Research ArticleSee related commentary by Ignatenko and Gerner, p. 161
Abdominal Obesity, Independent from Caloric Intake,Accounts for the Development of Intestinal Tumors inApc1638N/þ Female Mice
Derek M. Huffman1,4, Leonard H. Augenlicht1,3, Xueying Zhang1,5, John J. Lofrese1,4, Gil Atzmon1,2,4,John P. Chamberland6,7, and Christos S. Mantzoros6,7
AbstractTo determine whether visceral fat (VF), independent of other confounders, is causally linked to
intestinal tumorigenesis, we surgically removed visceral fat in Apc1638/Nþ mice. At 15 weeks of age,
male and female Apc1638/Nþmice were randomized to one of three groups: ad libitum, visceral fat removal
(VF-) and ad libitum fed, or caloric restriction, and were studied for effects on tumorigenesis and survival.
As compared with ad libitum, VF� and caloric restriction reducedmacroadenomas to a similar extent (P <0.05), but only caloric restriction significantly improved survival (P < 0.05). Given that a significant group
� gender interaction was observed, we next examined males and females separately. In females,
macroadenomas were markedly attenuated by VF� (1.33 � 0.23 mean � SE; P < 0.05), but not by
caloric restriction (2.35� 0.25; P¼ 0.71), as compared with ad libitum (2.50� 0.34). In males, however,
caloric restriction (1.71� 0.26; P < 0.01), but not VF� (2.94� 0.42; P¼ 0.29), reduced macroadenomas,
as compared with ad libitummales (3.47� 0.30). In females, both VF� (P¼ 0.05) and caloric restriction
(P < 0.01) improved survival, but not in male mice (P ¼ 0.15). The benefits observed with caloric
restriction were consistent with favorable metabolic adaptations, but protection conferred in VF� females
was despite lower adiponectin levels (P < 0.05), and failure to reduce body mass, total adiposity, glucose,
insulin, leptin, and chemokine (C–X–C motif) ligand 1 (CXCL-1) levels. In conclusion, these data
provide the first causal evidence linking visceral fat to intestinal cancer risk, and suggest that factors, other
than known metabolic mediators, may impact tumor development. Furthermore, these data emphasize
that strategies designed to deplete visceral fat stores in humans should be considered in the prevention of
intestinal cancer. Cancer Prev Res; 6(3); 177–87. �2012 AACR.
IntroductionObesity rates are at epidemic levels in the United States
and other developed countries (1, 2). Obese individuals areat increased risk for developingmetabolic syndrome, type IIdiabetes, cardiovascular disease (3), and cancer at manysites, including colon, kidney, thyroid, endometrium, liver,and esophagus (4–10). Excess adiposity poses an evengreater risk for cancer mortality, an observation that is
nearly universal among the most common forms of cancer(9). For example, obesity dramatically increases the riskof death from colon cancer by 46% in women and 84%in men, breast cancer by 2-fold, liver and kidney cancerby more than 4-fold, and uterine cancer by more than 6-fold (9).
Since the obesity–cancer link first emerged, studies acrossdisciplines have worked toward establishing the biologicunderpinnings explaining this association. Epidemiologicstudies (4, 5, 9) strongly suggest a relationship via endocrineand other metabolic effects (11, 12), whereas in vitro (13)and in vivo studies (14, 15) have provided evidence withrespect to possible mechanisms. The most commonly stud-ied pathways linking obesity to cancer include the ability ofexcess adipose tissue to promote insulin resistance andcontribute to a chronic, low-grade, proinflammatory statethrough the secretion of adipokines (16). However, estab-lishing the independent contribution of these mediators tosite-specific cancer risk has been difficult, as their impor-tance seems to vary by cancer type and stage and is furthercomplicated by interactions with other hormones (13,17–19). A second unresolved issue is the lack of in vivoevidence causally linking adipose tissue per se to cancer. The
Authors' Affiliations: Departments of 1Medicine, 2Genetics, and 3CellBiology; 4Institute for Aging Research, Albert Einstein College of Medicine,Bronx, New York; 5Institute of Zoology, Chinese Academy of Sciences,Chaoyang, Beijing, China; 6Section of Endocrinology, Department of Inter-nal Medicine, VA Boston Healthcare System; and 7Division of Endocrinol-ogy, Diabetes and Metabolism, Department of Internal Medicine, BethIsrael Deaconess Medical Center, Harvard Medical School, Boston,Massachusetts
Corresponding Author: Derek M. Huffman, Department of Medicine,Division of Endocrinology, Institute for Aging Research, Albert EinsteinCollege of Medicine, 1300 Morris Park Avenue, Golding Building Room502, Bronx, NY 10461. Phone: 718-430-4278; Fax: 718-430-8557; E-mail:[email protected]
predominant method used in obesity–cancer studies useshigh-fat feeding in rodents to cause excess gains in weightand adiposity, but this strategy introduces other confoun-ders, including changes in dietary components, nutrientflux, and energy balance. Furthermore, it is known thatnutrients can interact with adipose tissue to provoke theexpression and secretion of proinflammatory cytokines(20), thus interventions that limit this influx of nutrients,such as caloric restriction or bariatric surgery, for example,may work in part by attenuating this interaction.
Since the relationship between obesity and disease firstemerged, a more careful examination of body fat distribu-tionhas revealed that the risk posed by obesity inhumans todisease andmortality is primarily harbored by the extent ofvisceral fat (VF) accretion (21). Using a surgical model ofvisceral fat removal (VF�) in rats, our group first showedthat this relationship in regards to insulin resistance, type IIdiabetes, and lifespan is causal (22–24). Given that abdom-inal obesity has been shown tomore strongly predict cancerrisk and mortality, including colon cancer (25, 26), thengeneral obesity, we addressed whether visceral fat is a directmodulator of intestinal tumor development.We specificallyused our surgical approach of depleting visceral fat in amouse model of intestinal cancer (Apc1638/Nþ), to distin-guish the contribution of visceral adiposity to tumorigen-esis, independent of other important confounders.Here, weshow that VF� independently protects against the devel-opment of macroadenomas in female Apc1638/Nþ mice.
