The authors have declared that no competing interests exist.
Conceived and designed the experiments: CB MB RC SM NDB. Performed the experiments: RC NG GR. Analyzed the data: RC DG AR MGM NG MF LA MT. Contributed reagents/materials/analysis tools: NG MGM DG. Wrote the paper: MB CB SM FC GR.
5HTTLPR polymorphism of serotonin transporter yields short (S) and long (L) alleles. SS and LS genotypes are associated with reduced expression of serotonin transporter. This cross-sectional study investigated the association of 5HTTLPR with symptom severity of irritable bowel syndrome (IBS). Patients with IBS (Rome III) and healthy controls were included. Genomic DNA was extracted from saliva, and 5HTTLPR alleles were assessed by polymerase chain reaction. IBS symptom severity was evaluated by means of IBS-SSS questionnaire. Two hundreds and four IBS patients (159 females; mean age: 39.6±12.3 years; 106 with constipation: C-IBS; 98 with diarrhea: D-IBS) and 200 healthy controls (154 females; mean age: 40.4±15.8 years) were enrolled. The overall IBS-SSS value was higher in LS/SS than LL patients (319.0±71.5 versus 283.8±62.3; P = 0.0006). LS/SS patients had also higher values of abdominal pain (59.7±21.0 versus 51.0±18.8; P = 0.020) and bowel dissatisfaction (80.1±23.9 versus 70.5±22.8; P = 0.035). The overall IBS-SSS values in C-IBS and D-IBS patients were 317.2±68.3 and 296.1±71.4, respectively (P = 0.192), with significantly higher values for abdominal distension (65.0±24.4 versus 51.4±24.8; P = 0.0006), but not for bowel dissatisfaction (80.5±21.7 versus 72.9±25.7; P = 0.138). Frequencies of 5HTTLPR genotypes did not differ significantly when comparing IBS patients (overall or upon stratification in C-IBS and D-IBS) with healthy controls. In conclusion, the LS and SS genotypes are significantly correlated with IBS symptom severity, although their possible direct causal role remains to be proven. In addition, the present findings do not support an association of 5HTTLPR with IBS or its clinical presentation in terms of bowel habit predominance.
Irritable bowel syndrome (IBS) is a functional digestive disorder characterized by abdominal pain/discomfort and changes in bowel habit (constipation and/or diarrhoea), which can be associated with alterations of gastrointestinal (GI) transit, visceral hypersensitivity and psychopathological comorbidities.
From a pathogenic standpoint, IBS is a complex, heterogeneous disorder, whose development is widely considered as multifactorial in nature.
The involvement of serotonin (5-HT) in the pathophysiology of IBS is also an area of active investigation, owing to its presence in both GI tract and brain. Indeed, 5-HT acts as a main regulator of GI motility, secretion and sensory signalling, and plays also a pivotal role in the control of mood at level of the central nervous system (CNS).
The promoter region of human SERT gene contains a polymorphism, designated as ‘5-HT transporter length polymorphic region’ (5HTTLPR), which consists of a 44-base pair deletion/insertion, resulting in a short (S) and long (L) allele.
The primary objective was to examine the association of 5HTTLPR with the overall symptom severity in IBS patients. The secondary objectives were: 1) to evaluate the association of 5HTTLPR with the severity of different symptoms; 2) to evaluate the overall symptom severity and the severity of different IBS symptoms in patients with predominance of constipation (C-IBS) or diarrhoea (D-IBS); 3) to evaluate the association of 5HTTLPR with IBS in all IBS patients, as well as in the subgroups of C-IBS or D-IBS, by comparisons with healthy controls.
This study was conducted on native Italians of Caucasian origin, gender of either sex, age equal or higher than 18 years, and diagnosis of IBS according to Rome III criteria,
At least two samples of saliva were collected from each IBS patient and healthy volunteer in different days and stored at −20°C. All subsequent analytical procedures were performed at the Division of Pharmacology and Chemotherapy. The severity of IBS symptoms was evaluated by the questionnaire IBS-Symptom Severity Score (IBS-SSS).
Genomic DNA was isolated from saliva by the alkaline method using QIAamp DNA blood MINI kit (Qiagen, Milano, Italy), quantified by spectrophotometry, and visualized by agarose gel electrophoresis. At the first run, success rates of 97.5% and 95.5% were obtained for DNA extraction in IBS patients and healthy controls, respectively. Additional samples of saliva were then employed or collected again from subjects with extraction failure, and success rates of 100% were achieved at the second run of genomic DNA extraction. In order to avoid and properly manage genotyping errors, for each batch the following cautions were taken as suggested by Pompanon et al.
