Research ArticleInflammation, Immunity, Fibrosis, and Infection

The serotonin reuptake transporter is reduced in the epithelium of active Crohn’s disease and ulcerative colitis

Published Online:https://doi.org/10.1152/ajpgi.00244.2020

Abstract

Serotonin is a highly conserved and ubiquitous signaling molecule involved in a vast variety of biological processes. A majority of serotonin is produced in the gastrointestinal epithelium, where it is suggested to act as a prominent regulatory molecule in the inflammatory bowel diseases (IBDs) Crohn’s disease (CD) and ulcerative colitis (UC). Extracellular and circulating serotonin levels are thought to be elevated during intestinal inflammation, but the underlying mechanisms have been poorly understood. The data on human material are limited, contradictory, and in need of further investigation and substantiating. In this study, we show a potent and significant downregulation of the dominant serotonin reuptake transporter (SERT) mRNA (SLC6A4) in the epithelium from active CD ileitis, CD colitis, and UC colitis, compared with healthy controls. The mRNA of tryptophan hydroxylase 1, the rate-limiting enzyme in serotonin synthesis, was unregulated. Immunohistochemistry showed expression of the SERT protein in both the epithelium and the lamina propria and localized the downregulation to the epithelial monolayer. Laser capture microdissection followed by RNA sequencing confirmed downregulation of SLC6A4 in the epithelial monolayer during intestinal inflammation. Patient-derived colon epithelial cell lines (colonoids) incubated with the proinflammatory cytokine tumor necrosis factor alpha (TNF-α) reduced SERT expression. In summary, these results show that intestinal inflammation potently reduces the expression of SERT in both CD and UC and that TNF-α alone is sufficient to induce a similar reduction in colonoids. The reduced serotonin reuptake capacity may contribute to the increased interstitial serotonin level associated with intestinal inflammation.

NEW & NOTEWORTHY The serotonin reuptake transporter is potently reduced in inflamed areas of Crohn’s ileitis, Crohn’s colitis, and ulcerative colitis. The changes are localized to the intestinal epithelium and can be induced by TNF-α. The serotonin synthesis through tryptophan hydroxylase 1 is unchanged. This regulation is suggested as a mechanism underlying the increased extracellular serotonin levels associated with intestinal inflammation.

INTRODUCTION

Inflammatory bowel disease (IBD) is a collective term for Crohn’s disease (CD) and ulcerative colitis (UC). Both are incurable inflammatory conditions with acute and chronic components mainly involving the intestines. The pathogenesis of the diseases remains poorly understood, but it is multifactorial with involvement of genetic susceptibility, complex and partly unknown environmental factors, the microbiota, and immune dysregulation (25). Epithelial barrier dysfunction and intestinal permeability have long been seen as central to IBD pathogenesis (13). However, in the recent years, it has become clear that the epithelium is involved in many other aspects of IBD pathophysiology through alterations in immune regulation, regeneration, and interplay with the microbiome (36).

Serotonin, also known as 5-hydroxytryptamine (5-HT), is a highly preserved and ubiquitous endogenous monoamine signaling molecule involved in a vast variety of biological processes, including inflammation. Although serotonin can be found in the central nervous system (CNS), lungs, blood, and liver, the majority of serotonin is synthesized in the gastrointestinal (GI) tract, principally by the enterochromaffin cells (EC cells) of the intestinal epithelium (15, 26).

Tryptophan hydroxylase 1 (TPH-1) is the rate-limiting enzyme in the biosynthesis of serotonin, which involves hydroxylation of tryptophan and decarboxylation of the aromatic ring (9). There is no degradation of interstitial serotonin, which is removed by the serotonin reuptake transporter (SERT or 5-HTT). SERT transports serotonin from the extracellular fluid to the intracellular compartment of mucosal epithelial cells, mast cells, enteric neurons, thrombocytes, and hepatocytes. Intracellularly, serotonin participates in biological processes and is stored or degraded to the less biologically active metabolite 5-hydroxyindoleacetic acid (5-HIAA) through monoamine oxidase A (MAO-A) and, to a lesser extent, monoamine oxidase B (MAO-B) (35, 44, 48).

SERT is encoded by the gene SLC6A4 and is the only high-affinity and efficient transporter of 5-HT (40). Transgenic mice with SERT deletion (Slc6a4−/−) compensate to some extent by increasing the abundance of the low-affinity transporters and by minor desensitizing of some serotonin receptors. The compensation is, however, incomplete, and although the mice are viable until adult life, they have developmental defects, disturbed intestinal motility, and exacerbated inflammation in chemical colitis (2,4,6-trinitrobenzene sulfonic acid, TNBS) (4, 5). Furthermore, Stavely et al. (42) have shown that Winnie mice, a UC model that spontaneously develops chronic colitis, have increased mucosal serotonin levels as a result of reduced SERT and unchanged TPH-1. The increased serotonin levels and inadequate compensation in the absence of SERT illustrate its irreplaceability and key role in regulating the interstitial serotonin levels. As reviewed by Coates et al. (7), studies have shown that inflammation may affect SERT and TPH-1, and in return be affected by the serotonin levels, but that sufficient human data are sorely missing.

In our previously published whole genome gene expression meta-analysis of endoscopic pinch biopsies from CD, UC, and healthy controls, SLC6A4 (SERT) was one of the most downregulated genes during inflammation (16). In contrast to the data from animal studies, which is united in reporting SERT downregulation during inflammation (3, 30, 31, 37, 43), SERT and TPH-1 regulation in IBD in human material have been disputed. There are few previous studies, the reports are contradictory, and the designs have made them not strictly comparable with each other. Some have not detected SLC6A4 in the colon (49), others have only investigated the rectum of a pediatric population (8), and the reports have been inconsistent regarding the importance of inflammation in human data (6, 43, 50).

Recent years have seen the introduction of organoids, which are advanced, three-dimensional, and multicellular in vitro models that resemble the tissues of the stem cells origin (1). Organoids derived from the colon are called colonoids, or mini-guts, as they under the right conditions are able to reproduce the different cell populations found in the epithelial monolayer of the colon. They are, therefore, an advantageous model for in vitro experiments on human material, compared with more traditional studies using monocellular cancer cell lines (20).

