Plasma and Urine Metabolite Profiles Impacted by Increased Dietary Navy Bean Intake in Colorectal Cancer Survivors: a Randomized-Controlled Trial

Published OnlineFirst December 24, 2020; DOI: 10.1158/1940-6207.CAPR-20-0270

CANCER PREVENTION RESEARCH | RESEARCH ARTICLE

Plasma and Urine Metabolite Proﬁles Impacted by Increased Dietary Navy Bean Intake in Colorectal Cancer Survivors: A Randomized-Controlled Trial A C Iman Zarei1, Bridget A. Baxter1, Renee C. Oppel1, Erica C. Borresen1, Regina J. Brown2, and Elizabeth P. Ryan1

ABSTRACT ◥ Navy beans contain bioactive phytochemicals with colon (3.89-fold), and urine S-adenosylhomocysteine (2.09-fold) cancer prevention properties as demonstrated in carcino- and cysteine (1.60-fold) represent metabolites with cancer- gen-induced animal models. Human studies support that protective actions following navy bean consumption. Diet- dietary navy bean intake modulates metabolism by the gut derived metabolites were detected in plasma or urine and microbiome. This study investigated the effect of navy bean confirmed for presence in the navy bean intervention ingestion on plasma and urine metabolite profiles of over- meals and snacks. These included 3-(4-hydroxyphenyl) weight and obese colorectal cancer survivors. Twenty parti- propionate, betaine, pipecolate, S-methylcysteine, choline, cipants completed a single-blinded, randomized-controlled eicosapentaenoate (20:5n3), benzoate, S-adenosylhomo- dietary intervention with precooked navy beans (35 g bean cysteine, N-delta-acetylornithine, cysteine, 3-(4-hydroxy- powder/day) or control (0 g/day) for 4 weeks. Plasma and phenyl)lactate, gentisate, hippurate, 4-hydroxyhippurate, urine were collected at baseline, 2 weeks, and 4 weeks and salicylate. The navy bean dietary intervention for following consumption. Nontargeted metabolomics was 4 weeks showed changes to pathways of metabolic impor- applied to study meals and snacks, navy beans, plasma, and tance to colorectal cancer prevention and merit continued urine. Increased navy bean consumption was hypothesized attention for dietary modulation in future high-risk cohort to (i) delineate dietary biomarkers and (ii) promote meta- investigations. bolic shifts relevant for cancer protection in the plasma and urine metabolome. At 4 weeks, 16 plasma and 16 urine Prevention Relevance: This clinical study suggests that metabolites were significantly different in the navy bean increased consumption of navy beans would deliver bioac- intervention group compared with placebo control (P < tive metabolites to individuals at high risk for colorectal 0.05). Increased plasma 2,3-dihydroxy-2-methylbutyrate cancer recurrence and produce metabolic shifts in plasma (1.34-fold), S-methylcysteine (1.92-fold), and pipecolate and urine profiles.

Introduction prevention (4–7), and additionally promote lower risk of chronic diseases such as obesity (8–13), cardiovascular dis- Legumes are globally recognized as high-quality sources ease (14, 15), and type II diabetes (16). Hughes and colleagues of dietary fibers with prebiotic functions and are sustainable reported that 59 g of pinto beans per 100 g of diet reduced colon sources of essential amino acids and many other bioactive tumor (azomethane-induced) incidence by 50% in rats com- plant secondary metabolites (1–3). Epidemiologic studies pared with 24% by casein-fed rats (17). Similar to the findings and multiple clinical cohort investigations now support of previous studies, Hangen and Bennink found that diets fed to regular consumption of common beans for colorectal cancer rats containing 75% navy beans resulted in 44% decreased colon tumor and adenocarcinoma incidence (18). The con- 1 1Department of Environmental and Radiological Health Sciences, College of sumption of /2 cup of cooked dry beans per day for adults was Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort practical and significantly increased intake of fiber, protein, and Collins, Colorado. 2University of Colorado School of Medicine, Aurora, Colorado. minerals such as folate, zinc, iron, and magnesium, while Note: Supplementary data for this article are available at Cancer Prevention lowers the intake of saturated fats and total fat (19–21). Research Online (http://cancerprevres.aacrjournals.org/). Emerging evidence exists for diet-derived small molecules to Iman Zarei and Bridget A. Baxter contributed equally to this article. signal as modifiers of cancer risk and tumor behavior. Reliable Corresponding Author: Elizabeth P. Ryan, Colorado State University, 200 West dietary intake assessment methods are required to understand Lake Street, 1680 Campus Delivery, Fort Collins, CO 80523-1680. Phone: 970- the associations between diet, nutritional status, and colon 491-1536; Fax: 970-491-7569; E-mail: [email protected] cancer risk. Nontargeted metabolomics is a well-established Cancer Prev Res 2021;14:497–508 and sensitive tool for the comprehensive and concomitant doi: 10.1158/1940-6207.CAPR-20-0270 characterization of multiple metabolites within a biological 2020 American Association for Cancer Research. matrix, and is increasingly applied to cancer research to screen

AACRJournals.org | 497

Downloaded from cancerpreventionresearch.aacrjournals.org on September 28, 2021. © 2021 American Association for Cancer Research. Published OnlineFirst December 24, 2020; DOI: 10.1158/1940-6207.CAPR-20-0270

Zarei et al.

