Interaction of Fatty Acid Genotype and Diet on Changes inColonic Fatty Acids in a Mediterranean Diet InterventionStudy
Shannon R. Porenta1, Yi-An Ko2, Leon Raskin5, Stephen B. Gruber1, Bhramar Mukherjee2, Ana Baylin3,Jianwei Ren4, and Zora Djuric4
AbstractAMediterranean diet increases intakes of n-3 and n-9 fatty acids and lowers intake of n-6 fatty acids. This
can impact colon cancer risk as n-6 fatty acids aremetabolized to proinflammatory eicosanoids. The purpose
of this studywas to evaluate interactions of polymorphisms in the fatty aciddesaturase (FADS) genes,FADS1
and FADS2, and changes in diet on fatty acid concentrations in serumand colon. A total of 108 individuals at
increased riskof colon cancerwere randomized to either aMediterraneanor aHealthy Eatingdiet. Fatty acids
were measured in both serum and colonic mucosa at baseline and after six months. Each individual was
genotyped for four single-nucleotide polymorphisms in the FADS gene cluster. Linear regressionwas used to
evaluate the effects of diet, genotype, and the diet by genotype interaction on fatty acid concentrations in
serum and colon. Genetic variation in the FADS genes was strongly associated with baseline serum
arachidonic acid (n-6) but serum eicosapentaenoic acid (n-3) and colonic fatty acid concentrations were
not significantly associated with genotype. After intervention, there was a significant diet by genotype
interaction for arachidonic acid concentrations in colon. Subjectswhohad allmajor alleles for FADS1/2 and
were following a Mediterranean diet had 16% lower arachidonic acid concentrations in the colon after six
months of intervention than subjects following the Healthy Eating diet. These results indicate that FADS
genotype couldmodify the effects of changes in dietary fat intakes on arachidonic acid concentrations in the
colon. Cancer Prev Res; 6(11); 1212–21. �2013 AACR.
IntroductionMany studies have suggested that aMediterranean diet, as
compared with a typical Western diet, may decrease the riskof various chronic diseases including colorectal cancer(1, 2). Rates of colorectal cancer were very low in Greeceand have increased as diet has drifted away from thetraditional eating pattern (3). The traditional Greek diet,relative to a Western diet, had lower intakes of n-6 poly-unsaturated fatty acids (PUFA) and red meat, but higherintakes of plant-based foods, fish, and monounsaturatedfatty acids (MUFA) chiefly fromolive oil (2). The fat contentof the Mediterranean diet is of particular interest for coloncancer prevention, as in intervention studies, increasing
fiber alone does not seem to be preventive, and increasedintakes of fruit and vegetables have had modest preventiveeffects (4–6). In particular, we hypothesized that lowerintakes of n-6 linoleic acid and higher intakes of n-3 fattyacids have implications for preventing colon cancer as n-6fatty acids are metabolized to eicosanoids such as prosta-glandin E2 (PGE2) that are proinflammatory in the colon(7). PGE2 is formed from arachidonic acid (20:4 n-6) byCOXs in the colonic mucosa, and it plays an important rolein colonic crypt cellular expansion and subsequent adeno-ma formation (8).
In addition to the possible effects of dietary intakes,genetic variation in fatty acid desaturase (FADS) genes hasbeen shown to influence serum and tissue arachidonic acidconcentrations (9–15). Delta-5 desaturase (FADS1) anddelta-6 desaturase (FADS2) are key desaturase enzymesinvolved in the synthesis of arachidonic acid and eicosa-pentaenoic acid (EPA; 20:5, n-3) from 18 carbon precursorfatty acids. Dietary intake of arachidonic acid is low inhumans; however, arachidonic acid comprises between5%and 10%of the phospholipids in cells due to elongationand desaturation of linoleic acid (18:2 n-6) to arachidonicacid (16).
Polymorphisms in the FADS1 and FADS2 genes havebeen identified, and these significantly affect PUFA concen-trations in serum. The minor alleles are associated with
Authors' Affiliations: 1Departments of Internal Medicine, 2Biostatistics,3Epidemiology, and 4Family Medicine, University of Michigan, Ann Arbor,Michigan, and 5Division of Epidemiology, Department of Medicine, Van-derbilt University, Nashville, Tennessee
Current address for S.B. Gruber: Keck School of Medicine, Norris Com-prehensive Cancer Center, University of Southern California, Los Angeles,California.
Corresponding Author: Zora Djuric, University of Michigan, 1500 E.Medical Center Drive, Room 2150 Cancer Center, Ann Arbor, MI 48109-5930. Phone: 734-615-6210; Fax: 734-647-9817; E-mail:[email protected]
lower desaturase activity and lower concentrations of ara-chidonic acid in blood (9–15). Analogous associations forEPA and docosahexaenoic acid (DHA) have not been con-sistent across studies, perhaps because certain types of fishcan supply high amounts of preformed EPA and DHA.Dietary intakes are important to consider as conversion ofdietary linolenic acid to longer chain n-3 fatty acids com-petes with the analogous process for n-6 fatty acids (17). Inaddition todiet, desaturase activity seems tobe important incardiovascular health, and presence of the minor allele inFADS1/2 has been associated with improved measures ofblood lipids, C-reactive protein, insulin, and fasting glucose(18–21.) This indicates that lower arachidonic acid levelsare associated with lower proinflammatory states. The prev-alence ofminor alleles seems to have evolved in response toWestern diets that are plentiful in n-6 fatty acids, and theyare more prevalent in persons of European descent than ofAfrican descent (11, 22).Much less research is available on how FADS polymorph-
isms might affect changes in fatty acids in response tochanges in diet, and the available studies have generallyfocused on n-3 fatty acid supplementation. Flaxseed sup-plementation, which provides linolenic acid (18:3, n-3),was less effective in increasing EPA concentrations inminorallele carriers of either FADS1 or FADS2, resulting in sig-nificant diet by genotype interactions on plasma concentra-tions of EPA and arachidonic acid (23). Dietary n-3 fattyacids also may interact with FADS genotype in affectingconcentrations of blood cholesterol and triglycerides, withsignificant beneficial effects for carriers of all minor allelesbeing found in some but not all studies (20, 24–26).The goal of this present study was to assess potential
interactions of polymorphisms in FADS1 and FADS2 withchanges in diet on levels of arachidonic acid and EPA in theserumand in the colonicmucosa of persons at increased riskfor colon cancer. This was a secondary analysis of a ran-domized clinical trial that evaluated changes in fatty acidsand carotenoids elicited by 6 months of intervention witheither a Mediterranean or a standard Healthy Eating diet.In that study, we observed that dietary changes had littleeffect on colon fatty acids, which led to the hypothesis thatmetabolic factors may be limiting for changes in fattyacids (27). The randomized study obtained both blood andcolon biopsies. Here, the relationships of FADS polymor-phisms with serum and colonic fatty acid concentrationswere evaluated at baseline and after 6 months of dietaryintervention.
