UNDERSTANDING THE ROLE OF BROCCOLI SPROUT BIOACTIVES AND THE GUT MICROBIOME IN THE SUPRESSION OF COLON TUMORIGENESIS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: AFRI COMPETITIVE GRANT
• Reporting Frequency: Annual
• Accession No.: 1028297
• Grant No.: 2018-67017-36797
• Cumulative Award Amt.: $134,293.80
• Proposal No.: 2021-12958
• Project Start Date: Sep 1, 2021
• Project End Date: Feb 28, 2023
• Grant Year: 2022
• Program Code: [A1341]- Food Safety, Nutrition, and Health: Function and Efficacy of Nutrients

Recipient Organization
UNIVERSITY OF MAINE
(N/A)
ORONO,ME 04469

Performing Department
(N/A)

Non Technical Summary
There is increasing evidence that diet and the gut microbiota have significant impact on an individual's colorectal cancer risk. Cruciferous vegetables have anti-tumor activities, and the microbiota can contribute to the generation of bioactive metabolites. As diet can also affect the composition of the gut microbiota, it is possible that exposure to cruciferous vegetables may induce changes in the gut microbiota that promote intestinal homeostasis. Our long-term goal is to understand the interrelationship between the gut microbiota and the tumor suppressive activities of broccoli sprouts. We propose three objectives. For Objective 1, we will examine the effects of different broccoli sprout diets and their major metabolites on colon tumorigenesis by using two different mouse models. Objective 2 will help determine the role of the gut microbiota on the metabolism of dietary glucosinolates (GSLs) derived from cruciferous vegetables. We will evaluate the effects of GSL-rich diets on the composition of the gut microbiome, and the effect of the microbiota on the metabolism of GSLs. Finally, objective 3 will determine whether the protective effect of broccoli sprouts is mediated by the gut microbiota through the use of germfree mice and microbiome transfer experiments. This project directly addresses the function and efficacy of nutrients in broccoli sprouts in reducing susceptibility to colon tumorigenesis and will lay the groundwork for the development of strategies that involve modulation of the diet and microbiota for colon cancer prevention.

Animal Health Component

0%

Research Effort Categories

Basic

100%

Applied

0%

Developmental

0%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
702	3840	1010	100%

Knowledge Area
702 - Requirements and Function of Nutrients and Other Food Components;

Subject Of Investigation
3840 - Laboratory animals;

Field Of Science
1010 - Nutrition and metabolism;

Keywords

broccoli

broccoli sprout

colon cancer

gut microbiome

inflammation

Goals / Objectives
Our long-term goalis to understand the interrelationship between the gut microbiota and the tumor suppressive activities of cruciferous vegetables, which will enable the development of novel strategies for colon cancer chemoprevention through dietary and microbial manipulations. The following specific objectives are proposed:Objective 1:To examine the effects of different broccoli sprout diets and major GSLs and ITCs on the development of colon tumors.1.1 To determine the tumor-reducing activities of major GSLs and ITCs found in broccoli/broccoli sprouts using two different mouse models of colon cancer.1.2 To investigate a whole food approach using differently prepared broccoli sprout diets in reducing colon tumor susceptibility.Objective 2:To determine the role of the gut microbiota on the bioactivity of dietary GSLs.2.1 To evaluate gut microbiota changes induced by GSL/ITC-rich diets.2.2 To examine whether changes in the microbiota induced by exposure to GSLs improve ITC generation.Objective 3:To investigate whether the protective effect of broccoli sprouts is mediated by the gut microbiota.