and discover shifts in metabolism (22, 23). In recent years, that each contained 17.5 g of cooked navy bean powder; ref. 36). metabolomics has also been used for identifying putative The control study arm received placebo meals and snacks biomarkers of food consumption and dietary patterns (24, 25). without the addition of navy bean powder as previously Less is known about whether bioactive food components reach described (34, 36). The study protocol and written informed effective cancer-protective concentrations in people due to consent was approved by the Colorado State University extensive interindividual variability that can occur in host Research Integrity and Compliance Review Board and the digestion and gut metabolism. Some mechanisms linking University Colorado Cancer Center–North Institutional common dry bean consumption and cancer prevention Review Board (Protocol # 09-1530H and 10-1038, respectively) include, but are not limited to, induction of apoptosis and in accordance with the 1964 Helsinki Declaration and its 2013 modulation of cell cycle and proliferation (2, 26, 27). Other amendments. The 20 individuals completed the 4-week mechanisms for common bean intake to inhibit colon carci- pilot dietary intervention trial between August 2010 and nogenesis involve changes to energy metabolism (28) and December 2014. phase II detoxification (29) as shown by tissue microarray analysis in animals. Metabolomics has utility for measuring Blood and urine sample collection for metabolomics downstream products of metabolic disturbances shown by Fasting blood samples were collected into 4-mL ethylene- gene expression and for associations with environmental diamine-tetra-acetic acid (EDTA) blood collection tubes from exposures involved in cancer risk (30, 31). the 20 participants who completed study visits at baseline, Plasma and urine are ideal biological fluids to examine the 2 weeks, and week 4 and kept immediately on ice until the physiologic effects of dietary exposures and have shown sensi- centrifugation step to extract the plasma (1,500 rpm for 10 tivity to detect subtle and major changes in metabolite abun- minutes). Plasma was further stored at 80 C until processed dance over time (32, 33). A nontargeted metabolite approach for metabolomics analysis using ultra-high-performance liquid was applied herein to assess metabolic modulation of colo- chromatography–tandem mass spectrometry (UHPLC-MS/ rectal cancer risk following increased navy bean intake. The MS). At each study visit (baseline, 2 weeks, and 4 weeks), purposeofthisstudywastoassesstheimpactofnavybean participants were given study-labeled containers for self- consumption on the plasma and urine metabolome for collection of first void of the morning urine. Urine was stored colorectal cancer prevention when compared with place- at 80 C until extracted for analysis by UHPLC-MS/MS. bo-control foods, and to identify a suite of metabolites that Urine samples were subjected to the osmolality normalization originated from navy beans. It was hypothesized that using a Fiske 210 Micro-Sample Osmometer to correct for the increased consumption of navy beans would deliver bioac- fluid intake variability between participants. tive metabolites to individuals at high risk for colorectal cancer recurrence and produce metabolic shifts in plasma Plasma and urine extraction for metabolomics and urine profiles. The global nontargeted metabolomics was performed through Metabolon, Inc. The extraction method has been described previously (33). Briefly, 80% methanol was added Materials and Methods on plasma and urine in a ratio of 300 mL solvent per 100 mg Study design and dietary interventions plasma or urine. The samples were shaken for 2 minutes, and Twenty overweight or obese colorectal cancer survivors that then centrifuged for 10 minutes at 4C (12,000 rpm). The were a minimum of 4 months after cancer treatment were metabolite extracts from plasma and urine were analyzed using recruited for this single-blinded, placebo-controlled, random- UHPLC-MS/MS with positive and negative ion mode electro- ized intervention trial as previously described (3, 34, 35). All spray ionization (ESI). the participants were blinded to the study arms, and only the study coordinator was aware of the participant study group UHPLC-MS/MS analysis allocation. The study design adhered to CONSORT guidelines Samples were analyzed using a Waters ACQUITY UHPLC (http://www.consort-statement.org/). The recruitment took system coupled with a Thermo Scientific Q-Exactive high place through the University of Colorado Health–North Can- resolution/accurate mass spectrometer and interfaced with a cer Center Network (Fort Collins, CO). Supplementary Fig. S1 heated ESI (HESI-II) source and Orbitrap mass analyzer. illustrates the eligibility criteria and the allocation of 20 colo- Prior to analysis, samples were resuspended in acidic or basic rectal cancer survivors that completed a study titled “Beans/ UHPLC-compatible solvents. The acidic and basic solutions Bran Enriching Nutritional Eating for Intestinal Health Trial were analyzed using positive and negative ESI, and both (BENEFIT)” (NCT01929122). Participants were randomized hydrophilic interaction chromatography (HILIC) and reverse by body mass index (between 25 and 35), sex, and daily caloric phase (RP) chromatography (37). In HILIC mode, the extracts intake (36) prior to dietary intervention. Supplementary were injected into a BEH amide column (2.1 mm 100 mm, Table S1 shows study population characteristics at baseline. 1.7 mm; Waters Corporation). The mobile phase delivered In this 4-week intervention, participants assigned to the navy at 0.35 mL/minute consists of water, acetonitrile, and bean intervention consumed 35 g of cooked whole navy 10 mmol/L ammonium formate at pH 10.8 (HILIC positive), beans in powder form (daily intake of one meal and one snack methanol þ water, and 6.5 mmol/L ammonium bicarbonate at

498 Cancer Prev Res; 14(4) April 2021 CANCER PREVENTION RESEARCH

Plasma and Urine Metabolites Following Dietary Intervention

pH 8.0 (HILIC negative). In RP-positive and -negative ion- navy bean intervention meals and snacks but not in the control. optimized conditions, extracts were gradient eluted from a C18 The navy bean source of metabolites in meals and snacks column (2.1 mm 150 mm, 1.7 mm; Corporation). In RP was determined by the overlapping presence of metabolite positive, extracts were eluted from the column with water þ in the navy bean powder metabolome. Supplementary methanol, containing 0.05% perfluoropentanoic acid and 0.1% Fig. S2B–S2D displays normalized relative abundance values formic acid, while in RP negative, water þ methanol, contain- in placebo-control and navy bean intervention meals and ing 0.05% perfluoropentanoic, 0.01% formic acid, and aceto- snacks for the following metabolites: pipecolate, S-methylcys- nitrile was used as mobile phase. For the MS data acquisition of teine, and N-delta-acetylornithine. Pipecolate, S-methylcys- study meals and snacks, plasma and urine, an extended teine, and N-delta-acetylornithine had higher abundance in dynamic exclusion was used with the instrumental ion acquire- the bean intervention meals and snacks compared with control ment of 70 to 1,000 m/z. The peak picking process was that was due to the high relative abundance shown for these performed by Metabolon through in-house peak detection and compounds in the navy bean powder. integration software. Compound identification (level 1 iden- Dietary modulation of plasma metabolome in colorectal tification) was completed using library of over 3,300 purified cancer survivors after 2 and 4 weeks standards and based on the experimentally MS-MS spectra fi ¼ matching the accurate mass the standard within 8 ppm, reten- There were 854 identi ed metabolites from plasma (N 20 tion time, and m/z ratio. participants) across all time points, including baseline, 2 weeks, The nontargeted metabolomics analysis of cooked navy bean and 4 weeks after intervention. Supplementary Table S3 shows fi powder has been previously described (3). The plasma and the relative abundance of plasma metabolites classi ed into urine metabolites from participants in both dietary groups were metabolic pathways for all participants. Table 1 lists 22 plasma fi cross-referenced with metabolites identified in the navy beans metabolites that were signi cantly modulated in either the and in the one meal and three snacks metabolome after 4 weeks control, navy bean, or both study groups for 20 participants at 2 of either control or navy bean consumption. and/or 4 weeks compared with baseline (P < 0.05, q < 0.2). All data are presented as the mean fold change in the median- Statistical analyses scaled relative abundance of each metabolite. After consuming Plasma and urine metabolite intensity levels underwent navy beans for 2 weeks, 21 metabolites including two gut median-scaled normalization as previously described (33). A microbial metabolites, four amino acids, one cofactors and fi Welch two-sample t test and two-way ANOVA were per- vitamins, 11 lipids, and three xenobiotics were signi cantly formed for between-group comparisons. Repeated-measures modulated in plasma when compared with baseline. fi ANOVA was used for comparisons at 2 and 4 weeks from Navy bean powder signi cantly increased three amino acid baseline and within each study group. An estimate of false metabolites in plasma of colorectal cancer survivors at week 2 discovery rate (q-value) was calculated. Statistical significance and at week 4 when compared with the baseline levels – was defined as P values of <0.05 and q-values below the (Table 1). Figure 1A C illustrates median-scaled relative threshold of <0.2 for within-group comparisons. abundance for pipecolate, S-methylcysteine, and N-delta-acetylornithine. Individual responses over time are shown in Supplementary Fig. S3. Notably, these three metabolites were Results highlighted for abundance in navy bean powder and study Metabolome profile of study-provided meals and snacks meals and snacks metabolomes. The metabolomes of one meal and three snacks used in this Table 2 lists 16 plasma metabolites with significant differ- study comprised of 653 metabolites with confirmed identifica- ences in the abundance between intervention and control tions. Eight groups were used for the overarching metabolite groups for the 20 participants at 4 weeks (P < 0.05, q < 0.2). classifications: amino acids, carbohydrates, cofactors and vita- For all participants, data are presented as the mean fold mins, energy metabolism, lipids, nucleotides, peptides, and difference between intervention and control groups at 4 weeks. xenobiotics. The meals and snacks metabolome contained The five plasma metabolites increased in abundance for 143 amino acids, 41 carbohydrates, 27 cofactors and vitamins, the navy bean group were 3-(4-hydroxyphenyl)propionate 14 energy metabolism-related metabolites, 226 lipids, 61 (4.48-fold), S-allylcysteine (4.08-fold), pipecolate (3.88-fold), nucleotides, 32 peptide, and 111 xenobiotics. Supplementary S-methylcysteine (1.92-fold), and 2,3-dihydroxy-2-methylbu- Table S2 shows the metabolite raw abundance for one meal and tyrate (1.34-fold). The 11 plasma metabolites with significant three snacks. Composition of meals and snacks was analyzed by decreased levels compared with control at 4 weeks were chemical classes for placebo-control (0 g/serving navy bean betaine (0.84-fold), 1-arachidonoyl-GPC (20:4n6) (0.80-fold), powder) and navy bean intervention (17.5 g/serving navy bean 5alpha-pregnan-3beta-20alpha-diol monosulfate (0.12-fold), powder). Supplementary Fig. S2A displays the meals and choline (0.80-fold), DHA (0.74-fold), docosahexaenoylcarni- snacks metabolite profiles for placebo-control and navy bean tine (C22:6) (0.63-fold), eicosapentaenoate (EPA) (0.77-fold), intervention, whereby each bar represents the total number of linoleoylcarnitine (C18:2) (0.71-fold), 5,6-dihydrothymine identified metabolites in the chemical class, and the shaded (0.71-fold), gamma-glutamyl-2-aminobutyrate (0.30-fild), and portion represents the number of metabolites that appeared in benzoate (0.77-fold). The eight metabolites identified in the

AACRJournals.org Cancer Prev Res; 14(4) April 2021 499

Zarei et al.