Materials and MethodsStudy design and eligibilityDetails of recruitment and conduct of the Healthy Eating
for Colon Cancer Prevention Study have been publishedpreviously (27, 28). The study was approved by the Univer-sity of Michigan Medical Internal Review Board and wasregistered at the ClinicalTrials.org (NCT00475722). Briefly,120 individuals at increased risk of colon cancer gaveinformed consent and were randomized to follow a mod-ified Mediterranean diet or to Healthy People 2010 diet for
6 months. Blood and colonic mucosal tissue samples werecollected at baseline and at 6 months by flexible sigmoid-oscopy without prior preparation of the bowels. Blood wasdrawn after an overnight fast. At baseline, a Health StatusQuestionnaire was filled out by participants that includedhealthanddemographicdata.Health informationwasaskedagain at 6 months. Dietary data were collected at 0 and 6monthsusing2daysof food records and two24-hour recalls.
The decision to genotype subjects with regard to FADSwasmade after the study began, and consent for genotypingcould not be obtained from 9 individuals, 2 of whomcompleted 6 months of study and 7 of whom had droppedout after enrolling. Three samples were not genotypedsuccessfully. The present analysis therefore included 108of 120 subjects enrolled in the study and randomized to 6months of counseling for either a Mediterranean or aHealthy Eating diet.
The frequency of counseling sessions was the same inboth study arms. The Healthy Eating diet had dietary goalsbased on the Healthy People 2010 diet. The goals were toinclude two servings/day of fruit, three servings/day ofvegetables with at least one of those servings being darkgreen or orange, six servings/day of grains with at least threefromwhole grains, less than 10% of calories from saturatedfat (SFA) and less than 30% of calories from total fat. TheMediterranean diet had goals for consumption of high n-3foods, such as fish or flax at least two times a week,consumption of foods in a manner to increase MUFA anddecrease n-6 PUFA intakes, six servings/day of grains with atleast three from whole grains, and seven to nine fruits andvegetable servings/day in a specified variety.
Serum and colonic fatty acidsFatty acid analysis was conducted by gas chromatogra-
phy—mass spectroscopy (GC-MS) of fatty acid methylesters. Total lipids were extracted from serum using a 1:1mixture of chloroform andmethanol, and 17:0 (1,2-dihep-tadecanoyl-sn-glycero-3-phosphocholine) was used as theinternal standard. For colon tissue, one biopsy of about 5mg was sufficient for analysis of fatty acids. The biopsy waspulverized in liquid nitrogen, sonicated in 150 mL of ice-cold PBS containing 0.1% butylated hydroxytoluene (BHT)and 1 mmol/L EDTA with an Ultrasonic processor (30seconds twice), and then total lipids were extracted with1 mL of chloroform and methanol (1:1). The organic layerin either case was used to prepare fatty acid methyl esterswith Meth-Prep II derivatization reagent (Alltech). The GC-MS analysis was conducted with A SupelcoSP2330 column,30m� 0.32mmX0.2 mmfilm thickness (Sigma-Aldrich), aHP 7673/5971 GC-MS, and helium as the carrier gas with avalidated assay (29). The following fatty acids in serum andcolon tissue weremeasured in 12 analytic different batches:12:0, 14:0, 16:0, 16:1, 18:0, 18:1, 18:2 (n6), 18:3 (n3), 20:0,20:1, 20:3 (n6), 20:4 (n6), 20:5 (n3), and 22:6 (n3).
DNA extraction and genotypingMany polymorphisms have been identified in the FADS1/2
gene cluster. Haplotypes have been constructed using three
Fatty Acid Genotype, Diet, and Colonic Fatty Acids
www.aacrjournals.org Cancer Prev Res; 6(11) November 2013 1213
to 18 single-nucleotide polymorphisms (SNP), and arachi-donic acid concentrations were typically about 30% higherin carriers of all major alleles (9, 12, 16, 30). This literaturehas indicated that there was little additional benefit fromgenotyping more than three SNPs, we therefore chose togenotype the three SNPs used in the study of the Rzehakand colleagues (30). A subsequent genome-wide associa-tion study indicated that another polymorphism in theFADS1/2 region explained 18% of the inter-individualvariation in arachidonic acid concentrations, we thereforeadded rs174537 to the present analysis (21).
DNA was extracted from the buffy coat of heparinizedblood samples. The buffy coat had been collected for eachblood sample and mixed with 1% SDS/1 mmol/L EDTAbefore freezing at �80�C. After all the samples had beencollected, they were treated with RNases A and heat-treatedRNase T1 followed by digestion with protease K, solventextraction, and precipitation ofDNA. TheDNAwas purifiedusing the MinElute Reaction Cleanup Kit (QIAGEN) toensure high-quality DNA for genotyping. Four SNPs weregenotyped: one in the FADS2 gene (rs3834458), two SNPslocated in FADS1 (rs174556 and rs174561), and one SNPlocated in the intragenic region between FADS1 and FADS2(rs174537).
Three of the SNPs (rs174556, rs174561, and rs174537)were genotyped using TaqMan SNP Genotyping Assays(Applied Biosystems). All assays were conducted both inreal time and after read mode for allelic discrimination onan AB7900 system. The rs3834458 polymorphisms weredetected by sequencing. For quality control, 10% of allsamples were regenotyped. All plates incorporated positiveand negative controls.