Tumor necrosis factor α (TNF-α) is a prominent proinflammatory cytokine in acute inflammation (12). Anti-TNF-α therapy is well established in the treatment of numerous inflammatory diseases, including CD and UC (21, 45), demonstrating its involvement in these diseases. TNF-α was, therefore, chosen for mechanistic investigations of inflammatory responses in colonoids.

We aim to identify the regulation of SERT and TPH-1 in patient material from a comprehensive IBD research biobank and to explore a possible mechanism of regulation in human colonoids.

MATERIALS AND METHODS

Ethical considerations.

All participants gave informed written consent, and the study was approved by the Central Norway Regional Committee for Medical and Health Research Ethics (Ref. No. 5.2007.910). The study was conducted in accordance with the general ethical principles of the Declaration of Helsinki.

Clinical material.

The St. Olav IBD research biobank is built on patient-derived material collected at the Gastrointestinal Endoscopy Unit at St. Olav’s University Hospital as described in a study by Granlund et al. (16). In short, four endoscopic pinch biopsies were taken from noninflamed mucosa at the hepatic flexure or ileum in patients with IBD and healthy controls and from maximally inflamed mucosa, if found, in patients with IBD. Biopsies were fixed on formalin for subsequent pathological evaluation or snap frozen and kept on liquid nitrogen for expression analysis. All biopsies were examined and classified by an expert pathologist. Healthy controls were classified as normal. Biopsies from patients with CD or UC with no signs of disease were classified as inactive CD or UC, and active disease was identified as presence of inflammatory infiltrate and diagnosed according to generally accepted histopathological criteria. Samples were randomly drawn from the biobank from each of the following groups of interest: normal control ileum (N ileum), inactive CD ileum (CDn ileum), active CD ileum (CDa ileum), normal control colon (N colon), inactive CD colon (CDn colon), active CD colon (CDa colon), inactive UC colon (UCn), and active UC colon (UCa). Human colonoid cultures were established from endoscopic biopsies collected from three individuals (one patient with UC and two healthy controls) based on optimized protocols from studies by Jung et al. (24) and Mahe et al. (33) as described in Østvik et al.’s study (38). Patient characteristics are given in Tables 1, 2, 3, and 4.

Table 1. Patient characteristics—gene expression microarray

OrganIleum
Colon
Sample groupNormalCDnCDaNormalCDnCDaUCnUCa
No666201894740
Age median (range)41 (26–57)35 (18–62)34 (21–47)45 (19–71)39 (20–63)31 (18–46)46 (21–71)38 (19–72)
Female sex2*438632421
5-ASA/S-ASA0100642826
Steroids01408248

Age is given as median; sex and medication, as numbers.

* Two individuals in control group ileum have unknown sex.

5-ASA, 5-aminosalicylic acid; CDa, Crohn’s disease with active inflammation; CDn, Crohn’s disease with no inflammation; Normal, healthy control; S-ASA, sulfasalazine; UCa, ulcerative colitis with active inflammation; UCn, ulcerative colitis with no inflammation.

Table 2. Patient characteristics—RT-qPCR of pinch biopsies

OrganIleum
Colon
Sample groupNormalCDnCDaNormalCDnCDaUCnUCa
No55555555
Age median (range)38 (27–63)23 (18–44)39 (21–66)52 (18–71)35 (23–59)35 (23–59)37 (29–76)37 (29–76)
Female sex33445511
5-ASA/S-ASA00000033
Steroids00003311

Age is given as median; sex and medication, as numbers. 5-ASA, 5-aminosalicylic acid; CDa, Crohn’s disease with active inflammation; CDn, Crohn’s disease with no inflammation; Normal, healthy control; S-ASA, sulfasalazine; UCa, ulcerative colitis with active inflammation; UCn, ulcerative colitis with no inflammation.

Table 3. Patient characteristics—immunohistochemistry

OrganIleum
Colon
Sample groupNormalCDaNormalCDaUCa
No66666
Age median (range)55 (45–70)49 (23–55)41 (17–71)28 (20–46)35 (25–73)
Female sex15445
5-ASA/S-ASA00003
Steroids01043

Age is given as median; sex and medication, as numbers. 5-ASA, 5-aminosalicylic acid; CDa, Crohn’s disease with active inflammation; CDn, Crohn’s disease with no inflammation; Normal, healthy control; S-ASA, sulfasalazine; UCa, ulcerative colitis with active inflammation; UCn, ulcerative colitis with no inflammation.

Table 4. Patient characteristics—RNA-Seq of laser capture microdissected (LCM) epithelium

Organ
Colon
Sample groupNormalCDnCDaUCnUCa
No65567
Age median (range)50 (35–68)42 (55–20)33 (20–46)47 (33–60)38 (21–66)
Female sex11245
5-ASA/S-ASA03232
Steroids03301

Age is given as median; sex and medication, as numbers. 5-ASA, 5-aminosalicylic acid; CDa, Crohn’s disease with active inflammation; CDn, Crohn’s disease with no inflammation; Normal, healthy control; S-ASA, sulfasalazine; UCa, ulcerative colitis with active inflammation; UCn, ulcerative colitis with no inflammation.

RNA isolation and RT-qPCR.

The sample preparation for microarray analysis and RT-qPCR of patient biopsies has previously been described by Granlund et al. (16). In brief, total RNA was extracted using the Ambion mirVana miRNA Isolation Kit (Applied Biosystems) according to the manufacturer’s protocol. RNA from human primary colonoids was isolated from frozen cell pellets using the RNeasy Mini Kit (Qiagen). The quality of extracted RNA was verified by using the NanoDrop Spectrophotometer (Thermo Scientific) and Bioanalyzer (Agilent Technologies). Samples were only used for subsequent analysis when RNA integrity number (RIN) > 7.

Microarray data are from studies by Granlund et al. (colon; 16) and Thorsvik et al. (ileum; 47), and datasets and sample characteristics are available through array express GSE number E-MTAB-184 and E-MTAB-6593, respectively. Sample site distribution for colon samples is summarized in Supplemental Table S1 (all Supplemental material is available at https://doi.org/10.6084/m9.figshare.12497399). All statistical analyses were done in R software for statistical computing (46). Quality assessment and analysis was performed using tools supplied with the Bioconductor package (14). Final differential expression analysis was done using limma (41). P values were corrected for multiple testing using Benjamini–Hochberg false discovery rate (FDR) correction.