Project Methods
Objective 1: To examine the effects of different broccoli sprout diets and major GSLs and ITCs on the development of colon tumors.Methods: GSL or ITC diets: Diets will be produced by Envigo and consist of 1) control 5001 diet (LabDiet), 2) 5001 blended with GLR and glucoerucin, and 3) 5001 blended with SFN and erucin. The final molar concentrations of GLR and glucoerucin will approximate levels found in the broccoli sprout diets described below. The molar concentrations of SFN and erucin to be used will be the same as GLR and glucoerucin. Broccoli sprouts diets: We will prepare three different broccoli sprout diets (Raw, Mildheat, Steam) to test their tumor suppressive properties. The mouse 5001 diet will serve as the control diet, and be used as the base for the addition of broccoli sprout preparations that contain either raw, mildly heated at 60oC, or 10-min steamed broccoli sprouts. Freeze-dried broccoli sprout preparations will then be added to the base control so that it constitutes 5%, 10%, or 20% of the diet by weight. Mouse models. We will utilize two different mouse models of colon tumorigenesis: 1) the inflammation-associated colon tumorigenesis model (AOM/DSS model), in which mice are treated with the carcinogen AOM followed by multiple rounds of water containing DSS, which induces inflammation in the colon, and 2) a mutation-driven mouse model of familial adenomatous polyposis (CDX2-Apc model) in which the Apc tumor suppressor gene is conditionally expressed in colon epithelial cells resulting in colon tumor development by 3-4 months of age. Assessment of inflammation and proliferation. For the AOM/DSS model, the kinetics of the development of inflammation and epithelial proliferation will be assessed on days 0, 10, 12, and 26. Colon sections will be stained with H&E for histologic scoring in which the extent of epithelial damage, inflammatory cell infilteration and epithelial hyperplasia are assessed. Levels of epithelial proliferation will be assessed by Ki67 staining of colonic sections. The other half of the colon will be processed for RNA extraction and cDNA synthesis. Gene expression profiling by qPCR will be performed examining the upregulation of various pro-inflammatory and protumorigenic mediators. For the CDX2-Apc model, tumor and normal adjacent tissue will be analyzed for histology and gene expression of proinflammatory and proliferative factors at time of sacrifice (4 months).Statistical Analyses: Statistical differences between groups of mice for tumor counts will be assessed using ANOVA models, two-sample t-tests or generalized linear models. Generalized linear models will be conducted to compare tumor counts between wildtype mice treated with different diets. For gene expression profiling and assessment of epithelial proliferation and apoptosis, 5 mice will be used per time point per group, and Wilcoxon-Mann-Whitney tests and two-way ANOVA models will be applied to test differences. Descriptive statistics of these endpoints will be calculated for each group at each timepoint.Objective 2: To determine the role of the gut microbiota on the bioactivity of dietary GSLs.Methods: 16S rRNA gene sequencing and analysis. Fecal samples will be collected longitudinally from wildtype B6 mice fed the various diets as described in Objective 1 for two weeks prior to tumor induction with AOM/DSS. Microbial genomic DNA will be extracted using the PowerSoil-htp 96 Well Soil DNA Isolation kit. The V4 region of the 16S rRNA gene from each sample will be amplified and sequenced with the Illumina MiSeq, which will be performed by the DNA Sequencing Core at the University of Michigan. To analyze the 16S rRNA sequences from the various samples, we will make use of the mothur software package, which enables one to compare samples based on phylotypes, operational taxonomic units (OTUs), and phylogenetic methods. Principle coordinate analysis will be