- Email: [email protected]
The serotonin transporter promoter polymorphism (5-HTTLPR) is associated with type 2 diabetes
Clinica Chimica Acta 411 (2010) 167–171
Contents lists available at ScienceDirect
Clinica Chimica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n c h i m
The serotonin transporter promoter polymorphism (5-HTTLPR) is associated with type 2 diabetes Maria Iordanidou a, Anna Tavridou a, Ioannis Petridis a,b, Kostas I. Arvanitidis a, Dimitrios Christakidis b, Vassilios Vargemezis c, Vangelis G. Manolopoulos a,⁎ a b c
Laboratory of Pharmacology and Clinical Pharmacology, Medical School, Democritus University of Thrace, Dragana Campus, 68100 Alexandroupolis, Greece Outpatient Diabetic Medical Clinic, Department of Internal Medicine, Academic General Hospital of Alexandroupolis, Alexandroupolis, Thrace, Greece Department of Nephrology, Academic General Hospital of Alexandroupolis, Alexandroupolis, Thrace, Greece
a r t i c l e
i n f o
Article history: Received 3 July 2009 Received in revised form 26 October 2009 Accepted 26 October 2009 Available online 2 November 2009 Keywords: Serotonin 5-HTTLPR polymorphism 5-HTT gene Type 2 diabetes mellitus Obesity
a b s t r a c t Background: The serotonergic system contributes substantially to the regulation of glucose homeostasis and feeding. 5-HTTLPR is a serotonin transporter (5-HTT) gene-linked polymorphic region that regulates the transcriptional activity of 5-HTT. Our aim was to investigate the possible association of 5-HTTLPR polymorphism with type 2 diabetes mellitus and obesity. Methods: Study population consisted of 252 subjects diagnosed with Type 2 DM and 211 non-diabetic subjects, all Caucasians of Greek ethnic origin. Genomic DNA was extracted from peripheral blood and analyzed for 5-HTTLPR polymorphism with a novel PCR protocol. Results: The frequency of SS and SL genotypes of HTTLPR was significantly higher in the diabetic group (77.0%) than in the non-diabetic group (61.6%) (P < 0.001). The genetic risk of Type 2 DM for subjects carrying at least one S allele was increased compared to non-diabetic subjects (OR = 2.08, 95% CI = 1.39– 3.12). When subjects were divided according to BMI status, the frequency of S allele carriers was similar in obese and non-obese subjects. Conclusions: The S allele of 5-HTTLPR is strongly associated with the presence of Type 2 DM. This association appears to be direct and not dependent on obesity status. Therefore, 5-HTTLPR LL genotype might be protective for development of Type 2 DM. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Type 2 diabetes mellitus is a common metabolic disorder affecting millions of people worldwide defined by insulin resistance and an insulin secretory defect resulting in impaired glucose tolerance and hyperglycaemia. Its pathogenesis appears to involve complex interactions between genetic and environmental factors. On the other hand, obesity is a major risk factor for Type 2 DM associated with insulin resistance. A growing number of studies suggest that central neural pathways may play an important role in regulation of glucose homeostasis. Serotonin (5-hydroxytryptamine, 5-HT) is one of the most important neurotransmitters implicated in the regulation of energy balance through both central modulation of the activity of various downstream neuropeptide systems and autonomic pathways and also peripheral mechanisms [1]. Serotonin transporter (5-HTT or SERT) is the major factor responsible for the inactivation of serotoninergic transmission after serotonin release in the synaptic cleft [2]. It is a sodium-dependent
⁎ Corresponding author. Tel./fax: +30 2551 030523. E-mail address: [email protected] (V.G. Manolopoulos). 0009-8981/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2009.10.022
transporter that regulates the entire serotoninergic system and its receptors as it is responsible for the active transport of serotonin into neurons, enterochromaffin cells, platelets, and other cells modulating extracellular fluid serotonin concentrations [3]. Human 5-HTT gene spans 37.8 kb on chromosome 17q11.2 and is composed of 14 exons encoding a protein of 630 amino acids [4]. Transcriptional activity of human 5-HTT gene is modulated by a repetitive element of varying length in the 5’ flanking region located ~ 1.4 kb upstream of the transcription start site, termed 5-HTT gene-linked polymorphic region (5-HTTLPR). 5-HTTLPR is only present in humans and higher non-human primates and a typical insertion/ deletion of 44 bp results in the presence of two alleles, the “long” (L) comprising 16 copies of a 20–23 bp repeat unit and the “short” (S) comprising 14 copies [5]. Human 5-HTT gene is differentially modulated by the allelic variants of the 5-HTT gene promoter at both transcriptional and translational levels modifying 5-HTT function [6]. S allele is dominant and its presence is associated with lower expression of 5-HTT gene, resulting in a reduced capability to take up and release 5-HT [5,7]. In contrast, the L variant was associated with an almost threefold increased transcription of the 5-HTT gene [5]. A possible association of 5-HTTLPR with Type 2 DM has been implied in diabetic animals that exhibit altered neurotransmitters in
168
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
brain monoaminergic systems. Particularly, it was suggested that altered monoamine transporter gene expression may contribute to the observed dysfunctions in brain monoamine transmission in chronic diabetes in mice [8]. Additionally, in a study with 264 Japanese women not on medication for diabetes, hypercholesterolemia or hypertension, subjects with SS genotype of 5-HTTLPR showed larger change in fasting blood glucose between beginning and end of the study in comparison with other genotypes [9]. Polymorphisms in 5-HTT gene have been shown to influence satiety and sensory aspects of energy balance [10]. To the best of our knowledge, no studies have been published on the association of 5-HTTLPR with Type 2 DM. Thus, the aim of the present study was to investigate the possible association of the 5-HTTLPR polymorphism with Type 2 DM in a human population of Caucasian origin. In addition we assessed the relation of 5-HTTLPR with obesity in this population. 2. Patients and methods A total of 463 subjects were included in this study. The diabetic study population consisted of 252 subjects diagnosed as having Type 2 DM (122 males and 130 females) aged 67.4 ± 8.8 years. The control (non-diabetic) study population consisted of 211 subjects (118 males and 93 females) aged 66.3 ± 12.8 years. Matching of subjects was performed on a frequency basis. All subjects were of Caucasian ethnic origin, Greek citizens residing in Alexandroupolis urban area but with origins from all parts of Greece. From November 2006 to December 2007, 314 consecutive diabetic patients were presented in the Outpatient Diabetic Clinic of the Internal Medicine Department of the Academic General Hospital of Alexandroupolis. From these patients, 43 did not meet inclusion criteria and 19 refused enrollment. Inclusion of the remaining 252 subjects in the diabetic group was based on the criteria recommended by WHO for Type 2 DM (http:// www.aafp.org/afp/981015ap/mayfield.html). According to these criteria, diagnosis of Type 2 DM was based on two measurements of fasting plasma glucose levels of 126 mg/dl (7.0 mmol/l) or higher or two casual glucose readings of 200 mg/dl (11.1 mmol/l) or higher or if Type 2 diabetes was controlled by medication. Subjects included in the control population had been at the Internal Medicine Department of the same hospital in the same period for reasons unrelated to Type 2 DM and had received a thorough medical examination, including specific evidence (medical and drug history taken by a specialist clinician, two measurements of fasting plasma glucose <126 mg/dl) to exclude the presence of Type 2 DM. They were age and sexmatched with the diabetic group. No subjects diagnosed with psychiatric diseases or on treatment with 5-HT agonists or antagonists (such as typical and atypical antipsychotic agents) were included in this study. Furthermore, patients with pulmonary hypertension or known cardiovascular disease were excluded from the study population. Subjects were also divided to obese and non-obese groups on the basis of their BMI. Of the total 463 subjects, 182 (80 males and 102 females) were classified as obese (BMI ≥ 30 kg/m2) and 281 (160 males and 121 females) were classified as non-obese (BMI < 30 kg/m2). Glucose levels, urea, creatinine and lipid profile (total cholesterol, HDL-cholesterol, and triglycerides) were calculated by standard enzymatic methods, while LDL was calculated by Friedewald equation. Blood samples were obtained after an overnight fast. Blood pressure was measured with a mercury sphygmomanometer. Hypertension was defined as known or newly diagnosed hypertension according to current national guidelines (Systolic BP / Diastolic BP > 140/90 mmHg) or if hypertension was controlled by medication. Body mass index (BMI) was calculated as weight divided by the square of height (kg/m2). The study was approved by the Scientific Council and the Ethics Committee of the Academic General Hospital of Alexandroupolis, Greece and was conducted according to