RNA from microdissected samples was isolated using the RNeasy Plus Micro Kit (Qiagen), and quality was assessed using a Bioanalyzer (Agilent Technologies) to measure DV200, the percentage of nucleotide fragments longer than 200 nucleotides. Only samples with DV200 > 70% were used in subsequent RNAseq. To confirm the microarray results, five samples from each of the groups of interest were randomly drawn from the biobank material and analyzed by RT-qPCR. cDNA was prepared using the High-Capacity RNA-to-cDNA Kit (Applied Biosystems). RT-qPCR was performed using TaqMan probes (GAPDH: Hs02758991_g1, SLC6A4: Hs00984349_m1, TPH1: Hs00188220_m1; Life Technologies) on a StepOnePlus Real-Time PCR System and StepOne software v.2.1 (Applied Biosystems). The ΔΔCt method was used to analyze the RT-qPCR results, normalizing across all samples against a common reference sample pool. Statistical analysis of RT-qPCR results was done using one-way ANOVA followed by Sidak multiple comparison correction in GraphPad Prism version 8.00 for Windows (GraphPad Software).

RNA sequencing of microdissected colon epithelium.

Samples were processed and sequenced as previously described by Østvik et al. (38). Sequencing libraries were generated using the TruSeq RNA Access Library Kit (Illumina) according to the manufacturer’s protocol. Libraries were normalized to 22 pM and subjected to clustering, and single-end sequencing was performed for 50 cycles on a HiSeq 2500 instrument (Illumina), according to the manufacturer’s instructions. Base calling was done on the HiSeq instrument by RTA 1.17.21.3. FASTQ files were generated using CASAVA 1.8.2 (Illumina).

The gene abundance quantification was done with featureCounts after alignment of the reads to the genome (GRCh38, v79) using STAR v2.4.1b (29). Duplicate reads were marked using STAR and ignored by featureCounts using the option “–ignoreDup.” Gene models were defined by protein coding genes in the Ensembl GTF file for the GRCh38 genome. Genes that did not meet the criterion of counts per million (CPM) above 1 in more than eight samples were filtered out. The matrix of filtered read counts was imported to R, normalized using trimmed mean of M-values (TMM), and subsequently voom normalized (27). Differential expression was identified by linear models using the edgeR/limma packages, and significance was decided by Benjamini–Hochberg FDR-adjusted P values ≤ 0.05.

Immunohistochemistry.

Six formalin-fixed endoscopic biopsies from each of the groups, N ileum, CDa ileum, N colon, CDa colon, and UCa colon, were stained after heat-induced antigen retrieval in Tris-EDTA buffer (pH9) using SERT-specific antibody (no. MAB5618, Merck Millipore, Sigma-Aldrich, AB_2190560), dilution 1:1200, for 1 h at room temperature. Detection was done using the Dako REAL EnVision Detection System, Peroxidase/DAB+, Rabbit/Mouse (no. K5007, Dako, Agilent Technologies, Inc.). Positive cells on all slides were quantified by two independent observers blinded to the pathologist’s classification, scoring each section on signal intensity from 0 to 3. One-way ANOVA followed by Dunnett’s multiple comparison test was applied on the average score for evaluation of statistically significant difference in anti-SERT positivity between the sample groups.

Human colonoid experiment.

The human colonoid experiment was performed as described by Østvik et al. (38), using protocols based on work by Mahe et al. (33) and Jung et al. (24). Full details are given in the Colonoid Experiment Protocol in the Supplemental Material (https://doi.org/10.6084/m9.figshare.12497399.v1). In brief, colonoids from three donors (one UC, two healthy controls) were expanded for 10–12 days, before inducing differentiation over 3 days by reducing Wnt-3A (WNT3a-producing cells; ATCC Cat. No. CRL-2647, RRID:CVCL_0635) concentration to 5%, removing nicotinamide (no. N3376-100G; MerckMillipore) and MAPK inhibitor SB202190 (no. S7067; Sigma-Aldrich) from the medium, and adding the pan-Notch inhibitor DAPT (4.324 µg/mL, no. 2634; Bio-Techne). On the fourth day, the experiment was performed with the differentiation medium. Fully differentiated colonoids display a polarized three-dimensional (3-D) structure, with all relevant epithelial cell types present as described previously (38).

The TNF-α-incubated cultures received 100 ng/mL TNF-α (no. 300-01A; PeproTech) for 24 h at 37°C, while differentiation medium was added to the controls. After 24 h, cells were collected and snap frozen for later RNA extraction. One of the donors was used to provide material for two organoid cultures, making a total of four organoid cultures used in the experiment. The experiment was repeated twice, aiming for eight pairs of TNF-α-incubated colonoid cultures with matched controls. However, RNA isolation failed for one of the replicates, resulting in seven replicates available for RT-qPCR. Significance was tested using paired t test, comparing the TNF response in each organoid culture with a matched control, to capture the culture-specific responses.

Data availability.

Gene expression microarray datasets for colon and ileum are available through array express GSE number E-MTAB-184 and E-MTAB-6593, respectively. Raw data from gene expression microarray, RT-qPCR, IHC, and RNA-Seq are given in the supplemental document Raw Data (https://doi.org/10.6084/m9.figshare.12497399.v1).

Statistics.

All significance measures are adjusted for multiple testing using appropriate methods and were deemed significant when adjusted P value was < 0.05. The adjustments vary between the analyses and are given in the appropriate materials and methods section.

The colonoid analysis was done as a paired analysis, where each TNF-stimulated sample was matched by unstimulated control of the same colonoid culture. In all other cases, the mean of each of the groups was compared with the mean of the appropriate control group, and this ratio is referred to as fold change (FC) with respect to case/control. An FC of 2 would describe a twofold higher value in the case group as compared with the control. Results are given as the binary logarithm of this ratio (log2FC) with 95% confidence intervals.

RESULTS

SLC6A4 is downregulated in IBD with active inflammation, whereas TPH-1 is unchanged.