used to determine similarities in microbial community structures between mice groups before and after the various diets. Analysis of molecular variance (AMOVA) and homogeneity of molecular variance (HOMOVA) will be used to compare bacterial communities between samples. Statistically significant differences in relative abundances of bacterial communities will be assessed using the Metastats function incorporated in and Lds effective size (Lefse) will be used to determine differentially expressed populations that are significantly correlated with specific diet preparations. PICRUST will be used to predict functional activities based on 16S rRNA data. Statistics for microbiome analyses are incorporated in the mothur or R software and will be performed as described above. Measurement of bioactive ITCs and metabolites. We will collect total stool and urine samples from the mice described in Objective 1 to measure the levels of GSLs, ITCs and their metabolites by LC-MS/MS, specifically GLR, SFN, SFN-GSH, SFN-cysteine (Cys), SFN-N-acetylcysteine (NAC) conjugates, glucoerucin, erucin, erucin-GSH, erucin-Cys, and erucin-NAC conjugates. Plasma will be collected as well on days 0, 7 and 14. In addition, concentrations of total ITCs and metabolites will also be quantified by the cyclocondensation reaction and HPLC. To determine if exposure to GSLs induces changes in the relative abundance of microbiota that increases the metabolism of GSLs, fresh bacterial suspensions will be prepared from the feces collected from the same mice on days 0, 1, 6 and 13. Bacteria will be collected from fecal homogenates in phosphate buffered saline after gentle centrifugation to remove debris. An aliquot of the bacterial suspension will be mixed with anaerobic medium containing GLR in an anaerobic chamber at 37oC for 0.5, 1, 2, 4, 6, 8, 12, and 24 hours. We will measure levels of ITCs and metabolites from cultures at these various time points to see if levels of ITCs increase with time.Statistical Analyses: Comparison of levels of GSLs, ITCs, and metabolites obtained from mouse samples will be analyzed by generalized linear models as well as by one-way ANOVA. Two-way ANOVA models will also be used, and results will be considered significantly different for p-values ≤ 0.05.Objective 3: To investigate whether the protective effect of broccoli sprouts is mediated by the gut microbiota.Methods: In order to determine whether diet-induced community-wide microbiome changes alone can lower tumor incidence in the colon, we will conventionalize germ-free mice with the microbiome from SPF WT C57BL/6 mice on the specialized diet that elicits the greatest tumor suppression in our mouse models as determined in Objective 1 or the microbiome of SPF mice fed a control diet. Conventionalization will be performed by transferring fresh feces and bedding from cages of donor SPF mice (after 2 weeks of microbial acclimatization to the selected tumor-suppressive GSL-rich diet identified in Objective 1) to GF mice. Stool samples will be collected and 16S rRNA sequences analyzed to determine how closely the gut microbiome of conventionalized GF mice recapitulates donor mice. Following conventionalization over 2 weeks, we will treat the conventionalized GF mice with AOM/DSS, and tumors will be counted. GF mice will be kept on standard control 5001 diets. Mice conventionalized with donor mice fed control diets will be compared with GF mice that harbor the microbiome established by the GSL/ITC-rich diet for the development of tumors after AOM/DSS treatment. Separate cohorts of GF mice conventionalized with diet-specific microbiomes will be used to obtain colon tissue on days 0, 10, 12, and 26 for evaluation of inflammation histologically, epithelial proliferation by Ki67 staining, and expression of proinflammatory and protumorigenic factors. Statistical significance will be determined using two-sample t-tests and ANOVA models.?

