Studying host genetic background effects on multimorbidity of intestinal cancer development, type 2 diabetes and obesity in response to oral bacterial infection and high‐fat diet using the collaborative cross (CC) lines

Abstract Background Multimorbidity of intestinal cancer (IC), type 2 diabetes (T2D) and obesity is a complex set of diseases, affected by environmental and genetic risk factors. High‐fat diet (HFD) and oral bacterial infection play important roles in the etiology of these diseases through inflammation and various biological mechanisms. Methods To study the complexity of this multimorbidity, we used the collaborative cross (CC) mouse genetics reference population. We aimed to study the multimorbidity of IC, T2D, and obesity using CC lines, measuring their responses to HFD and oral bacterial infection. The study used 63 mice of both sexes generated from two CC lines (IL557 and IL711). For 12 weeks, experimental mice were maintained on specific dietary regimes combined with co‐infection with oral bacteria Porphyromonas gingivalis and Fusobacterium nucleatum, while control groups were not infected. Body weight (BW) and results of a intraperitoneal glucose tolerance test (IPGTT) were recorded at the end of 12 weeks, after which length and size of the intestines were assessed for polyp counts. Results Polyp counts ranged between 2 and 10 per CC line. The combination of HFD and infection significantly reduced (P < .01) the colon polyp size of IL557 females to 2.5 cm2, compared to the other groups. Comparing BW gain, IL557 males on HFD gained 18 g, while the females gained 10 g under the same conditions and showed the highest area under curve (AUC) values of 40 000‐45 000 (min mg/dL) in the IPGTT. Conclusion The results show that mice from different genetic backgrounds respond differently to a high fat diet and oral infection in terms of polyp development and glucose tolerance, and this effect is gender related.


| INTRODUC TI ON
Multimorbidity is the existence of multiple chronic conditions and diseases associated with an elevated risk of death, disability, low quality of life, and environmental and genetic factors. The term intestinal cancer refers to a slowly developing cancer that begins as a tumor or tissue growth on the inner lining of different parts of the intestines. 1 Epidemiological and molecular evidence links obesity and metabolic status with inflammation and an increased risk of many cancers. leads to a hyperinflammatory response, which disrupts the balance of oral microbiota, subsequently resulting in inflammation, which accelerates intestinal cancer development. 4 Finally, studies of familial heritability and the genome-wide associations study (GWAS) have confirmed a significant role of genetic factors in colon cancer development. 5 Another study has shown that oral bacteria including Porphyromonas gingivalis (Pg) and Fusobacterium nucleatum (Fn) may colonize the host gut, which subsequently develops local inflammation that may trigger/initiate intestinal cancer development. [6][7][8] In addition, a previous study by Demmer et al reported that oral bacteria were related to inflammation and insulin resistance among diabetesfree adults. 9 Our recent study showed that different strains of mice respond differently to dietary and infection challenge-induced co-morbidity of T2D and obesity. 10 This variation is explained to a significant extent by genetic variability. A better understanding of how this genetic variation translates into different clinical manifestations will eventually highlight pathways through which dietary composition may initiate or accelerate inflammatory disease processes and indicate mechanisms through which disease can potentially be prevented.
Recently, it was demonstrated that genetically highly diverse sets of recombinant inbred mouse lines (RIL), collectively named the Collaborative Cross (CC), can be used as a tool for the identification of risk genes in complex human disease. 8 The CC was developed as the next generation of mouse genetic reference population, which allows time-and cost-efficient mapping of quantitative trait loci (QTLs) associated with complex traits. [11][12][13][14][15][16][17][18][19][20][21] The CC population is a large panel of recombinant inbred (RI) strains derived from a genetically diverse set of eight founder strains: A/J, C57BL/6J, 129S1/ SvImJ, NOD/LtJ, NZO/HiLtJ, CAST/Ei, PWK/PhJ, and WSB/EiJ. The key features of the CC genetic reference population with gene mapping are that a very large number of variants segregate in the population (there are over 36 million SNPs) 12 and the relatively high level of recombination present compared to other mouse RI sets.
Finally, our previous and current results support the use of the CC mouse genetic reference population as a unique and powerful platform for studying the multimorbidity of intestinal cancer, T2D, and obesity in response to a high-fat diet and oral bacterial co-infection, as a step towards identifying disease-related genetic factors.

