Intestinal microbiota regulates diabetes and cancer progression by IL-1β and NOX4 dependent signaling cascades

Abstract

Diabetes changes the host microbiota, a condition known as dysbiosis. Dysbiosis is an important factor for the pathogenesis of diabetes and colorectal cancer (CRC). We aimed at identifying the microbial signature associated with diabetes and CRC; and identifying the signaling mechanism altered by dysbiosis and leading to CRC progression in diabetes. MKR mice that can spontaneously develop type 2 diabetes were used. For CRC induction, another subset of mice was treated with azoxymethane and dextran sulfate sodium. To identify the role of microbiota, microbiota-depleted mice were inoculated with fecal microbial transplant from diabetic and CRC mice. Further, a mouse group was treated with probiotics. At the end of the treatment, 16S rRNA sequencing was performed to identify microbiota in the fecal samples. Blood was collected, and colons were harvested for molecular, anatomical, and histological analysis. Our results show that diabetes is associated with a microbial signature characterized by reduction of butyrate-forming bacteria. This dysbiosis is associated with gastrointestinal complications reflected by a reduction in colon lengths. These changes are reversed upon treatment with probiotics, which rectified the observed dysbiosis. Inoculation of control mice with diabetic or cancer microbiota resulted in the development of increased number of polyps. Our data also show that inflammatory cytokines (mainly interleukin (IL)-1β) and NADPH oxidase (NOX)4 are over-expressed in the colon tissues of diabetic mice. Collectively our data suggest that diabetes is associated with dysbiosis characterized by lower abundance of butyrate-forming bacteria leading to over-expression of IL-1β and NOX4 leading to gastrointestinal complications and CRC.

Keywords: Colorectal cancer; Diabetes; Dysbiosis; Inflammation; Oxidative stress.

Fig. 1

Illustration for the study design of the animal experiments. a Illustration for the study design of the first animal experiment. This experiment included four groups (n = 4 per group) of mice. The first group served as the control group. The second group are wild type FVB-NJ mice in which colorectal cancer (CRC) was chemically induced. The third group composed of MKR diabetic mice. The fourth group included MKR diabetic mice in which CRC was chemically induced. CRC was induced by intraperitoneally injecting the mice with the genotoxin, Azoxymethane (AOM), (10 mg/kg) then after one week, mice were supplied with dextran sulfate sdium (DSS), 2.5% solution instead of drinking water for one week. Mice were allowed to rest for 2 weeks then the DSS cycle was repeated for 3 more times. Fecal samples were collected at an early time point (age: 7 weeks) and a late time point (age: 31 weeks). b Illustration for the study design of the second animal experiment. In this experiment, FVB-NJ mice were depleted from their endogenous microbiota by using an antibiotic cocktail after weaning (age: 3 weeks). The antibiotic cocktail was supplied in the drinking water of mice on alternative days for 5 weeks. After the depletion of microbiota, FVB-NJ mice were divided into five groups. The first group was left without any microbial transplant. The second group was inoculated with fecal microbial transplant (FMT) from FVB-NJ mice from the first experiment. The third group was inoculated with FMT from “FVB CRC” mice from the first experiment. The fourth group was inoculated with FMT from MKR diabetic mice from the first experiment. The fifth group was inoculated with FMT from “MKR CRC” mice from the first experiment. After the FMT, colorectal cancer was chemically induced in all mice groups. CRC was induced by intraperitoneally injecting the mice with the genotoxin, Azoxymethane (AOM), (10 mg/kg) then after one week, mice were supplied with dextran sulfate sodium (DSS), 2.5% solution instead of drinking water for one week. Mice were allowed to rest for 2 weeks then the DSS cycle was repeated for 3 more times. Fecal samples were collected at an early time point (age: 7 weeks) and a late time point (age: 31 weeks). c Illustration for the study design of the third animal experiment. The first group included FVB-NJ mice that served as the control group. The second group included FVB-NJ mice in which colorectal cancer (CRC) was induced followed by treatment with vehicle (Phosphate-buffered saline “PBS”). The third group included FVB-NJ mice in which colorectal cancer (CRC) was induced followed by treatment with Probiotics formula (ProBioLife, 5 mg/kg once daily by oral gavage for 12 weeks). The fourth group included MKR diabetic mice treated with vehicle (Phosphate-buffered saline “PBS”). The fifth group included MKR diabetic mice treated with Probiotics formula (ProBioLife, 5 mg/kg once daily by oral gavage for 12 weeks). Treatment started at week 19 of age and ended at week 31. CRC colorectal cancer, AOM Azoxymethane, DSS Dextran sulfate sodium, FMT Fecal microbial transplant