PCR reactions for rs3834458 included 5 mL (20 pmol/mL)of both forward and reverse primers, 12 mL AmpliTaq GoldMaster Mix, 10 mg/mL genomic DNA, and Millipore waterfor a total volumeof 25mL. Primers used for rs3834458were50-TCCACGATTCCCAAAGAGAC-30 and 50-TCTGCAACCT-CCCTAGAGACA-30. Samples were covered in mineral oil,denatured for 10 minutes at 95�C, were passed through 40cycles of amplification consisting of 1 minute of denatur-ation at 95�C, 1 minute of primer annealing at 55�C, and 1minute of elongation at 72�C. The PCR products werechecked by running on a 2% agarose gel stained withethidiumbromide before sequencing. Sequencingwas doneon an ABI 3730 sequencer in the University of MichiganSequencing Core Facility.
Statistical analysesThe distributions of fatty acid variables were first checked
for normality and transformed to approximate normality asneeded before analyses. The transformations applied beforeanalysis are given in the table footnotes, and variables wereback transformed to calculate percentage of increases ordifferences. Untransformed means are shown in the tablesfor ease of interpretation. Deviations from Hardy–Wein-berg equilibrium for the genotypes of each SNP were testedusing x2 tests. Differences in baseline parameters betweendiet arms were assessed using independent t or x2 tests, as
appropriate (Tables 1 and 2). Genotype data for the fourSNPs were summarized to yield the count of minor alleles(the minimum and maximum counts were zero and eight,respectively). Linear regression was used to evaluate theeffect of number of minor alleles on fatty acid concentra-tions. Subsequently, a binary variable for genotype groupwas created by the presence/absence ofminor alleles, that is,all major alleles versus one or more minor alleles.
A linear mixed model was used to evaluate whether thepresence of any FADS variant affects baseline fatty acidconcentrations (arachidonic acid and EPA). Each of thebaseline fatty acids in both serum and colonic mucosa wasregressed on genotype group (Table 2). Batch number was arandom effect to account for heterogeneity as fatty acidswere measured in different batches. The covariates in themodel included age, gender, body mass index (BMI; inkg/m2), anddietary intakemeasuresof n-6PUFA,n-3PUFA,and long-chain n-3 PUFA (sum of the n-3 fatty acids 20:5,22:5, and 22:6) as a percentage of energy using 9 kcal/g.
Next, we used linear mixed models to evaluate thechanges in fatty acid concentrations after 6 months of dietintervention: dietary intake, serum, and colon fatty acidconcentrations were regressed on time (baseline, 6 month)with a random intercept for each individual. For serum andcolon fatty acids, batch numberwas included in the randomeffects. Separate analyses were conducted for the two dietgroups (Table 3). Finally, analyses were conducted to com-pare the changes in fatty acid composition over 6 monthsbetween the twodiet arms and to assesswhether the changesweremodified by the presence ofminor alleles in FADS. Forthese analyses, each of the outcome variables (arachidonicacid and EPA for both serum and colonic mucosa) at 6-month follow-up was regressed on genotype group, dietarm, and genotype group�diet assignment interaction by alinear mixed model (Table 4). The model was adjusted forage, BMI, and the concentration of each corresponding fattyacid at baseline. In all the models, batch number wasincorporated as a random effect when appropriate. AllreportedP valueswere two-tailed. The statistical significancewas set a ¼ 0.05 level. Analyses were conducted using SASversion 9.1 (SAS Institute).
ResultsBaseline characteristics and genotyping
The overall study consisted of 108 study participants afterexclusions for lack of genotyping consent (n ¼ 9) andincomplete genotype data (n ¼ 3). Genotyping success rateof the four SNPs chosen to define the FADS1/2 haplotype asdescribed in Materials and Methods, was between 96.7%and 98.3%. Minor allele frequencies were in the range of25.0% to32.9%. Thegenotypedistribution for eachSNPdidnot deviate from Hardy–Weinberg equilibrium (P > 0.05).
Baseline characteristics for the Healthy Eating diet group(n ¼ 54) and the Mediterranean diet group (n ¼ 54) weresummarized in Table 1. No significant differences werefound in minor allele frequency of any SNP, gender, race,age, or BMI between the two diet groups at baseline.Likewise, baseline measurements of arachidonic acid, EPA,
Porenta et al.
Cancer Prev Res; 6(11) November 2013 Cancer Prevention Research1214
and long-chain n-3 fatty acids (the sum of EPA and DHA)did not differ significantly in the serum or the colonicmucosa between the two diet groups (Table 1).
Baseline measuresLinear regression analysis indicated that the number of
minor alleles was a significant predictor of baseline serumarachidonic acid concentration (P < 0.001) and almostsignificant for colonic arachidonic acid concentration(P ¼ 0.058). A greater number of minor alleles was signif-icantly associated with lower arachidonic acid concentra-tion in serum. Dietary arachidonic acid intakes were not asignificant predictor of either serum of colon concentra-tions. For long-chain n-3 fatty acids, however, the situationwas the reverse. Dietary intake of long-chain n-3 fatty acidswas a significant predictor of baseline serum long-chain n-3concentration (P < 0.001) and colonic long-chain n-3concentration (P ¼ 0.044), whereas the number of minoralleles was not a significant predictor of either.Subsequent analyses were conducted categorizing sub-
jects into two groups by presence or absence of any minoralleles in the FADS gene cluster. The only dietary or demo-graphic factor to differ by genotype at baseline was BMI,which was lower in carriers of any minor alleles (mean of
27.8, SD3.7, in allmajor allele carriers andmeanof 26.1, SD3.6, in carriers of anyminor alleles;P¼0.02by the two-sidedt test). Age was not significantly different (P¼ 0.11) but wasretained as a covariate. No significant difference was foundfor other demographic characteristics (race, gender, smok-ing, andcommonmedicationuse)betweenminor allele andall major allele carriers. Results were similar when using anyone SNP individually versus all minor SNPS (not shown).