Table 1. Plasma metabolites following consumption of control foods or navy bean–based foods for 2 and 4 weeks. Post intervention metabolite proﬁles were compared with baseline (0 weeks).

Fold changed Control group Navy bean group 2wk/ P q 4wk/ P q 2wk/ P q 4wk/ P q Metabolite 0wk value value 0wk value value 0wk value value 0wk value value

Amino acids Leucine, isoleucine, 2,3-dihydroxy-2- 1.15 0.676 1 1.16 0.415 1 1.71 0.000 0.025 1.73 0.000 0.014 and valine methylbutyrate Lysine Pipecolatea,b 1.22 0.550 1 1.10 0.677 1 2.59 0.000 0.025 2.67 0.000 0.014 Methione, cysteine, S-methylcysteinea,b 1.20 0.548 1 1.41 0.059 1 1.88 0.000 0.070 1.88 0.001 0.173 SAM, and taurine S-methylcysteine 1.53 0.032 1 1.66 0.008 1 1.86 0.001 0.123 2.07 0.001 0.173 sulfoxide Polyamine Tryptophan betaine 0.98 0.629 1 1.31 0.183 1 0.74 0.002 0.138 0.68 0.000 0.017 Urea cycle; arginine N-d-acetylornithinea,b 1.15 0.725 1 1.28 0.091 1 1.55 0.002 0.145 1.69 0.002 0.255 and proline Cofactors and vitamins Tocopherol g-CEHC 0.81 0.047 1 1.12 0.37 1 0.66 0.002 0.124 0.72 0.016 0.567 Lipids Diacylglycerol Diacylglycerol (16:1/18:2 1.04 0.961 1 0.92 0.283 1 1.72 0.002 0.124 1.44 0.021 0.64 [2], 16:0/18:3 [1])c Linoleoyl-linolenoyl- 1.02 0.671 1 0.95 0.412 1 1.73 0.011 0.296 1.50 0.09 0.698 glycerol (18:2/18:3) [2]c Oleoyl-linoleoyl-glycerol 0.95 0.483 1 0.90 0.192 1 1.47 0.001 0.124 1.28 0.063 0.688 (18:1/18:2) [1] Oleoyl-linoleoyl-glycerol 1.01 0.890 1 0.90 0.231 1 1.49 0.004 0.176 1.38 0.019 0.608 (18:1/18:2) [2] Oleoyl-oleoyl-glycerol 1.02 0.930 1 0.88 0.055 1 1.35 0.003 0.145 1.14 0.228 0.825 (18:1/18:1) [1]c Oleoyl-oleoyl-glycerol 1.06 0.654 1 0.89 0.087 1 1.42 0.000 0.106 1.21 0.082 0.698 (18:1/18:1) [2]c Palmitoyl-linoleoyl- 1.05 0.93 1 0.98 0.643 1 1.59 0.005 0.196 1.46 0.01 0.567 glycerol (16:0/18:2) [2]b,c Palmitoyl-oleoyl-glycerol 1.07 0.681 1 0.96 0.348 1 1.45 0.002 0.139 1.24 0.093 0.698 (16:0/18:1) [1]c Palmitoyl-oleoyl-glycerol 1.10 0.518 1 0.96 0.38 1 1.50 0.000 0.106 1.30 0.028 0.678 (16:0/18:1) [2]b,c Palmitoyl-palmitoyl- 1.15 0.513 1 1.00 0.603 1 1.83 0.002 0.124 1.61 0.015 0.567 glycerol (16:0/16:0) [2]c Steroid Cortisone 0.83 0.022 1 0.92 0.267 1 1.43 0.005 0.19 1.29 0.073 0.688 Xenobiotics Benzoate metabolism Hippuratea,b 1.34 0.860 1 1.19 0.563 1 1.29 0.39 0.964 2.91 0.002 0.174 Food component/ Piperinea 1.94 0.451 1 1.91 0.901 1 13.77 0.000 0.036 3.72 0.001 0.174 plant 4-allylphenol sulfate 4.45 0.000 0.009 7.58 0.000 0.000 3.59 0.000 0.025 4.54 0.000 0.011 Drug 4-hydroxycoumarin 6.82 0.030 1 2.22 0.225 1 8.04 0.001 0.124 5.68 0.104 0.698

Abbreviation: CEHC, carboxyethyl-hydroxychromans. aMetabolite also identified from the navy bean metabolome. bMetabolite also identified from the study food metabolome. cIndicates that the metabolite identified was not made against a purified chemical standard. dValues presented are fold change of the mean relative abundance within control, navy bean group at 2 and 4 weeks compared with their baselines (P < 0.05, are bold) and (q < 0.2). Statistically significant fold changes are bolded and shaded. Light shaded represent metabolites and pathways that showed trends toward significance (0.05 < P < 0.10).

plasma that were also present in the cooked navy beans include illustrates median-scaled relative abundance for 3-(4-hydro- pipecolate, S-methylcysteine, betaine, EPA, docosahexaenoate xyphenyl)propionate. This phenolic acid derivative was signif- (DHA), choline, gamma-glutamyl-2-aminobutyrate, and ben- icantly increased after consuming navy beans for 4 weeks when zoate (see Table 2, note “a”) and 6 metabolites were identiﬁed compared with the control group. This phenolic acid was also in the study meals and snacks metabolome including 3-(4- found in the study meals and snacks metabolome hydroxyphenyl)propionate, betaine, pipecolate, S-methylcys- (Table 2). Figure 1E and F shows the median-scaled relative teine, choline, and benzoate (see Table 2, note “b”). Figure 1D abundances for EPA and benzoate. These endogenous and/or

500 Cancer Prev Res; 14(4) April 2021 CANCER PREVENTION RESEARCH

Plasma and Urine Metabolites Following Dietary Intervention

A BCb b a 10 10 a 6 a a 8 8

4 6 a 6 a

4 4 2

MSRA of pipecolate MSRA 2 2 of S-methylcysteine MSRA MSRA of N-delta-acetylornithine MSRA

0 0 0 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk Control Navy bean Control Navy bean Control Navy bean D EF b b b 10 2.0 35

30 8 1.5 25 6 20 1.0 10 4

MSRA of benzoate MSRA 0.5 5 2

MSRA of eicosapentaenoate (EPA) MSRA 0.0 0 0 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk MSRA of 3-(4-hydroxypheny)propionate MSRA 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk Control Navy bean Control Navy bean Control Navy bean