| Ethical statement
All the experiments and mouse usage described in this study were compatible with the standards for care and use of laboratory animals and approved by the Institutional Animal Care and Use Committee (IACUC) of Tel Aviv University (TAU), Israel (IACUC no. 01-19-013).

| Study cohort
63 mice were used in this study from two different CC lines, IL711 and IL557, which have different genetic backgrounds; both sexes were represented in each study group. 6,12,14,18,19,22 High molecular weight genomic DNA for the CC lines was initially genotyped using the mouse diversity array (MDA), which consists of 620 000 SNPs, and re-genotyped by mouse universal genotype array (MUGA-7500 markers) and eventually with MegaMuga (77 800 markers) SNP arrays to confirm their genotype status and variations in their genetic structure. 12 These mice were maintained at our animal facility at TAU under suitable and agreed ethical conditions of temperature (21-23°C), humidity and daily supervision. Mice were weaned at the age of 3 weeks old, and then maintained separately by line and sex, with a maximum of five mice in an open-top cage and free access to water and rodent chow diet. Summary of the used mice in this study and their assignments in the different experimental groups are presented in (Table 1).

| Study design
The experiments started when mice were 8 weeks old and lasted for a period of 12 weeks, with the mice on either a high-fat dietary

| Dietary challenge
Mice were weaned at the age of 3 weeks and maintained until 8 weeks of age on a standard rodent chow diet (CHD; Altromin 1324 IRR, Altromin Spezialfutter GmbH & Co., Germany), which provides 11% Kcal from fat, 24% from protein, and 65% from carbohydrates. The dietary challenge started when the mice were 8 weeks old, when they were maintained on either CHD or HFD for the 12-week period of the experiment. The high-fat diet (HFD; TD.88137) was considered equivalent to a western diet, and was supplied by Teklad Global (Harlan Inc, Madison, WI, USA). The HFD provided 42.0% Kcal from fat, 15.3% from protein, and 42.7% from carbohydrates.

| Oral infection challenge
Before the infection challenge, mice were treated with antibiotics to standardize the oral microbiota status of the different mice, using sulfamethoxazole (10/500 mL) water administered for 10 days, followed by 3 days' recovery (antibiotic-free). The mice were then orally infected with 400 μL per mouse of the mixed oral bacteria (Pg and Fn). The infection procedure was repeated 3 times, on days 1, 3 and 5 of week 5. In parallel, control groups of the placebo infection were treated with 400 µL of 2% CMC in distilled water and 1% PBS (CMC:PBS ratio of 2:1).

| Body weight measurements
During the 12 weeks of the experiment, the body weight (BW) of the mice was recorded bi-weekly using an electronic scale with 0.1 g accuracy.

| Intraperitoneal glucose tolerance test (IPGTT)
The IPGTT was performed at the end of the experiment to detect disturbances in glucose tolerance ability as an indicator of the development of T2D. On the morning of the IPGTT, the mice were fasted for 6 hours (6:00-12:00 am), with free access to water. Fasting blood glucose levels (time 0) were then determined from tail blood, and subsequently a solution of glucose (2.5 mg glucose per g mouse body mass) was administered by intraperitoneal (IP) injection. Thereafter, blood glucose levels were monitored at different time points during the following 180 minutes (15, 30, 60, 120, and 180 minutes after glucose injection). The mice were then returned to their cages with free access to food and water for overnight recovery.

| The area under the curve (AUC)
IPGTT results were corrrelated to glucose tolerance ability by calculating the AUC of the 180-minute glucose tolerance clearance process.
Calculation of AUC was conducted according to the trapezoid rule from time 0 to 180 min after glucose injection to quantitatively evaluate glucose clearance activity. AUC between any two time points was calculated as (time difference in minutes between sequential reads) × (glucose level at 1st time point + glucose level at 2nd time point)/2). The total AUC value of the 180-minute IPGTT was calculated as the sum of the AUC between each set of two time points, total AUC0-180 = AUC0-15 + AUC15-30 + AUC30-60 + AUC60-120 + AUC120-180.

| Intestines collection
At the end point of the experiment (at 20 weeks old), the mice were sacrificed by CO 2 inhalation and the intestines collected.