Progress 09/01/21 to 02/28/23

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Some of the results have been published in journals, and other results are being analyzed for submission of new manuscripts. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Objective 1: 1.1We did oral gavage of GLR with SPF C57BL/6 mice for 14 d and observed increasing levels of SFN, the production of which is presumably dependent on the gut microbiota. We gavaged mice with GLR or SFN and evaluated inflammatory responses in the colon after one round of DSS, and did not observe beneficial effect. However, administration of broccoli-enriched diets that are high in GLR reduced the morbid effects of DSS, suggesting that the bioavailability of SFN may be improved when delivered as part of a diet. We evaluated SFN in the CDX2-Apc model of mutation-driven tumorigenesis and found no significant protective effects. 1.2We developed three broccoli sprout preparations (raw, mild heat, and steam). Broccoli sprouts that were mildly heated yielded 10 times more SFN compared to the raw and steamed sprouts, likely due to the inactivation of ESP while myrosinase remains intact. The steamed broccoli sprouts contain high levels of GLR due to the deactivation of GLR-metabolizing enzymes. We examined the protective effects of broccoli sprouts against DSS-induced colitis. The 3% DSS resulted in 10% loss of body weight within 5 d, while all of the broccoli sprout diets prevented the weight loss. The average colon length was shorter in DSS treated mice compared to no-DSS mice, and broccoli sprout feeding (except 5% raw) maintained colon length to normal. The plasma levels of IL-6 that was raised by DSS were significantly reduced in groups fed broccoli sprout diets (except 5% raw diet), as compared to the group on base diet. Using the AOM/DSS model of inflammation-associated tumorigenesis, we found that mice fed mild heat diet had consistently, significantly reduced tumor numbers compared to mice fed the control diet. Steamed diet (5%) also resulted in reduced tumor counts, but not consistently, perhaps due to the dependence on the microbiota to generate SFN from GLR. We did not see consistent reduction in tumors with the raw diet. Interestingly, we observed amelioration of DSS-induced colonic inflammation based on weights and serum cytokine levels with all three preparations, suggesting that the antitumor properties of the mildly heated preparation are not entirely due to anti-inflammatory effect. With the mildly heated diet, we found a dose-dependent relationship with greater reduction in tumor numbers with 10% mildly heated diet compared to 5%. This was also associated with greater resistance to DSS-induced weight loss, suggesting an effect on suppressing inflammation.Using the CDX2-Apc model, we saw no significant effect of 5% steamed broccoli sprout on tumor number, although a trend in reduction of tumor size with steamed broccoli.We also saw no significant difference in tumor numbers with 10% mildly heated broccoli in two separate experiments. It is likely that the primary effect of broccoli diets is its anti-inflammatory effect. Objective 2: 2.1We analyzed the gut microbiota of B6 mice that received oral gavage of GLR. 16S rRNA sequencingrevealed significant differences in the overall community structure before and after 2 wks. At the phylum level, there was significant differences in the relative abundance of Verrucobacteria and Bacteroidetes, but several of the most differentially abundant bacterial populations belonged to Firmicutes. We then examined the microbiome of mice fed 5% raw, steamed and mildly heated diets. For all three diets, the composition of the microbiome changed significantly with increased richness and diversity; but he microbial community structure was different between different diets. We observed that there were common bacterial populations significantly depleted after ingestion of broccoli regardless of the preparation, such as Akkermansia, a mucus-degrader that increases colitis susceptibility in genetically-susceptible hosts, and Erypsipelotrichaceae. The gut microbiome of control diet-fed mice was enriched for Akkermansia. Members of the Porphyromonadaceae were also more differentially abundant in the control diet-fed mice. There were specific bacterial populations that were more abundant after ingestion of 5% mildly heated diet that did not occur with the raw or to the same extent as in the steamed preparation, including several spp of Lachnospiraceae, which are butyrate producers and associated with anti-inflammatory activities and reported to decrease in abundance in patients with colorectal cancers. Steamed diet resulted in relative abundance changes in multiple families including reductions in Verrucobacteriaceae, Porphyromonodaceae, Lactobacilaceae, and Erysipelotrichaceae and increases in Lachnospiraceae. 2.2We used two different base diets. In the LD5001 set, increasing levels of SFN was observed in the feces of mice that were on raw or steam diet for a month. We observed decreasing levels of GLR in the feces of mice that were on raw or mild heat diet, as well as a relatively constant fecal GLR levels in the steam diet group. The decreasing levels of GLR might be correlated with the increasing levels of SFN. In the AIN93G set, increasing levels of SFN and a relatively constant GLR levels were seen in the feces of mice that were on steam or raw diet, similar to the LD5001 set. We also measured the concentrations of GLR and SFN in sections of GI tissues. The SFN content increased along the GI tract, reaching the highest in colon tissue. The GLR was high in the small intestines but low in the colon, supporting the conversion of GLR to SFN by the gut microbiota locally in the GI tract, particularly in the colon. Additionally, we have collected feces from B6 mice that were on steamed diet and are doing anaerobic culturing and isolating bacteria responsible for converting GSLs to ITCs. Objective 3: GF mice were fed control or 5% steamed diet and treated with 1% DSS (a lower concentration given the high mortality of GF to higher concentrations of DSS). As shown in Objective 1, SPF mice on 5% steamed broccoli developed less inflammation in the colon and less weight loss. However, there was no significant difference in the colon length of GF mice on steamed or control diet. Histologic scoring confirmed a lack of significant difference in colon inflammation between steamed and control diet-fed GF mice. These results suggest that the protective effects of the steamed diet are mediated, in part, by the gut microbiota. We also treated GF mice on 5% steamed, mildly heated, and raw broccoli diets with AOM/DSS to induce tumors, and found that whereas tumors were reduced in SPF control mice fed 10% mildly heated broccoli, there was no differences in tumor numbers in GF mice fed control or 10% mildly heated broccoli, strongly suggesting an important role for the gut microbiota in mediating the effects of the mildly heated broccoli despite relatively high levels of SFN. We performed another experiment with the DSS model only to look at the role of the microbiota in mediating the protective effects of 10% mildly heated broccoli using GF mice. We again had high mortality in both groups and the 10% mildly heated broccoli group had greater weight loss, consistent with a potential role for the microbiota in mediating the protective effects. We then performed fecal microbiome transplants (one time gavage) in GF mice using control or 10% mildly heated broccoli fed mice as donors and after stable colonization over a month then treated mice with AOM/DSS to induce tumors. We observed no differences in tumor counts. We performed another fecal microbiome transplant with DSS only, but this time gavaging repeatedly every few days. However, we did not observe significant differences in weight loss, suggesting that the broccoli-induced changes in the microbiome was insufficient for protection against DSS-induced colitis or AOM/DSS-induced tumors. Another possibility is that the fecal microbiome transfers were not complete and did not fully recapitulate the donor microbiome. This possibility is currently being evaluated.

