Long-term iron deficiency and iron supplementation exacerbate acute DSS-induced colitis and are associated with significant dysbiosis

Patients taking oral iron supplementation often suffer from gastrointestinal side effects. We have previously shown that acute alterations in oral iron exacerbate dextran sodium sulphate (DSS) induced colitis and are associated with dysbiosis. As patients take iron supplementation for long periods, we asked whether this too would influence colitis and the microbiome. We assessed the impact of long-term changes in dietary iron, by feeding chow containing 100ppm, 200ppm and 400ppm (reflecting a deficient, normal or supplemented diet, respectively) for up to 9 weeks to female wild-type C57BL/6 (WT) mice in presence or absence of chronic colitis, or acute colitis induced after 8 weeks, induced by DSS. Assessment was made based on (i) clinical and histological severity of colitis, and (ii) faecal microbial diversity, as assessed by sequencing the V4 region of 16S rRNA. In mice with long term changes to their dietary iron, reduced iron intake (100ppm iron diet) was associated with increased weight loss and histology scoring in the acute colitis model. Chronic colitis was not influenced by altering dietary iron however there was a clear change in the faecal microbiome in the 100 and 400ppm iron DSS-treated groups and in controls consuming the 400ppm iron diet. Proteobacteria levels increased significantly at day-63 compared to baseline and Bacteroidetes levels decreased in the 400ppm iron DSS group at day-63 compared to baseline; mirroring our previously published work in acute colitis. Long term dietary iron alterations clearly affects gut microbiota signatures but do not appear to exacerbate chronic colitis. However, acute colitis is exacerbated by changes in dietary iron. More work is needed to understand the impact of iron supplementation of the pathologenesis of IBD and rise that possiblity that the change in the microbiome, in patients with colitis, is a consequence of the increase in luminal iron and not simply the presence of colitis.

Abstract: 49 Patients taking oral iron supplementation often suffer from gastrointestinal side effects. We 50 have previously shown that acute alterations in oral iron exacerbate dextran sodium sulphate 51 (DSS) induced colitis and are associated with dysbiosis. As patients take iron supplementation 52 for long periods, we asked whether this too would influence colitis and the microbiome. We  172 were purified using Axygen SPRI Beads. The second-round PCR was performed to the same conditions. Samples were re-purified then quantified using Qubit and assessed using 178 the Fragment Analyser. Successfully-generated amplicon libraries were sequenced (25).

180
The final libraries were pooled in equimolar amounts using the Qubit and Fragment Analyser (25). The sequences passing the above filters for each sample were pooled into a single file.

197
A metadata file was created to describe each sample. These two files were analysed using

218
Chronic DSS-induced colitis induces C57BL/6 weight loss 219 Colitis was reproducibly induced by 1.25% DSS. All mice lost body weight from day-6 and 220 maximal weight loss occurred at day-8 of each cycle. Mice receiving the 100ppm iron diet 221 appeared to lose more weight than other groups, but this difference was not significant (Fig.   222 1-a). All control mice, irrespective of the iron dosing, showed a steady increase in body weight.

236
Weight loss began earlier (day-3) in the 100ppm iron group than in the 200 and 400ppm 237 iron DSS-treated groups ( Fig. 1-b). During this acute DSS cycle, mice fed 100ppm iron lost 238 significantly (P<0.001) more weight than the other treated groups. 200ppm iron and all the mice that received cycles of 1.25% DSS (Fig. 2-b). 200ppm treated groups that received DSS both showed significant differences at day-1 vs day-21, 42 and 63 ( Fig. 3-a) consistent with the presence of luminal iron from bleeding resulting from colitis. Faecal iron concentration increased significantly in control mice (63 days 295 on diet alone) taking 200 and 400ppm diets, but did not change with time in those mice 296 consuming 100ppm iron (Fig. 3-a).

343
The analysis of faecal samples from mice in the 400ppm iron DSS-treated group showed highly abundant among all groups while the lowest abundance phylum was Actinobacteria.

362
However, 100ppm iron and 400ppm iron chronic DSS groups showed seven different genera 363 apart from the three genera (Bacteroides, Lactobacillus and Bilophila) that they shared.

364
STAMP encourages the use of effect sizes and confidence intervals (29). The results of the 365 relative abundances of various phyla and identified genera are summarised in Table 1: a-c.  Bacteroides related were greatly reduced. The authors concluded that all significant 423 differences in bacterial abundance in wild-type mice appeared as a result of the interaction 424 between treatment and host-mediated inflammation (37, 38). There are several key 425 differences between that paper and our own: they investigated caecal contents, not faeces; 426 they induced ileitis, not colitis and they did not measure faecal iron concentration. Thus, their 427 paper and our data cannot be directly compared.

429
Our data analysis showed that seven genera were significantly different. In the half standard 430 iron diets (100ppm) DSS-treated group, we found reductions in Lactobacillus (P<0.002), Lactobacillus (P<0.01). The only control group in which significant differences were found was  459 All data files uploaded in supporting information file. method in the induction of reliable experimental acute and chronic ulcerative colitis in mice.