Fig. 1

Illustration for the study design of the animal experiments. a Illustration for the study design of the first animal experiment. This experiment included four groups (n = 4 per group) of mice. The first group served as the control group. The second group are wild type FVB-NJ mice in which colorectal cancer (CRC) was chemically induced. The third group composed of MKR diabetic mice. The fourth group included MKR diabetic mice in which CRC was chemically induced. CRC was induced by intraperitoneally injecting the mice with the genotoxin, Azoxymethane (AOM), (10 mg/kg) then after one week, mice were supplied with dextran sulfate sdium (DSS), 2.5% solution instead of drinking water for one week. Mice were allowed to rest for 2 weeks then the DSS cycle was repeated for 3 more times. Fecal samples were collected at an early time point (age: 7 weeks) and a late time point (age: 31 weeks). b Illustration for the study design of the second animal experiment. In this experiment, FVB-NJ mice were depleted from their endogenous microbiota by using an antibiotic cocktail after weaning (age: 3 weeks). The antibiotic cocktail was supplied in the drinking water of mice on alternative days for 5 weeks. After the depletion of microbiota, FVB-NJ mice were divided into five groups. The first group was left without any microbial transplant. The second group was inoculated with fecal microbial transplant (FMT) from FVB-NJ mice from the first experiment. The third group was inoculated with FMT from “FVB CRC” mice from the first experiment. The fourth group was inoculated with FMT from MKR diabetic mice from the first experiment. The fifth group was inoculated with FMT from “MKR CRC” mice from the first experiment. After the FMT, colorectal cancer was chemically induced in all mice groups. CRC was induced by intraperitoneally injecting the mice with the genotoxin, Azoxymethane (AOM), (10 mg/kg) then after one week, mice were supplied with dextran sulfate sodium (DSS), 2.5% solution instead of drinking water for one week. Mice were allowed to rest for 2 weeks then the DSS cycle was repeated for 3 more times. Fecal samples were collected at an early time point (age: 7 weeks) and a late time point (age: 31 weeks). c Illustration for the study design of the third animal experiment. The first group included FVB-NJ mice that served as the control group. The second group included FVB-NJ mice in which colorectal cancer (CRC) was induced followed by treatment with vehicle (Phosphate-buffered saline “PBS”). The third group included FVB-NJ mice in which colorectal cancer (CRC) was induced followed by treatment with Probiotics formula (ProBioLife, 5 mg/kg once daily by oral gavage for 12 weeks). The fourth group included MKR diabetic mice treated with vehicle (Phosphate-buffered saline “PBS”). The fifth group included MKR diabetic mice treated with Probiotics formula (ProBioLife, 5 mg/kg once daily by oral gavage for 12 weeks). Treatment started at week 19 of age and ended at week 31. CRC colorectal cancer, AOM Azoxymethane, DSS Dextran sulfate sodium, FMT Fecal microbial transplant

Fig. 2

Colorectal cancer polyps in different mice groups. a Gross characteristics for colorectal cancer (CRC) and a scatter plot representing the number of polyps in FVB-NJ CRC and MKR CRC mice groups. Values are means ± SEM of 3 animals per group (n = 3/group). *p < 0.05 vs. FVB CRC mice. b A scatter plot representing the polyp numbers of mice inoculated with fecal microbial transplant and representative images of the polyps. Values are means ± SEM of 3 animals per group (n = 3/group). ^$p < 0.05 vs. No microbiota and ^#p < 0.05 vs. FVB-FMT group. Results are expressed as mean ± SD. c Representative images of hematoxylin and eosin (H&E)‐stained colon sections illustrating the histological changes in FVB control, FVB CRC, CRC probiotics, MKR diabetic, MKR CRC and MKR probiotics groups. The worst alterations were encountered in the MKR CRC group. Note the presence of adenomatous development (square), hyperplasia(circle), inflammatory cells invading the edematous submucosa (star), crypt abscess (black triangle), as well as crypt architecture disarray (arrow)

Fig. 3

Microbiota composition as revealed by 16S rRNA sequencing. a A box plot representing alpha diversity in FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). b The principal coordinate analysis (PCoA) plot for FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). c A bar graph representing the relative abundance of each bacterial phyla in different samples obtained from FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). d A box plot representing alpha diversity in mice with no microbiota, FVB FMT (fecal microbial transplant), MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). e Rarefaction curve of samples obtained from mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). f A bar graph representing the relative abundance of each bacterial phyla in different samples obtained from mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). g PCoA plot for mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n > 2/group). CRC colorectal cancer, FMT fecal microbial transplant, PCoA principal coordinate analysis

Fig. 3

Microbiota composition as revealed by 16S rRNA sequencing. a A box plot representing alpha diversity in FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). b The principal coordinate analysis (PCoA) plot for FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). c A bar graph representing the relative abundance of each bacterial phyla in different samples obtained from FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). d A box plot representing alpha diversity in mice with no microbiota, FVB FMT (fecal microbial transplant), MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). e Rarefaction curve of samples obtained from mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). f A bar graph representing the relative abundance of each bacterial phyla in different samples obtained from mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). g PCoA plot for mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n > 2/group). CRC colorectal cancer, FMT fecal microbial transplant, PCoA principal coordinate analysis