Materials and MethodsAnimals
Male Apc1638/Nþ mice on a C57BL/6 background werebred with C57BL6/J female mice (Jackson Laboratory).Genotyping of Apc1638/Nþ offspring was conducted asdescribed (27). At weaning (3-weeks old), mice were grouphoused in same-sex cages and fed a purified 45% high-fatdiet (cat#D12451, Research Diets Inc.) to induce weightgain. Animals were housed at standard temperature(�22�C) and humidity-controlled conditions under a stan-dard light/dark photoperiod (14L:10D). All experimentswere approved by the Institutional Animal Care and UseCommittee at the Albert Einstein College of Medicine(Bronx, NY).
Experimental designAt 15weeks of age,whichwe foundwas an age that fat can
be ablated without significant regrowth, approximately twothirds of mice were randomly assigned to sham abdominalsurgery and approximately one third to VF�. Following a 1-week recovery, sham operated and ad libitum-fed animalswere further randomized to either ad libitum or caloric-restricted groups, resulting in 3 experimental groups (adlibitum;n¼45 total,males¼21, females¼24), VF� and adlibitum fed (VF�; n¼ 40 total, males¼ 21, females¼ 19), orad libitumand40%caloric restriction (n¼38 total,males¼18, females¼ 20).Micewere thenmonitored daily for up to12 weeks (28 weeks of age) for tumor development. Ani-mals were removed early from the study and sacrificed if
evidence of decreased food intake and weight loss (>20%),coupled with signs of sickness and lethargy were observed.In all cases, mice removed early from the experiment due tothese criteria were found to harbor intestinal tumors.
Sham and fat removal surgeryFor each surgery, mice were anesthetized with isoflurane
(2% induction and maintenance) and a small midlineincision was made in the lower abdomen under asepticconditions, as previously described (24). For VF�mice, allvisible gonadal (epididymal in males, periovarian infemales) and perinephric adipose tissue was carefullyremoved by blunt dissection, weighed, and recorded. Forthe sham operation, a similar incision was made and softtissueswere disturbed in a similarmanner as VF�, but no fatwas removed. Analgesic was given (buprenorphine, 0.1mg/kg s.c.) perioperatively and as needed for up to 72 hourspostoperatively. Recovery was more than 90% for both thesham and fat removal procedures.
Food intake, body weight, and body compositionFollowing surgery, mice were singly housed for the
duration of the experiment. Body weight and food intakewas monitored weekly. Animals assigned to the caloric-restricted group were provided a weighed food pellet dailybetween 12:00 and 17:00 hours. Because ad libitummale mice consumed slightly more food than ad libitumfemale mice, the amount of food provided to male andfemale caloric-restricted mice, respectively, was calculatedas approximately 60% of the intake consumed by same-sexad libitum controls. Body composition was assessed byquantitative magnetic resonance (qMR) for determinationof fat and leanmass (EchoMedical Systems) at baseline (�1week), 4, 8, and 12 weeks in study. At sacrifice, visceral fatpads (perinephric, mesenteric, and epididymal/periovar-ian) were removed, weighed, and recorded.
Necropsy and tissue processingFor determinationof tumors,micewere sacrificed and the
gastrointestinal tract was separated from the mesenteric fatdepot and divided into 4 segments: duodenum, ileum,jejunum, and colon. Each segment was opened longitudi-nally, rinsed in PBS, and laid out flat for examination oftumor multiplicity with the aid of a dissecting magnifyinglens. Macroadenomas (�>0.5-mm diameter) were countedin the proximal and distal region of each segment ofintestinal tissue and recorded. Tissues were then rolled andfixed overnight in 10% neutral-buffered formalin at 4�C,processed through a series of alcohols and xylenes, andembedded in paraffin. Sections (5 mm) were stained withhematoxylin and eosin and assessed by a pathologist forhistologic changes following consensus recommendationsfor assessing intestinal tumors in rodents (28).
Four-week cohort for insulin-tolerance tests andserumcollection
Tominimize the confounding effects of cancermorbidityon metabolic outcomes, a separate cohort of mice were
Huffman et al.
Cancer Prev Res; 6(3) March 2013 Cancer Prevention Research178
included and used for insulin-tolerance tests (ITT) at 3weeks in study and sacrificed for serum collection at 4weeksin study (i.e., 20 weeks of age). ITTs were conducted inrandom-fed mice, early in their light cycle (�07:00–08:00hours), as described (29). Briefly, following a baselineglucose measurement (One Touch Ultra, LifeScan, Inc.),mice were intraperitoneally injected with 0.75 U/kg insulinand blood glucose was measured at 15, 30, 45, and 60minutes later. For serum collection, food was removed atapproximately 06:00 hours and mice were killed 5- to 6hours later by decapitation without anesthesia. Blood wasallowed to clot at room temperature, and serum was sep-arated by centrifugation and stored at�80�C until analysis.
Serum measuresAt sacrifice, glucose was measured in whole blood with a
handheld glucose analyzer (One Touch Ultra). Serum insu-linwasmeasured in duplicate using a high-sensitivitymouseELISA (Alpco Inc.). 17b-Estradiol levels in serum were mea-sured in duplicate by enzyme-linked immunosorbent assay(EIA) in female mice (Cayman Inc.). Serum leptin andadiponectin were measured in duplicate by ELISA (AlpcoInc.) as described (15). Other cytokines and chemokines,including interleukin (IL)-1b, IL-10, IL-12-p70, IL-6, IFN-g ,and chemokine (C–X–C motif) ligand 1 (CXCL-1) weremeasured in duplicate using an electrochemiluminescenceassay with sensitivities of 0.75, 11.0, 35.0, 4.5, 0.38, and3.3 pg/mL, respectively (Meso Scale Discovery).