Publications

• Type: Journal Articles Status: Published Year Published: 2022 Citation: 2. Tao Zhang, Johanna Holman, Delaney McKinstry, Bruno Caetano Trindade, Kathryn A. Eaton, Jonny Mendoza-Castrejon, Sharon Ho, Emily Wells, Hebao Yuan, Bo Wen, Duxin Sun, Grace Y. Chen, Yanyan Li. A steamed broccoli sprout diet preparation that reduces colitis via the gut microbiota. Journal of Nutritional Biochemistry. 2022 Nov 9; 112:109215. doi: 10.1016/j.jnutbio.2022.109215.
• Type: Journal Articles Status: Published Year Published: 2022 Citation: 3. Johanna Holman, Molly Hurd, Peter L. Moses, Gary M. Mawe, Tao Zhang, Suzanne L. Ishaq, Yanyan Li. Interplay of broccoli/broccoli sprout bioactives with gut microbiota in reducing inflammation in inflammatory bowel diseases. Journal of Nutritional Biochemistry. 2022 Nov 25;109238. doi: 10.1016/j.jnutbio.2022.109238.
• Type: Conference Papers and Presentations Status: Published Year Published: 2022 Citation: 4. Yanyan Li, Tao Zhang, Johanna Holman, Delaney McKinstry, Bruno Caetano Trindade, Jonny Mendoza-Castrejon, Sharon Ho, Duxin Sun, Emily Wells, Grace Y. Chen. A specific broccoli sprout preparation reduces chemically-induced colitis via gut microbiota. Current Developments in Nutrition. 2022, 6(S1): 307. https://doi.org/10.1093/cdn/nzac053.048.
• Type: Journal Articles Status: Under Review Year Published: 2023 Citation: 1. Johanna Holman, Louisa Colucci, Dorien Baudewyns, Joe Balkan, Timothy Hunt, Benjamin Hunt, Lola Holcomb, Grace Y. Chen, Peter Moses, Gary Mawe, Tao Zhang, Yanyan Li, Suzanne L. Ishaq. Steamed broccoli sprouts alleviate gut inflammation and retain gut microbiota against DSS-induced symptoms. Revision submitted. doi: 10.1101/2023.01.27.522641.