the Declaration of Helsinki. All subjects participated after being informed about the study by their attending clinician and giving written consent. Genomic DNA was extracted from white blood cells in peripheral venous blood by Puregene DNA Purification System (Gentra, Minnesota, MI, USA) and analyzed for the 5-HTTLPR polymorphism, taking into account previous reports presenting the difficulties in genotyping this polymorphism concerning Mg concentrations [11,12]. A novel set of primers was designed with the aid of the Oligo-6 software (NBI, Plymouth, USA) at very high stringency conditions. Using polymerase chain reaction (PCR), DNA was amplified with the following primer set: [5 -GTTTTGTGTTGCCCTTGCCTAT - 3 ] and [5 - CACCGCCCCTTGTACTTG - 3] to generate 705- or 749-bp fragments. The PCR reaction was performed in a 50 μl aliquot of the reaction mixture with 0.5 μg genomic DNA; 0.2 mM dNTP mix; 60 pmol of each primer; 1.2 mM MgCl2 and 2.5 IU of Taq polymerase. All reagents for PCR amplification were supplied by Invitrogen (Carlsbad, CA, USA). After an initial denaturation step at 94 °C for 10 min, the cycling parameters were 42 cycles with denaturation at 94 °C for 1 min, annealing at 60 °C for 1 min, extension at 72 °C for 1 min, and a final extension step at 72 °C for 5 min. All PCR amplifications were carried in the PCR-engine apparatus PTC-200 of MJ Research (Watertown, Mass., USA). The PCR products were visualized on a 1.5% agarose gel stained with ethidium bromide to determine each subject's genotype. The L allele of 5-HTTLPR was denoted by the presence of 749 bp fragment and the S allele by the presence of 705 bp fragment. All genotype determinations were carried out in duplicate with identical results for all 463 subjects genotyped. All quantitative data are presented as mean ± standard deviation (SD). Relative frequencies of genotypes and alleles were calculated for each group and a chi-square analysis was conducted comparing the distribution of genotypes and alleles between diabetic and nondiabetic subjects as well as between obese and non-obese subjects. Comparisons for continuous or categorical data between two groups were conducted using an independent t-test or chi-square test, respectively. To estimate the risk of Type 2 DM or obesity associated with 5-HTTLPR polymorphism, odds ratios (OR) were calculated using logistic regression analysis before and after adjustment for other factors known to affect these conditions. A P < 0.05 was considered statistically significant. Analyses were carried out with the use of SPSS software package (15.0 for Windows). 3. Results Demographic and clinical variables of diabetic and non-diabetic subjects, as well as obese and non-obese subjects are shown in Table 1. Subjects were age matched and the frequency of male and female subjects did not differ significantly between study groups. As expected, subjects in the diabetic group had significantly higher blood glucose levels than subjects in the non-diabetic group. Furthermore, in the diabetic and obese groups, subjects had higher body weight and BMI than subjects in the non-diabetic and non-obese group, respectively. Regarding other clinical variables, subjects in the nondiabetic group had significantly higher total cholesterol and LDL levels than subjects in the diabetic group (P < 0.005). Similarly, subjects in the non-obese group had lower triglycerides and higher HDL levels (P < 0.05) compared to the subjects in the obese group. The higher total cholesterol and LDL levels observed in non-diabetic vs. diabetic subjects could possibly be attributed to their controlled diet and treatment for dyslipidemia of diabetic patients. However, no associations were found between frequencies of 5-HTTLPR alleles and these clinical variables (data not shown). Frequencies of 5-HTTLPR genotypes and alleles in diabetic and non-diabetic group are shown in Table 2. Genotype and allele frequencies, tested using conventional x2 test, were in HardyWeinberg equilibrium both in diabetic and non-diabetic group. The
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
169
Table 1 Demographic and clinical variables of groups of patients included in the study (diabetic, non-diabetic, obese and non-obese group).
Age (years) Sex (M/F) Presence of T2DM Presence of obesity Body weight (kg) BMI (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Glucose (mg/dl) Urea (mg/dl) Creatinine (mg/dl) Total cholesterol (mg/dl) Triglycerides (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl)
Non-diabetic group (n = 211)
Diabetic group (n = 252)
P value
Non-obese group (BMI < 30) (n = 281)
Obese group (BMI ≥ 30) (n = 182)
P value
66.3 ± 12.8 118 (55.9%)/93 (44.1%)
67.4 ± 8.8 122 (48.4%)/130 (51.6%)
ns ns
67.3 ± 11.7 160(56.9%)/ 121(43.1%) 46.3%
66.3 ± 9.4 80 (44.0%)/102 (56.0%) 67.6%
ns ns <0.001
28.0 % 78.2 ± 15.8 28.4 ± 4.7 141.1 ± 18.4 81.9 ± 12.1 104.2 ± 24.6 40.6 ± 13.5 1.0 ± 0.2 215.2 ± 52.2 145.2 ± 73.9 51.8 ± 10.8 140.8 ± 86.8
48.4% 80.5 ± 13.2 30.5 ± 4.8 147.4 ± 19.5 80.1 ± 10.1 150.6 ± 53.1 42.6 ± 17.9 1.1 ± 0.3 200.0 ± 45.8 159.8 ± 102.5 53.7 ± 13.9 116.1 ± 38.4
<0.001 ns <0.001 0.005 ns <0.001 ns ns 0.002 0.086 ns <0.001
72.4 ± 9.8 26.6 ± 2.5 144.0 ± 19.1 79.9 ± 10.0 124.0 ± 47.9 40.8 ± 14.5 1.0 ± 0.2 207.0 ± 49.5 143.2 ± 5.1 54.0 ± 13.4 130.4 ± 61.1
89.4 ± 13.0 34.1 ± 4.1 146.2 ± 19.5 81.8 ± 11.8 138.6 ± 48.6 42.8 ± 18.2 1.0 ± 0.3 206.3 ± 49.0 166.9 ± 7.7 51.4 ± 11.3 122.9 ± 42.1
<0.001 <0.001 ns ns 0.002 ns ns ns 0.011 0.040 ns
ns = non-significant statistically data. Values are mean ± SD.
frequency of 5-HTTLPR S allele was significantly higher in diabetic (53.2%) than in the non-diabetic group (40.5%) (P < 0.001). Our study captured a statistically significant (P = 0.05) difference of 12.7 % in the frequency of L allele between the diabetic and non-diabetic group with 80 % power. Regarding 5-HTTLPR genotypes, in the diabetic group subjects had a significantly lower frequency of LL genotype (23%) compared with the non-diabetic group (38.4%) (P < 0.001). We also grouped all LS and SS homozygous subjects in a dominant model as carriers of the S allele because this allele is known to exert a dominant influence [7]. The frequency of S allele carriers (SL and SS genotypes) was significantly higher in the diabetic group (77%) than in the non-diabetic group (61.6%) (P < 0.001) [Table 2]. Two models were used to assess the genetic risk of Type 2 DM for patients carrying the S allele. In these subjects, the genetic risk was more than more than doubled compared with subjects not carrying the S allele (OR = 2.08, 95% CI = 1.39–3.12) before adjustment for other factors known to affect Type 2 DM. After adjustment for age, sex, and BMI, the genetic risk for Type 2 DM was slightly higher (OR = 2.14, 95% CI = 1.41–3.25); BMI was also found to be a weak but significant predictor of Type 2 DM (OR = 1.11, 95% CI = 1.06–1.16) [Table 3]. The same subjects used in our diabetes study were grouped also according to BMI and used for assessing the possible association of 5-HTTLPR with obesity status. Specifically, we analyzed the frequency of 5-HTTLPR genotypes in obese and non-obese subjects as well as the mean BMI in subjects according to their 5-HTTLPR
Table 2 Frequencies 5-HTTLPR genotypes of serotonin transporter gene in the diabetic and the non-diabetic group. Non-diabetic group
Diabetic group
n
n
%
95% CI
%
95% CI
P value
5-HTTLPR genotypes Total 211 252 LL 81 38.4 32.02–45.07 58 23.0 18.15–28.50 < 0.001 LS 89 42.2 35.66–48.91 120 47.6 41.51–53.78 SS 41 19.4 14.53–25.17 74 29.4 24.00–35.20 Carriers of S allele Total 211 252 Presence of S allele 130 61.6 54.93–67.98 194 77.0 71.50–81.85 Absence of S allele 81 38.4 32.02–45.07 58 23.0 18.15–28.50 < 0.001 5-HTTLPR alleles Total L S
422 504 251 59.5 54.74–64.08 236 46.8 42.50–51.19 < 0.001 171 40.5 35.92–45.26 268 53.2 48.81–57.50
CI: Confidence Interval.
