prostate cancer. Am J Pathol 2010;177:3180–91.

53. KimEJ, ChoiMR, ParkH, KimM,Hong JE, Lee JY, et al. Dietary fat increasessolid tumor growth and metastasis of 4T1 murine mammary carcinomacells and mortality in obesity-resistant BALB/c mice. Breast Cancer Res2011;13:R78.

54. Landskron G, De la Fuente M, Thuwajit P, Thuwajit C, Hermoso MA.Chronic inflammation and cytokines in the tumor microenvironment.J Immunol Res 2014;2014:149185.

55. Maywald RL, Doerner SK, Pastorelli L, De Salvo C, Benton SM, Dawson EP,et al. IL-33 activates tumor stroma to promote intestinal polyposis. ProcNatl Acad Sci U S A 2015;112:E2487–96.

56. Chehoud C, Rafail S, Tyldsley AS, Seykora JT, Lambris JD, Grice EA.Complement modulates the cutaneous microbiome and inflammatorymilieu. Proc Natl Acad Sci U S A 2013;110:15061–6.

57. Phieler J, Chung KJ, Chatzigeorgiou A, Klotzsche-von Ameln A, Garcia-Martin R, Sprott D, et al. The complement anaphylatoxin C5a receptorcontributes to obese adipose tissue inflammation and insulin resistance.J Immunol 2013;191:4367–74.

58. UnsworthDJ,Wurzner R, BrownDL, LachmannPJ. Extracts ofwheat glutenactivate complement via the alternative pathway. Clin Exp Immunol1993;94:539–43.

59. Doerner SK, Berger NA. Dietary fats as mediators of obesity, inflammationand cancer. In Dannenberg AJ, Berger NA,editors. Obesity, inflammationand cancer. New York: Springer;2013. p.99–132.

60. Reis ES, Mastellos DC, Yancopoulou D, Risitano AM, Ricklin D, LambrisJD. Applying complement therapeutics to rare diseases. Clin Immunol2015;161:225–40.






2016;14:953-965. Published OnlineFirst August 17, 2016.Mol Cancer Res Stephanie K. Doerner, Edimara S. Reis, Elaine S. Leung, et al. Inflammation and Neoplasia, Independent of ObesityHigh-Fat Diet-Induced Complement Activation Mediates Intestinal

Updated version

10.1158/1541-7786.MCR-16-0153doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2016/10/06/1541-7786.MCR-16-0153.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/14/10/953.full#ref-list-1

This article cites 57 articles, 13 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/14/10/953.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/14/10/953To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-16-0153

http://mcr.aacrjournals.org/content/suppl/2016/10/06/1541-7786.MCR-16-0153.DC1

http://mcr.aacrjournals.org/content/14/10/953.full#ref-list-1

http://mcr.aacrjournals.org/content/14/10/953.full#related-urls

http://mcr.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mcr.aacrjournals.org/content/14/10/953


high-fat diet-induced complement activation mediates...

Documents