The IBS-SSS questionnaire takes into account the following items: a) presence and severity of abdominal pain or discomfort; b) frequency of abdominal pain or discomfort; c) presence and severity of abdominal distension; d) degree of satisfaction of defecatory behaviour; e) degree of interference of IBS symptoms with daily lifestyle. Each of the above items generates a maximum score of 100, leading to an overall score of 500.
Primary endpoint: overall IBS-SSS value, to compare LL versus LS/SS genotypes in patients. Secondary endpoints: 1) value of each IBS-SSS item, to compare LL versus LS/SS genotypes in patients; 2) overall IBS-SSS value, to compare C-IBS versus D-IBS patients; 3) value of each IBS-SSS item, to compare C-IBS versus D-IBS patients; 4) frequencies of LL and LS/SS genotypes, to compare IBS patients with healthy volunteers; 5) frequencies of LL and LS/SS genotypes, to compare C-IBS patients, D-IBS patients and healthy volunteers.
Data were expressed as mean ± standard deviation (SD) for continuous variables, and as percentage for qualitative variables. Comparisons between groups were performed by analysis of variance (ANOVA) and the non-parametric Wilcoxon unpaired test for continuous variables, and by the Pearson’s chi-square or the Fisher’s exact test for categorical variables. Adjustments for multiple comparisons were carried out by the Bonferroni correction. Comparisons between groups for IBS-SSS values were also performed by a two-factor (genotype and IBS subgroups) ANOVA, and the two-factor interaction was evaluated. The patients sample size was determined to evaluate a between-groups (LL versus LS/SS genotype) difference of 30 in the mean overall IBS-SSS value, with a standard deviation of 65, a first type error of 0.05 and a power of 0.80. A 2-sided P value <0.05 was considered to be statistically significant. Statistical analysis was performed by JMP 4.0 (SAS Institute Inc., Cary, NC, U.S.A.) and StatXact-4 (Cytel Software Corporation, Cambridge, MA, U.S.A.).
Two hundreds and four consecutive patients (45 males, 159 females; mean age: 39.6±12.3 years), with a diagnosis of C-IBS (6 males, 100 females; mean age: 41.2±12.9 years) or D-IBS (39 males, 59 females; mean age: 38.0±11.5 years), were enrolled to assess the relationship between 5HTTLPR genotypes and IBS symptom severity. The C-IBS and D-IBS subgroups differed significantly for male/female proportion (P<0.0001), but not for age. For the purpose of analyzing the primary and secondary endpoints, the subjects with SS and LS genotypes were pooled (LS/SS), since the S allele has been reported to be dominant and both SS and LS genotypes have been associated with a lower SERT expression and reduced efficiency of 5-HT uptake, as compared with the LL genotype.
The overall mean symptom severity, as assessed by IBS-SSS, was significantly higher in patients with LS/SS than LL genotype (
LL (n = 69) | LS/SS (n = 135) | P | P (*) | |
Abdominal pain severity | 51.0±18.8 | 59.7±21.0 | 0.004 | 0.020 |
Occurrence of pain over the preceding 10 days | 50.4±26.4 | 55.4±29.8 | 0.089 | 0.445 |
Abdominal distension severity | 56.4±24.7 | 59.5±25.9 | 0.401 | 1,000 |
Bowel dissatisfaction | 70.5±22.8 | 80.1±23.9 | 0.007 | 0.035 |
Interference with quality of life | 55.7±25.9 | 62.1±28.3 | 0.116 | 0.580 |
|
283.8±62.3 | 319.0±71.5 | 0.0006 |
(*) P values after Bonferroni correction.
Symptom severity was assessed also in C-IBS and D-IBS patients. The overall mean IBS-SSS values were found to be significantly higher in the C-IBS than D-IBS subgroup, irrespectively of their 5HTTLPR genotype (
C-IBS (n = 106) | D-IBS (n = 98) | P | P (*) | |
Abdominal pain severity | 57.5±20.8 | 55.9±20.6 | 0.563 | 1,000 |
Occurrence of pain over the preceding 10 days | 56.6±30.5 | 53.3±27.9 | 0.424 | 1,000 |
Abdominal distension severity | 65.0±24.4 | 51.4±24.8 | 0.0001 | 0.0006 |
Bowel dissatisfaction | 80.5±21.7 | 72.9±25.7 | 0.023 | 0.138 |
Interference with quality of life | 57.6±27.6 | 62.5±27.5 | 0.210 | 1.000 |
Overall score | 317.2±68.3 | 296.1±71.4 | 0.032 | 0.192 |
(*) P values after Bonferroni correction.