StatisticsTumor multiplicity was analyzed by two-way ANOVA
(group� gender) and post hoc comparisons were conductedwhen appropriate. A total of 5 statistical outliers (>2 SDfrom the group mean) were identified and excluded fromthe macroadenoma dataset (1 each from ad libitum male,VF�, caloric-restricted and ad libitum female and VF�groups, respectively). Histologic outcomes were analyzedby the Kruskal–Wallis procedure and post hoc comparisonsconducted with the Mann–Whitney U test when appropri-ate. Survival to 12 weeks in study was conducted using theKaplan–Meier method and significant differences in surviv-al distribution among groups was tested by a log-rank test.Longitudinal measures were assessed by repeated-measuresANOVA and cross-sectionalmeasures were assessed by one-way ANOVA. When significance was detected for the maineffect, planned contrastswere carried outwhen appropriate.All analyses were conducted using either SPSS (SPSS Inc.) orJMP software version 9 (SAS Institute Inc.). A P � 0.05 wasconsidered statistically significant.
ResultsCaloric restriction or visceral fat removal resulted in asimilar and significant reduction in macroadenomasAt sacrifice, tumormultiplicitywas determined inmice by
counting macroadenomas in the small and large intestine.Macroadenomas were primarily confined to the duodenumand jejunum, less frequently in the ileum, and only rarelyobserved in the colon. No differences in macroadenoma
number by intestinal region (i.e., proximal or distal duo-denum, ileum, jejunum, and colon) were detected amonggroups (data not shown), but a significant group effect wasobserved for totalmacroadenomas (P<0.01).Comparisonsamong groups revealed a similar and significant reductionin tumor multiplicity in VF� and caloric-restricted mice, ascompared with ad libitum mice (Fig. 1A; P < 0.05).
Using criteria described in Materials and Methods, wenext examined the effect of interventions on survival to 12weeks in study (28 weeks of age). Analysis of survivorshiprevealed that caloric restriction significantly improved sur-vival (89.5% remaining; P < 0.05), but no significantdifference was observed between VF� and ad libitum ani-mals (80.0% vs. 62.2% remaining, respectively; Fig. 1B; P¼0.10). In addition, a significant main effect for gender (P <0.05) and a group � gender interaction (P < 0.01) weredetected for tumor multiplicity, leading to a secondaryanalysis within female and male groups, respectively.
Visceral fat removal, but not caloric restriction,reduced the number of macroadenomas in Apc1638/Nþ
female miceIn females, macroadenoma numbers were significantly
reduced by VF� (P < 0.05), but not by caloric-restriction(P ¼ 0.71), as compared with ad libitum females (Fig. 1C).When intestinal tissue sections were evaluated histological-ly by a pathologist, no differences were observed in theseverity of crypt hyperplasia or the frequency of dysplasiaamong female groups, but a marked increase in microade-nomas was observed in VF� femalemice, as compared withad libitum females (Table 1; P < 0.05), whereas caloric-restricted female mice had an intermediary level, relative toad libitum and VF� females. Taken together, these findingssuggest that in VF� females, but not in caloric-restrictedfemales, the progression of microadenomas to macroade-nomas was significantly attenuated.
Caloric restriction and visceral fat removalsignificantly improved survival in Apc1638/Nþ femalemice
When a survival analysis, truncated at 12 weeks, wasconducted in female mice, the percentage of both caloric-restricted (100% remaining; P < 0.01) and VF� femalemice(94.7% remaining; P ¼ 0.05) surviving to 12 weeks, wassignificantly greater than ad libitum females (79.2%remaining; Fig. 1D).
Caloric restriction, but not visceral fat removal,reduced the number of macroadenomas in Apc1638/Nþ
male miceIn contrast to females, caloric-restricted male mice
(P < 0.01), but not VF�malemice (P¼ 0.29), hadmarkedlyfewer totalmacroadenomas than ad libitummales (Fig. 1E).In addition, caloric-restricted males had significantly fewermacroadenomas than VF�mice (P < 0.05). Histopathology
Visceral Fat and Colon Cancer
www.aacrjournals.org Cancer Prev Res; 6(3) March 2013 179
revealed that the severity of crypt hyperplasia was lowest inVF�male mice (Table 2; P < 0.05), whereas caloric-restrict-ed males had the lowest incidence of microadenomas, butthe greatest occurrence of crypt dysplasia (Table 2; P <0.05).Collectively, these findings suggest that the progression ofdysplastic cells toward adenoma formation was abrogatedonly in caloric-restrictedmalemice, leading to developmentof fewer micro- and macroadenomas in this group. Whenassessing survivorship in males, the percentage of ad libi-tum male mice still remaining at 12 weeks was 42.9%,whereas the percentage of VF� and caloric-restricted malemice surviving to 12 weeks were 66.7% and 77.8%, respec-
tively, but no significant model effect was observed (Fig. 1F;P ¼ 0.15).