Serum and colon fatty acid concentrations at baseline bygenotype group are shown in Table 2. Linear mixed modelswere used to evaluate differences between genotype groups.The presence of any minor alleles was highly significantlyassociated with baseline serum 20:4, n-6 concentrations(P < 0.0001) and 18:3, n-3 concentration (P ¼ 0.01), andmarginally significant for colonic 20:4, n-6 concentration(P ¼ 0.07), with adjustment for age, BMI, and dietaryintakes of n-6 PUFA, n-3 PUFA, and/or long-chain n-3PUFA as a percentage of energy. Specifically, mean serum20:4, n-6 concentration (percentage of total fatty acids) forminor allele carriers was estimated to be 2% [95% confi-dence interval (CI), 1%–3%] lower, whereas mean serum18:3, n-3 concentration for minor allele carriers was esti-mated to be 21% (95% CI, 4%–41%) higher, comparedwith those individualswith allmajor alleles in the four SNPs
Table 1. Characteristics of study subjects at baseline by diet arm
Characteristic Healthy Eating arm (n ¼ 54) Mediterranean arm (n ¼ 54) Pa
Serum fatty acids, as % of total20:4, n-6 8.7 (1.9) 9.2 (2.1) 0.17Long-chain n-3 PUFAc 2.98 (1.20) 3.04 (1.37) 0.82
Colon fatty acids, as % of total20:4, n-6 10.5 (3.1) 10.1 (3.0) 0.53Long-chain n-3 PUFAc 3.52 (1.53) 3.18 (1.62) 0.26
NOTE: The data shown are mean (SD) or number (%).aP values for differences at baseline between diet arms are from two-sided t tests for continuous variables or x2 tests for proportions.Natural log transformations were used to normalize the data before analysis except for dietary n-3 PUFA and serum 20:4, n-6, whichdid not require transformation. Data are shown untransformed for ease of interpretation.bFor dietary intakes, n-6PUFAwas the sumof 18:2, and 20:4, n-3PUFAwas the sumof 18:3, 20:5, 22:5, and22:6, and "long-chain n-3"was the sum of 20:5, 22:5 and 22:6.cLong-chain n-3 PUFA in serum and colon were the sum of 20:5, n-3 and 22:6, n-3.
Fatty Acid Genotype, Diet, and Colonic Fatty Acids
www.aacrjournals.org Cancer Prev Res; 6(11) November 2013 1215
in FADS. There was neither any significant association ofgenotype with EPA nor with long-chain n-3 fatty acids (thesum of EPA and DHA). Genotype group also had nosignificant effects on total cholesterol, Low density lipopro-tein (LDL), high density lipoprotein (HDL), triglycerides,insulin, glucose, and C-reactive protein (CRP) with P > 0.11in each case (not shown).
Effects of dietary intervention on fatty acid intakes andfatty acid concentrations in serum and colon
We first evaluated changes in fatty acids by diet groupassignment alone without considering the genotypegroups. Table 3 displays dietary intakes, serum, and colonfatty acid concentrations for the two diet arms at baselineand after 6 months of intervention. On the basis of datafrom food records and 24-hour recalls, dietary intakes ofSFA andMUFA were significantly reduced (P < 0.0001) andlong-chain n-3 PUFA was significantly increased (P ¼
0.004) in the Healthy Eating group after 6 months. Thedecrease in mean SFA resulted in an increased PUFA:SFAratio from 0.60 to 0.92 in the Healthy Eating group (P ¼0.008 from mixed linear regression models controlling forage). In theMediterranean group, dietary intakes of SFA andn-6 PUFA both significantly decreased (P < 0.0001), where-as MUFA and long-chain n-3 PUFA significantly increased(P < 0.0001), in accordwith the counseling goals. ThemeanPUFA:SFA ratio increased nonsignificantly from 0.72 to0.77 in the Mediterranean group.
Serum 18:2 n-6 significantly decreased (P ¼ 0.02), andboth MUFA and n-3 PUFA significantly increased (P ¼0.0005 and 0.01, respectively) in the Mediterranean armonly (Table 3). There was little change in colon fatty acidconcentrations. The only significant change was for long-chain n-3 PUFA that significantly increased in bothHealthy Eating (P ¼ 0.01) and Mediterranean groups(P ¼ 0.01).
NOTE: The data shown are mean and SD. Significant p-values are bolded.aP valueswereobtained from linearmixedmodelswith eachof thebaseline fatty acids regressedongenotypegroupwith batchnumberas a random effect. The covariates of the model included age, baseline BMI, and baseline dietary intake measures of n-6 PUFA,n-3 PUFA, and/or long-chain n-3 PUFA as a percentage of energy. An inverse squared root transformation was applied to colon 18:2,n-6 and colon 22:6, n-3 to approximate normality, respectively. Natural logarithm transformations were used to normalize the rest fattyacid data except for serum 20:4, n-6 and 18:2, n-6, which did not require transformation. Mean and SDs are shown untransformed forease of interpretation.bLong-chain n-3PUFA in serumand colonwere the sumof 20:5, n-3 and 22:6, n-3. In the diet, g/day of the n-3 fatty acids 20:5, 22:5 and22:6 were summed and expressed as a percentage of energy using 9 kcal/g.
Porenta et al.
Cancer Prev Res; 6(11) November 2013 Cancer Prevention Research1216
Interactions of genotype and diet interventionFigures 1 and 2 show the raw means in each group over
time. Table 4 shows the linear mixed model results for theanalysis of the genotype by diet interaction. There was asignificant interaction of genotype by diet for 20:4, n-6(arachidonic acid) concentrations in the colon (P ¼0.004). Significant genotype-by-diet interactions were nei-ther found for arachidonic acid in serum nor for EPA.Among subjects with no minor alleles, mean colon arachi-donic acid concentrations were estimated to be 16% (95%CI, 5%–26%) lower for the Mediterranean arm than theHealthy Eating arm at 6 months. These results indicate thatafter adjusting for baseline arachidonic acid concentrations,mean colon arachidonic acid concentrations at 6 monthswere significantly different between diet arms only in per-sonswith nominor alleles in the FADS1/2 gene cluster. Thiswas mainly due to an increase in colon arachidonic acid intheHealthy Eating diet arm,whereas colon arachidonic acidconcentrations remained fairly constant in the Mediterra-nean group.