genotype. Frequencies of 5-HTTLPR genotypes and alleles in the obese and non-obese groups are shown in Table 4. The frequency of the S allele carriers in obese group (72.5%) was similar to that in the nonobese group (68.4%). Further analysis in male and female subjects showed no significant difference of the 5-HTTLPR S allele carriers frequency between obese and non-obese subjects (data not shown). Because there was a significant difference in frequency of Type 2 DM between non-obese and obese subjects (46.3% vs. 67.6%, P < 0.001), logistic regression analysis was performed with obesity as the dependent variable and presence of S allele, as contributing variable. However, the presence of the polymorphism was not found to be a significant predictor of obesity neither in an unadjusted model (OR = 1.22, 95% CI = 0.81–1.85, P = 0.340) nor in a model adjusted for age, sex, and Type 2 DM [Table 5]. Furthermore, the average BMI in the presence or absence of the S allele in the total number of subjects participating in the study was identical in the diabetic and nondiabetic group [Table 6]. Finally, and despite the fact that BMI was higher in diabetic subjects, no difference was found in the frequency of S allele carriers between obese and non-obese diabetic subjects and obese and non-obese non-diabetic subjects (P > 0.05, data not shown). 4. Discussion It is already known that 5-HT is an important regulator of energy balance and serotonin-sensitive neurons recruit various downstream systems to exert their effects on energy homeostasis [1]. Furthermore, we recently found that − 759C/T polymorphism of 5-HT2C receptor gene is strongly associated with the development of Type 2 DM (OR = 2.34) but not with obesity [13]. On the basis of the above, we hypothesized that variations in other elements of the 5-HT pathways may also influence the susceptibility to Type 2 DM and obesity. 5-HTTLPR appears to be a good such candidate since it is a 5-HTT geneTable 3 Logistic regression analysis with T2DM as the dependent variable and 5-HTTLPR genotype as contributing variable.
Model 1 (unadjusted) Presence of S allele
OR
95% CI
P value
2.08
1.39–3.12
< 0.001
1.41–3.25 1.06–1.16
< 0.001 < 0.001
Model 2 (Adjusted for age, sex and BMI)a Presence of S allele 2.14 BMI 1.11
OR: Odds ratio. a Age and sex were found not to be significant in model 2.
170
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
Table 4 Frequencies of 5-HTTLPR genotypes and alleles in the non-obese and in the obese group. Non-obese group
Obese group
n
95% CI
n
26.44–37.28 39.80–51.40 18.17–27.94
182 50 81 51
50.32–58.53 41.47–49.68
364 181 183
%
5-HTTLPR genotypes Total 281 LL a 89 31.7 a LS 128 45.6 SS a 64 22.8 5-HTTLPR alleles Total 562 La 306 54.4 a S 256 45.6 a
%
Table 6 BMI in the presence or absence of the S allele of 5-HTTLPR in diabetic and non-diabetic subjects. Absence of S
95%CI
P
BMI
n
BMI
139
29.5 ± 5.3
324
29.6 ± 4.7
Non-diabetic group Total 81
28.6 ± 5.2
87
28.3 ± 4.3
Diabetic group Total
30.8 ± 5.1
194
30.4 ± 4.8
All subjects Total 27.5 44.5 28.0
48.4 51.6
21.38–34.28 37.42–51.76 21.88–34.86
44.61–54.85 45.15–55.39
Presence of S
n
0.387
0.160
58
Values are mean ± SD (kg/m2).
No significant differences between non-obese and obese groups.
linked polymorphic region that regulates the transcriptional activity of the 5-HTT gene and has been associated with various diseases. Consequently, the aim of our study was to examine the possible association of 5-HTTLPR with Type 2 DM and obesity in humans. We found that higher frequency of 5-HTTLPR LS and SS genotypes is strongly associated with Type 2 DM (OR = 2.14 after adjustment of age, sex and BMI). No such association was found between 5-HTTLPR and obesity. Taken together these findings suggest that the lower activity of 5-HTTLPR S allele confers vulnerability to Type 2 DM independently of obesity status, probably provoking an impairment of brain serotonin transmission. Alternatively, basal expression in the presence of 5-HTTLPR LL genotype may be protective against the development of Type 2 DM. In humans, the only study that associates 5-HTTLPR with glucose levels was performed in healthy Japanese women who were not on medication for diabetes, hypercholesterolemia or hypertension. The difference in fasting blood glucose levels between beginning and end of the study in subjects with the SS genotype was significantly larger than the difference in subjects carrying LL and LS genotypes [9]. Indeed, Type 2 DM is accompanied by altered monoamine neurotransmission in the brain, mostly manifested by an increase in content and a decrease in turnover rate [14]. In 1997, Salkovic Petrisic and coworkers reported that diabetic rats exhibit altered neurotransmission in brain monoaminergic systems and that the altered monoamine transporter gene expression may possibly contribute to the observed dysfunctions in brain monoamine transmission in chronic diabetes [8]. 5-HTTLPR could be one of the genetic causes of this condition as the S allele results in lower expression of 5-HTT mRNA that might account for the features of brain insulin system observed in Type 2 DM. Our findings in humans correlating serotonin with T2DM are supported by several studies in animals which have shown that serotonin can alter not only release but also synthesis of insulin in pancreatic islets [15] and that a perturbation of brain serotonin systems can predispose to Type 2 DM in young adult mice [16]. Most studies have reported that 5-HT dose-dependently increases serum glucose in mice [17] whereas 5-HT receptor antagonists induce hyperglycaemia [18].
Table 5 Logistic regression analysis with obesity as the dependent variable and 5-HTTLPR genotype as contributing variable.
Model 1 (unadjusted) Presence of S allele
OR
95% CI
1.22
0.81–1.85
0.34
1.62–3.55 1.10–2.37
< 0.001 0.010
Model 2 (Adjusted for age, sex and T2DM) Presence of T2DM 2.40 Sex 1.62
OR: Odds ratio. Presence of S allele and age were not found to be significant in model 2.