| Intestinal preparation for polyp counts
Intestines were collected and soaked in a small plate filled with phosphate buffered saline (PBS) until wash (5-10 minutes), which makes the tissue easier to use. Small and large intestines were then washed and cleaned at least twice with PBS. The small intestines were divided into three equal segments; proximal, middle, and distal. Each segment including the colon was horizontally opened/cut by razor and spread over 154 cm 2 Whatman paper and fixed overnight in 10% neutral buffered formalin (NBF). The length and width of each was measured by ruler.
The intestines were washed with 70% ethanol and then stained with 0.02% methylene blue (1.5 minutes for each paper) and kept overnight in PBS on the shaker. The number of polyps, length in cm, and size (width × length) in cm 2 of each segment of intestines were recorded.

| RE SULTS
Herein, we present a data analysis of 63 mice generated from two genetically different CC lines, with both sexes represented. These mice

| Polyp numbers vary between the CC lines in response to dietary and infection challenge
Our In both small and large intestines, separately, female mice on HFD developed more polyps compared to control mice, but the numbers were not significantly different, averaging 5 ± 1 polyps in IL557 and 5.4 ± 0.244 polyps in IL711 in the small intestines, and 3 ± 0.00 and 2.40 ± 0.40 in the colon of IL557 and IL711, respectively ( Figure 2C,E). Interestingly, in the male population, the different lines responded differently to the dietary challenge as shown in Figure 2B,D,F. Across the whole of the intestines, IL557 on HFD developed more polyps, averaging 9.8 ± 1.53, compared to the control mice group with an average of 7.67 ± 1.76 polyps, while IL711 on HFD developed fewer polyps compared to control mice. The same pattern was also seen in the small intestines but the opposite pattern was observed in the colon, where IL557 developed in average 3 ± 0.5 polyps in both groups while IL711 on HFD developed more polyps (1.5 ± 0.5) compared to the control group, with 1 ± 0.5 polyps on average.
The effect of the infection on polyp development, along the whole intestines was highly significant in the female population.

| The effect of the dietary and infection challenges on the length of the intestines varies between IL577 and IL711
In the female population, the same pattern of changes was ob- Interestingly, the combination of HFD and infection resulted in a significant shortening (P < .05) of the intestines compared to other groups, especially in the length of the colon, which in IL557 and IL711 averaged 6.5 ± 0.0 cm and 5.84 ± 0.42 cm, respectively, significantly different (P < .05 and P < .01, respectively) from other groups. In the male population, the whole intestines of infected IL557 mice on CHD averaged 50.5 ± 0.6 cm, significantly different from the control group (CHD/no-infection) (P < .01), and from infected mice on HFD (P < .05) ( Figure 3B). The length of the small intestines was, significantly different, in control mice compared to non-infected mice on HFD (P < .05), and also significantly different in infected mice from other groups ( Figure 3D), while the length of the colon was close to that of other groups ( Figure 3F).
In IL557, there was a significant difference (P < .05) in the length of the colon between infected and non-infected mice on HFD, with an average of 1.42 cm.

| Significant variations in the size of intestines between IL557 and IL711
The IL711 male population did not show any significant variations in the size of intestines in response to HFD and an infection challenge. In contrast, the size of the intestines, and in particular the small intestines, of infected male IL557 mice on CHD was significant larger (P < .01) than other groups and challenges ( Figure 4B), with an average size of 48.09 ± 2.5 cm 2 for the whole intestines, and an average of 40.2 ± 0.66 cm 2 for the small intestines ( Figure 4D).
In the female population, the infected IL711 mice maintained on CHD presented the largest size of whole intestines, which was significantly different (P < .01) from other groups ( Figure 4A).
The females in the control group presented the smallest intes-

| Male and female IL557 mice gained more weight than IL711 mice under the same conditions of diet and infection
In our previous studies, it was shown that line and sex have major effects on body weight gain in response to dietary and infection challenge. 12,13,23 In this study, while in IL711 females there were no significant changes in body weight gain between the different groups, body weights of infected IL711 males on HFD were significantly higher than in the control group (P < .05), with an average weight gain of 5.5 ± 0.37 g. IL557 males were susceptible to HFD consumption, and they gained on average 17.72 ± 0.88 g in weight, significantly more (P < .01) than the other three groups ( Figure 6B), while infected IL557 females on CHD reached an average weight gain of 1053 ± 1.03 g, significantly higher (P < .05) than control groups.