DiscussionThis randomized, dietary intervention study afforded the
opportunity to evaluate the impact of FADS genotype and
diet on fatty acid concentrations in both serum and colonicmucosa of individuals at increased risk for colon cancer. Thenumber of minor alleles in the FADS gene cluster, but notdiet, predicted serum arachidonic acid concentrations. Thisagrees well with results of previous studies, namely thatcarriers of minor alleles have lower arachidonic acid con-centrations (9–15). For EPA concentrations in serum, geno-type had no effect, whereas diet did have a significant effect,likely because n3 fatty acid intakes were fairly low andlimiting in this study population. It should, however, benoted that diet in this studywas assessedusing self-report onfour separate days. In addition to the possibility of mis-reporting of intakes, those 4 days might not represent usualintakes over the last month of study and therefore willweaken any apparent associations with diet.
In epidemiologic studies, relatively higher dietary intakesof both n-3 and n-9 fatty acids are thought to be protective,whereashigh intakesof n-6 fatty acids increase riskof severalcancers including that of the colon (31). This has beenconfirmed in experimentalmodels of colon cancer, and lowversus high n6 fatty acid diets are associated with decreasedtumors and lower production of certain eicosanoids such asPGE2 (32, 33). In the colon, PGE2 has been tightly linkedwith colon cancer risk (34). Increased n-3 fatty acid intakes
Table 3. Effects of dietary intervention on changes in dietary intakes and fatty acid concentrations
aThe transformations used for dietary variables were natural log both for 18:3, n-3 and for long-chain n-3 fatty acids (sum of 20:5 and22:6). The transformations used for serum concentrations were squared for SFA, and natural log both for 18:3, n-3 and long-chain n-3fatty acids. The transformations used for colon fatty acids were: squared for SFA, inverse squared root for 18:2, n-6 and natural log forMUFA, 20:4, n-6, 18:3, n-3, and long-chain n-3 fatty acids. Mean and SDs are shown untransformed for ease of interpretation.bSignificant different from baseline for that diet arm (P < 0.05).
Fatty Acid Genotype, Diet, and Colonic Fatty Acids
www.aacrjournals.org Cancer Prev Res; 6(11) November 2013 1217
also reduce PGE2 production (35–39). Interestingly, areduction in n-6 fatty acid intakes can augment increasesin EPA after n-3 fatty acid supplementation (40). Bartoli andcolleagues observed inhibition of aberrant crypt foci, ade-nocarcinomas, decreasedmucosal arachidonate (20:4), anddecreased PGE2 in rats fed either n-9 or n-3 diets relative torats fed diets high in n-6 fatty acids (41). The levels of colonmucosal PGE2 were directly proportional to arachidonatelevels in the colon in that study (41). These data make itimportant to better understand factors that could affectarachidonic acid and EPA levels in the human colon.
Unlike serum fatty acids, genotype had no significanteffects on fatty acid concentrations in the colon at baseline(Table 2). It may be the case that serum concentrations offatty acids are affected by first pass livermetabolismmore sothan tissues. After absorption of fatty acids, mainly in thesmall intestine, the liver is the initial site of fatty acidmetabolism. The subsequent distributionof fatty acids fromthe circulation to tissues will be dependent on lipoproteinlipase activity in each tissue site and on tissue-specific
metabolic conversions. In a well-controlled study in pigs,increased dietary intakes of linolenic acid and/or linoleicacid significantly affectedmetabolism of each other to long-chain fatty acids in the liver, but the effect was minimal inbrain cortex (42). In a human lipodomic study, FADSactivity of blood reflected activity in the liver but not inadipose tissue (43). Serum and colon fatty acid concentra-tions therefore reflect not only diet and genotype, but anytissue-specific regulation of fatty acid metabolism as well.
Because the present study was a randomized clinical trial,we then evaluated the effects of the two dietary interven-tions on changes in fatty acid intakes and levels over time.Both dietary interventions decreased SFA intakes andincreased n-3 PUFA intakes. Only the Mediterranean inter-vention resulted in increased MUFA and decreased n-6PUFA intakes. Serum fatty acids in the Mediterranean armreflected these changes in diet (Table 3). In the colon,however, the only significant change was an increase inn-3 PUFA. This indicates that tissue-specific processes maylimit the impact of dietary changes in colon fatty acids. The
Table 4. Interactions of diet group assignment with genotype on fatty acid concentrations in serum andcolonic mucosa at 6 monthsa
aAnalyseswere conducted by linearmixedmodels by including each fatty acid concentration at 6months as the response variable andcovariates: diet arm (Mediterranean vs.HealthyEating), genotype (presenceof anyminor alleles versus allmajor alleles at the four SNPsin the FADS gene), interaction of diet and genotype, baseline age, baselineBMI, and baseline concentration of each fatty acid. Batch ofsample analysis was treated as a random effect.bNatural log transformation was applied to fatty acid variables approximate normality in analysis. Significant p-values are bolded
Porenta et al.
Cancer Prev Res; 6(11) November 2013 Cancer Prevention Research1218
increase in colon n-3 PUFA is interesting, however, as theincreases in dietary n-3 PUFAweremodest in each diet arm.The effect of FADS genotype on fatty acid concentrations
in colonwas only evident after intervention (Table 4). Studysubjects who were carriers of all major alleles and random-ized to the Healthy Eating intervention had higher colonarachidonic acid concentrations after 6 months than sub-jects with all major alleles in the Mediterranean group. It is
not entirely clear why this should be the case, but theHealthy Eating intervention did result in a higher relativeamount of n-6 PUFA to other dietary fats. This could havehelped increase the percentage of arachidonic acid in thecolon fatty acids after the Healthy Eating intervention. Inaddition to polymorphisms in FADS, other factors could beoperative to affect fatty acid desaturation, such as diet-induced changes in the expression and the activity of FADS,
Figure 2. Colon mucosaconcentrations of arachidonic acid(AA) andEPAat baseline and after 6months of intervention. Subjectswere grouped by presence/absence of any minor alleles in theFADS1/FADS2 gene cluster(rs3834458, rs174556, rs174561,and rs174537). There was asignificant gene�diet interaction forarachidonic acid in personswith nominor alleles in any of the SNPs inlinear mixed models. The datashown are untransformed meanand SE. Solid lines represent datafor the Healthy Eating arm anddashed lines represent data for theMediterranean arm.