G Plasma Fold change Fold difference

S-methylcysteine sulfoxide Tryptophan betaine 3-(4-hydorxyphenyl)propionate N-δ-acetylornithine Betaine γ-CEHC 1-arachidonoy1-GPC (20:4n6 Diacylglycerol (16:1/18:2 [2], 16:0/18:3 [1]) 5alpha-pregnan-3beta, 20alpha-diol Linoleoyl-linolenoyl-glycerol (18:2/18:3) [2] monosulfate Oleoyl-linoleoyl-glycerol (18:1/18:2) [1] Choline Docosahexaenoate (22:6n3) (DHA) Oleoyl-oleoyl-glycerol (18:1/18:1) [1] 2, 3-dihydroxy-2- Docosahexaenoylcarnitine (C22:6) Oleoyl-oleoyl-glycerol (18:1/18:1) [2] methylbuyrate Eicosapentaenoate (20:5n3) (EPA) Palmitoyl-oleoyl-glycerol (16:0/18:1) [1] Pipecolate Linoleoylcarnitine (C18:2) Palmitoyl-oleoyl-glycerol (16:0/18:1) [2] S-methylcysteine Palmitoyl-palmitoyl-glycerol (16:0/16:0) [2] 5,6-dihydrothymine Palmitoyl-linoleoyl-glycerol (16:0/18:2) [2] gamma-glutamy1-2-aminobutyrate Cortisone Benzoate Hippurate, Piperine S-allylcysteine 4-allylphenol sulfate 4-hydroxycoumarin

Figure 1. Median-scaled relative abundance (MSRA) of selected amino acids, gut microbial metabolites, lipid and phytochemicals for control or navy bean at 0, 2, and 4 weeks after dietary intervention in plasma. A, Pipecolate; B, S-methylcysteine; C, N-delta-acetylornithine; D, 3–4(hydroxyphenyl)propionate; E, eicosapentaenate; F, benzoate. a, significant fold change; b, significant fold difference (P < 0.05). G, Venn diagram summarizes plasma metabolites with statistical significance to navy bean intake. The left circle includes metabolites with fold changes at two or four weeks of navy bean intake compared with baseline, and the right circle shows metabolites with fold differences between the control and bean groups (4-week time point). Overlapping metabolites of statistical significance in each analysis are listed in the center.

AACRJournals.org Cancer Prev Res; 14(4) April 2021 501

Zarei et al.

Table 2. Navy bean consumption revealed distinct plasma at 2 and/or 4 weeks compared with baseline (P < 0.05, q < 0.2). metabolites after 4 weeks compared with a control intervention All data are presented as the mean fold change in the median- for 4 weeks. scaled relative abundance of each metabolite. Figure 2A–D Fold differencec illustrates median-scaled relative abundance for gentisate, Navy hippurate, salicylate, and N2,N5-diacetlyornithine that were Metabolite HMDB bean P significantly increased at week 4 when compared with their baseline in the intervention group. Individual changes over Gut microbial metabolites 2,3-dihydroxy-2-methylbutyrate HMDB29576 1.34 0.036 time are shown in Supplementary Fig. S4. These metabolites 3-(4-hydroxyphenyl)propionateb HMDB02199 4.48 0.031 were chosen for their potential colorectal cancer protection, Amino acids and the four metabolites were identified in navy bean and study Betaineb HMDB00043 0.84 0.032 a,b meals and snacks metabolomes. Pipecolate HMDB00070 3.88 0.000 fi S-methylcysteinea,b HMDB02108 1.92 0.006 Table 4 lists 16 urine metabolites with signi cant differences Lipids in the abundance between intervention and control for the 20 1-arachidonoyl-GPC (20:4n6) HMDB10395 0.80 0.045 participants at 4 weeks (P < 0.05, q < 0.2). For all participants, 5alpha-pregnan-3beta,20alpha-diol — 0.12 0.043 data are presented as the mean fold difference between inter- monosulfate vention and control at 4 weeks. The 14 urine metabolites Cholinea,b HMDB00097 0.80 0.008 Docosahexaenoate (22:6n3) (DHA)a HMDB02183 0.74 0.014 increased in abundance for the navy bean group were Docosahexaenoylcarnitine (C22:6) — 0.63 0.043 1-methylhistamine (1.72-fold), cysteine (1.6-fold), N2,N5-dia- Eicosapentaenoate (20:5n3) (EPA)a HMDB01999 0.77 0.039 cetylornithine (1.55-fold), N-acetyltaurine (1.57-fold), pyro- Linoleoylcarnitine (C18:2) HMDB06469 0.71 0.037 glutamine (1.74-fold), S-adenosylhomocysteine (SAH) (2.09- Nucleotide fold), N6-carbamoylthreonyladenosine (1.44-fold), guanosine 5,6-dihydrothymine HMDB00079 0.71 0.022 Peptide (2.81-fold), 2-keto-3-deoxy-gluconate (2.18-fold), 4-hydroxy- gamma-glutamyl-2-aminobutyrateb — 0.30 0.005 hippurate (1.92-fold), alliin (3.34-fold), O-sulfo-L-tyrosine Xenobiotics (1.33-fold), and thioproline (2.66-fold). The three urine a,b Benzoate HMDB01870 0.77 0.029 metabolites that decreased in abundance for the navy S-allylcysteine HMDB34323 4.08 0.019 bean group were dopamine 3-O-sulfate (0.45-fold), indolepro- Abbreviations: GPC, glycerophosphocholine; HMDB, Human Metabolome pionate (0.35-fold), and isovalerylglutamine (0.53-fold). Database. Among these 17 urine metabolites, eight metabolites were aMetabolite also identified from the navy bean metabolome. identified in the navy beans, including cysteine, N2,N5-diace- b fi Metabolite also identi ed from the study food metabolome. tylornithine, pyroglutamine, SAH, N6-carbamoylthreonylade- cValues presented are fold difference of the mean relative abundance between navy bean compared with control at 4 weeks (P < 0.05). Statistically significant nosine, guanosine, 4-hydroxyhippurate, and O-sulfo-L-tyrosin fold differences are bolded. (see Table 4, note “a”) and five urine metabolites were identified in the study meals and snacks metabolome, including cysteine, pyroglutamine, SAH, guanosine, and O-sulfo-L- fi diet-derived metabolites were signi cantly decreased after tyrosine (see Table 4, note “b”). Navy bean–derived metabo- consuming navy beans for 4 weeks when compared with the lites were distinguished for presence and abundance in urine of fi control group (Table 2). Both metabolites were identi ed in the the intervention group compared with control. Figure 2D–F fi navy bean metabolome, while benzoate was also identi ed in illustrates the median-scaled relative abundance for amino the study meals and snacks. Figure 1G shows a Venn diagram acids with anti-inflammatory potential—N2, N5-diacetylor- of 33 plasma metabolites with fold change or fold differences nithine, cysteine, and SAH—which were identified in the navy fi identi ed at 2 or 4 weeks after intervention. The center bean metabolome. The Venn diagram shown in Fig. 2G high- metabolites in the Venn diagram illustrate those with over- lights 23 urine metabolites (statistical significance by fold fi lapping signi cance from each of the statistical comparisons change or fold difference) in urine at 2 or 4 weeks after dietary applied, and were pipecolate, S-methylcyteine, and 2,3-dihy- bean intervention. N2,N5-diacetylornithine was the only urine droxy-2methylbutyrate. metabolite with overlap in the analyses at 4 weeks in the intervention group. Dietary modulation of urine metabolome in colorectal cancer survivors after 2 and 4 weeks There were 703 identified metabolites from urine of 20 Discussion participants at all time points. The urine metabolites were This study established that increased navy bean consump- classified in the metabolic pathways as shown for plasma. tion for 4 weeks produced metabolic shifts in plasma and urine Supplementary Table S4 lists the relative abundance of all the profiles of overweight and obese colorectal cancer survivors at identified urine metabolites and the respective metabolic path- risk for recurrence. The plasma and urine metabolites that ways for the control and navy bean groups. Table 3 lists 10 showed changes in abundance over time or differences between urine metabolites that were significantly modulated in either the two study groups were also compared with the phytochem- the control, navy bean, or both study groups for 20 participants ical compounds that originate from navy beans (including

502 Cancer Prev Res; 14(4) April 2021 CANCER PREVENTION RESEARCH

Downloaded from cancerpreventionresearch.aacrjournals.org on September 28, 2021. © 2021 American Association for Cancer Research. Downloaded from AACRJournals.org Published OnlineFirstDecember24,2020;DOI:10.1158/1940-6207.CAPR-20-0270 cancerpreventionresearch.aacrjournals.org Table 3. Urine metabolites following consumption of control foods or navy bean–based foods for 2 and 4 weeks. Post-intervention metabolite proﬁles were compared with baseline (0 weeks).