P value
Serotonin is also well distributed in normal and diabetic pancreatic tissues and has stimulatory effects on insulin secretion from normal pancreas and glucagon secretion from diabetic pancreas in rats [19]. Furthermore, both metabolic syndrome and insulin resistance are associated with blunted central serotonergic responsiveness [20]. However, it is not clear whether insulin resistance causes decreased central serotonergic responsivity or the S allele of 5-HTTLPR which is found in higher frequency in patients with insulin resistance causes decreased serotonergic activity. Serotonin is also known to exert direct peripheral effects. Both serotonin and exogenous agonists of 5-HT2A/2C receptor increase glucose uptake in isolated myocytes by increasing recruitment of glucose transporters to the plasma membrane [21]. Recently, Wade and coworkers demonstrated in mice lacking 5-HT2C receptor and the adipocyte hormone leptin that the improvement in glucose tolerance in wild-type mice treated with the serotonin releaser and reuptake inhibitor fenfluramine was blunted in 5-HT2C receptor mutant mice indicating that 5-HT2c receptor has direct effects on glucose homeostasis independently of obesity status [22]. We recently provided clinical evidence for a direct role of serotonergic system in the development of Type 2 DM by showing that the functional -759C/T polymorphism of 5-HT2C receptor gene was strongly associated with Type 2 DM in a Caucasian population directly, independently of obesity status [13]. Several studies have investigated the association of 5-HTTLPR polymorphism with obesity. An early study had found that patients with anorexia nervosa have allele frequencies for 5-HTTLPR polymorphism not significantly different to those observed for obese and underweight individuals [23]. Two recent studies by Sookoian et al. showed that the S allele of 5-HTTLPR is a risk factor for obesity in adolescents [24] and in male adults with mean age of 34 years [25]. Furthermore, Lan et al. found that the SS genotype was associated with increased BMI only in non-elderly stroke subjects (<65 years) [26], and Mergen et al. showed a weak, but significant, association of 5-HTTLPR with BMI status in 262 obese (BMI > 30 kg/m2) and 138 control (BMI < 25 kg/m2) subjects (their age was not reported) [27]. Finally, Fuemmeler et al. found that the prevalence of obese and overweight plus obese combined was significantly lower among carriers of the L variant compared with subjects having SS genotype only in men adolescents [28]. In our study, which found no association between 5-HTTLPR and obesity, both male and female subjects have been included, with mean age of 67 years old and were divided according to BMI in obese (BMI > 30 kg/m2) and non-obese (BMI< 30kg/m2). Apparently, the association of 5-HTTLPR with obesity is agedepended and is gradually reduced or “masked” by other factors as subjects become older. In conclusion, we report for the first time that the S allele of 5-HTTLPR is strongly associated with the presence of Type 2 DM in a human population. This association is direct and not related to obesity status. The frequency of LL genotype was significantly lower in diabetic vs. nondiabetic group suggesting that this genotype might be protective from development of Type 2 DM or, conversely, the lower activity of
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
the 5-HTTLPR S allele that impairs 5-HT transmission may confer vulnerability to this disease. This observation, taken together with our recent finding of an association of the -759C/T polymorphism of 5-HT2C receptor gene with Type 2 DM but not with obesity [13], indicates that the serotonergic system may be involved in the pathogenesis of Type 2 DM directly, with a mechanism not related to the regulation of food intake or obesity. Overall, our findings suggest that the 5-HTTLPR polymorphism might be a useful prognostic marker for Type 2 DM. In addition, this and our previous study [13] provide evidence in humans for a direct role of 5-HT in development of Type 2 DM and might open the road for better insights into the pathogenesis of Type 2 DM and novel approaches in the quest for pharmacological targets for its treatment. However, it should be kept in mind that no evaluation of population stratification was done in our study, thus these results should be considered as preliminary. Further studies in other populations are necessary in order to confirm and solidify these findings. References [1] Lam DD, Heisler LK. Serotonin and energy balance: molecular mechanisms and implications for type 2 diabetes. Expert Rev Mol Med 2007;9:1–24. [2] Smeraldi E, Serretti A, Artioli P, Lorenzi C, Catalano M. Serotonin-transporter genelinked polymorphic region: possible pharmacogenetic implications of rare variants. Psychiatr Gen 2006;16:153–8. [3] Ni W, Watts SW. 5-Hydroxytryptamine in the cardiovascular system: focus on the serotonin transporter (SERT). Clin Exp Pharmacol Physiol 2006;33:575–83. [4] Lesch KP, Balling U, Gross J, et al. Organization of the human serotonin transporter gene. J Neural Transm 1994;95:157–62. [5] Heils A, Teufel A, Petri S, Stöber G, Riedezez P, Bengel D, et al. Allelic variation of human serotonin transporter gene expession. J Neurochem 1996;66:2621–4. [6] Lesch KP, Wolozin BL, Estler HC, Murphy DL, Riederer P. Isolation of a cDNA encoding the human brain serotonin transporter. J Neural Transm 1993;91:67–73. [7] Mortensen OV, Thomassen M, Larsen MB, Whittemore SR, Wiborg O. Functional analysis of a novel human serotonin transporter gene promoter in immortalized raphe cells. Brain Res Mol Brain Res 1999;68:141–8. [8] Salkovic Petrisic M, Augood SJ, Bicknell J. Monoamine transporter gene expression in the central nervous system in diabetes mellitus. J Neurochem 1997;68:2435–41. [9] Yamakawa M, Fukushima A, Sakuma K, Yanagisawa Y, Kagawa Y. Serotonin transporter polymorphisms affect human blood glucose control. Biochem Biophys Res Commun 2005;334:1165–71. [10] Huang XF, Huang X, Han M, Chen F, Storlien L, Lawrence AJ. 5-HT2A/2C receptor and 5-HT transporter densities in mice prone or resistant to chronic high-fat dietinduced obesity: a quantitative autoradiography study. Brain Res 2004;1018: 227–35.
171
[11] Kaiser R, Tremblay PB, Roots I, Brockmoller J. Validity of PCR with emphasis on variable number of tandem repeat analysis. Clin Biochem 2002;35:49–56. [12] Yonan AL, Palmer AA, Gilliam TC. Hardy-Weinberg disequilibtium identified genotyping error of the serotonin transporter (SLC6A4) promoter polymorphism. Psychiatr Gen 2006;16:31–4. [13] Iordanidou M, Tavridou A, Vasiliadis MV, Azvanitidis KI, Petridis J, Chzistakidis D, et al. The -759C/T polymorphism of the 5-HT2C receptor is associated with type 2 diabetes in male and female Caucasians. Pharmacogenetics Genom 2008;18: 153–9. [14] MacCall AL. The impact of diabetes on the CNS. Diabetes 1992;41:557–70. [15] Peschke E, Peschke D, Hammer T, Csernus V. Influence of melatonin and serotonin on glucose-stimulated insulin release from perifused rat pancreatic islets in vitro. J Pineal Res 1997;23:156–63. [16] Nonogaki K, Strack AM, Dallman MF, Tecott LH. Leptin-independent hyperphagia and type 2 diabetes in mice with a mutated serotonin 5-HT2C receptor gene. Nat Med 1998;4:1152–6. [17] Sugimoto Y, Kimura I, Yamada J, Watanabe Y, Takeuchi N, Horisaka K. Effects of serotonin on blood glucose and insulin levels of glucose- and streptozotocintreated mice. Jpn J Pharmacol 1990;54:93–6. [18] Wonziak KM, Linnoila M. Hypergycemic properties of serotonin receptor antagonists. Life Sci 1991;49:101–9. [19] Adeghate E, Ponery AS, Pallot D, Parvez SH, Singh J. Distribution of serotonin and its effect on insulin and glucagons secretion in normal and diabetic pancreatic tissues in rat. Neuro Endocrinol Lett 1999;20:315–22. [20] Muldoon MF, Mackey RH, Korytkowski MT, Flory JD, Pollock BG, Manuck SB. The metabolic syndrome is associated with reduced central serotonergic responsivity in healthy community volunteers. J Clin Endocrinol Metab 2006;91:718–21. [21] Hajduch E, Rencurel F, Balendran A, Batty IH, Downes CP, Hundal HS. Serotonin (5Hydroxytryptamine), a novel regulator of glucose transport in rat skeletal muscle. J Biol Chem 1999;274:13563–8. [22] Wade JM, Juneja P, MacKay AW, Gzaham J, Havel PJ, Tecott LH, et al. Synergistic impairment of glucose homeostasis in ob/ob mice lacking functional serotonin 2C receptors. Endocrinology 2008;149:955–61. [23] Hinney A, Barth N, Ziegler A, von Prittwitz S, Hamann A, Hennighausen K, et al. Serotonin transporter gene-linked polymorphic region: allele distributions in relationship to body weight and in anorexia nervosa. Life Sci 1997;6:295–303. [24] Sookoian S, Gemma C, Garcia SI, Gianotti TF, Dieuzeide G, Roussos A, et al. Short allele of serotonin transporter gene promoter is a risk factor for obesity in adolescents. Obesity 2007;15:271–6. [25] Sookoian S, Gianotti TF, Gemma C, Burgueño A, Pirola CJ. Contribution of functional 5-HTTLPR variant of the SLC6A4 gene to obesity risk in male adults. Obesity 2008;16:488–91. [26] Lan MY, Chang YY, Chen WH, Kao YF, Lin HS, Liu JS. J Endocrinol Invest 2009;32: 119–22. [27] Mergen H, Karaaslan C, Mergen M. Deniz Ozsoy E, Ozata M. LEPR, ADBR3, IRS-1 and 5-HTT genes polymorphisms do not associate with obesity. Endocr J 2007;54: 89–94. [28] Fuemmeler BF, Agurs-Collins TD, Mccleron FJ, Kollins SH, Kail MF, Bergen AW, et al. Obesity 2008;16:348–55.