Caloric restriction, but not visceral fat removal,significantly altered phenotypic characteristics inApc1638/Nþ female mice
On average, 730.4� 61.7 mg periovarian fat and 331.5�61.7 mg perinephric fat (1,061.9 mg total) was surgicallyremoved from VF� female mice. As expected, no significantdifferences were detected for body weight (Fig. 2A), foodintake (Fig. 2B), lean mass (Fig. 2C), or total adiposity(Fig. 2D) over the course of the study between ad libitum
Figure 1. Tumor multiplicity and survival to 12 weeks in ad libitum-fed (AL), VF�, and caloric-restricted (CR) mice. A, at necropsy, the number ofmacroadenomas in the intestinal tract were counted with the assistance of a dissecting magnifying lens. VF� and caloric-restriction interventions ledto a significant reduction inmacroadenomas, as comparedwith ad libitummice (P<0.05). B, survivorship to 12weekswasdeterminedusing theKaplan–Meiermethod. Events were defined as the early removal of an animal based on predefined criteria (see Materials and Methods). As compared with ad libitum mice(62.2% remaining), caloric restriction significantly improved survival (89.5% remaining;P < 0.05), whereas a tendency toward improved survival was observedwith VF� (80.0% remaining; P ¼ 0.10). C, in females, VF�, but not caloric restriction, led to a significant reduction in macroadenomas, as comparedwith ad libitum mice (P < 0.05). D, when assessing survivorship to 12 weeks in study for female mice, the percentage of caloric-restricted mice (100%remaining;P<0.01) andVF�mice (94.7% remaining;P¼0.05) surviving to 12weeks,was significantly greater than ad libitum females (79.2%remaining). E, inmales, caloric restriction significantly reduced macroadenomas, as compared with ad libitum (P < 0.01) and VF� male mice (P < 0.05). F, in males, 42.9%of ad libitummice survived to 12 weeks in study, as compared with 66.7% of VF�mice and 77.8% of caloric-restricted mice, but the overall model effect didnot reach significance (P ¼ 0.15). Bars are mean � SE. Different letters denote a significant difference between groups (P < 0.05).
Huffman et al.
Cancer Prev Res; 6(3) March 2013 Cancer Prevention Research180
and VF� females. In contrast, caloric-restricted females con-sumed nearly 40% fewer calories, weighed significantly less,and had reduced amounts of lean and fat mass, as comparedwithad libitumandVF�females (Fig.2A–D;P<0.001).At12weeks in study, individual fat pads were weighed, whichconfirmed both the successful surgical ablation of visceral fatpads in VF� females, and the significant depletion in thesesame depots by caloric restriction, as compared with adlibitum female mice (Fig. 2E; P < 0.05).
Visceral fat removal resulted in hyperinsulinemia,hyperleptinemia, and lower adiponectin levels inApc1638/Nþ female miceSerum measures in females at 4 weeks in study are
presented in Table 3 and Fig. 2F. Glucose and leptin levelswere significantly reduced in caloric-restricted females, ascompared with ad libitum and VF� females (Table 3; P <0.01). In contrast, VF� females were hyperinsulinemic,hyperleptinemic, and had significantly lower levels of adi-ponectin, as compared with ad libitum and caloric-restrict-ed female mice (Table 3; P < 0.05). Insulin sensitivity, asdetermined by ITTs, was greatest in caloric-restricted femalemice, but no distinguishable difference was observedbetween ad libitum and VF� females (Fig. 2F). In addition,CXCL-1, which is thought to play a role in angiogenesis andmetastasis, was markedly reduced in caloric-restrictedfemales (Table 3; P < 0.01), but no significant differenceswere observed among groups for other serum cytokines and
chemokines, including IFN-g , IL-6, IL-10, IL-12-p70, and IL-1b. Furthermore, no differences were observed in estradiollevels among female groups (Table 3).
Caloric restriction, but not visceral fat removal,significantly altered phenotypic characteristics inApc1638/Nþ male mice
Similar to females, VF�males had an average of 1,237.9�120.0 mg of epididymal fat and 377.0 � 38.1 mg of peri-nephric fat surgically removed. Despite removing approxi-mately 1.6 g of fat, no significant differences were observedover the course of the study for body weight (Fig. 3A), foodintake (Fig. 3B), lean mass (Fig. 3C), or total adiposity (Fig.3D) between ad libitum and VF� male mice. As in females,males caloric restricted by approximately 40% weighed sig-nificantly less (Fig. 3A) andhad reducedamounts of lean andfat mass (Fig. 3C and D), as compared with ad libitum andVF� males (P < 0.001). We confirmed at 12 weeks thatepididymal and perinephric fat pads did not return in VF�males, and these fat pads were markedly lower in both VF�and caloric-restricted males, as compared with ad libitummale mice (Fig. 3E; P < 0.05).
Caloric restriction, but not visceral fat removal, led tobeneficial metabolic adaptations in Apc1638/Nþ malemice
Serum measures of hormones, cytokines, chemokines,and metabolites in male mice from the 4-week cohort are
Table 1. Histopathology of the gastrointestinal tract in ad libitum, VF�, and caloric-restricted female mice
NOTE: Data are mean � SE. Different letters (a, b, ab) denote a significant difference between groups, P < 0.05.cValue based on the pathologic severity using a 1–4 scale, with 4 being most severe.dValue indicates the number of identified foci per section.
Table 2. Histopathology of the gastrointestinal tract in ad libitum, VF�, and caloric-restricted male mice
NOTE: Data are mean � SE. Different letters (a, b) denote a significant difference between groups, P < 0.05.cDenotes the severity score, based on a 1–4 scale (4 being most severe).dDenotes the number of foci per section.
Visceral Fat and Colon Cancer
www.aacrjournals.org Cancer Prev Res; 6(3) March 2013 181
presented in Table 4. Caloric-restricted male mice hadlower glucose, insulin, and leptin levels (P < 0.05),whereas adiponectin levels tended to be greater in thisgroup, as compared with ad libitum (P ¼ 0.27) and VF�
males (P ¼ 0.03; Table 4). In contrast to the observedtrends for VF� female mice (see Table 3 for female data),VF� in male mice did not significantly alter insulin (P ¼0.55) or leptin levels (P ¼ 0.45), as compared with ad
Figure 2. Phenotypiccharacteristics of ad libitum-fed(AL), VF�, and caloric-restricted(CR) female mice. A, body weight,and (B) food intake were monitoredon a weekly basis throughout thecourse of the study. No significantdifferences were observedbetween ad libitum-fed and VF�females, but caloric-restrictedmiceweighed less and consumedapproximately 40% fewer calories.Body composition was assessedby qMR at baseline (�1 week) andat 4- to 5-week intervalsthroughout the study fordetermination of (C) fat mass and(D) lean mass. Caloric-restrictedfemales had significantly less fatand lean mass, as compared withad libitum and VF� mice. E,periovarian and retroperitoneal fatpads were weighed and at 12weeks in study. These fat pads didnot return in VF� mice, whereascaloric restriction also significantlyreduced fat pad weights, ascompared with ad libitum females.F, insulin sensitivity was measuredby ITT in female mice. Nodifferences were found between adlibitum and VF� mice, whereasglucose levels remainedsignificantly lowest in caloric-restricted mice. Values are mean� SE. Different letters denote asignificant difference betweengroups (P < 0.05).