Fold changed Control group Navy bean group 2 wk/ 4wk/ 2wk/ 4wk/ Metabolite 0wk P value q value 0wk P value q value 0wk P value q value 0wk P value q value Amino acids Histidine N-acetyl-1-methylhistidinec 0.78 0.072 0.791 0.62 0.000 0.147 1.00 0.998 1 1.00 0.975 1 Lecine, isoleucine, and valine N6,N6,N6-trimethyllysine 0.93 0.561 1 0.64 0.000 0.147 0.98 0.869 1 0.95 0.650 1 Methionine, cysteine, N-methyltaurine 2.70 0.018 0.584 2.05 0.084 0.534 4.20 0.001 0.111 2.29 0.048 0.927 SAM, and taurine b 1.19

Cancer Research. Urea cycle; arginine and proline N2,N5-diacetylornithine 1.07 0.511 1 0.093 0.544 1.38 0.003 0.315 1.62 0.000 0.007 pro-hydroxy-pro 0.86 0.207 1 0.59 0.000 0.020 0.98 0.880 1 1.13 0.296 1 Carbohydrate Pentose Ribonatea,b 1.52 0.033 0.684 1.58 0.022 0.398 2.15 0.000 0.049 1.83 0.002 0.324 Cofactors and vitamins on September 28, 2021. © 2021American Association for Tocopherol gamma-CEHC glucuronidec 0.76 0.137 1 0.86 0.423 0.817 0.52 0.000 0.109 0.58 0.004 0.405 c alpha-CEHC glucuronide 0.79 0.196 1 0.97 0.867 0.996 0.49 0.000 0.049 0.71 0.061 0.927 Intervention Dietary Following Metabolites Urine and Plasma alpha-CEHC sulfate 0.71 0.061 0.751 0.74 0.089 0.536 0.50 0.000 0.049 0.73 0.083 0.935 Xenobiotics Food component/plant 4-allylphenol sulfate 3.95 0.000 0.000 5.09 0.000 0.000 4.07 0.000 0.000 4.25 0.000 0.000

Abbreviation: CEHC, carboxyethyl-hydroxychromans. aMetabolite also identified from the navy bean metabolome. bMetabolite also identified from the study food metabolome. cIndicates that the metabolite identified was not made against a purified chemical standard. dMetabolite also identified from the study food metabolome. Values presented are fold change of the mean relative abundance within control and navy bean at 2 and 4 weeks compared with their baselines (P < 0.05) and (q < 0.2). Statistically significant fold changes are bold and shaded. Light shaded represent metabolites and pathways that showed trends toward significance (0.05 < P < 0.10). acrPe e;1()Arl2021 April 14(4) Res; Prev Cancer 503 Published OnlineFirst December 24, 2020; DOI: 10.1158/1940-6207.CAPR-20-0270

Zarei et al.

A BC 5 3 40

4 a 30 a 2 a 3 20 2 1 MSRA of salicylate MSRA MSRA of gentisate MSRA MSRA of hippurate MSRA 10 1

0 0 0 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk Control Navy bean Control Navy bean Control Navy bean

DEFbb 5 b 3 a 3 4

2 3 2

2 1

MSRA of cysteine MSRA 1 1

MSRA N2, N5-diacetylornithine MSRA 0 0 0 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk

0 wk 2 wk 4 wk 0 wk 2 wk 4 wk of S-adenosylhomocysteine (SAH) MSRA 0 wk 2 wk 4 wk 0 wk 2 wk 4 wk Control Navy bean Control Navy bean Control Navy bean G Urine

Fold change Fold difference

1-methylhistamine S-adenosylhomocysteine (SAH) Cysteine N-methyltaurine Dopamine 3-O-sulfate pro-hydroxy-pro Indolepropionate Ribonate Isovalerylglutamine gamma-CEHC glucuronide N2,N5- N-acetyltaurine alpha-CEHC glucuronide diacetylornithinea N-methylleucineb alpha-CEHC sulfate Pyroglutamine 4-allylphenol sulfate Guanosine 2-keto-3-deoxy-gluconate 4-hydroxyhippuratea N6-carbamoylthreony ladenosine Alliin O-sulfo-L-tyrosine Thioproline

Figure 2. Median-scaled relative abundance (MSRA) of selected amino acids and phytochemicals for control or navy bean at 0, 2, and 4 weeks after dietary intervention in urine. A, Gentisate; B, hippurate; C, salicylate; D, N2,N5-diacetlyornithine; E, cysteine; F, S-adenosylhomocysteine. a, significant fold change; b, significant fold difference (P < 0.05). G, Venn diagram summarizes urine metabolites with statistical significance to navy bean intake. The left circle includes metabolites with fold changes at two or four weeks of navy bean intake compared with baseline, and the right circle shows metabolites with fold differences between the control and bean groups (4-week time point). Overlapping metabolites of statistical significance in each analysis are listed in the center.

the meals and snacks that contained navy beans). This inte- biomarkers in adults. There are also a number of metabolites gration of the navy bean metabolites with the results from discussed below from distinct metabolic pathways that support changes in plasma and urine has established novel relationships novel associations for bean intake with anti-inﬂammatory, between bean exposure in the diet and cancer prevention antioxidant, and anticancer mechanisms that may reduce