Contents lists available at ScienceDirect
Clinica Chimica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n c h i m
The serotonin transporter promoter polymorphism (5-HTTLPR) is associated with type 2 diabetes Maria Iordanidou a, Anna Tavridou a, Ioannis Petridis a,b, Kostas I. Arvanitidis a, Dimitrios Christakidis b, Vassilios Vargemezis c, Vangelis G. Manolopoulos a,⁎ a b c
Laboratory of Pharmacology and Clinical Pharmacology, Medical School, Democritus University of Thrace, Dragana Campus, 68100 Alexandroupolis, Greece Outpatient Diabetic Medical Clinic, Department of Internal Medicine, Academic General Hospital of Alexandroupolis, Alexandroupolis, Thrace, Greece Department of Nephrology, Academic General Hospital of Alexandroupolis, Alexandroupolis, Thrace, Greece
a r t i c l e
i n f o
Article history: Received 3 July 2009 Received in revised form 26 October 2009 Accepted 26 October 2009 Available online 2 November 2009 Keywords: Serotonin 5-HTTLPR polymorphism 5-HTT gene Type 2 diabetes mellitus Obesity
a b s t r a c t Background: The serotonergic system contributes substantially to the regulation of glucose homeostasis and feeding. 5-HTTLPR is a serotonin transporter (5-HTT) gene-linked polymorphic region that regulates the transcriptional activity of 5-HTT. Our aim was to investigate the possible association of 5-HTTLPR polymorphism with type 2 diabetes mellitus and obesity. Methods: Study population consisted of 252 subjects diagnosed with Type 2 DM and 211 non-diabetic subjects, all Caucasians of Greek ethnic origin. Genomic DNA was extracted from peripheral blood and analyzed for 5-HTTLPR polymorphism with a novel PCR protocol. Results: The frequency of SS and SL genotypes of HTTLPR was significantly higher in the diabetic group (77.0%) than in the non-diabetic group (61.6%) (P < 0.001). The genetic risk of Type 2 DM for subjects carrying at least one S allele was increased compared to non-diabetic subjects (OR = 2.08, 95% CI = 1.39– 3.12). When subjects were divided according to BMI status, the frequency of S allele carriers was similar in obese and non-obese subjects. Conclusions: The S allele of 5-HTTLPR is strongly associated with the presence of Type 2 DM. This association appears to be direct and not dependent on obesity status. Therefore, 5-HTTLPR LL genotype might be protective for development of Type 2 DM. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Type 2 diabetes mellitus is a common metabolic disorder affecting millions of people worldwide defined by insulin resistance and an insulin secretory defect resulting in impaired glucose tolerance and hyperglycaemia. Its pathogenesis appears to involve complex interactions between genetic and environmental factors. On the other hand, obesity is a major risk factor for Type 2 DM associated with insulin resistance. A growing number of studies suggest that central neural pathways may play an important role in regulation of glucose homeostasis. Serotonin (5-hydroxytryptamine, 5-HT) is one of the most important neurotransmitters implicated in the regulation of energy balance through both central modulation of the activity of various downstream neuropeptide systems and autonomic pathways and also peripheral mechanisms [1]. Serotonin transporter (5-HTT or SERT) is the major factor responsible for the inactivation of serotoninergic transmission after serotonin release in the synaptic cleft [2]. It is a sodium-dependent
⁎ Corresponding author. Tel./fax: +30 2551 030523. E-mail address: [email protected] (V.G. Manolopoulos). 0009-8981/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2009.10.022
transporter that regulates the entire serotoninergic system and its receptors as it is responsible for the active transport of serotonin into neurons, enterochromaffin cells, platelets, and other cells modulating extracellular fluid serotonin concentrations [3]. Human 5-HTT gene spans 37.8 kb on chromosome 17q11.2 and is composed of 14 exons encoding a protein of 630 amino acids [4]. Transcriptional activity of human 5-HTT gene is modulated by a repetitive element of varying length in the 5’ flanking region located ~ 1.4 kb upstream of the transcription start site, termed 5-HTT gene-linked polymorphic region (5-HTTLPR). 5-HTTLPR is only present in humans and higher non-human primates and a typical insertion/ deletion of 44 bp results in the presence of two alleles, the “long” (L) comprising 16 copies of a 20–23 bp repeat unit and the “short” (S) comprising 14 copies [5]. Human 5-HTT gene is differentially modulated by the allelic variants of the 5-HTT gene promoter at both transcriptional and translational levels modifying 5-HTT function [6]. S allele is dominant and its presence is associated with lower expression of 5-HTT gene, resulting in a reduced capability to take up and release 5-HT [5,7]. In contrast, the L variant was associated with an almost threefold increased transcription of the 5-HTT gene [5]. A possible association of 5-HTTLPR with Type 2 DM has been implied in diabetic animals that exhibit altered neurotransmitters in
168
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
brain monoaminergic systems. Particularly, it was suggested that altered monoamine transporter gene expression may contribute to the observed dysfunctions in brain monoamine transmission in chronic diabetes in mice [8]. Additionally, in a study with 264 Japanese women not on medication for diabetes, hypercholesterolemia or hypertension, subjects with SS genotype of 5-HTTLPR showed larger change in fasting blood glucose between beginning and end of the study in comparison with other genotypes [9]. Polymorphisms in 5-HTT gene have been shown to influence satiety and sensory aspects of energy balance [10]. To the best of our knowledge, no studies have been published on the association of 5-HTTLPR with Type 2 DM. Thus, the aim of the present study was to investigate the possible association of the 5-HTTLPR polymorphism with Type 2 DM in a human population of Caucasian origin. In addition we assessed the relation of 5-HTTLPR with obesity in this population. 2. Patients and methods A total of 463 subjects were included in this study. The diabetic study population consisted of 252 subjects diagnosed as having Type 2 DM (122 males and 130 females) aged 67.4 ± 8.8 years. The control (non-diabetic) study population consisted of 211 subjects (118 males and 93 females) aged 66.3 ± 12.8 years. Matching of subjects was performed on a frequency basis. All subjects were of Caucasian ethnic origin, Greek citizens residing in Alexandroupolis urban area but with origins from all parts of Greece. From November 2006 to December 2007, 314 consecutive diabetic patients were presented in the Outpatient Diabetic Clinic of the Internal Medicine Department of the Academic General Hospital of Alexandroupolis. From these patients, 43 did not meet inclusion criteria and 19 refused enrollment. Inclusion of the remaining 252 subjects in the diabetic group was based on the criteria recommended by WHO for Type 2 DM (http:// www.aafp.org/afp/981015ap/mayfield.html). According to these criteria, diagnosis of Type 2 DM was based on two measurements of fasting plasma glucose levels of 126 mg/dl (7.0 mmol/l) or higher or two casual glucose readings of 200 mg/dl (11.1 mmol/l) or higher or if Type 2 diabetes was controlled by medication. Subjects included in the control population had been at the Internal Medicine Department of the same hospital in the same period for reasons unrelated to Type 2 DM and had received a thorough medical examination, including specific evidence (medical and drug history taken by a specialist clinician, two measurements of fasting plasma glucose <126 mg/dl) to exclude the presence of Type 2 DM. They were age and sexmatched with the diabetic group. No subjects diagnosed with psychiatric diseases or on treatment with 5-HT agonists or antagonists (such as typical and atypical antipsychotic agents) were included in this study. Furthermore, patients with pulmonary hypertension or known cardiovascular disease were excluded from the study population. Subjects were also divided to obese and non-obese groups on the basis of their BMI. Of the total 463 subjects, 182 (80 males and 102 females) were classified as obese (BMI ≥ 30 kg/m2) and 281 (160 males and 121 females) were classified as non-obese (BMI < 30 kg/m2). Glucose levels, urea, creatinine and lipid profile (total cholesterol, HDL-cholesterol, and triglycerides) were calculated by standard enzymatic methods, while LDL was calculated by Friedewald equation. Blood samples were obtained after an overnight fast. Blood pressure was measured with a mercury sphygmomanometer. Hypertension was defined as known or newly diagnosed hypertension according to current national guidelines (Systolic BP / Diastolic BP > 140/90 mmHg) or if hypertension was controlled by medication. Body mass index (BMI) was calculated as weight divided by the square of height (kg/m2). The study was approved by the Scientific Council and the Ethics Committee of the Academic General Hospital of Alexandroupolis, Greece and was conducted according to