Figure 1. Serum concentrations ofarachidonic acid (AA) and EPA atbaseline and after 6 months ofintervention. Subjects weregrouped by presence/absence ofany minor alleles in the FADS1/FADS2 gene cluster (rs3834458,rs174556, rs174561, andrs174537). The data shown areuntransformedmean and SE. Solidlines represent data for the HealthyEating arm and dashed linesrepresent data for theMediterranean arm.
Fatty Acid Genotype, Diet, and Colonic Fatty Acids
www.aacrjournals.org Cancer Prev Res; 6(11) November 2013 1219
and to changes in substrate competition (44). In carriers ofall major alleles randomized to the Mediterranean inter-vention, arachidonic acid levels stayed relatively low at bothtime points andwere estimated to be 16% lower than in theHealthy arm after 6 months of intervention.
Limitations of this study include the small sample size,the relatively short intervention length, and the self-reportof diet that is known to be subject to biases. It may takelonger for a change in diet to be fully manifest, especially intissues. In addition, the measurement of fatty acids wasdone as a percentageof total fatty acids such that increases inone fatty acid on a volume basis would result in decreases inother fatty acids. An additional consideration is that ara-chidonic acid concentrations are not easily modifiable bychanges in n6 fatty acids in the diet, especially if arachidonicacid is not elevated at the outset (45). Strengths of the studyinclude that it was a randomized study, and measures wereavailable before and after diet change in both serum andcolonic mucosa of individuals at increased risk for coloncancer.
In conclusion, this study showed that those subjects withno minor alleles in the FADS1/2 cluster had higher con-centrations of arachidonic acid in serum. Polymorphism inFADS1/2 had no effect on concentrations of EPA, perhapsbecause concentrations of this fatty acid are more highlydriven by dietary intakes. The trends were similar in colontissue fatty acids but not significant. After randomization toMediterraneanorHealthy Eating intervention for 6months,there was a significant gene�diet interaction for colon ara-chidonic acid concentrations. Subjects who had all majoralleles for FADS1/2 had significantly lower arachidonic acidconcentrations in the colon after 6 months if they were inthe Mediterranean diet arm. Because arachidonic acid is thesubstrate for PGE2 production, these results indicate that aMediterranean diet could be especially favorable for reduc-
ing colon cancer risk in the subset of subjects with all majoralleles in FADS1/2. Future work should evaluate the effectsof these FADS polymorphisms on colonic proinflammatorystates.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: S.B. Gruber, B. Mukherjee, Z. DjuricDevelopment of methodology: Y.-A. Ko, B. Mukherjee, J. RenAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): S.R. Porenta, S.B. GruberAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): S.R. Porenta, Y.-A. Ko, S.B. Gruber,B. Mukherjee, A. Baylin, J. Ren, Z. DjuricWriting, review, and/or revision of the manuscript: S.R. Porenta, Y.-A.Ko, S.B. Gruber, B. Mukherjee, A. Baylin, Z. DjuricStudy supervision: S.B. Gruber, Zora Djuric
AcknowledgmentsThe authors thank all the individuals who volunteered for the Healthy
Eating Study for Colon Cancer Prevention. The parent study was designedand conducted in collaboration with Drs. Dean E. Brenner, Mack T. Ruffin,D. Kim Turgeon, and Ananda Sen. Mary Rapai was the coordinator for thestudy and Maria Cornellier was the study dietitian.
Grant SupportThis study was supported by a grant from the Ronald P. and Joan M.
Nordgren Cancer Research Fund, NIH grant RO1 CA120381, and CancerCenter Support Grant P30 CA046592. The study used core resources sup-ported by a Clinical Translational Science Award, NIH grant UL1RR024986and NIH grant P50 CA130810 (the Michigan Clinical Research Unit), theMichigan Diabetes Research Center NIH grant 5P60 DK020572 (ChemistryLaboratory), and the Michigan Nutrition and Obesity Research Center NIHgrant P30 DK089503.
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 April 10, 2013; revised July 16, 2013; accepted August 30, 2013;published OnlineFirst September 10, 2013.
References1. Verberne L, Bach-Faig A, Buckland G, Serra-Majem L. Association
between the Mediterranean diet and cancer risk: a review of obser-vational studies. Nutr Cancer 2010;62:860–70.
2. Giacosa A, Barale R, Bavaresco L, Gatenby P, Gerbi V, Janssens J,et al. Cancer prevention in Europe: the Mediterranean diet as aprotective choice. Eur J Cancer Prev 2013;22:90–5.
3. Frenandez E, Vecchia CL, Gonzales JR, Lucchini F, Negri E, Levi F.Coverging patterns of colorectal cancer mortality in Europe. Eur JCancer 2005;41:430–7.
4. Koushik A, Hunter DJ, Spiegelman D, BeesonWL, van den Brandt PA,Buring JE, et al. Fruits, vegetables, and colon cancer risk in a pooledanalysis of 14 cohort studies. J Natl Cancer Inst 2007;99:1471–83.
5. Bingham SA, Day NE, Luben R, Ferrari P, Slimani N, Norat T, et al.Dietary fiber in food and protection against colorectal cancer in theEuropean Prospective Investigation into Cancer and Nutrition (EPIC):an observational study. Lancet 2003;361:1496–501.
6. Alberts DS, Einspahr J, Ritenbaugh C, Aickin M, Rees-McGee S,Atwood J, et al. The effect of wheat bran fiber and calcium supple-mentation on rectal mucosal proliferation rates in patients withresected adenomatous colorectal polyps. Cancer Epidemiol Biomar-kers Prev 1997;6:161–9.