Microarray analyses showed downregulation of SLC6A4 in active CD ileitis (log2FC = −1.26, 95% confidence interval (CI) [−1.87, −0.65], P = 0.0001), as well as in active CD colitis (log2FC = −0.95, 95% CI [−1.13, −0.77], P < 0.0001) and active UC colitis (log2FC = −0.96, 95% CI [−1.09, −0.83], P < 0.0001) (Fig. 1). RT-qPCR also showed SLC6A4 as downregulated when compared with healthy controls during active CD ileitis (log2FC = −2.15, 95% CI [−3.34, −0.96], P = 0.001), as well as in active CD colitis (log2FC = −6.03, 95% CI [−8.08, −3.97], P < 0.0001) and active UC colitis (log2FC = −8.37, 95% CI [−10.43, −6.31], P < 0.0001) (Fig. 2).

Fig. 1.

Fig. 1.Gene expression microarray of endoscopic pinch biopsies from the ileum and colon. N ileum = healthy control ileum (n = 6), CDn ileum = Crohn’s disease (CD) ileum with no inflammation (n = 6), CDa ileum = CD ileum with active inflammation (n = 6), N colon = healthy control colon (n = 20), CDn colon = CD colon with no inflammation (n = 18), CDa colon = CD colon with active inflammation (n = 9), UCn colon = ulcerative colitis (UC) colon with no inflammation (n = 47), UCa colon = UC colon with active inflammation (n = 40). A: gene expression microarray showed significant and potent downregulation of SLC6A4 in active inflammatory bowel disease: CDa ileum log2FC −1.26, 95% CI [−1.87, −0.65], P = 0.0001, CDa colon log2FC −0.95, 95% CI [−1.13, −0.77], P < 0.0001, and UCa colon log2FC −0.96, 95% CI [−1.09, −0.83], P < 0.0001. B: TPH1 was unregulated. Differential expression analysis was done using limma linear models with least squares regression and empirical Bayes moderated t statistics. ****P value < 0.0001. CDa,  Crohn’s disease with active inflammation; CDn, Crohn’s disease with no inflammation; N, healthy control; UCn, ulcerative colitis with no inflammation; UCa, ulcerative colitis with active inflammation.


Fig. 2.

Fig. 2.RT-qPCR of endoscopic pinch biopsies from the ileum and colon. A: results verified significant downregulation of SLC6A4 in active inflammatory bowel disease: CDa ileum (log2FC −2.15, 95% CI [-3.34, −0.96], P = 0.001), CDa colon (log2FC −6.03, 95% CI [-8.08, −3.97], P < 0.0001), and UCa colon (log2FC −8.37, 95% CI [-10.43, −6.31], P < 0.0001). B: TPH1 was unregulated. **P value < 0.01; ****P value < 0.0001. Statistical analysis was done with one-way ANOVA followed by Sidak multiple comparison correction. n = 5 for all groups. CDa, Crohn’s disease with active inflammation; CDn, Crohn’s disease with no inflammation; N, healthy control; UCn, ulcerative colitis with no inflammation; UCa, ulcerative colitis with active inflammation.


Immunohistochemistry confirmed that SERT downregulation also occurred at the protein level and that this regulation mainly takes place in the epithelial monolayer. Evaluation of staining intensity showed a significantly decreased anti-SERT reactivity level during inflammation, active CD ileitis (mean difference = −1.50, 95% CI [−2.81, −0.19], P = 0.025), as well as in active CD colitis (mean difference = −1.25, 95% CI [−2.44, −0.06], P = 0.040) and active UC colitis (mean difference = −1.67, 95% CI [−2.86, −0.48], P = 0.007) (Fig. 3).

Fig. 3.

Fig. 3.Immunohistochemistry of serotonin reuptake transporter (SERT). A: immunohistochemistry images. All images are of the same magnification and the bar represents 50 µm. Insets show a magnified portion of the epithelium where there was a significant reduction in anti-SERT signal positivity in active Crohn’s disease (CD) and ulcerative colitis (UC). B: anti-SERT signal intensity scoring. The epithelial anti-SERT signal intensity in six sections from each group were given an intensity score between 0 and 3 by two independent observers, blinded to each other and the pathologist’s classification. One-way ANOVA followed by Dunnett’s multiple comparison test showed a significant difference in anti-SERT intensity in the contrasts CDa ileum versus N ileum (P = 0.03), CDa colon versus N colon (P = 0.04) and UCa colon versus N colon (P = 0.007). *P value < 0.05; **P value < 0.01. n = 6 for all groups. N, healthy control. CDa , Crohn’s disease with active inflammation. UCa, ulcerative colitis with active inflammation.


Microdissection allowed us to isolate the colonic epithelial monolayer. The isolated material was analyzed with RNA-Seq, which showed that SLC6A4 mRNA was very potently downregulated during active inflammation in the colon of both CD (log2FC = −4.64, 95% CI [−6.24, −3.04], P < 0.001) and UC (log2FC = −5.40, 95% CI [−6.93, −3–88], P < 0.001) compared with healthy controls (Fig. 4). No regulation of SLC6A4 was seen in inactive disease. TPH1 was analyzed for the same contrasts as SLC6A4, with no significant changes found during active or inactive IBD.

RT-qPCR analysis of human colonoids incubated with TNF-α for 24 h revealed that TNF-α induced a significant downregulation of SLC6A4 (log2FC −2.78, 95% CI [−3.89, −1.66], P < 0.001) (Fig. 5).

Fig. 4.

Fig. 4.RNA-sequencing of laser capture microdissected colon epithelium N = Healthy control (n = 6). CDn, Crohn’s disease (CD) with no inflammation (n = 5). CDa = CD with active inflammation (n = 5). UCn, ulcerative colitis (UC) with no inflammation (n = 6). UCa, UC with active inflammation (n = 7). Differential expression was identified using limma voom and corrected for multiple testing by Benjamini–Hochberg false discovery rate adjusted P values. A: RNA-sequencing showed potent and significant downregulation of SLC6A4 in the colon epithelium of active inflammatory bowel disease: CDa [log2FC −4.64, 95% CI (−6.24, −3.04), P < 0.001] and UCa [log2FC −5.40, 95% CI (−6.93, −3–88), P < 0.001]. B: TPH-1 was unregulated. ****P value < 0.0001.