the Declaration of Helsinki. All subjects participated after being informed about the study by their attending clinician and giving written consent. Genomic DNA was extracted from white blood cells in peripheral venous blood by Puregene DNA Purification System (Gentra, Minnesota, MI, USA) and analyzed for the 5-HTTLPR polymorphism, taking into account previous reports presenting the difficulties in genotyping this polymorphism concerning Mg concentrations [11,12]. A novel set of primers was designed with the aid of the Oligo-6 software (NBI, Plymouth, USA) at very high stringency conditions. Using polymerase chain reaction (PCR), DNA was amplified with the following primer set: [5 -GTTTTGTGTTGCCCTTGCCTAT - 3 ] and [5 - CACCGCCCCTTGTACTTG - 3] to generate 705- or 749-bp fragments. The PCR reaction was performed in a 50 μl aliquot of the reaction mixture with 0.5 μg genomic DNA; 0.2 mM dNTP mix; 60 pmol of each primer; 1.2 mM MgCl2 and 2.5 IU of Taq polymerase. All reagents for PCR amplification were supplied by Invitrogen (Carlsbad, CA, USA). After an initial denaturation step at 94 °C for 10 min, the cycling parameters were 42 cycles with denaturation at 94 °C for 1 min, annealing at 60 °C for 1 min, extension at 72 °C for 1 min, and a final extension step at 72 °C for 5 min. All PCR amplifications were carried in the PCR-engine apparatus PTC-200 of MJ Research (Watertown, Mass., USA). The PCR products were visualized on a 1.5% agarose gel stained with ethidium bromide to determine each subject's genotype. The L allele of 5-HTTLPR was denoted by the presence of 749 bp fragment and the S allele by the presence of 705 bp fragment. All genotype determinations were carried out in duplicate with identical results for all 463 subjects genotyped. All quantitative data are presented as mean ± standard deviation (SD). Relative frequencies of genotypes and alleles were calculated for each group and a chi-square analysis was conducted comparing the distribution of genotypes and alleles between diabetic and nondiabetic subjects as well as between obese and non-obese subjects. Comparisons for continuous or categorical data between two groups were conducted using an independent t-test or chi-square test, respectively. To estimate the risk of Type 2 DM or obesity associated with 5-HTTLPR polymorphism, odds ratios (OR) were calculated using logistic regression analysis before and after adjustment for other factors known to affect these conditions. A P < 0.05 was considered statistically significant. Analyses were carried out with the use of SPSS software package (15.0 for Windows). 3. Results Demographic and clinical variables of diabetic and non-diabetic subjects, as well as obese and non-obese subjects are shown in Table 1. Subjects were age matched and the frequency of male and female subjects did not differ significantly between study groups. As expected, subjects in the diabetic group had significantly higher blood glucose levels than subjects in the non-diabetic group. Furthermore, in the diabetic and obese groups, subjects had higher body weight and BMI than subjects in the non-diabetic and non-obese group, respectively. Regarding other clinical variables, subjects in the nondiabetic group had significantly higher total cholesterol and LDL levels than subjects in the diabetic group (P < 0.005). Similarly, subjects in the non-obese group had lower triglycerides and higher HDL levels (P < 0.05) compared to the subjects in the obese group. The higher total cholesterol and LDL levels observed in non-diabetic vs. diabetic subjects could possibly be attributed to their controlled diet and treatment for dyslipidemia of diabetic patients. However, no associations were found between frequencies of 5-HTTLPR alleles and these clinical variables (data not shown). Frequencies of 5-HTTLPR genotypes and alleles in diabetic and non-diabetic group are shown in Table 2. Genotype and allele frequencies, tested using conventional x2 test, were in HardyWeinberg equilibrium both in diabetic and non-diabetic group. The
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
169
Table 1 Demographic and clinical variables of groups of patients included in the study (diabetic, non-diabetic, obese and non-obese group).
Age (years) Sex (M/F) Presence of T2DM Presence of obesity Body weight (kg) BMI (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Glucose (mg/dl) Urea (mg/dl) Creatinine (mg/dl) Total cholesterol (mg/dl) Triglycerides (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl)
Non-diabetic group (n = 211)
Diabetic group (n = 252)
P value
Non-obese group (BMI < 30) (n = 281)
Obese group (BMI ≥ 30) (n = 182)
P value
66.3 ± 12.8 118 (55.9%)/93 (44.1%)
67.4 ± 8.8 122 (48.4%)/130 (51.6%)
ns ns
67.3 ± 11.7 160(56.9%)/ 121(43.1%) 46.3%
66.3 ± 9.4 80 (44.0%)/102 (56.0%) 67.6%
ns ns <0.001
28.0 % 78.2 ± 15.8 28.4 ± 4.7 141.1 ± 18.4 81.9 ± 12.1 104.2 ± 24.6 40.6 ± 13.5 1.0 ± 0.2 215.2 ± 52.2 145.2 ± 73.9 51.8 ± 10.8 140.8 ± 86.8
48.4% 80.5 ± 13.2 30.5 ± 4.8 147.4 ± 19.5 80.1 ± 10.1 150.6 ± 53.1 42.6 ± 17.9 1.1 ± 0.3 200.0 ± 45.8 159.8 ± 102.5 53.7 ± 13.9 116.1 ± 38.4
<0.001 ns <0.001 0.005 ns <0.001 ns ns 0.002 0.086 ns <0.001
72.4 ± 9.8 26.6 ± 2.5 144.0 ± 19.1 79.9 ± 10.0 124.0 ± 47.9 40.8 ± 14.5 1.0 ± 0.2 207.0 ± 49.5 143.2 ± 5.1 54.0 ± 13.4 130.4 ± 61.1
89.4 ± 13.0 34.1 ± 4.1 146.2 ± 19.5 81.8 ± 11.8 138.6 ± 48.6 42.8 ± 18.2 1.0 ± 0.3 206.3 ± 49.0 166.9 ± 7.7 51.4 ± 11.3 122.9 ± 42.1
<0.001 <0.001 ns ns 0.002 ns ns ns 0.011 0.040 ns
ns = non-significant statistically data. Values are mean ± SD.
frequency of 5-HTTLPR S allele was significantly higher in diabetic (53.2%) than in the non-diabetic group (40.5%) (P < 0.001). Our study captured a statistically significant (P = 0.05) difference of 12.7 % in the frequency of L allele between the diabetic and non-diabetic group with 80 % power. Regarding 5-HTTLPR genotypes, in the diabetic group subjects had a significantly lower frequency of LL genotype (23%) compared with the non-diabetic group (38.4%) (P < 0.001). We also grouped all LS and SS homozygous subjects in a dominant model as carriers of the S allele because this allele is known to exert a dominant influence [7]. The frequency of S allele carriers (SL and SS genotypes) was significantly higher in the diabetic group (77%) than in the non-diabetic group (61.6%) (P < 0.001) [Table 2]. Two models were used to assess the genetic risk of Type 2 DM for patients carrying the S allele. In these subjects, the genetic risk was more than more than doubled compared with subjects not carrying the S allele (OR = 2.08, 95% CI = 1.39–3.12) before adjustment for other factors known to affect Type 2 DM. After adjustment for age, sex, and BMI, the genetic risk for Type 2 DM was slightly higher (OR = 2.14, 95% CI = 1.41–3.25); BMI was also found to be a weak but significant predictor of Type 2 DM (OR = 1.11, 95% CI = 1.06–1.16) [Table 3]. The same subjects used in our diabetes study were grouped also according to BMI and used for assessing the possible association of 5-HTTLPR with obesity status. Specifically, we analyzed the frequency of 5-HTTLPR genotypes in obese and non-obese subjects as well as the mean BMI in subjects according to their 5-HTTLPR
Table 2 Frequencies 5-HTTLPR genotypes of serotonin transporter gene in the diabetic and the non-diabetic group. Non-diabetic group
Diabetic group
n
n
%
95% CI
%
95% CI
P value
5-HTTLPR genotypes Total 211 252 LL 81 38.4 32.02–45.07 58 23.0 18.15–28.50 < 0.001 LS 89 42.2 35.66–48.91 120 47.6 41.51–53.78 SS 41 19.4 14.53–25.17 74 29.4 24.00–35.20 Carriers of S allele Total 211 252 Presence of S allele 130 61.6 54.93–67.98 194 77.0 71.50–81.85 Absence of S allele 81 38.4 32.02–45.07 58 23.0 18.15–28.50 < 0.001 5-HTTLPR alleles Total L S
422 504 251 59.5 54.74–64.08 236 46.8 42.50–51.19 < 0.001 171 40.5 35.92–45.26 268 53.2 48.81–57.50
CI: Confidence Interval.