There was a lack of significant interaction between bowel habits (i.e., C-IBS and D-IBS) and 5HTTLPR genotypes, indicating no effect modification and independent effects of bowel habits and genotypes on symptom severity (interaction P values: 0.297 for overall IBS-SSS value; 0.078 for abdominal pain severity; 0.481 for occurrence of pain over the preceding 10 days; 0.811 for abdominal distension severity; 0.379 for bowel dissatisfaction; 0.406 for interference with quality of life).
In order to evaluate whether 5HTTLPR polymorphism is associated with IBS in the present group of patients, a comparison was made with a group of healthy controls. For this purpose, 200 healthy volunteers (46 males, 154 females; mean age: 40.4±15.8 years) were enrolled in the study. They did not differ significantly in terms of male/female proportion and mean age from the group of IBS patients.
The frequencies of 5HTTLPR genotypes in IBS patients and healthy volunteers are displayed in
IBS (n = 204) | HV (n = 200) | P | |
|
69 (33.8%) | 60 (30.0%) | 0.657 (a) |
|
101 (49.5%) | 102 (51.0%) | |
|
34 (16.7%) | 38 (19.0%) | |
|
135 (66.2%) | 140 (70.0%) | 0.455 (b) |
(a) comparison by LL, LS and SS subgroups.
(b) comparison by LL and LS/SS subgroups.
C-IBS (n = 106) | D-IBS (n = 98) | HV (n = 200) | P | |
|
36 (33.9%) | 33 (33.7%) | 60 (30.0%) | 0.594 (a) |
|
56 (52.8%) | 45 (45.9%) | 102 (51.0%) | |
|
14 (13.3%) | 20 (20.4%) | 38 (19.0%) | |
|
70 (66.1%) | 65 (66.3%) | 140 (70.0%) | 0.711 (b) |
(a) comparison by LL, LS and SS subgroups.
(a) comparison by LL and LS/SS subgroups.
Great attention is currently revolving around the pathogenic bases and pathophysiological mechanisms accounting for the clinical manifestations of IBS,
The contention that the LS/SS genotypes can increase the severity of IBS symptoms, with particular regard for abdominal pain, is supported by previous investigations. In particular our findings are in line with data from Camilleri et al.
Whether the influence of 5HTTLPR on abdominal pain results mainly from peripheral or central mechanisms, or both, is presently unknown and remains open to future investigations. However, some evidence suggests that central mechanisms may contribute to the enhanced pain perception in IBS patients with LS/SS genotypes: 1) specific brain regions of IBS patients and healthy controls are subjected to differential activation upon application of rectal stimuli;
In the present study, when the P values were adjusted for multiple comparisons, the analysis for secondary endpoints revealed that the overall symptom severity did not differ in C- and D-IBS patients, while the analysis of single severity items displayed significant differences only for the severity of abdominal distension. Moreover, our analysis showed a lack of significant interaction between bowel habits and 5HTTLPR genotypes. These findings were not completely unexpected, since C-IBS patients are known to complain of abdominal distension more than D-IBS patients.
The influence of 5HTTLPR on IBS symptom severity and, particularly, on abdominal pain severity, might have interesting implications for the effectiveness and/or safety of drugs proposed for the therapeutic management of this syndrome. SERT is a specific target of ‘selective serotonin reuptake inhibitors’ (SSRIs) and there is evidence linking 5HTTLPR to the efficacy and safety of these drugs in patients with depression.
To date, most of the studies on the impact of 5-HT genetics on IBS have investigated the associations of 5HTTLPR with its pathogenesis or clinical presentation. In this regard, the present study was conducted also with the secondary objective of evaluating whether any of such associations could be highlighted in our IBS patients. For this purpose, the frequencies of their 5HTTLPR genotypes were compared with those of healthy controls, selected with similar demographic characteristics, and no significant association was found either for the overall IBS group or for the subgroups of C-IBS and D-IBS patients. Our results are consistent with the majority of previous studies
In conclusion, the present study suggests that the LS and SS genotypes are significantly correlated with IBS symptom severity, although their possible direct causal role remains to be proven. In addition, the present findings do not support an association of 5HTTLPR with IBS or its clinical presentation in terms of bowel habit predominance.