Fig. 5.

Fig. 5.RT-qPCR of tumor necrosis factor α (TNF-α)-incubated colonoids. ***P value < 0.001. n = 7. Each TNF-α-incubated colonoid culture was paired with control cultures from the same host. After 24 h of incubation with 100 ng/mL TNF-α, the cultures were collected, and isolated RNA was analyzed with RT-qPCR. Paired two-tailed t test showed a significant downregulation of SLC6A4 in response to TNF-α with log2FC −2.78, 95% CI [−3.89, −1.66], P < 0.001.


DISCUSSION

Our data comprehensively demonstrate for the first time that SLC6A4 mRNA and SERT protein are downregulated in active ileal and colonic CD, as well as in UC. The regulation is mainly attributed to downregulation in the epithelial monolayer. In addition, we demonstrate that incubation with the highly IBD-relevant cytokine TNF-α induces SLC6A4 mRNA downregulation in human colonoids, representing an interesting potential activator of the downregulation observed in the biopsies.

TPH-1, the rate-limiting enzyme in serotonin synthesis, was not differentially expressed between the compared groups, suggesting an unchanged serotonin production. Our research group and others have previously found that there are few differences at mRNA levels between inactive IBD and healthy controls (16). As expected, the analyses did not show differential expression of SLC6A4 between inactive IBD and healthy controls.

With SERT being the only efficient transporter of serotonin, and the intestinal serotonin synthesis seemingly unchanged, it is plausible that the extensive reduction of epithelial SERT leads to an ineffective elimination of serotonin with a proportional increase in interstitial serotonin levels, as observed in the murine models by Stavely et al. (42). This is further corroborated by a study performed in guinea pigs by Linden et al. (30), where it was demonstrated that the selective SERT inhibitor fluoxetine significantly decreased elimination of 5-HT through reuptake by SERT upon mechanical stimulation of 5-HT release. Other studies have previously shown that the colon adenocarcinoma cell line Caco-2 downregulates SERT as a response to TNF and/or IFN-γ exposure (2, 10). Our data from three independent fully differentiated primary cell lines derived from healthy individuals show a similar downregulation of SLC6A4 in response to 24 h of TNF (100 ng/mL) exposure, suggesting that this response is relevant to the nonmalignant epithelium. The pleiotropic nature of TNF action in the gut, including a number of intracellular signaling pathways, is complex and cannot be elucidated by data generated in this study (28).

Consequences of reduced SERT can be observed in mice with SERT deficiency (SERT−/−), and these knockout mice have lifelong increased extracellular serotonin levels accompanied by cardiac valve and placental fibrosis (11, 19). Rats exposed to supraphysiological levels of serotonin developed similar cardiac valve fibrosis (18). In humans, high serotonin levels are associated with fibrosis in the mesentery, cardiac valves, the respiratory tract, and a wide range of other tissues in patients with serotonin-producing tumors (17, 32, 39). Although animal studies strongly suggest that abolished SERT activity increases extracellular serotonin levels resulting in exacerbated colitis and increased fibrosis, the effect of the epithelial downregulation of SERT as described here is more difficult to assess. It could be argued that an attenuated epithelial SERT activity is compensated by thrombocyte sequestration of 5-HT, thus preventing mucosal exposure to increased serotonin levels. However, as demonstrated by the known role of EC cell serotonin on sensory neurons of the enteric nervous system, at least a paracrine effect should exist increasing the local exposure to serotonin. Moreover, a recent study from Manzella et al. (34) shows that serum serotonin levels are increased in active CD (but not UC) and that levels are highly correlated with disease status. SERT downregulation in the gastrointestinal mucosa of active CD is suggested as a possible mechanism for this increase in serum serotonin, further supporting the notion that changes in serotonin reabsorption in IBD can result in an increased extracellular serotonin level.

A limitation to our study is the imbalance by sex within the PCR and IHC analyses. To evaluate whether the data could suggest a sex difference with regard to SLC6A4 expression, we repeated the microarray data analyses in sex-stratified sample groups, detecting significant downregulation of SLC6A4 in inflamed samples independent of sex (Supplemental Fig. S1). Because of our choice to standardize the sampling location of uninflamed samples to the hepatic flexure, there is also an inherent location imbalance in the sample pools as shown in Supplemental Table S2. The SLC6A4 downregulation in inflamed colon samples is unison regardless of sample location, but the average expression is higher in ileal samples. This was addressed by using uninflamed ileal samples as control for ileal analyses.

The increasing evidence of serotonin’s complex involvement and modulation of inflammation suggests a potential role of serotonin as a prominent regulatory molecule of importance in IBD and calls for further research to substantiate its regulation during inflammation. By combining gene expression and protein analyses of patient material from a comprehensive IBD research biobank with in vitro experiments on human colonoids, we consider our study to be a small, but well-controlled, robust, targeted, and precise set of human data providing novel and needed insight in the regulation of SERT and TPH-1 in IBD.

In conclusion, our data demonstrate that SERT is strongly downregulated during both ileal and colonic inflammation in CD and UC and that the regulation is located to the epithelial monolayer. We also suggest TNF-α as a possible activator of this regulation, demonstrating a decrease in SLC6A4 expression in human colonoids in response to TNF-α incubation. SERT is the only high-affinity and efficient transporter of serotonin in the GI tract, and the inadequate compensation in its absence illustrates its irreplaceability. Intestinal expression of the serotonin synthesis rate-limiting enzyme TPH-1 is unchanged. Consequently, we would argue that a reduction of SERT, as observed in active CD and UC, contributes to the increased interstitial serotonin level associated with intestinal inflammation.

GRANTS

J. W. Jørandli is the recipient of a PhD Grant from the Norwegian University of Science and Technology (NTNU). A. van B. Granlund is funded by the Norwegian Research Council - FRIPRO (262549) and NTNU Outstanding Academic Fellows Programme. The laboratory work was funded by the Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology. The RNA sequencing was performed in close collaboration with the Genomics Core Facility (GCF), Norwegian University of Science and Technology (NTNU). GCF is funded by the Faculty of Medicine and Health Sciences at NTNU and Central Norway Regional Health Authority. The graphical abstract was created with BioRender.com.