504 Cancer Prev Res; 14(4) April 2021 CANCER PREVENTION RESEARCH

Plasma and Urine Metabolites Following Dietary Intervention

Table 4. Navy bean consumption revealed distinct urine on clusters of abnormal tube-like glands in the lining of the metabolites after 4 weeks compared with a control intervention colon and rectum called aberrant crypt foci. A decreased level of for 4 weeks. urinary cysteine warranted discussion as cysteine is a unique Fold differencec thiol containing amino acid that can undergo oxidation/reduc- Navy tion (redox) reactions with antioxidant properties and may Metabolite HMDB bean P reduce inflammation (49, 50). These metabolite changes col- lectively may reduce cell division to decrease risk for colorectal Amino acid 1-Methylhistamine HMDB00898 1.72 0.034 cancer recurrence. Cysteinea,b HMDB00574 1.60 0.048 Pipecolate represents another candidate dietary biomarker Dopamine 3-O-sulfate HMDB06275 0.45 0.025 of bean consumption (38) that has functional relevance to this Indolepropionate HMDB02302 0.35 0.038 prospective cohort investigation because this phytochemical Isovalerylglutamine — 0.53 0.006 N2,N5-diacetylornithineb — 0.020 serves as a precursor to gut microbial metabolism and has anti- 1.55 fl – N-acetyltaurine — 1.57 0.050 in ammatory, antitumor, and antibiotic properties (51 53). 3- N-methylleucineb — 19.60 0.000 (4-hydroxyphenyl)propionate was also increased following Pyroglutaminea,b — 1.74 0.006 navy bean intake and is a small phenolic acid produced from a,b S-adenosylhomocysteine (SAH) HMDB00939 2.09 0.027 biotransformation of large polyphenols by human colonic Nucleotide fl N6-carbamoylthreonyladenosinea HMDB41623 1.44 0.037 micro ora (54). Phenolic acids exert effects on carcinogen Guanosinea,b HMDB00133 2.81 0.034 bioactivation, cell signaling, cell-cycle regulation, angiogenesis, Xenobiotics oxidative stress, and inflammation (55, 56). 2,3-dihydroxy-2- 2-keto-3-deoxy-gluconate HMDB01353 2.18 0.018 methylbutyrate is a normal organic acid constituent of the gut a 4-hydroxyhippurate HMDB13678 1.92 0.036 and the increase measured herein for plasma of navy bean Alliin HMDB33592 3.34 0.034 a,b fi fl O-sulfo-L-tyrosine — 1.33 0.026 consumers suggested bene cial roles against in ammation and Thioproline — 2.66 0.009 alterations to microbiota. Polyunsaturated fatty acids (PUFA), such as EPA (20:5n3) that is a navy bean metabolite, were Abbreviation: HMDB, Human Metabolome Database. shown to exert anticancer activities in various stages of cancer a fi Metabolite also identi ed from the navy bean metabolome. progression, such as cell proliferation, metastasis, inflamma- bMetabolite also identified from the study food metabolome. cValues presented are fold difference of the mean relative abundance between tion, cell survival, and angiogenesis (57, 58). navy bean compared with control at 4 weeks (P < 0.05). Statistically significant This next group of metabolites has mechanistic links to increased fold differences are in bold. diverse epigenetic modifications involved in cancer progression. Benzoic acid (navy bean metabolite) has naturally occur- cancer risk. Thus, plasma and urine metabolites have the ring derivatives with potential to suppress cancer cell growth by potential to dually serve as biomarkers of dietary navy blocking histone deacetylase, a key enzyme involved in tumor beans exposure and metabolic health to prevent and control regulatory gene expression (59). The decreased urinary SAH colon cancer. following daily bean intake is important as an inhibitor of Six plasma metabolites modulated by navy bean consump- methyltransferases, the key enzymes for DNA methylation that tion and that have cancer control and prevention relevance may cause aberrant DNA methylation patterns in colon can- from this study are S-methylcysteine, pipecolate, 3-(4-hydro- cer (60). Higher abundance of SAH could be important to block xyphenyl)propionate, N-delta-acetylornithine, S-allylcysteine, aberrant DNA methylation and dysregulation of histones and and 2,3-dihydroxy-2-methylbutyrate (Tables 1 and 2; Fig. 2). gene expression that stimulate carcinogenesis (61). Both cho- S-methylcysteine, a water-soluble organosulfur compound, line and betaine were decreased in plasma from this study, and may be a biomarker of dietary exposure as it was previously the roles in colorectal cancer merit attention as they are identified from common bean (Phaseolus vulgaris; refs. 38, 39). essential to one-carbon metabolism (62–64). The lack of Cysteine-related components have also demonstrated suppres- choline was shown to promote fatty liver disease, DNA hypo- sion of chemical induced carcinogenesis (40). S-methylcysteine methylation, and tumor development (in the absence of carci- was reported to inhibit the formation of glutathione S- nogens; ref. 65) and an epidemiologic study showed that the transferase placental form–positive foci (GST-P-positive foci), risk of colorectal adenomas statistically increased with increas- which are markers for preneoplastic lesions (41) and was also ing choline intake (65). Betaine is notably an oxidation product found to inhibit ornithine decarboxylase (ODC), an enzyme of choline, and a methyl donor that has nutrient contributions that is upregulated in a wide variety of cancers (42). ODC is for the prevention of chronic diseases (66). essential in the biosynthesis of polyamines (43), namely, anti- There were multiple metabolites identified in this study with oxidants important for stabilizing DNA structure (44). Lack of associated mechanisms to colorectal cancer prevention that ODC has been found to cause cell apoptosis in embryonic mice, relate to metabolism by the gut microbiota. These may include, induced by DNA damage (45). S-allylcysteine was also but were not limited to, 3-(4-hydroxyphenyl)lactate, N2,N5- increased in plasma (Table 2) and has antioxidant proper- diacetylornithine, hippurate, and salicylate. High urinary levels ties (46, 47), cholesterol-lowering (48), and anticancer effects of 3-(4-hydroxyphenyl)lactate were found in individuals with

AACRJournals.org Cancer Prev Res; 14(4) April 2021 505

Zarei et al.

tyrosine malabsorption and bacterial overgrowth (67, 68), and Ethics Approval and Consent to Participate our data suggest improved tyrosine absorption and balanced All procedures performed in this study were in accordance with the gut microﬂora with navy bean intake. Dietary bean intake ethical standards of the Colorado State University and University of inﬂuences on metabolism by gut microbiota were also shown Colorado Health–North Institutional Review Board, and with the 1964 by changes to indole propionate, 4-hydroxyhippurate, and 3- Helsinki Declaration and its 2013 amendments. Informed consent was (4-hydroxyphenyl)propionate that have been reported previ- obtained from all individual participants before included in the study. This ously in stool (3, 35). trial was registered at clinicaltrials.gov under NCT01929122. The data sets used and/or analyzed during the current study are available as Supporting The limitations of this study include the small cohort size. Information along with this article. The acute (one-month) study duration of the diet intervention may also be considered a limitation, yet these metabolite Authors’ Disclosures changes detected early-on following dietary change may have long-term impacts. The study participants were free-living No disclosures were reported. individuals and the study-provided meals and snacks contributed 35% of daily intake. Follow-up investigations will utilize Authors’ Contributions a larger sample size, and a longer duration of elevated navy bean I. Zarei: Data curation, writing–original draft. B.A. Baxter: Data curation, exposure may be needed to observe colon tissue metabolic writing–review and editing. R.C. Oppel: Writing–review and editing. changes that will improve cancer control, treatment, and E.C. Borresen: Conceptualization, methodology, and project administra- prevention outcomes. tion. R.J. Brown: Conceptualization, methodology, and project administration. E.P. Ryan: Conceptualization, data curation, visualization, methodology, writing–original draft, project administration, and writing–review Conclusion and editing.

This study supports that an elaborate array of interactions Acknowledgments between food, host digestion, and gut microbial metabolism Funding for this study was provided by NIH–NCI (1R21CA161472), occurs in overweight and obese colorectal cancer survivors. National Institute of Food and Agriculture (NIFA; 2016-67001-24538), Plasma and urine metabolites that merit attention for valida- the University of Colorado Cancer Center–Division of Cancer Preven- tion after the daily consumption of navy beans in adults are 2,3- tion and Control pilot award program, and the Dry Bean Health dihydroxy-2-methylbutyrate, pipecolate, and S-methylcysteine Research Group. The authors appreciate the collaboration of University – (in plasma) and N2,N5-diacetylornithine and 4-hydroxyhip- of Colorado Health North Cancer Center and the Oncology Clinical purate (in urine). The metabolite results shown for blood and Research team. The authors wish to thank the study subjects for their participation. urine were integrated with the analysis of phytochemicals originating from the navy beans, and revealed the amino acid The costs of publication of this article were defrayed in part by the and lipid metabolism pathways as mechanistic targets to reduce payment of page charges. This article must therefore be hereby marked colorectal cancer recurrence following dietary bean interven- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate tion. Assessing the bioactivity and bioavailability of navy bean– this fact. derived phytochemicals in human colon tissue is one future analytical approach to validate food-exposure biomarkers with Received May 28, 2020; revised August 28, 2020; accepted December 21, primary and secondary colorectal cancer prevention outcomes. 2020; published ﬁrst December 24, 2020.