genotype. Frequencies of 5-HTTLPR genotypes and alleles in the obese and non-obese groups are shown in Table 4. The frequency of the S allele carriers in obese group (72.5%) was similar to that in the nonobese group (68.4%). Further analysis in male and female subjects showed no significant difference of the 5-HTTLPR S allele carriers frequency between obese and non-obese subjects (data not shown). Because there was a significant difference in frequency of Type 2 DM between non-obese and obese subjects (46.3% vs. 67.6%, P < 0.001), logistic regression analysis was performed with obesity as the dependent variable and presence of S allele, as contributing variable. However, the presence of the polymorphism was not found to be a significant predictor of obesity neither in an unadjusted model (OR = 1.22, 95% CI = 0.81–1.85, P = 0.340) nor in a model adjusted for age, sex, and Type 2 DM [Table 5]. Furthermore, the average BMI in the presence or absence of the S allele in the total number of subjects participating in the study was identical in the diabetic and nondiabetic group [Table 6]. Finally, and despite the fact that BMI was higher in diabetic subjects, no difference was found in the frequency of S allele carriers between obese and non-obese diabetic subjects and obese and non-obese non-diabetic subjects (P > 0.05, data not shown). 4. Discussion It is already known that 5-HT is an important regulator of energy balance and serotonin-sensitive neurons recruit various downstream systems to exert their effects on energy homeostasis [1]. Furthermore, we recently found that − 759C/T polymorphism of 5-HT2C receptor gene is strongly associated with the development of Type 2 DM (OR = 2.34) but not with obesity [13]. On the basis of the above, we hypothesized that variations in other elements of the 5-HT pathways may also influence the susceptibility to Type 2 DM and obesity. 5-HTTLPR appears to be a good such candidate since it is a 5-HTT geneTable 3 Logistic regression analysis with T2DM as the dependent variable and 5-HTTLPR genotype as contributing variable.
Model 1 (unadjusted) Presence of S allele
OR
95% CI
P value
2.08
1.39–3.12
< 0.001
1.41–3.25 1.06–1.16
< 0.001 < 0.001
Model 2 (Adjusted for age, sex and BMI)a Presence of S allele 2.14 BMI 1.11
OR: Odds ratio. a Age and sex were found not to be significant in model 2.
170
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
Table 4 Frequencies of 5-HTTLPR genotypes and alleles in the non-obese and in the obese group. Non-obese group
Obese group
n
95% CI
n
26.44–37.28 39.80–51.40 18.17–27.94
182 50 81 51
50.32–58.53 41.47–49.68
364 181 183
%
5-HTTLPR genotypes Total 281 LL a 89 31.7 a LS 128 45.6 SS a 64 22.8 5-HTTLPR alleles Total 562 La 306 54.4 a S 256 45.6 a
%
Table 6 BMI in the presence or absence of the S allele of 5-HTTLPR in diabetic and non-diabetic subjects. Absence of S
95%CI
P
BMI
n
BMI
139
29.5 ± 5.3
324
29.6 ± 4.7
Non-diabetic group Total 81
28.6 ± 5.2
87
28.3 ± 4.3
Diabetic group Total
30.8 ± 5.1
194
30.4 ± 4.8
All subjects Total 27.5 44.5 28.0
48.4 51.6
21.38–34.28 37.42–51.76 21.88–34.86
44.61–54.85 45.15–55.39
Presence of S
n
0.387
0.160
58
Values are mean ± SD (kg/m2).
No significant differences between non-obese and obese groups.
linked polymorphic region that regulates the transcriptional activity of the 5-HTT gene and has been associated with various diseases. Consequently, the aim of our study was to examine the possible association of 5-HTTLPR with Type 2 DM and obesity in humans. We found that higher frequency of 5-HTTLPR LS and SS genotypes is strongly associated with Type 2 DM (OR = 2.14 after adjustment of age, sex and BMI). No such association was found between 5-HTTLPR and obesity. Taken together these findings suggest that the lower activity of 5-HTTLPR S allele confers vulnerability to Type 2 DM independently of obesity status, probably provoking an impairment of brain serotonin transmission. Alternatively, basal expression in the presence of 5-HTTLPR LL genotype may be protective against the development of Type 2 DM. In humans, the only study that associates 5-HTTLPR with glucose levels was performed in healthy Japanese women who were not on medication for diabetes, hypercholesterolemia or hypertension. The difference in fasting blood glucose levels between beginning and end of the study in subjects with the SS genotype was significantly larger than the difference in subjects carrying LL and LS genotypes [9]. Indeed, Type 2 DM is accompanied by altered monoamine neurotransmission in the brain, mostly manifested by an increase in content and a decrease in turnover rate [14]. In 1997, Salkovic Petrisic and coworkers reported that diabetic rats exhibit altered neurotransmission in brain monoaminergic systems and that the altered monoamine transporter gene expression may possibly contribute to the observed dysfunctions in brain monoamine transmission in chronic diabetes [8]. 5-HTTLPR could be one of the genetic causes of this condition as the S allele results in lower expression of 5-HTT mRNA that might account for the features of brain insulin system observed in Type 2 DM. Our findings in humans correlating serotonin with T2DM are supported by several studies in animals which have shown that serotonin can alter not only release but also synthesis of insulin in pancreatic islets [15] and that a perturbation of brain serotonin systems can predispose to Type 2 DM in young adult mice [16]. Most studies have reported that 5-HT dose-dependently increases serum glucose in mice [17] whereas 5-HT receptor antagonists induce hyperglycaemia [18].
Table 5 Logistic regression analysis with obesity as the dependent variable and 5-HTTLPR genotype as contributing variable.
Model 1 (unadjusted) Presence of S allele
OR
95% CI
1.22
0.81–1.85
0.34
1.62–3.55 1.10–2.37
< 0.001 0.010
Model 2 (Adjusted for age, sex and T2DM) Presence of T2DM 2.40 Sex 1.62
OR: Odds ratio. Presence of S allele and age were not found to be significant in model 2.