DISCLAIMER

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.W.J., B.I.G., A.K.S., and A.v.G. conceived and designed research; J.W.J., S.T., H.K.S., B.K., S.S., and A.v.G. performed experiments; J.W.J., A.K.S., and A.v.G. analyzed data; J.W.J., A.K.S., and A.v.G. interpreted results of experiments; J.W.J. prepared figures; J.W.J. drafted manuscript; J.W.J., A.K.S., and A.v.G. edited and revised manuscript; J.W.J., S.T., H.K.S., B.K., S.S., B.I.G., A.K.S., and A.v.G. approved final version of manuscript.

ENDNOTES

At the request of the authors, readers are herein alerted to the fact that additional materials related to this manuscript may be found at https://doi.org/10.6084/m9.figshare.12497399.v1: the document titled Raw Data contains gene expression microarray, RT-qPCR, IHC, and RNA-Seq data. These materials are not a part of this manuscript and have not undergone peer review by the American Physiological Society (APS). APS and the journal editors take no responsibility for these materials, website address, or for any links to or from it.

REFERENCES

  • 1. Method of the year 2017: organoids. Nat Methods 15: 1, 2018. doi:10.1038/nmeth.4575.
    Crossref | Web of Science | Google Scholar
  • 2. Barbaro MR, Di Sabatino A, Cremon C, Giuffrida P, Fiorentino M, Altimari A, Bellacosa L, Stanghellini V, Barbara G. Interferon-γ is increased in the gut of patients with irritable bowel syndrome and modulates serotonin metabolism. Am J Physiol Gastrointest Liver Physiol 310: G439–G447, 2016. doi:10.1152/ajpgi.00368.2015.
    Link | Web of Science | Google Scholar
  • 3. Bertrand PP, Barajas-Espinosa A, Neshat S, Bertrand RL, Lomax AE. Analysis of real-time serotonin (5-HT) availability during experimental colitis in mouse. Am J Physiol Gastrointest Liver Physiol 298: G446–G455, 2010. doi:10.1152/ajpgi.00318.2009.
    Link | Web of Science | Google Scholar
  • 4. Bischoff SC, Mailer R, Pabst O, Weier G, Sedlik W, Li Z, Chen JJ, Murphy DL, Gershon MD. Role of serotonin in intestinal inflammation: knockout of serotonin reuptake transporter exacerbates 2,4,6-trinitrobenzene sulfonic acid colitis in mice. Am J Physiol Gastrointest Liver Physiol 296: G685–G695, 2009. doi:10.1152/ajpgi.90685.2008.
    Link | Web of Science | Google Scholar
  • 5. Chen JJ, Li Z, Pan H, Murphy DL, Tamir H, Koepsell H, Gershon MD. Maintenance of serotonin in the intestinal mucosa and ganglia of mice that lack the high-affinity serotonin transporter: Abnormal intestinal motility and the expression of cation transporters. J Neurosci 21: 6348–6361, 2001. doi:10.1523/JNEUROSCI.21-16-06348.2001.
    Crossref | PubMed | Web of Science | Google Scholar
  • 6. Coates MD, Mahoney CR, Linden DR, Sampson JE, Chen J, Blaszyk H, Crowell MD, Sharkey KA, Gershon MD, Mawe GM, Moses PL. Molecular defects in mucosal serotonin content and decreased serotonin reuptake transporter in ulcerative colitis and irritable bowel syndrome. Gastroenterology 126: 1657–1664, 2004. doi:10.1053/j.gastro.2004.03.013.
    Crossref | PubMed | Web of Science | Google Scholar
  • 7. Coates MD, Tekin I, Vrana KE, Mawe GM. Review article: the many potential roles of intestinal serotonin (5-hydroxytryptamine, 5-HT) signalling in inflammatory bowel disease. Aliment Pharmacol Ther 46: 569–580, 2017. doi:10.1111/apt.14226.
    Crossref | PubMed | Web of Science | Google Scholar
  • 8. Faure C, Patey N, Gauthier C, Brooks EM, Mawe GM. Serotonin signaling is altered in irritable bowel syndrome with diarrhea but not in functional dyspepsia in pediatric age patients. Gastroenterology 139: 249–258, 2010. doi:10.1053/j.gastro.2010.03.032.
    Crossref | PubMed | Web of Science | Google Scholar
  • 9. Fitzpatrick PF. Tetrahydropterin-dependent amino acid hydroxylases. Annu Rev Biochem 68: 355–381, 1999. doi:10.1146/annurev.biochem.68.1.355.
    Crossref | PubMed | Web of Science | Google Scholar
  • 10. Foley KF, Pantano C, Ciolino A, Mawe GM. IFN-γ and TNF-α decrease serotonin transporter function and expression in Caco2 cells. Am J Physiol Gastrointest Liver Physiol 292: G779–G784, 2007. doi:10.1152/ajpgi.00470.2006.
    Link | Web of Science | Google Scholar
  • 11. Fox MA, Jensen CL, French HT, Stein AR, Huang S-J, Tolliver TJ, Murphy DL. Neurochemical, behavioral, and physiological effects of pharmacologically enhanced serotonin levels in serotonin transporter (SERT)-deficient mice. Psychopharmacology (Berl) 201: 203–218, 2008. doi:10.1007/s00213-008-1268-7.
    Crossref | PubMed | Web of Science | Google Scholar
  • 12. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 340: 448–454, 1999. doi:10.1056/NEJM199902113400607.
    Crossref | PubMed | Web of Science | Google Scholar
  • 13. Gardiner KR, Anderson NH, Rowlands BJ, Barbul A. Colitis and colonic mucosal barrier dysfunction. Gut 37: 530–535, 1995. doi:10.1136/gut.37.4.530.
    Crossref | PubMed | Web of Science | Google Scholar
  • 14. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, Ellis B, Gautier L, Ge Y, Gentry J, Hornik K, Hothorn T, Huber W, Iacus S, Irizarry R, Leisch F, Li C, Maechler M, Rossini AJ, Sawitzki G, Smith C, Smyth G, Tierney L, Yang JYH, Zhang J. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 5: R80, 2004. doi:10.1186/gb-2004-5-10-r80.