References 1. Bennink MR. Consumption of black beans and navy beans (Phaseolus 6. Zhang X, Browman G, Siu W, Basen-Engquist KM, Hanash vulgaris) reduced azoxymethane-induced colon cancer in rats. SM, Hoffman KL, et al. The BE GONE trial study protocol: a Nutr Cancer 2002;44:60–5. randomized crossover dietary intervention of dry beans target- 2. Moreno-Jimenez MR, Lopez-Barraza R, Cervantes-Cardoza V, Perez- ing the gut microbiome of overweight and obese patients with Ramírez IF, Reyna-Rojas JA, Gallegos-Infante JA, et al. Mechanisms a history of colorectal polyps or cancer. BMC Cancer 2019;19: associated to apoptosis of cancer cells by phenolic extracts from two 1233. canned common beans varieties (Phaseolus vulgaris L.). J Food Bio- 7. Zhu B, Sun Yu, Qi Lu, Zhong R, Miao X. Dietary legume consumption chem 2019;43:e12680. reduces risk of colorectal cancer: evidence from a meta-analysis of 3. Baxter BA, Oppel RC, Ryan EP. Navy beans impact the stool meta- cohort studies. Sci Rep 2015;5:8797. bolome and metabolic pathways for colon health in cancer survivors. 8. Slavin JL. Dietary fiber and body weight. Nutrition 2005;21: Nutrients 2018;11:28. 411–8. 4. Howe GR, Benito E, Castelleto R, Cornee J, Esteve J, Gallagher RP, et al. 9. Du H, van der A DL, Boshuizen HC, Forouhi NG, Wareham NJ, Dietary intake of fiber and decreased risk of cancers of the colon and Halkjaer J, et al. Dietary fiber and subsequent changes in body weight rectum: evidence from the combined analysis of 13 case-control and waist circumference in European men and women. Am J Clin studies. J Natl Cancer Inst 1992;84:1887–96. Nutr 2009;91:329–36. 5. Cummings JH, Bingham SA, Heaton KW, Eastwood MA. Fecal 10. Pereira MA, Ludwig DS. Dietary fiber and body-weight regulation: weight, colon cancer risk, and dietary intake of nonstarch polysac- observations and mechanisms. Pediatr Clin North Am 2001;48: charides (dietary fiber). Gastroenterology 1992;103:1783–9. 969–80.

506 Cancer Prev Res; 14(4) April 2021 CANCER PREVENTION RESEARCH

Plasma and Urine Metabolites Following Dietary Intervention

11. Luhovyy BL, Mollard RC, Panahi S, Nunez MF, Cho F, Anderson GH. among never-smoking women in China. JAMA Netw Open 2019;2: Canned navy bean consumption reduces metabolic risk factors asso- e1911970. ciated with obesity. Can J Diet Pract Res 2015;76:33–7. 32. Playdon MC, Sampson JN, Cross AJ, Sinha R, Guertin KA, Moy KA, 12. Ferreira H, Vasconcelos M, Gil AM, Pinto E. Benefits of pulse et al. Comparing metabolite profiles of habitual diet in serum and consumption on metabolism and health: a systematic review of urine. Am J Clin Nutr 2016;104:776–89. randomized controlled trials. Crit Rev Food Sci Nutr 2020;61:85–96. 33. Zambrana LE, McKeen S, Ibrahim H, Zarei I, Borresen EC, Doumbia L, 13. McGinley JN, Fitzgerald VK, Neil ES, Omerigic HM, Heuberger AL, et al. Rice bran supplementation modulates growth, microbiota and Weir TL, et al. Pulse crop effects on gut microbial populations, metabolome in weaning infants: a clinical trial in Nicaragua and Mali. intestinal function, and adiposity in a mouse model of diet-induced Sci Rep 2019;9:13919. obesity. Nutrients 2020;12:593. 34. Borresen EC, Brown DG, Harbison G, Taylor L, Fairbanks A, O’Malia 14. Ludwig DS, Pereira MA, Kroenke CH, Hilner JE, Van Horn L, Slattery J, et al. A randomized controlled trial to increase navy bean or rice bran ML, et al. Dietary fiber, weight gain, and cardiovascular disease risk consumption in colorectal cancer survivors. Nutr Cancer 2016;68: factors in young adults. JAMA 1999;282:1539–46. 1269–80. 15. King DE. Dietary fiber, inflammation, and cardiovascular disease. 35. Sheflin AM, Borresen EC, Kirkwood JS, Boot CM, Whitney AK, Lu S, Mol Nutr Food Res 2005;49:594–600. et al. Dietary supplementation with rice bran or navy bean alters gut 16. Curran J. The nutritional value and health benefits of pulses in bacterial metabolism in colorectal cancer survivors. Mol Nutr Food relation to obesity, diabetes, heart disease and cancer. Br J Nutr Res 2017;61:10.1002/mnfr.201500905. 2012;108:S1–S2. 36. Borresen E, Gundlach K, Wdowik M, Rao S, Brown R, Ryan E. 17. Hughes JS, Ganthavorn C, Wilson-Sanders S. Dry beans inhibit Feasibility of increased navy bean powder consumption for primary azoxymethane-induced colon carcinogenesis in F344 rats. J Nutr and secondary colorectal cancer prevention. Curr Nutr Food Sci 2014; 1997;127:2328–33. 10:112–9. 18. Hangen L, Bennink MR. Consumption of black beans and navy beans 37. Zarei I, Brown DG, Nealon NJ, Ryan EP. Rice bran metabolome (Phaseolus vulgaris) reduced azoxymethane-induced colon cancer in contains amino acids, vitamins & cofactors, and phytochemicals with rats. Nutr Cancer 2002;44:60–5. medicinal and nutritional properties. Rice 2017;10:24. 19. Mitchell DC, Lawrence FR, Hartman TJ, Curran JM. Consumption of 38. Perera T, Young MR, Zhang Z, Murphy G, Colburn NH, Lanza E, et al. dry beans, peas, and lentils could improve diet quality in the US Identification and monitoring of metabolite markers of dry bean population. J Am Diet Assoc 2009;109:909–13. consumption in parallel human and mouse studies. Mol Nutr Food 20. Bobe G, Barrett KG, Mentor-Marcel RA, Saffiotti U, Young MR, Res 2015;59:795–806. Colburn NH, et al. Dietary cooked navy beans and their fractions 39. Thompson JF, Morris CJ, Zacharius RM. Isolation of () S-methyl-L- attenuate colon carcinogenesis in azoxymethane-induced ob/ob mice. cysteine from beans (Phaseolus vulgaris). Nature 1956;178:593. Nutr Cancer 2008;60:373–81. 40. Fukushima S, Takada N, Hori T, Min W, Wanibuchi H, Yamamoto S. 21. Mentor-Marcel RA, Bobe G, Barrett KG, Young MR, Albert PS, Suppression of chemical carcinogenesis by water-soluble organosulfur Bennink MR, et al. Inflammation-associated serum and colon markers compounds. J Nutr 2001;131:1049S–53S. as indicators of dietary attenuation of colon carcinogenesis in ob/ob 41. Takada N, Yano Y, Wanibuchi H, Otani S, Fukushima S. S- mice. Cancer Prev Res 2009;2:60–9. methylcysteine and cysteine are inhibitors of induction of glutathione 22. Armitage EG, Ciborowski M. Applications of metabolomics in cancer S-transferase placental form-positive foci during initiation and pro- studies. Adv Exp Med Biol 2017;965:209–34. motion phases of rat hepatocarcinogenesis. Jpn J Cancer Res 1997;88: 23. Spratlin JL, Serkova NJ, Eckhardt SG. Clinical applications of meta- 435–42. bolomics in oncology: a review. Clin Cancer Res 2009;15:431–40. 42. Vijayaraghavan M, Wanibuchi H, Takada N, Yano Y, Otani S, 24. O’Gorman A, Brennan L. The role of metabolomics in determination Yamamoto S, et al. Inhibitory effects of S-methylcysteine and cysteine of new dietary biomarkers. Proc Nutr Soc 2017;76:295–302. on the promoting potential of sodium phenobarbital on rat liver 25. Gibbons H, Brennan L. Metabolomics as a tool in the identification of carcinogenesis. Jpn J Cancer Res 2000;91:780–5. dietary biomarkers. Proc Nutr Soc 2017;76:42–53. 43. Kern AD, Oliveira MA, Coffino P, Hackert ML. Structure of mam- 26. Thompson MD, Thompson HJ, Brick MA, McGinley JN, Jiang W, Zhu malian ornithine decarboxylase at 1.6 Å resolution: stereochemical Z, et al. Mechanisms associated with dose-dependent inhibition of rat implications of PLP-dependent amino acid decarboxylases. Structure mammary carcinogenesis by dry bean (Phaseolus vulgaris, L.). J Nutr 1999;7:567–81. 2008;138:2091–7. 44. Rato C, Amirova SR, Bates DG, Stansfield I, Wallace HM. Transla- 27. Campos-Vega R, Guevara-Gonzalez RG, Guevara-Olvera BL, Dave tional recoding as a feedback controller: systems approaches reveal Oomah B, Loarca-Pina~ G. Bean (Phaseolus vulgaris L.) polysacchar- polyamine-specific effects on the antizyme ribosomal frameshift. ides modulate gene expression in human colon cancer cells (HT-29). Nucleic Acids Res 2011;39:4587–97. Food Res Int 2010;43:1057–64. 45. Pendeville H, Carpino N, Marine J-C, Takahashi Y, Muller M, Martial 28. Rondini EA, Bennink MR. Microarray analyses of genes differentially JA, et al. The ornithine decarboxylase gene is essential for cell survival expressed by diet (Black Beans and Soy Flour) during azoxymethane- during early murine development. Mol Cell Biol 2001;21:6549–58. induced colon carcinogenesis in rats. J Nutr Metab 2012;2012:351796. 46. Numagami Y, Ohnishi ST. S-allylcysteine inhibits free radical pro- 29. Daniell EL, Ryan EP, Brick MA, Thompson HJ. Dietary dry bean duction, lipid peroxidation and neuronal damage in rat brain ischemia. effects on hepatic expression of stress and toxicity-related genes in rats. J Nutr 2001;131:1100S–5S. Br J Nutr 2012;108:S37–45. 47. Colín-Gonzalez AL, Santana RA, Silva-Islas CA, Chanez-Cardenas 30. Auslander N, Yizhak K, Weinstock A, Budhu A, Tang W, Wang XW, ME, Santamaría A, Maldonado PD. The antioxidant mechanisms et al. A joint analysis of transcriptomic and metabolomic data uncovers underlying the aged garlic extract- and S-allylcysteine-induced pro- enhanced enzyme-metabolite coupling in breast cancer. Sci Rep 2016; tection. Oxid Med Cell Longev 2012;2012:16. 6:29662. 48. Liu L, Yeh Y-Y. Cholesterol-lowering effect of garlic extracts and 31. Seow WJ, Shu X-Ou, Nicholson JK, Holmes E, Walker DI, Hu W, et al. organosulfur compounds: human and animal studies. J Nutr 2001;131: Association of untargeted urinary metabolomics and lung cancer risk 989S–93S.

