For cell culture work we utilised Caco-2 human epithelial colorectal adenocarcinoma cells [15] which are maintained by the APC Microbiome Institute cell culture collection and originally obtained from the American Type Culture Collection (Rockville). Caco-2 cells were grown in Dulbecco’s modified eagle medium (DMEM)(Sigma) supplemented with 10% foetal bovine serum (FBS)(Sigma) and 100 U/ml penicillin and 100 μg/ml streptomycin (Sigma) and 1% non-essential amino acids (Sigma). For in vitro circadian experiments, the Caco-2 cells were plated on 6 well plates and at ~80% confluency the cells were incubated in serum free DMEM-F12 (50:50) medium for 24 hours. At t = 0 the cells were serum shocked by exposing to DMEM-F12 (50:50) medium containing 50% FBS [4] and bile acid 100μM for 2 hours. The medium was then replaced with serum free DMEM and Ham F12 (50:50) medium with bile acids at a final concentration of 100μM. At the indicated times, the cells were washed once with PBS, lysed directly in lysis buffer (Fisher Scientific) and the lysates were stored at -70°C until used for RNA Isolation. All bile acids used in this study (Deoxycholic acid (DCA), Chenodeoxycholic acid (CDCA), Sodium taurodeoxycholate (TDCA) and Sodium taurochenodeoxycholate (TCDCA)) were obtained from Sigma. Cell viability was measured using the CellTiter-Glo Luminescent Cell Viability Assay (Promega) according to the manufacturer's instructions.

Caco-2 cells were seeded on collagen-coated polycarbonate membrane cell culture inserts of diameter, 0.33 cm² area with 0.4 μm pore size (Corning) at a density 250,000 cells/cm² of cells in basal medium (BD biosciences). The following day the medium was changed to a differentiation medium (BD biosciences) and cells were allowed to differentiate for 2 days. Bile acid was added at the apical side when absolute TEER reached >350 ohms, which indicates an intact monolayer. The TEER was measured before and 24 hours after bile acid treatment using an EVOM—Epithelial Voltohmmeter (World precision instrument).

Eight week old male C57BL/6 mice used in this study were obtained from Harlan UK. Mice were acclimatized for 7 days in a controlled environment with a 12h: 12h light-dark cycle and divided into 3 groups. The first group of 8 mice was dosed by oral gavage with corn oil. A second and third group of 4 mice each were dosed by oral gavage with DCA (9 μmol/kg body weight) and CDCA (9 μmol/kg body weight) dissolved in corn oil respectively [16]. The mice were gavaged three times (t = 0, t = 24h and t = 48h) with corn oil or bile acids, were fasted for 3h to minimise inter-subject variation in bile acid profiles due to food intake during this time and euthanised to collect organs. All mice were euthanized within the same 45 minute period during the light phase. Liver, ileum and colon samples were stored in RNA later at -80°C.

This study was approved by the University Animal Experimental Ethics Committee (AEEC) and all animal experimental procedures used in this study were performed following institutional review and approval by the University Animal Experimental Ethics Committee (AEEC) (approval ID 2011/017) and under licence from the Irish Department of Health. All efforts were made to minimize suffering and to utilise to lowest possible number of animal subjects. None of the animals used in this study showed signs of pain or discomfort. At the relevant time-point mice were euthanized by cervical dislocation in accordance with institutional guidelines.

RNA was isolated from tissues using the RNeasy kit (Qiagen) according to the manufactures instructions and total RNA from cultured cells was isolated using a Nucleospin RNA kit (Fisher Scientific). cDNA synthesis was performed using 1 μg of total RNA isolated from tissues or from cells and mastermix containing a blend of random primer (Roche), RT enzyme (Roche) and dNTP (Roche)and real-time PCR amplification was performed as described previously [13] using a LightCycler^® 480 Real-Time PCR System (Roche, Dublin, Ireland). The results were normalised to the β actin gene and relative expression was calculated based on the 2^-ΔΔCt method as previously described [17]. All gene-specific primer sequences used in the study are listed in S1 and S2 Tables.

Data analysis was carried out using MS excel and Graphpad prism 5 software (California, USA). Two technical replicates were carried out for each PCR reaction and for cell culture experiments values were calculated as standard error of the mean (SEM) from three independent biological experiments [18]. Multiple groups were analysed by one-way analysis of variance (ANOVA), followed by Dunnett’s multiple comparison post hoc test.

Recent work has demonstrated that the microbiota is necessary to entrain the peripheral circadian clock [6]. Based upon previous findings from our group [13] we focused upon bile acids as potential signalling molecules in the regulation of expression of host circadian genes using a synchronised Caco-2 cell system that has been described previously [19]. As our previous work implicated the bacterial enzyme BSH in the expression of host clock-regulated genes [13] we tested the effects of unconjugated bile acids (DCA and CDCA) and corresponding conjugated bile acids (TDCA and TCDCA) upon expression of clock-regulated genes in the Caco-2 system. We exposed cells to concentrations of bile acids (100 μM) that have been proposed to be physiologically relevant and are non-toxic [20]. The CellTiter-Glo luminescent cell viability assay demonstrated that this concentration of bile acids has no cytotoxic effect on Caco-2 cells in our system (S1A Fig). The integrity of polarised Caco-2 monolayers as assessed by TEER measurements was also unaffected by bile acids at the concentrations used in this study (S1B Fig).