Table 3. Serum measures in ad libitum, VF�, and caloric-restricted female mice at 4 weeks in study
libitum male controls, whereas total adiponectin levelswere numerically lowest in this group (P ¼ 0.16), similarto what was found in VF� females. ITTs in male micerevealed a tendency for improved insulin sensitivity in
VF� males, as compared with ad libitum mice (Fig. 3F; P¼ 0.09), but glucose levels during the ITT were onlysignificantly improved in caloric-restricted male mice(Fig. 3F; P < 0.001). Similar to female caloric-restricted
Figure 3. Phenotypiccharacteristics of ad libitum-fed (AL),VF�, andcaloric-restricted (CR)malemice. A, body weight and (B) foodintake were monitored weeklythroughout the course of theexperiment. No significantdifferences were observed betweenad libitum-fed and VF� males, butcaloric-restricted mice weighed lessas a result of consumingapproximately 40% fewer calories.Body composition was assessed byqMR at baseline (�1 week) and at 4-to 5-week intervals throughout thestudy for determination of (C) fatmass and (D) lean mass. Caloric-restricted males had significantlyless fat and lean mass, as comparedwith ad libitum-fed and VF�mice. E,epididymal and retroperitoneal fatpads were weighed and at 12 weeksin study and found to not return inVF�mice, andwere also significantlyreduced in caloric-restrictedmice, ascompared with ad libitum-fed males.F, insulin sensitivity was measuredby ITTs at 3 weeks in study. Whilecaloric-restrictedmicewere themostinsulin sensitive, VF� also tended toimprove insulin action in males(P ¼ 0.09). Values are mean � SE.Different letters denote a significantdifference between groups (P <0.05).
Table 4. Serum measures in ad libitum, VF�, and caloric-restricted male mice at 4 weeks in study
mice, CXCL-1 levels were markedly reduced in caloric-restricted males (Table 4; P < 0.01), but no significantdifferences were observed among groups for IFN-g , IL-1b,IL-6, IL-10, and IL-12-p70 levels (Table 4).
DiscussionThis study establishes the importanceof visceral adiposity
in obesity-associated intestinal tumorigenesis in theApc1638/Nþ mouse model. Numerous epidemiologic andpreclinical studies, including a prior effort in this model(30), have linked the obese state to increased colon cancerrisk and/or mortality. However, obesity is a complex phe-notype, characterized not only by excess weight gain andadiposity, but also bydietary factors anda sedentary lifestyle(31). Collectively, this has made isolating the individualcontribution of adipose tissue to cancer risk challenging forthe field. Here, using a surgical approach to deplete visceralfat, we have circumvented these issues, providing causalevidence linking visceral adiposity per se to intestinal tumor-igenesis, independent of other important confoundersrelated to energy balance. Similarly, Lu and colleaguesfound that partial removal of visceral fat protected femaleSKH-1 mice against UVB-induced skin carcinogenesis (32).Importantly, these data provide yet another line of evidencedirectly linking visceral fat to the etiology of aging (24) andage-related diseases (22, 23, 32, 33).
Remarkably, we found that VF� was not only effective atattenuating macroadenoma development, but this reduc-tionwas comparablewith that observed in caloric-restrictedmice. This was accompanied by improved survival in calo-ric-restricted mice, with a similar, albeit nonsignificanttrend observed for increased survival in VF� mice, ascomparedwith ad libitummice. However, we also observeda clear effect of sex differences in the efficacy of theseinterventions on tumor initiation, promotion, and survival.Perhaps most striking was the marked protection conferredby VF� for development ofmacroadenomas in females, butnot inmales. While it is possible that the inability of VF� toprotect against macroadenomas inmales was simply due toinherent sex differences, we also observed a marked shift infat distribution among VF� males that could also explainthis difference. Indeed, when we evaluated mesenteric fatmass in females (Fig. 4A) andmales (Fig. 4B), we observed adistinct increase in mesenteric fat in VF� males, but not inVF� females. Given the hazardous nature of the mesentericfat depot, coupled with anatomic location of this depotlying in close proximity to the intestine, it is plausible thatmesenteric fat accretion abrogated the benefits of removingthe epididymal and perinephric fat depots via endocrineand/or paracrine mechanisms.
Interestingly, although VF� conferred protectionagainst macroadenoma development in female mice, itwas unexpectedly accompanied by a greater incidence ofmicroadenomas. We believe the most likely reason forthis finding is that VF� led to a systemic change in factor(s) that blocked the progression of tumor development atthe microadenoma–macroadenoma transition. In con-trast, caloric restriction was associated with increased
dysplasia in males but was protective against the devel-opment of both micro- and macroadenomas. However,caloric restriction had no apparent effect on tumorigen-esis in females. This suggests that energy availability mayplay a unique role in tumor initiation and early promo-tion in male mice, but caloric restriction was also bene-ficial for females, perhaps at later stages, as shown by theirimproved survival (see Fig. 1D). Future studies are neededto explore the mechanisms whereby abdominal obesityand nutrient availability act independently during stagesof initiation, promotion, and progression, and how theseinteractions are modulated by gender.