    Crossref | PubMed | Web of Science | Google Scholar
  • 15. Gershon MD. 5-Hydroxytryptamine (serotonin) in the gastrointestinal tract. Curr Opin Endocrinol Diabetes Obes 20: 14–21, 2013. doi:10.1097/MED.0b013e32835bc703.
    Crossref | PubMed | Web of Science | Google Scholar
  • 16. Granlund A, Flatberg A, Østvik AE, Drozdov I, Gustafsson BI, Kidd M, Beisvag V, Torp SH, Waldum HL, Martinsen TC, Damås JK, Espevik T, Sandvik AK. Whole genome gene expression meta-analysis of inflammatory bowel disease colon mucosa demonstrates lack of major differences between Crohn’s disease and ulcerative colitis. PLoS One 8: e56818, 2013. doi:10.1371/journal.pone.0056818.
    Crossref | PubMed | Web of Science | Google Scholar
  • 17. Gustafsson BI, Hauso O, Drozdov I, Kidd M, Modlin IM. Carcinoid heart disease. Int J Cardiol 129: 318–324, 2008. doi:10.1016/j.ijcard.2008.02.019.
    Crossref | PubMed | Web of Science | Google Scholar
  • 18. Gustafsson BI, Tømmerås K, Nordrum I, Loennechen JP, Brunsvik A, Solligård E, Fossmark R, Bakke I, Syversen U, Waldum H. Long-term serotonin administration induces heart valve disease in rats. Circulation 111: 1517–1522, 2005. doi:10.1161/01.CIR.0000159356.42064.48.
    Crossref | PubMed | Web of Science | Google Scholar
  • 19. Hadden C, Fahmi T, Cooper A, Savenka AV, Lupashin VV, Roberts DJ, Maroteaux L, Hauguel-de Mouzon S, Kilic F. Serotonin transporter protects the placental cells against apoptosis in caspase 3-independent pathway. J Cell Physiol 232: 3520–3529, 2017. doi:10.1002/jcp.25812.
    Crossref | PubMed | Web of Science | Google Scholar
  • 20. In JG, Foulke-Abel J, Estes MK, Zachos NC, Kovbasnjuk O, Donowitz M. Human mini-guts: new insights into intestinal physiology and host-pathogen interactions. Nat Rev Gastroenterol Hepatol 13: 633–642, 2016. doi:10.1038/nrgastro.2016.142.
    Crossref | PubMed | Web of Science | Google Scholar
  • 21. Järnerot G, Hertervig E, Friis-Liby I, Blomquist L, Karlén P, Grännö C, Vilien M, Ström M, Danielsson A, Verbaan H, Hellström PM, Magnuson A, Curman B. Infliximab as rescue therapy in severe to moderately severe ulcerative colitis: a randomized, placebo-controlled study. Gastroenterology 128: 1805–1811, 2005. doi:10.1053/j.gastro.2005.03.003.
    Crossref | PubMed | Web of Science | Google Scholar
  • 24. Jung P, Sato T, Merlos-Suárez A, Barriga FM, Iglesias M, Rossell D, Auer H, Gallardo M, Blasco MA, Sancho E, Clevers H, Batlle E. Isolation and in vitro expansion of human colonic stem cells. Nat Med 17: 1225–1227, 2011. doi:10.1038/nm.2470.
    Crossref | PubMed | Web of Science | Google Scholar
  • 25. Kaser A, Zeissig S, Blumberg RS. Inflammatory bowel disease. Annu Rev Immunol 28: 573–621, 2010. doi:10.1146/annurev-immunol-030409-101225.
    Crossref | PubMed | Web of Science | Google Scholar
  • 26. Kim DY, Camilleri M. Serotonin: a mediator of the brain-gut connection. Am J Gastroenterol 95: 2698–2709, 2000. doi:10.1111/j.1572-0241.2000.03177.x.
    Crossref | PubMed | Web of Science | Google Scholar
  • 27. Law CW, Chen Y, Shi W, Smyth GK. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol 15: R29, 2014. doi:10.1186/gb-2014-15-2-r29.
    Crossref | PubMed | Web of Science | Google Scholar
  • 28. Leppkes M, Roulis M, Neurath MF, Kollias G, Becker C. Pleiotropic functions of TNF-α in the regulation of the intestinal epithelial response to inflammation. Int Immunol 26: 509–515, 2014. doi:10.1093/intimm/dxu051.
    Crossref | PubMed | Web of Science | Google Scholar
  • 29. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30: 923–930, 2014. doi:10.1093/bioinformatics/btt656.
    Crossref | PubMed | Web of Science | Google Scholar
  • 30. Linden DR, Chen JX, Gershon MD, Sharkey KA, Mawe GM. Serotonin availability is increased in mucosa of guinea pigs with TNBS-induced colitis. Am J Physiol Gastrointest Liver Physiol 285: G207–G216, 2003. doi:10.1152/ajpgi.00488.2002.
    Link | Web of Science | Google Scholar
  • 31. Linden DR, Foley KF, McQuoid C, Simpson J, Sharkey KA, Mawe GM. Serotonin transporter function and expression are reduced in mice with TNBS-induced colitis. Neurogastroenterol Motil 17: 565–574, 2005. doi:10.1111/j.1365-2982.2005.00673.x.
    Crossref | PubMed | Web of Science | Google Scholar
  • 32. Lundin L, Norheim I, Landelius J, Oberg K, Theodorsson-Norheim E. Carcinoid heart disease: relationship of circulating vasoactive substances to ultrasound-detectable cardiac abnormalities. Circulation 77: 264–269, 1988. doi:10.1161/01.CIR.77.2.264.
    Crossref | PubMed | Web of Science | Google Scholar
  • 33. Mahe MM, Sundaram N, Watson CL, Shroyer NF, Helmrath MA. Establishment of human epithelial enteroids and colonoids from whole tissue and biopsy. J Vis Exp (97): 2015. doi:10.3791/52483.
    Crossref | PubMed | Web of Science | Google Scholar