We demonstrate that the unconjugated bile acids DCA and CDCA significantly increase the amplitude of the expression of cellular clock genes relative to conjugated moieties (TDCA and TCDCA) or control cells. The expression of CLOCK and ARNTL displayed a cyclical expression pattern that was amplified by the presence of unconjugated bile acids. PER1 and PER3 expression was also shown to cycle rhythmically but in a pattern antiphase to CLOCK and ARNTL in the presence of unconjugated bile acids, peaking at 24 h. Amplification of gene transcription was clearly associated with the presence of unconjugated bile acids and was not seen when cells were exposed to the corresponding conjugated bile acid moieties at equimolar concentrations (Fig 1).

Using a single sample time-point our previous study in germ free mice suggested that microbial BSH activity (which produces elevated concentrations of unconjugated bile acids) may affect the phase of oscillation of the circadian cycle [13]. Here we demonstrate that exposure of Caco-2 cells to unconjugated bile acids very subtly alters the phase of the circadian gene expression cycle. For instance expression of PER2 in control synchronised cells peaks at 12, and 36 hours (see Fig 2 for baseline circadian oscillations in CACO-2 cells). In the presence of either DCA or CDCA this phase shifts towards a peak expression at 18 and 42 hours (Fig 1). Similar shifts in early peak expression occur for PER1 and PER3 from peak expression at 12/18 hours (Fig 2) to highest expression at 24 hours in the presence of unconjugated bile acids (Fig 1). The ability of unconjugated bile acids to affect the expression patterns of input regulators into the system (such as RORα or E4BP4 (Fig 1) or DBP (S2 Fig)) may explain these alterations to the transcriptional cycle. Whilst the precise details surrounding the control of the cellular circadian gene expression cycle remain to be elucidated we clearly show that unconjugated bile acids influence the expression levels of genes involved in this cycle in cultured gut epithelial cells.

Given that unconjugated bile acids influence the oscillation of circadian genes in cell culture we investigated whether oral administration of bile acids would influence expression of circadian genes in mice. We gavaged C57BL/6 mice with either DCA or CDCA (at 9 μmol/kg of body weight) dissolved in corn oil or with vehicle (corn oil) with three doses over a 48 hour period as per previous studies [16]. We did not administer mice with conjugated bile acids in this study as these would be subjected to in vivo deconjugation by the resident microbiota [21]. Mice were euthanized at the same single time point in order to allow direct comparison of patterns of gene expression across the different groups.

Across all peripheral tissues (ileum, colon and liver) we noted that bile acid administration induces significant changes to circadian gene expression profiles which were broadly similar. In the ileum administration of DCA or CDCA bile acids significantly increased the expression of Per1, Per2, Per3 and Cry2 genes as compared to the vehicle (Fig 3A and 3B). As expected the expression of Per genes was antiphasic to Clock and Arntl which are on a different limb of the circadian gene expression cycle. Interestingly regulation of input regulators of the system was consistent with the observed changes in circadian gene expression (Fig 4A and 4B). We determined elevated expression of Dbp and reduced expression of E4bp4, changes that are predicted to increase expression of Per and related genes [1,3]. Both bile acids also stimulated expression of Nr1d1 which encodes a factor that is known to block expression of Clock/Arntl [1,3].

Broadly similar patterns were seen in the colon where we determined significantly elevated levels of Per1 and Cry2 expression in mice administered DCA or CDCA with elevated levels of Per2 expression in mice receiving DCA (Fig 3C and 3D). Again we determined that bile acid administration significantly elevated expression of the gene encoding the regulator Dbp with a trend towards reduction in E4bp4 which may explain the enhanced expression of Per genes (Fig 4C and 4D).

As bile acids are efficiently absorbed in the terminal ileum we hypothesized that orally administered bile acids may influence the circadian expression of genes in the liver. Indeed we found that oral administration of DCA or CDCA to mice significantly altered the expression of hepatic circadian genes (Figs 3E and 3F and 4E and 4F). Bile acid administration increased expression of hepatic Per2, Per3 and Cry2 whilst DCA induced elevated Per1 and CDCA induced elevated Cry1 (Fig 3E and 3F). As expected these were in antiphase to the expression of Arntl which was significantly reduced in the liver. Again in this organ administration of DCA or CDCA significantly increased expression of Dbp a finding which potentially correlates with elevated Per/Cry expression profiles [1,3]. We appreciate that the work was carried out on a single timepoint in the murine circadian cycle. Further work in our laboratory is ongoing to determine the effects of bile acid and microbiota alterations across the wider circadian cycle in a murine model.