| Heatmap of multimorbidity of polyp counts, AUC, BW, and length and size changes in the development of intestines in IL557 and IL711 mice
To be better visualize and understand the phenotype patterns of  Figure 7, and shows which traits were found to be non-significant or significant at levels P < .05 and P < .01 in female and male mice of the studied lines in the different experimental groups.
The results presented here, show that body weight gain was highly significantly (P < .01) increased in female IL557 mice com-

| D ISCUSS I ON
A previous study by our group, using the CC mouse model population, confirmed that different CC lines respond differently to HFD, with males and females of the different lines varying significantly in T2D development and progression in response to two diet challenges, CHD and HFD. 8,23-25 Furthermore, a recent study from our lab has reported comorbidity between T2D and obesity. 10 This study also showed that the response of the mice to the interaction between infection and the diet challenges varied between the lines and between sexes of the same lines. 10  infectious diseases. 8,[22][23][24][25][29][30][31] In the present study, the results confirm that the host genetic background is an important factor for defining the severity of the multimorbidity of the assessed and presented phenotypes.
In order to investigate the link between intestinal cancer, obesity and T2D, here we presented a comprehensive picture of the susceptibility status and profiles of two CC lines, IL557 and IL711, to diet and infection challenges, separately and together. 10,[23][24][25] The heatmap analysis emphasizes the strong linear profiles between these phenotypes, especially between the BW and AUC results. The findings demonstrate the diversity between different strains under the same environmental conditions, and how the patterns and levels of the development of more than one disease can be studied at the same time in each strain. This multimorbidity analysis approach will help us to better understand the mechanism of each disease separately and how it is related to other diseases, under the same conditions. Identifying the gene(s) underlying these patterns will help us to better predict the development of many diseases, which will offer a new set of opportunities to prevent the early stages of each individual disease and its associated diseases.
In this study, we used a chow diet (CHD), which provides of 11% Since HFD-induced free fatty acids, which damage the intestines, we may hypothesis is that the cytotoxicity of the abundant HFD-derived free fatty acids in the intestinal lumen impairs the intestinal immune system. 34 Our results showed that a high number of polyps developed in HFD-fed mice, but interestingly, this effect varied between mice with different genetic backgrounds.
This proves the influence of the host genetic background on the development of these diseases, whether considered separately, or as co-or multimorbidities. A recent study indicated that maintaining mice on HFD induced shortening of the small intestine and colon, and thinning of the ileum, which was defined as intestinal lipotoxicity. In metabolic syndromes, lipotoxicity usually indicates cytotoxicity to pancreatic islet β-cells, causing abnormal glucose tolerance. 34 In our study, the CC lines exhibited tremendous variation in glucose sensibility, body weight gain and polyp counts.
We believe that by using this genetic reference population and expanding the number of the lines studied to 30 or more and using genome-wide association studies (GWAS), we will be able to identify modifier genes associated with the multimorbidity of these diseases, as was successfully achieved in our previous studies. 12,14-21, 23,35-37 In conclusion, it is believed that with providing more data on different CC lines, will promise to elucidate the components of the host genetic background that are involved in resistance to and rate of development of the multimorbidities of intestinal cancer, T2D and obesity induced by high fat and oral infection. Once obtained, such data can be used to predict the individual genetic risk factors underlying the development of these diseases and provide a platform for utilizing early prediction, prevention and treatment strategies.

ACK N OWLED G M ENTS
The present work is part of an MSc thesis by Asal Milhem. The authors declare no competing financial interests or other associations that may pose a conflict of interest (eg pharmaceutical stock ownership, consultancy). This report was supported by Binational Science

CO N FLI C T O F I NTE R E S T
The authors have declared no conflicts of interest. F I G U R E 8 Comparisons of polyp counts, glucose tolerance, body weight gain, and length and size of intestines at week 12 in male and female mice from the IL557 and IL711 CC lines in response to a 42% high-fat diet (HFD) or chow diet (CHD), and with (+Inf.) or without (No-Inf.) oral bacterial infection challenges in female and male populations. Number of recorded polyps in the whole intestines in each experimental group (HFD + Inf., HFD + No-Inf., CHD + Inf. and CHD + No-Inf.) in female and male mice from IL557 and IL711 CC lines are presented in (A) and (B), respectively, while length of the intestines in the same groups in female and male mice are presented in (C) and (D), respectively. (E) and (F) present intestines sizes for females and males, respectively, in these two lines. AUC values in female and male mice in these studied groups are presented in (G) and (H), respectively, and body weight gain traits (ΔBW) for female and male mice are presented in (I) and (J), respectively. * and **indicate significant P values of <.05 and <.01, respectively