We also noted that while both ad libitummales (24.3%fat) and females (26.3% fat) were obese, due to
Figure 4. Mesenteric fatmass inmale and female ad libitum-fed (AL), VF�,and caloric-restricted (CR)mice.Mesenteric fat is believed to be themostclosely related VF depot to humans, both anatomically and due to itsportal access. However, unlike other VF depots, this fat pad cannot besurgically removed because of the extensive vascularization and neuralinnervation to the tissue. We recovered and weighed this fat depot atsacrifice. A, in females, caloric restriction significantly decreasedmesenteric fat mass, whereas no differences were observed between adlibitum and VF� mice. B, in males however, while caloric restrictionsimilarly decreased mesenteric fat, VF removal led to a markedredistribution of fat to this depot, as compared with ad libitum-fedmales.Bars are mean � SE. Different letters denote a significant differencebetween groups (P < 0.05).
Huffman et al.
Cancer Prev Res; 6(3) March 2013 Cancer Prevention Research184
consuming a 45% high-fat diet, ad libitum females devel-oped fewer tumors (P < 0.05) and had improved survival(P < 0.05), as compared with ad libitum males. Indeed,consistent evidence from human studies shows a strongerrisk posed by obesity and visceral obesity to colon cancerincidence and mortality in men, as compared with wom-en (9, 26). The reason for this sex difference is not clear,but may be related to reproductive hormone status, asindicated by protection conferred from oral contraceptiveuse in women (34), and evidence that ovariectomizedfemale mice injected with colon cancer cells haveincreased fat mass, insulin resistance, and tumor growth(29). Females in our study also had approximately 50%lower insulin levels than males, and nearly 2-fold greateradiponectin levels, which may also have also contributedto these differences.Another important observation from this study was the
observed hyperinsulinemia, hyperleptinemia, and reducedadiponectin levels with VF� in females, which seems to beat odds with the reduction in macroadenomas. Indeed, themodest, but significant increase in fasting insulin was unex-pected given that our grouphas previously shown improvedhepatic insulin action with VF� in chow-fedmale rats (23),whereas Shi and colleagues (33) showed that removing justa single visceral fat depot (periovarian) was sufficient toimprove glucose tolerance in high-fat fed female mice. Itshould be pointed out, however, that any potential differ-ence in insulin action here between ad libitum and VF�femalemicewas likely limited to basal conditions, aswe didnot detect any difference in response to an insulin challenge(ITTs).We also observed that VF� in males and females resulted
in an approximately 21% reduction in adiponectin, whichwas not unexpected, given that a significant amount of fattissue was removed. While there is in vitro and in vivoevidence to support a protective role for increased adipo-nectin levels in colon cancer (15), rodent studies linkinglow adiponectin levels with colon cancer risk are mostlyderived frommodels of constitutive adiponectin deficiency(adiponectin knockout mice; ref. 35). Thus, a much moremodest 21% reduction in adiponectinmay not be sufficientto predispose to tumorigenesis, at least in females, in whichlevels are 2-fold higher than in male mice. However, anadverse effect in males cannot be ruled out and along withthe compensatory increase in mesenteric fat, may explainwhy VF� was ineffective in male mice.Leptin has also been widely implicated in linking
obesity to colon as well as other cancers (31), and weobserved that VF� females had greater leptin levels,which may be indicative of leptin resistance in thesemice. However, this may be a complex effect as evidencefor leptin as a stimulator of proliferation is clear only forcolon cancer cells in vitro (31, 36). We also measuredseveral other inflammatory mediators in serum andobserved a reduction in the chemokine, CXCL-1, whichhas been linked to angiogenesis and metastasis, in bothmale and female caloric-restricted mice, but no otherconsistent patterns emerged. It is important to note that
Apc1638/Nþ mice presented with splenomegaly, which iscommonly observed in these mice (37), and corre-sponded with elevated cytokine levels in some animals,making it difficult to distinguish the contribution ofadipose tissue to the proinflammatory milieu.
The evidence implicating obesity and its related sequel-ae to cancer risk spans a wide array of models andsystems. Yet, efforts to elucidate mechanism(s) linkingthe obese state to site-specific cancers such as breast,colon, prostate, and skin, particularly in vivo, haverevealed a more complex biology then perhaps was ini-tially anticipated. For example, studies using the fatless A-ZIP/F-1 mouse model showed that adipokines are notabsolutely required for tumor development. Indeed,despite the absence of adipose tissue and adipose-derivedpeptides, such as leptin, these mice present with severalother features of the obese phenotype, including hyper-glycemia, hyperinsulinemia, and elevated levels of proin-flammatory cytokines. When subjected to a 2-stage skincarcinogenesis procedure, these mice develop more skinpapillomas (38, 39), but this effect is not seen in theseverely obese, ob/ob mouse model (39).
A-ZIP/F-1 mice also have accelerated development ofmammary tumors when crossed with C3(1)/T-Ag transgen-ic mice (38), whereas a similar effect was observed whenMKR mice, which are lean but diabetic, were crossed withMMTV-PyVmT mice (40). In contrast, MMTV-TGFa/db/db,which are obese and insulin resistant, but leptin-receptordeficient, fail to develop mammary tumors (19), but haveincreased incidence of intestinal neoplasms (30, 41). Thus,our finding that protection conferred from intestinaltumors in VF� females occurs despite a failure to producefavorable changes in factors suggested to play an importantrole in obesity-associated tumor growth is not withoutprecedent. Collectively, these data show the complexity ofthe obesity–cancer interface and emphasize the need forcontinued efforts to delineate how specific perturbations toendocrine and other factors (42) by obesity, contribute tosite-specific cancer risk.