Fig. 3

Microbiota composition as revealed by 16S rRNA sequencing. a A box plot representing alpha diversity in FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). b The principal coordinate analysis (PCoA) plot for FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). c A bar graph representing the relative abundance of each bacterial phyla in different samples obtained from FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice at early and late time points (n = 3/group). d A box plot representing alpha diversity in mice with no microbiota, FVB FMT (fecal microbial transplant), MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). e Rarefaction curve of samples obtained from mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). f A bar graph representing the relative abundance of each bacterial phyla in different samples obtained from mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n = 3/group). g PCoA plot for mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT at early and late time points (n > 2/group). CRC colorectal cancer, FMT fecal microbial transplant, PCoA principal coordinate analysis

Fig. 4

Pie charts of microbiota composition of different mice groups at early and late time points showing mean relative abundance of major bacterial phyla as well as the fraction of the butyrate-producing bacteria

Fig. 5

Protein expression of interleukin (IL)-1β and NADPH oxidase (NOX)4 in different mice groups. a A scatter plot representing the quantification of protein expression of IL-1β in colon tissues of FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. b A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. c a scatter plot representing the quantification of protein expression of IL-1β in colon tissues of for mice with no microbiota, FVB FMT(fecal microbial transplant), MKR FMT, FVB CRC FMT and MKR CRC FMT. Values are means ± SEM of 3 animals per group (n = 3/group). d A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of for mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT. Values are means ± SEM of 3 animals per group (n = 3/group). e Colon lengths of FVB control, MKR diabetic treated with vehicle (PBS), FVB CRC treated with vehicle (PBS), MKR diabetic treated with probiotics for 12 weeks and FVB CRC treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. ^$ statistically significant at p < 0.05 vs FVB CRC. f A scatter plot representing the polyp numbers of FVB CRC mice treated with vehicle (PBS) or probiotics for 12 weeks and representative images of the polyps. Values are means ± SEM of 3 animals per group (n = 3/group). ^$Statistically significant at p < 0.05 vs FVB CRC. g A scatter plot representing the quantification of protein expression of IL-1β in colon tissues of FVB control, MKR diabetic, and MKR mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. h A scatter plot representing the quantification of protein expression of IL-1β in colon tissues of FVB control, FVB CRC mice, FVB CRC mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. i A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of FVB control, MKR diabetic, and MKR mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^# statistically significant at p < 0.05 vs MKR Diabetic. j A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of FVB control, FVB CRC mice, FVB CRC mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. CRC Colorectal cancer, FMT Fecal microbial transplant, IL interleukin, NOX NADPH oxidase, PBS phosphate-buffered saline

Fig. 5

Protein expression of interleukin (IL)-1β and NADPH oxidase (NOX)4 in different mice groups. a A scatter plot representing the quantification of protein expression of IL-1β in colon tissues of FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. b A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of FVB control, MKR diabetic, FVB with colorectal cancer (CRC) and MKR CRC mice. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. c a scatter plot representing the quantification of protein expression of IL-1β in colon tissues of for mice with no microbiota, FVB FMT(fecal microbial transplant), MKR FMT, FVB CRC FMT and MKR CRC FMT. Values are means ± SEM of 3 animals per group (n = 3/group). d A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of for mice with no microbiota, FVB FMT, MKR FMT, FVB CRC FMT and MKR CRC FMT. Values are means ± SEM of 3 animals per group (n = 3/group). e Colon lengths of FVB control, MKR diabetic treated with vehicle (PBS), FVB CRC treated with vehicle (PBS), MKR diabetic treated with probiotics for 12 weeks and FVB CRC treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. ^$ statistically significant at p < 0.05 vs FVB CRC. f A scatter plot representing the polyp numbers of FVB CRC mice treated with vehicle (PBS) or probiotics for 12 weeks and representative images of the polyps. Values are means ± SEM of 3 animals per group (n = 3/group). ^$Statistically significant at p < 0.05 vs FVB CRC. g A scatter plot representing the quantification of protein expression of IL-1β in colon tissues of FVB control, MKR diabetic, and MKR mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. h A scatter plot representing the quantification of protein expression of IL-1β in colon tissues of FVB control, FVB CRC mice, FVB CRC mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. i A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of FVB control, MKR diabetic, and MKR mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^# statistically significant at p < 0.05 vs MKR Diabetic. j A scatter plot representing the quantification of protein expression of NOX4 in colon tissues of FVB control, FVB CRC mice, FVB CRC mice treated with probiotics for 12 weeks. Values are means ± SEM of 3 animals per group (n = 3/group). *Statistically significant at p < 0.05 vs FVB control group. ^#Statistically significant at p < 0.05 vs MKR Diabetic. CRC Colorectal cancer, FMT Fecal microbial transplant, IL interleukin, NOX NADPH oxidase, PBS phosphate-buffered saline