P value
Serotonin is also well distributed in normal and diabetic pancreatic tissues and has stimulatory effects on insulin secretion from normal pancreas and glucagon secretion from diabetic pancreas in rats [19]. Furthermore, both metabolic syndrome and insulin resistance are associated with blunted central serotonergic responsiveness [20]. However, it is not clear whether insulin resistance causes decreased central serotonergic responsivity or the S allele of 5-HTTLPR which is found in higher frequency in patients with insulin resistance causes decreased serotonergic activity. Serotonin is also known to exert direct peripheral effects. Both serotonin and exogenous agonists of 5-HT2A/2C receptor increase glucose uptake in isolated myocytes by increasing recruitment of glucose transporters to the plasma membrane [21]. Recently, Wade and coworkers demonstrated in mice lacking 5-HT2C receptor and the adipocyte hormone leptin that the improvement in glucose tolerance in wild-type mice treated with the serotonin releaser and reuptake inhibitor fenfluramine was blunted in 5-HT2C receptor mutant mice indicating that 5-HT2c receptor has direct effects on glucose homeostasis independently of obesity status [22]. We recently provided clinical evidence for a direct role of serotonergic system in the development of Type 2 DM by showing that the functional -759C/T polymorphism of 5-HT2C receptor gene was strongly associated with Type 2 DM in a Caucasian population directly, independently of obesity status [13]. Several studies have investigated the association of 5-HTTLPR polymorphism with obesity. An early study had found that patients with anorexia nervosa have allele frequencies for 5-HTTLPR polymorphism not significantly different to those observed for obese and underweight individuals [23]. Two recent studies by Sookoian et al. showed that the S allele of 5-HTTLPR is a risk factor for obesity in adolescents [24] and in male adults with mean age of 34 years [25]. Furthermore, Lan et al. found that the SS genotype was associated with increased BMI only in non-elderly stroke subjects (<65 years) [26], and Mergen et al. showed a weak, but significant, association of 5-HTTLPR with BMI status in 262 obese (BMI > 30 kg/m2) and 138 control (BMI < 25 kg/m2) subjects (their age was not reported) [27]. Finally, Fuemmeler et al. found that the prevalence of obese and overweight plus obese combined was significantly lower among carriers of the L variant compared with subjects having SS genotype only in men adolescents [28]. In our study, which found no association between 5-HTTLPR and obesity, both male and female subjects have been included, with mean age of 67 years old and were divided according to BMI in obese (BMI > 30 kg/m2) and non-obese (BMI< 30kg/m2). Apparently, the association of 5-HTTLPR with obesity is agedepended and is gradually reduced or “masked” by other factors as subjects become older. In conclusion, we report for the first time that the S allele of 5-HTTLPR is strongly associated with the presence of Type 2 DM in a human population. This association is direct and not related to obesity status. The frequency of LL genotype was significantly lower in diabetic vs. nondiabetic group suggesting that this genotype might be protective from development of Type 2 DM or, conversely, the lower activity of
M. Iordanidou et al. / Clinica Chimica Acta 411 (2010) 167–171
the 5-HTTLPR S allele that impairs 5-HT transmission may confer vulnerability to this disease. This observation, taken together with our recent finding of an association of the -759C/T polymorphism of 5-HT2C receptor gene with Type 2 DM but not with obesity [13], indicates that the serotonergic system may be involved in the pathogenesis of Type 2 DM directly, with a mechanism not related to the regulation of food intake or obesity. Overall, our findings suggest that the 5-HTTLPR polymorphism might be a useful prognostic marker for Type 2 DM. In addition, this and our previous study [13] provide evidence in humans for a direct role of 5-HT in development of Type 2 DM and might open the road for better insights into the pathogenesis of Type 2 DM and novel approaches in the quest for pharmacological targets for its treatment. However, it should be kept in mind that no evaluation of population stratification was done in our study, thus these results should be considered as preliminary. Further studies in other populations are necessary in order to confirm and solidify these findings. References [1] Lam DD, Heisler LK. Serotonin and energy balance: molecular mechanisms and implications for type 2 diabetes. Expert Rev Mol Med 2007;9:1–24. [2] Smeraldi E, Serretti A, Artioli P, Lorenzi C, Catalano M. Serotonin-transporter genelinked polymorphic region: possible pharmacogenetic implications of rare variants. Psychiatr Gen 2006;16:153–8. [3] Ni W, Watts SW. 5-Hydroxytryptamine in the cardiovascular system: focus on the serotonin transporter (SERT). Clin Exp Pharmacol Physiol 2006;33:575–83. [4] Lesch KP, Balling U, Gross J, et al. Organization of the human serotonin transporter gene. J Neural Transm 1994;95:157–62. [5] Heils A, Teufel A, Petri S, Stöber G, Riedezez P, Bengel D, et al. Allelic variation of human serotonin transporter gene expession. J Neurochem 1996;66:2621–4. [6] Lesch KP, Wolozin BL, Estler HC, Murphy DL, Riederer P. Isolation of a cDNA encoding the human brain serotonin transporter. J Neural Transm 1993;91:67–73. [7] Mortensen OV, Thomassen M, Larsen MB, Whittemore SR, Wiborg O. Functional analysis of a novel human serotonin transporter gene promoter in immortalized raphe cells. Brain Res Mol Brain Res 1999;68:141–8. [8] Salkovic Petrisic M, Augood SJ, Bicknell J. Monoamine transporter gene expression in the central nervous system in diabetes mellitus. J Neurochem 1997;68:2435–41. [9] Yamakawa M, Fukushima A, Sakuma K, Yanagisawa Y, Kagawa Y. Serotonin transporter polymorphisms affect human blood glucose control. Biochem Biophys Res Commun 2005;334:1165–71. [10] Huang XF, Huang X, Han M, Chen F, Storlien L, Lawrence AJ. 5-HT2A/2C receptor and 5-HT transporter densities in mice prone or resistant to chronic high-fat dietinduced obesity: a quantitative autoradiography study. Brain Res 2004;1018: 227–35.
171
[11] Kaiser R, Tremblay PB, Roots I, Brockmoller J. Validity of PCR with emphasis on variable number of tandem repeat analysis. Clin Biochem 2002;35:49–56. [12] Yonan AL, Palmer AA, Gilliam TC. Hardy-Weinberg disequilibtium identified genotyping error of the serotonin transporter (SLC6A4) promoter polymorphism. Psychiatr Gen 2006;16:31–4. [13] Iordanidou M, Tavridou A, Vasiliadis MV, Azvanitidis KI, Petridis J, Chzistakidis D, et al. The -759C/T polymorphism of the 5-HT2C receptor is associated with type 2 diabetes in male and female Caucasians. Pharmacogenetics Genom 2008;18: 153–9. [14] MacCall AL. The impact of diabetes on the CNS. Diabetes 1992;41:557–70. [15] Peschke E, Peschke D, Hammer T, Csernus V. Influence of melatonin and serotonin on glucose-stimulated insulin release from perifused rat pancreatic islets in vitro. J Pineal Res 1997;23:156–63. [16] Nonogaki K, Strack AM, Dallman MF, Tecott LH. Leptin-independent hyperphagia and type 2 diabetes in mice with a mutated serotonin 5-HT2C receptor gene. Nat Med 1998;4:1152–6. [17] Sugimoto Y, Kimura I, Yamada J, Watanabe Y, Takeuchi N, Horisaka K. Effects of serotonin on blood glucose and insulin levels of glucose- and streptozotocintreated mice. Jpn J Pharmacol 1990;54:93–6. [18] Wonziak KM, Linnoila M. Hypergycemic properties of serotonin receptor antagonists. Life Sci 1991;49:101–9. [19] Adeghate E, Ponery AS, Pallot D, Parvez SH, Singh J. Distribution of serotonin and its effect on insulin and glucagons secretion in normal and diabetic pancreatic tissues in rat. Neuro Endocrinol Lett 1999;20:315–22. [20] Muldoon MF, Mackey RH, Korytkowski MT, Flory JD, Pollock BG, Manuck SB. The metabolic syndrome is associated with reduced central serotonergic responsivity in healthy community volunteers. J Clin Endocrinol Metab 2006;91:718–21. [21] Hajduch E, Rencurel F, Balendran A, Batty IH, Downes CP, Hundal HS. Serotonin (5Hydroxytryptamine), a novel regulator of glucose transport in rat skeletal muscle. J Biol Chem 1999;274:13563–8. [22] Wade JM, Juneja P, MacKay AW, Gzaham J, Havel PJ, Tecott LH, et al. Synergistic impairment of glucose homeostasis in ob/ob mice lacking functional serotonin 2C receptors. Endocrinology 2008;149:955–61. [23] Hinney A, Barth N, Ziegler A, von Prittwitz S, Hamann A, Hennighausen K, et al. Serotonin transporter gene-linked polymorphic region: allele distributions in relationship to body weight and in anorexia nervosa. Life Sci 1997;6:295–303. [24] Sookoian S, Gemma C, Garcia SI, Gianotti TF, Dieuzeide G, Roussos A, et al. Short allele of serotonin transporter gene promoter is a risk factor for obesity in adolescents. Obesity 2007;15:271–6. [25] Sookoian S, Gianotti TF, Gemma C, Burgueño A, Pirola CJ. Contribution of functional 5-HTTLPR variant of the SLC6A4 gene to obesity risk in male adults. Obesity 2008;16:488–91. [26] Lan MY, Chang YY, Chen WH, Kao YF, Lin HS, Liu JS. J Endocrinol Invest 2009;32: 119–22. [27] Mergen H, Karaaslan C, Mergen M. Deniz Ozsoy E, Ozata M. LEPR, ADBR3, IRS-1 and 5-HTT genes polymorphisms do not associate with obesity. Endocr J 2007;54: 89–94. [28] Fuemmeler BF, Agurs-Collins TD, Mccleron FJ, Kollins SH, Kail MF, Bergen AW, et al. Obesity 2008;16:348–55.