In summary, these data provide causal evidence linkingvisceral fat to intestinal cancer risk. The protection conferredby VF�was preferentially seen in femalemice, despite a lackof favorable changes in leptin, insulin, adiponectin, andseveral proinflammatory cytokines and chemokines, sug-gesting that other unknown mechanisms may underlie theobesity–colon cancer link. However, as the genetic modelused here develops tumors predominantly in the smallintestine, rather than the colon, further work on the under-lying mechanisms will need to focus on a model in whichthe colon is the principal site of tumor development. Giventhat visceral fat accrual and subcutaneous fat depletionrepresent a common hallmark of aging (3), it is temptingto conclude that the nearly logarithmic increase in cancerincidence and mortality with age is driven in part bythese unfavorable changes in fat redistribution. Therefore,strategies designed to deplete visceral fat stores in abdom-inally obese individuals, including pharmacologic and/orbehavioral strategies, such as diet and exercise, may be an
Visceral Fat and Colon Cancer
www.aacrjournals.org Cancer Prev Res; 6(3) March 2013 185
important cancer prevention strategy as well as an adju-vant therapy for improving outcomes following a cancerdiagnosis.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: D.M. Huffman, L.H. AugenlichtDevelopment of methodology: D.M. HuffmanAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): D.M. Huffman, J.J. Lofrese, J.P. ChamberlandAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): D.M. Huffman, L.H. Augenlicht, X.Zhang, G. Atzmon, J.P. Chamberland, C.S. MantzorosWriting, review, and/or revision of themanuscript:D.M. Huffman, L.H.Augenlicht, X. Zhang, J.J. Lofrese, G. Atzmon, J.P. Chamberland, C.S.MantzorosAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): D.M. Huffman, J.J. Lofrese, J.P.Chamberland, C.S. MantzorosStudy supervision: D.M. Huffman, C.S. Mantzoros
AcknowledgmentsThe authors thank Dr. Rani Sellers for assistance with the histopath-
ologic evaluation of tissue sections and Youngmei Zhao and HongquianLiang for technical support.
Grant SupportThis work has been supported by grants from the Prevent Cancer Foun-
dation and National Institute on Aging (AG037574) to D.M. Huffman, andthe Einstein Nathan Shock Center Healthy Aging Physiology Core(P30AG038072) and theDiabetes Research and Training Center (DK20541)at theAlbert EinsteinCollege ofMedicine. C.S.Mantzoros is supportedby theNational Institute of Diabetes and Digestive and Kidney Diseases grants58785, 79929, and 81913 as well as Award Number 1I01CX000422-01A1from the Clinical Science Research andDevelopment Service of the VAOfficeof Research and Development.
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 October 2, 2012; revised December 2, 2012; accepted December3, 2012; published online March 6, 2013.
References1. Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of obesity and
trends in the distribution of body mass index among US adults, 1999–2010. JAMA 2012;307:491–7.
2. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity andtrends in bodymass index amongUSchildren and adolescents, 1999–2010. JAMA 2012;307:483–90.
3. Huffman DM, Barzilai N. Role of visceral adipose tissue in aging.Biochim Biophys Acta 2009;1790:1117–23.
4. Larsson SC, Permert J, Hakansson N, Naslund I, Bergkvist L, Wolk A.Overall obesity, abdominal adiposity, diabetes and cigarette smokingin relation to the risk of pancreatic cancer in two Swedish population-based cohorts. Br J Cancer 2005;93:1310–5.
5. Rapp K, Schroeder J, Klenk J, Stoehr S, Ulmer H, Concin H, et al.Obesity and incidence of cancer: a large cohort study of over 145,000adults in Austria. Br J Cancer 2005;93:1062–7.
6. Frezza EE,WachtelMS, Chiriva-Internati M. Influence of obesity on therisk of developing colon cancer. Gut 2006;55:285–91.
7. Kaaks R, Lukanova A, Kurzer MS. Obesity, endogenous hormones,and endometrial cancer risk: a synthetic review. Cancer EpidemiolBiomarkers Prev 2002;11:1531–43.
8. Pischon T, Nothlings U, Boeing H. Obesity and cancer. Proc Nutr Soc2008;67:128–45.
9. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight,obesity, andmortality from cancer in a prospectively studied cohort ofU.S. adults. N Engl J Med 2003;348:1625–38.
10. Carroll KK. Obesity as a risk factor for certain types of cancer. Lipids1998;33:1055–9.
11. Hursting SD, Hursting MJ. Growth signals, inflammation, and vas-cular perturbations: mechanistic links between obesity, metabolicsyndrome, and cancer. Arterioscler Thromb Vasc Biol 2012;32:1766–70.
12. Dalamaga M, Diakopoulos KN, Mantzoros CS. The role of adipo-nectin in cancer: a review of current evidence. Endocr Rev 2012;33:547–94.
14. Barzilai N, Gupta G. Revisiting the role of fat mass in the life extensioninduced by caloric restriction. J Gerontol A Biol Sci Med Sci 1999;54:B89–96.
15. Moon HS, Liu X, Nagel JM, Chamberland JP, Diakopoulos KN,Brinkoetter MT, et al. Salutary effects of adiponectin on coloncancer: in vivo and in vitro studies in mice. Gut. 2012 Jun 26. [Epubahead of print].
17. Nunez NP, Perkins SN, Smith NC, Berrigan D, Berendes DM,Varticovski L, et al. Obesity accelerates mouse mammary tumorgrowth in the absence of ovarian hormones. Nutr Cancer 2008;60:534–41.
18. Fenton JI, Birmingham JM, Hursting SD, Hord NG. Adiponectin blocksmultiple signaling cascades associated with leptin-induced cell pro-liferation in Apc Min/þ colon epithelial cells. Int J Cancer 2008;122:2437–45.