Progress 09/01/21 to 08/31/22

Outputs
Target Audience:We presented our research findings from this project at a national conference, American Society for Nutrition 2022, in June 2022, and pubshed the abstract in Current Developments in Nutrition. Our research reached the audience in the professional society. Changes/Problems:Due to PI's change in institution (from Husson University to University of Maine), the transfer of grant took several months from September 2021 to January 2022. Research activity was interrupted during that gap of four months. The PI has received the approval for no-cost one-year extension, which allows for enough time to complete the research activity. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?The results have been published in journals and presented in national conference to reach the scientific community. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? There is increasing evidence that diet and the gut microbiome impacts colorectal cancer risk. Cruciferous vegetables have anti-tumor activities, and the microbiota can contribute to the generation of bioactive metabolites. As diet can affect the composition of the gut microbiota, it is possible that exposure to cruciferous vegetables may induce changes in the gut microbiota that promote intestinal homeostasis. Our goal is to understand the interrelationship between the gut microbiota and the tumor suppressive activities of broccoli sprouts. This will enable the development of novel strategies for colon cancer chemoprevention. Objective 1:To examine the effects of different broccoli sprout diets and major GSLs and ITCs on development of colon tumors. 1.1 To determine the tumor-reducing activities of major GSLs and ITCs found in broccoli sprouts using two different mouse models of colon cancer. 1.2 To investigate a whole food approach using differently prepared broccoli sprout diets in reducing colon tumor susceptibility. We had completed this objective before this reporting period. Objective 2: To determine the role of gut microbiota on the bioactivity of dietary GSLs. 2.1 To evaluate gut microbiota changes induced by GSL/ITC-rich diets. 2.2 To examine whether changes in the microbiota induced by exposure to GSLs improve ITC generation. We have completed anaerobic culturing and sequencing, and currently analyzing sequencing data to identify bacteria that may be responsible for converting GSLs to ITCs. We are also working on searching for primers for genes encoding myrosinase-like enzymes and will do PCR with those primers. Objective 3: To investigate whether the protective effect of broccoli sprouts is mediated by the gut microbiota. We have fed SPF and GF mice with 5% steamed broccoli sprout diet or control diet, treated them with 5 days of DSS, and assessed inflammation. The results showed that the full protection of broccoli sprout diet requires intact gut microbiota.

Publications

• Type: Journal Articles Status: Published Year Published: 2022 Citation: Zhang T, Holman J, McKinstry D, Trindade BC, Eaton KA, Mendoza-Castrejon J, Ho S, Wells E, Yuan H, Wen B, Sun D, Chen GY, Li Y. A steamed broccoli sprout diet preparation that reduces colitis via the gut microbiota. J Nutr Biochem. 2022 Nov 9;112:109215. doi: 10.1016/j.jnutbio.2022.109215. Epub ahead of print. PMID: 36370930.
• Type: Journal Articles Status: Published Year Published: 2022 Citation: Holman J, Hurd M, Moses PL, Mawe GM, Zhang T, Ishaq SL, Li Y. Interplay of Broccoli/Broccoli Sprout Bioactives with Gut Microbiota in Reducing Inflammation in Inflammatory Bowel Diseases. J Nutr Biochem. 2022 Nov 25:109238. doi: 10.1016/j.jnutbio.2022.109238. Epub ahead of print. PMID: 36442719.
• Type: Conference Papers and Presentations Status: Published Year Published: 2022 Citation: Yanyan Li, Tao Zhang, Johanna Holman, Delaney McKinstry, Bruno Caetano Trindade, Jonny Mendoza-Castrejon, Sharon Ho, Duxin Sun, Emily Wells, Grace Chen, A Specific Broccoli Sprout Preparation Reduces Chemically-Induced Colitis Via Gut Microbiota, Current Developments in Nutrition, Volume 6, Issue Supplement_1, June 2022, Page 307, https://doi.org/10.1093/cdn/nzac053.048