AACRJournals.org Cancer Prev Res; 14(4) April 2021 507

Zarei et al.

49. Iyer SS, Accardi CJ, Ziegler TR, Blanco RA, Ritzenthaler JD, Rojas M, 59. Anantharaju PG, Reddy BD, Padukudru MA, Kumari Chitturi ChM, et al. Cysteine redox potential determines pro-inflammatory IL-1beta Vimalambike MG, Madhunapantula SV. Naturally occurring benzoic levels. PLoS One 2009;4:e5017. acid derivatives retard cancer cell growth by inhibiting histone dea- 50. Hatono S, Jimenez A, Wargovich MJ. Chemopreventive effect of S- cetylases (HDAC). Cancer Biol Ther 2017;18:492–504. allylcysteine and its relationship to the detoxification enzyme gluta- 60. Lao VV, Grady WM. Epigenetics and colorectal cancer. Nat Rev thione S-transferase. Carcinogenesis 1996;17:1041–4. Gastroenterol Hepatol 2011;8:686–700. 51. He M. Pipecolic acid in microbes: biosynthetic routes and enzymes. 61. Shrubsole MJ, Wagner C, Zhu X, Hou L, Loukachevitch LV, Ness RM, J Ind Microbiol Biotechnol 2006;33:401–7. et al. Associations between S-adenosylmethionine, S-adenosylhomo- 52. Natarajan SK, Muthukrishnan E, Khalimonchuk O, Mott JL, Becker cysteine, and colorectal adenoma risk are modified by sex. Am J Cancer DF. Evidence for pipecolate oxidase in mediating protection against Res 2014;5:458–65. hydrogen peroxide stress. J Cell Biochem 2017;118:1678–88. 62. Ziegler RG, Lim U. One-carbon metabolism, colorectal carcinogenesis, 53. Broquist HP. Lysine-pipecolic acid metabolic relationships in chemoprevention—with caution. J Natl Cancer Inst 2007;99:1214–5. microbes and mammals. Annu Rev Nutr 1991;11:435–48. 63. Zeisel SH, da Costa KA. Choline: an essential nutrient for public health. 54. Rios LY, Gonthier M-P, Remesy C, Mila I, Lapierre C, Lazarus SA, et al. Nutr Rev 2009;67:615–23. Chocolate intake increases urinary excretion of polyphenol-derived 64. Craig SA. Betaine in human nutrition. Am J Clin Nutr 2004;80:539–49. phenolic acids in healthy human subjects. Am J Clin Nutr 2003;77: 65. Cho E, Willett WC, Colditz GA, Fuchs CS, Wu K, Chan AT, et al. 912–8. Dietary choline and betaine and the risk of distal colorectal adenoma in 55. Ward NC, Croft KD, Puddey IB, Hodgson JM. Supplementation with women. J Natl Cancer Inst 2007;99:1224–31. grape seed polyphenols results in increased urinary excretion of 66. Ueland PM. Choline and betaine in health and disease. J Inherit Metab 3-hydroxyphenylpropionic acid, an important metabolite of Dis 2011;34:3–15. proanthocyanidins in humans. J Agric Food Chem 2004;52:5545–9. 67. Spaapen LJM, Ketting D, Wadman SK, Bruinvis L, Duran M. Urinary 56. Le Marchand L. Cancer preventive effects of flavonoids—a review. D-4-hydroxyphenyllactate, D-phenyllactate and D-2-hydroxyiso- Biomed Pharmacother 2002;56:296–301. caproate, abnormalities of bacterial origin. J Inherit Metab Dis 57. Volpato M, Hull MA. Omega-3 polyunsaturated fatty acids as adjuvant 1987;10:383–90. therapy of colorectal cancer. Cancer Metastasis Rev 2018;37:545–55. 68. Beloborodova N, Bairamov I, Olenin A, Shubina V, Teplova V, 58. Park J-M, Kwon S-H, Han Y-M, Hahm Ki-B, Kim E-H. Omega-3 Fedotcheva N. Effect of phenolic acids of microbial origin on pro- polyunsaturated Fatty acids as potential chemopreventive agent for duction of reactive oxygen species in mitochondria and neutrophils. gastrointestinal cancer. J Cancer Prev 2013;18:201–8. J Biomed Sci 2012;19:89.

508 Cancer Prev Res; 14(4) April 2021 CANCER PREVENTION RESEARCH

Plasma and Urine Metabolite Profiles Impacted by Increased Dietary Navy Bean Intake in Colorectal Cancer Survivors: A Randomized-Controlled Trial

Iman Zarei, Bridget A. Baxter, Renee C. Oppel, et al.

Cancer Prev Res 2021;14:497-508. Published OnlineFirst December 24, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/1940-6207.CAPR-20-0270

Supplementary Access the most recent supplemental material at: Material http://cancerpreventionresearch.aacrjournals.org/content/suppl/2020/12/24/1940-6207.CAPR-20-0270.DC 1

Cited articles This article cites 67 articles, 10 of which you can access for free at: http://cancerpreventionresearch.aacrjournals.org/content/14/4/497.full#ref-list-1

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerpreventionresearch.aacrjournals.org/content/14/4/497. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.