  • 34. Manzella CR, Jayawardena D, Pagani W, Li Y, Alrefai WA, Bauer J, Jung B, Weber CR, Gill RK. Serum serotonin differentiates between disease activity states in Crohn’s patients. Inflamm Bowel Dis 26: 1607–1618, 2020. doi:10.1093/ibd/izaa208.
    Crossref | PubMed | Web of Science | Google Scholar
  • 35. Martel F, Monteiro R, Lemos C. Uptake of serotonin at the apical and basolateral membranes of human intestinal epithelial (Caco-2) cells occurs through the neuronal serotonin transporter (SERT). J Pharmacol Exp Ther 306: 355–362, 2003. doi:10.1124/jpet.103.049668.
    Crossref | PubMed | Web of Science | Google Scholar
  • 36. Martini E, Krug SM, Siegmund B, Neurath MF, Becker C. Mend your fences: the epithelial barrier and its relationship with mucosal immunity in inflammatory bowel disease. Cell Mol Gastroenterol Hepatol 4: 33–46, 2017. doi:10.1016/j.jcmgh.2017.03.007.
    Crossref | PubMed | Web of Science | Google Scholar
  • 37. O’Hara JR, Ho W, Linden DR, Mawe GM, Sharkey KA. Enteroendocrine cells and 5-HT availability are altered in mucosa of guinea pigs with TNBS ileitis. Am J Physiol Gastrointest Liver Physiol 287: G998–G1007, 2004. doi:10.1152/ajpgi.00090.2004.
    Link | Web of Science | Google Scholar
  • 38. Østvik AE, Svendsen TD, Granlund AVB, Doseth B, Skovdahl HK, Bakke I, Thorsvik S, Afroz W, Walaas GA, Mollnes TE, Gustafsson BI, Sandvik AK, Bruland T. Intestinal epithelial cells express immunomodulatory ISG15 during active ulcerative colitis and Crohn’s disease. J Crohn’s Colitis 14: 920–934, 2020. doi:10.1093/ecco-jcc/jjaa022.
    Crossref | PubMed | Google Scholar
  • 39. Rodríguez Laval V, Pavel M, Steffen IG, Baur AD, Dilz LM, Fischer C, Detjen K, Prasad V, Pascher A, Geisel D, Denecke T. Mesenteric fibrosis in midgut neuroendocrine tumors: functionality and radiological features. Neuroendocrinology 106: 139–147, 2018. doi:10.1159/000474941.
    Crossref | PubMed | Web of Science | Google Scholar
  • 40. Schömig E, Lazar A, Gründemann D. Extraneuronal monoamine transporter and organic cation transporters 1 and 2: a review of transport efficiency. Handb Exp Pharmacol 175: 151–180, 2006. doi:10.1007/3-540-29784-7_8.
    Crossref | PubMed | Google Scholar
  • 41. Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 3: e3, 2004. doi:10.2202/1544-6115.1027.
    Crossref | PubMed | Google Scholar
  • 42. Stavely R, Fraser S, Sharma S, Rahman AA, Stojanovska V, Sakkal S, Apostolopoulos V, Bertrand P, Nurgali K. The onset and progression of chronic colitis parallels increased mucosal serotonin release via enterochromaffin cell hyperplasia and downregulation of the serotonin reuptake transporter. Inflamm Bowel Dis 24: 1021–1034, 2018. doi:10.1093/ibd/izy016.
    Crossref | PubMed | Web of Science | Google Scholar
  • 43. Tada Y, Ishihara S, Kawashima K, Fukuba N, Sonoyama H, Kusunoki R, Oka A, Mishima Y, Oshima N, Moriyama I, Yuki T, Ishikawa N, Araki A, Harada Y, Maruyama R, Kinoshita Y. Downregulation of serotonin reuptake transporter gene expression in healing colonic mucosa in presence of remaining low-grade inflammation in ulcerative colitis. J Gastroenterol Hepatol 31: 1443–1452, 2016. doi:10.1111/jgh.13268.
    Crossref | PubMed | Web of Science | Google Scholar
  • 44. Takayanagi S, Hanai H, Kumagai J, Kaneko E. Serotonin uptake and its modulation in rat jejunal enterocyte preparation. J Pharmacol Exp Ther 272: 1151–1159, 1995.
    PubMed | Web of Science | Google Scholar
  • 45. Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, DeWoody KL, Schaible TF, Rutgeerts PJ. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 Study Group. N Engl J Med 337: 1029–1035, 1997. doi:10.1056/NEJM199710093371502.
    Crossref | PubMed | Web of Science | Google Scholar
  • 46. Team RC. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2013.
    Google Scholar
  • 47. Thorsvik S, Bakke I, van Beelen Granlund A, Røyset ES, Damås JK, Østvik AE, Sandvik AK. Expression of neutrophil gelatinase-associated lipocalin (NGAL) in the gut in Crohn’s disease. Cell Tissue Res 374: 339–348, 2018. doi:10.1007/s00441-018-2860-8.
    Crossref | PubMed | Web of Science | Google Scholar
  • 48. Wade PR, Chen J, Jaffe B, Kassem IS, Blakely RD, Gershon MD. Localization and function of a 5-HT transporter in crypt epithelia of the gastrointestinal tract. J Neurosci 16: 2352–2364, 1996. doi:10.1523/JNEUROSCI.16-07-02352.1996.
    Crossref | PubMed | Web of Science | Google Scholar
  • 49. Wojtal KA, Eloranta JJ, Hruz P, Gutmann H, Drewe J, Staumann A, Beglinger C, Fried M, Kullak-Ublick GA, Vavricka SR. Changes in mRNA expression levels of solute carrier transporters in inflammatory bowel disease patients. Drug Metab Dispos 37: 1871–1877, 2009. doi:10.1124/dmd.109.027367.
    Crossref | PubMed | Web of Science | Google Scholar
  • 50. Yu FY, Huang SG, Zhang HY, Ye H, Chi HG, Zou Y, Lv RX, Zheng XB. Comparison of 5-hydroxytryptophan signaling pathway characteristics in diarrhea-predominant irritable bowel syndrome and ulcerative colitis. World J Gastroenterol 22: 3451–3459, 2016. doi:10.3748/wjg.v22.i12.3451.
    Crossref | PubMed | Web of Science | Google Scholar

AUTHOR NOTES