19. Cleary MP, Juneja SC, Phillips FC, Hu X, Grande JP, Maihle NJ. Leptinreceptor-deficient MMTV-TGF-alpha/Lepr(db)Lepr(db) female micedo not develop oncogene-induced mammary tumors. Exp Biol Med(Maywood) 2004;229:182–93.
20. Einstein FH, Atzmon G, Yang XM, Ma XH, Rincon M, Rudin E, et al.Differential responses of visceral and subcutaneous fat depots tonutrients. Diabetes 2005;54:672–8.
21. Matsuzawa Y, Shimomura I, Nakamura T, Keno Y, Tokunaga K.Pathophysiology and pathogenesis of visceral fat obesity. Ann N YAcad Sci 1995;748:399–406.
22. Gabriely I, Ma XH, Yang XM, Atzmon G, Rajala MW, Berg AH, et al.Removal of visceral fat prevents insulin resistance and glucose intol-erance of aging: an adipokine-mediated process? Diabetes 2002;51:2951–8.
23. Barzilai N, She L, Liu BQ, Vuguin P, Cohen P, Wang J, et al. Surgicalremoval of visceral fat reverses hepatic insulin resistance. Diabetes1999;48:94–8.
25. Pischon T, Lahmann PH, Boeing H, Friedenreich C, Norat T, Tjonne-land A, et al. Body size and risk of colon and rectal cancer in theEuropean Prospective Investigation into Cancer and Nutrition (EPIC).J Natl Cancer Inst 2006;98:920–31.
26. Larsson SC, Wolk A. Obesity and colon and rectal cancer risk: ameta-analysis of prospective studies. Am J Clin Nutr 2007;86:556–65.
27. Fodde R, EdelmannW, Yang K, van Leeuwen C, Carlson C, Renault B,et al. A targeted chain-termination mutation in the mouse Apc generesults in multiple intestinal tumors. Proc Natl Acad Sci U S A 1994;91:8969–73.
28. Boivin GP, Washington K, Yang K, Ward JM, Pretlow TP, Russell R,et al. Pathology of mouse models of intestinal cancer: consensusreport and recommendations. Gastroenterology 2003;124:762–77.
Huffman et al.
Cancer Prev Res; 6(3) March 2013 Cancer Prevention Research186
29. Yakar S, Nunez NP, Pennisi P, Brodt P, Sun H, Fallavollita L, et al.Increased tumor growth in mice with diet-induced obesity: impact ofovarian hormones. Endocrinology 2006;147:5826–34.
30. Gravaghi C, Bo J, Laperle KM,Quimby F, Kucherlapati R, EdelmannW,et al. Obesity enhances gastrointestinal tumorigenesis in Apc-mutantmice. Int J Obes (Lond) 2008;32:1716–9.
31. Drew JE. Molecular mechanisms linking adipokines to obesity-relatedcolon cancer: focus on leptin. Proc Nutr Soc 2012;71:175–80.
32. Lu YP, Lou YR, Bernard JJ, PengQY, Li T, Lin Y, et al. Surgical removalof the parametrial fat pads stimulates apoptosis and inhibits UVB-induced carcinogenesis in mice fed a high-fat diet. Proc Natl Acad SciU S A 2012;109:9065–70.
33. Shi H, Strader AD, Woods SC, Seeley RJ. The effect of fat removal onglucose tolerance is depot specific in male and female mice. AmJ Physiol Endocrinol Metab 2007;293:E1012–20.
34. Bosetti C, Bravi F, Negri E, La Vecchia C. Oral contraceptives andcolorectal cancer risk: a systematic review and meta-analysis. HumReprod Update 2009;15:489–98.
35. Fujisawa T, Endo H, Tomimoto A, Sugiyama M, Takahashi H, Saito S,et al. Adiponectin suppresses colorectal carcinogenesis under thehigh-fat diet condition. Gut 2008;57:1531–8.
36. Aparicio T, Kotelevets L, Tsocas A, Laigneau JP, Sobhani I, Chastre E,et al. Leptin stimulates the proliferation of human colon cancer cellsin vitro but does not promote the growth of colon cancer xenografts in
nude mice or intestinal tumorigenesis in Apc(Min/þ) mice. Gut2005;54:1136–45.
37. Kucherlapati M, Nguyen A, Kuraguchi M, Yang K, Fan K, Bronson R,et al. Tumor progression in Apc(1638N) mice with Exo1 and Fen1deficiencies. Oncogene 2007;26:6297–306.
38. Nunez NP, Oh WJ, Rozenberg J, Perella C, Anver M, Barrett JC,et al. Accelerated tumor formation in a fatless mouse withtype 2 diabetes and inflammation. Cancer Res 2006;66:5469–76.
39. Ablamunits V, Cohen Y, Brazee IB, Gaetz HP, Vinson C, Klebanov S.Susceptibility to induced and spontaneous carcinogenesis isincreased in fatless A-ZIP/F-1 but not in obese ob/ob mice. CancerRes 2006;66:8897–902.
40. Novosyadlyy R, Lann DE, Vijayakumar A, Rowzee A, Lazzarino DA,Fierz Y, et al. Insulin-mediated acceleration of breast cancer devel-opment and progression in a nonobese model of type 2 diabetes.Cancer Res 2010;70:741–51.
41. Hata K, Kubota M, Shimizu M, Moriwaki H, Kuno T, Tanaka T, et al.C57BL/KsJ-db/db-Apc Mice Exhibit an Increased Incidence of Intes-tinal Neoplasms. Int J Mol Sci 2011;12:8133–45.
42. Klopp AH, Zhang Y, Solley T, Amaya-Manzanares F, Marini F, AndreeffM, et al. Omental adipose tissue-derived stromal cells promote vas-cularization and growth of endometrial tumors. Clin Cancer Res2012;18:771–82.
Visceral Fat and Colon Cancer
www.aacrjournals.org Cancer Prev Res; 6(3) March 2013 187