7. Wang D, Dubois RN. Eicosanoids and cancer. Nat Rev Cancer 2010;10:181–93.
8. Wu WK, Sung JJ, Lee CW, Yu J, Cho CH. Cyclooxygenase-2 intumorigenesis of gastrointestinal cancers: an update on the molecularmechanisms. Cancer Lett 2010;295:7–16.
9. Malerba G, Schaeffer L, Xumerle L, Klopp N, Trabetti E, Biscuola M,et al. SNPs of the FADS gene cluster are associated with polyunsat-urated fatty acids in a cohort of patients with cardiovascular disease.Lipids 2008;43:289–99.
10. Glaser C, Heinrich J, Koletzko B. Role of FADS1 and FADS2 poly-morphisms in polyunsaturated fatty acid metabolism. Metabolism2010;59:993–9.
11. Sergeant S, Hugenschmidt CE, Rudock ME, Ziegler JT, Ivester P,Ainsworth HC, et al. Differences in arachidonic acid levels and fattyacid desaturase (FADS) gene variants in African Americans and Euro-pean Americans with diabetes or the metabolic syndrome. Br J Nutr2012;107:547–55.
12. Schaeffer L, Gohlke H,Muller M, Heid IM, Palmer LJ, Kompauer I, et al.Common genetic variants of the FADS1 FADS2 gene cluster and theirreconstructed haplotypes are associated with the fatty acid compo-sition in phospholipids. Hum Mol Genet 2006;15:1745–56.
13. Aslibekyan S, Jensen MK, Campos H, Linkletter CD, Loucks EB,Ordovas JM, et al. Fatty Acid desaturase gene variants, cardiovascularrisk factors, and myocardial infarction in the costa rica study. FrontGenet 2012;3:72.
Porenta et al.
Cancer Prev Res; 6(11) November 2013 Cancer Prevention Research1220
14. Baylin A, Ruiz-Narvaez E, Kraft P, Campos H. alpha-Linolenic acid,Delta6-desaturase gene polymorphism, and the risk of nonfatal myo-cardial infarction. Am J Clin Nutr 2007;85:554–60.
15. MathiasRA, FuW, Akey JM, Ainsworth HC, Torgerson DG, Ruczinski I,et al. Adaptive evolution of the FADS gene cluster within Africa. PLoSONE 2012;7:e44926.
16. Lattka E, Illig T, Koletzko B, Heinrich J. Genetic variants of the FADS1FADS2 gene cluster as related to essential fatty acid metabolism. CurrOpin Lipidol 2010;21:64–9.
17. Hussein N, Ah-Sing E, Wilkinson P, Leach C, Griffin BA, Millward DJ.Long-chain conversion of [13C]linoleic acid and alpha-linolenic acid inresponse to marked changes in their dietary intake in men. J Lipid Res2005;46:269–80.
18. Dupuis J, LangenbergC, Prokopenko I, SaxenaR,SoranzoN, JacksonAU, et al. New genetic loci implicated in fasting glucose homeostasisand their impact on type 2 diabetes risk. Nat Genet 2010;42:105–16.
19. Martinelli N,Girelli D,MalerbaG,Guarini P, Illig T, Trabetti E, et al. FADSgenotypes and desaturase activity estimated by the ratio of arachi-donic acid to linoleic acid are associated with inflammation andcoronary artery disease. Am J Clin Nutr 2008;88:941–9.
20. Standl M, Lattka E, Stach B, Koletzko S, Bauer CP, von Berg A, et al.FADS1 FADS2 gene cluster, PUFA intake and blood lipids in children:results from the GINIplus and LISAplus studies. PLoS ONE 2012;7:e37780.
21. Tanaka T, Shen J, Abecasis GR, Kisialiou A, Ordovas JM, Guralnik JM,et al. Genome-wide association study of plasma polyunsaturated fattyacids in the InCHIANTI Study. PLoS Genet 2009;5:e1000338.
22. Mathias RA, Sergeant S, Ruczinski I, Torgerson DG, HugenschmidtCE, Kubala M, et al. The impact of FADS genetic variants on omega6polyunsaturated fatty acid metabolism in African Americans. BMCGenet 2011;12:50.
23. Gillingham LG, Harding SV, Rideout TC, Yurkova N, Cunnane SC, EckPK, et al. Dietary oils and FADS1-FADS2 genetic variants modulate[13C]alpha-linolenic acid metabolism and plasma fatty acid compo-sition. Am J Clin Nutr 2013;97:195–207.
24. Lu Y, Feskens EJ, Dolle ME, Imholz S, VerschurenWM, Muller M, et al.Dietary n-3 and n-6 polyunsaturated fatty acid intake interacts withFADS1 genetic variation to affect total and HDL-cholesterol concen-trations in the Doetinchem Cohort Study. Am J Clin Nutr 2010;92:258–65.
25. Hellstrand S, Sonestedt E, Ericson U, Gullberg B,Wirfalt E, Hedblad B,et al. Intake levels of dietary long-chain PUFAs modify the associationbetween genetic variation in FADS and LDL-C. J Lipid Res 2012;53:1183–9.
26. Cormier H, Rudkowska I, Paradis AM, Thifault E, Garneau V, LemieuxS, et al. Association between polymorphisms in the fatty acid desa-turase gene cluster and the plasma triacylglycerol response to an n-3PUFA supplementation. Nutrients 2012;4:1026–41.
27. Sen A, Ren J, Ruffin MT, Turgeon DK, Brenner DE, Sidahmed E, et al.Relationships between serum and colon concentrations of carote-noids and fatty acids in a randomized dietary intervention trial. CancerPrev Res 2013;6:558–65.
28. Djuric Z, RuffinMTt, Rapai ME, Cornellier ML, Ren J, Ferreri TG, et al. AMediterranean dietary intervention in persons at high risk of coloncancer: recruitment and retention to an intensive study requiringbiopsies. Contemp Clin Trials 2012;33:881–8.
29. Ren J, Mozurkewich EL, Sen A, Vahratian AM, Ferreri TG, Morse AN,et al. Total serum fatty acid analysis by GC–MS: assay validation andserum sample stability. Curr Pharm Anal 2013;9:331–9.
30. Rzehak P, Heinrich J, Klopp N, Schaeffer L, Hoff S, Wolfram G, et al.Evidence for an association between genetic variants of the fatty aciddesaturase 1 fatty acid desaturase 2 (FADS1 FADS2) gene cluster andthe fatty acid composition of erythrocyte membranes. Br J Nutr 2009;101:20–6.
31. Bartsch H, Nair J, Owen RW. Dietary polyunsaturated fatty acids andcancers of the breast and colorectum: emerging evidence for their roleas risk modifiers. Carcinogenesis 1999;20:2209–18.
32. Rao CV, Hirose Y, Indranie C, Reddy BS. Modulation of experimentalcolon tumorigenesis by types and amounts of dietary fatty acids.Cancer Res 2001;61:1927–33.
33. Rao CV, Simi B, Wynn TT, Garr K, Reddy BS. Modulating effect ofamount and types of dietary fat on colonic mucosal phospholipase A2,phosphatidylinositol-specific phospholipase C activities, and cyclo-oxygenasemetabolite formation during different stages of colon tumorpromotion in male F344 rats. Cancer Res 1996;56:532–7.
34. BacklundMG,Mann JR, Dubois RN.Mechanisms for the prevention ofgastrointestinal cancer: the role of prostaglandin E2. Oncology2005;69(Suppl 1):28–32.
35. Dommels YE, HaringMM, Keestra NG, Alink GM, vanBladeren PJ, vanOmmen B. The role of cyclooxygenase in n-6 and n-3 polyunsaturatedfatty acid mediated effects on cell proliferation, PGE(2) synthesis andcytotoxicity in human colorectal carcinoma cell lines. Carcinogenesis2003;24:385–92.
36. Bartram HP, Gostner A, Scheppach W, Reddy BS, Rao CV, Dusel G,et al. Effects of fish oil on rectal cell proliferation, mucosal fatty acids,and prostaglandin E2 release in healthy subjects. Gastroenterology1993;105:1317–22.
37. Nieto N, Fernandez MI, Torres MI, Rios A, Suarez MD, Gil A. Dietarymonounsaturated n-3 and n-6 long-chain polyunsaturated fatty acidsaffect cellular antioxidant defense system in rats with experimentalulcerative colitis induced by trinitrobenzene sulfonic acid. Dig Dis Sci1998;43:2676–87.
38. Kuratko CN, Becker SA. Dietary lipids alter fatty acid composition andPGE2production in colonic lymphocytes.Nutr Cancer 1998;31:56–61.
39. Broughton KS, Wade JW. Total fat and (n-3):(n-6) fat ratios influenceeicosanoid production in mice. J Nutr 2002;132:88–94.
41. Bartoli R, Fernandez-Banares F, Navarro E,Castella E,Mane J, AlvarezM, et al. Effect of olive oil on early and late events of colon carcino-genesis in rats: modulation of arachidonic acid metabolism and localprostaglandin E(2) synthesis. Gut 2000;46:191–9.
42. Smink W, Gerrits WJ, Gloaguen M, Ruiter A, van Baal J. Linoleic andalpha-linolenic acid as precursor and inhibitor for the synthesis of long-chain polyunsaturated fatty acids in liver and brain of growing pigs.Animal 2012;6:262–70.
43. Kotronen A, Seppanen-Laakso T, Westerbacka J, Kiviluoto T, Arola J,Ruskeepaa AL, et al. Comparison of lipid and fatty acid composition ofthe liver, subcutaneous and intra-abdominal adipose tissue, andserum. Obesity 2010;18:937–44.
44. Tu WC, Cook-Johnson RJ, James MJ, Muhlhausler BS, Gibson RA.Omega-3 long chain fatty acid synthesis is regulated more by sub-strate levels thangeneexpression.Prostaglandins LeukotEssent FattyAcids 2010;83:61–8.
45. Rett BS, Whelan J. Increasing dietary linoleic acid does not increasetissue arachidonic acid content in adults consuming Western-typediets: a systematic review. Nutr Metab 2011;8:36.
Fatty Acid Genotype, Diet, and Colonic Fatty Acids
www.aacrjournals.org Cancer Prev Res; 6(11) November 2013 1221
Correction: Interaction of Fatty Acid Genotypeand Diet on Changes in Colonic Fatty Acids in aMediterranean Diet Intervention Study
In this article (Cancer Prev Res 2013;6:1212–21), which appeared in theNovember 2013 issue of Cancer Prevention Research (1), an author was omittedfrom the article byline. Leon Raskin (Division of Epidemiology, Department ofMedicine, Vanderbilt University, Nashville, TN) should appear in the byline asthe third author. The correct author listing is given below. The authors regret theerror. The electronic version has been corrected and does not reflect the printversion.
Shannon R. Porenta, Yi-An Ko, Leon Raskin, Stephen B. Gruber, BhramarMukherjee, Ana Baylin, Jianwei Ren, and Zora Djuric.
Reference1. Porenta SR, Ko YA, Gruber SB, Mukherjee B, Baylin A, Ren J, et al. Interaction of fatty acid
genotype and diet on changes in colonic fatty acids in a Mediterranean diet intervention study.Cancer Prev Res 2013;6:1212–21.
Published OnlineFirst February 18, 2014.doi: 10.1158/1940-6207.CAPR-14-0014�2014 American Association for Cancer Research.
CancerPreventionResearch
Cancer Prev Res; 7(3) March 2014372
2013;6:1212-1221. Published OnlineFirst September 10, 2013.Cancer Prev Res Shannon R. Porenta, Yi-An Ko, Leon Raskin, et al. Fatty Acids in a Mediterranean Diet Intervention StudyInteraction of Fatty Acid Genotype and Diet on Changes in Colonic
Updated version
10.1158/1940-6207.CAPR-13-0131doi:
Access the most recent version of this article at:
