Alteration of serotonin transporter density and activity in fibromyalgia

The aim of the study was to evaluate the kinetic parameters of a specific serotonin transporter (SERT) and serotonin uptake in a mentally healthy subset of patients with fibromyalgia. Platelets were obtained from 40 patients and 38 healthy controls. SERT expression and functionality were evaluated through the measurement of [3H]paroxetine binding and the [3H]serotonin uptake itself. The values of maximal membrane binding capacity (Bmax) were statistically lower in the patients than in the healthy volunteers, whereas the dissociation constant (Kd) did not show any statistically significant variations. Moreover, a decrease in the maximal uptake rate of SERT (Vmax) was demonstrated in the platelets of patients, whereas the Michaelis constant (Km) did not show any statistically significant variations. Symptom severity score (tiredness, tender points index and Fibromyalgia Impact Questionnaire) were negatively correlated with Bmax and with Vmax, and positively correlated with Km. A change in SERT seems to occur in fibromyalgic patients, and it seems to be related to the severity of fibromyalgic symptoms.


Introduction
Fibromyalgia syndrome (FMs) is a chronic pain syndrome characterized by widespread pain and stiffness, multiple tender points, and fatigue [1]. This pain syndrome has an incidence of 2% in the general population and occurs with higher frequency among women in middle age [2]. FM is often associated with increased prevalence of depressive symptoms, major depression and anxiety [3]. The cause and pathophysiology of FMs is unclear; pathophysiological hypotheses include impairment in the functioning of the hypothalamic-pituitary axis and alterations in specific neurotransmitters such as substance P, N-methyl-D-aspartate, noradrenaline (norepinephrine) and serotonin (5-HT). However, interest has been growing in a possible involvement of 5-HT in FM. Indeed, strong evidence has accumulated to support the hypothesis that a deficiency in serotonergic neuronal functioning might be related to the pathophysiology of FM [5][6][7][8]. Patients with FM have been found to have decreased concentrations of 5-HT and tryptophan  in serum and cerebrospinal fluid. 5-HT, in particular, is theorized to have a function in stage 4 sleep and in the pain threshold [9]. This neurotransmitter is implicated in psychiatric disorders such as depression, anxiety, and obsessive compulsive disorder. Stressful experiences lead to depression in some people who are already genetically predisposed, and increase the probability of FM exordium. The 5-HT gene could moderate the serotonergic response to stress [4].
As a mediator, 5-HT exerts its actions by means of interaction with distinct receptors, which are differentiated on the basis of structures, molecular mechanisms and pharmacological profiles [9]. On this basis, drugs acting on 5-HT receptors, in particular on 5-HT 2 and 5-HT 3 , are being used or investigated for the clinical management of FM [10,11]. Among drugs 5-HT = serotonin; B max = maximal membrane binding capacity; FIQ = Fibromyalgia Impact Questionnaire; FM = fibromyalgia; K d = dissociation constant; K m = Michaelis constant; SERT = specific serotonin transporter; SSRI = selective 5-HT reuptake inhibitor; TPi = tender point index; V max = maximal uptake rate of SERT. targeting 5-HT receptors, ketanserin is a selective 5-HT 2 antagonist that can reduce the hyperalgesia, spontaneous pain, sleep disorders and other symptoms of FM [12], and granisetron and tropisetron are selective 5-HT 3 receptor antagonists that show clinical efficacy in FM [11].
In an analogous manner to other transmitters, the endogenous activity of 5-HT is controlled by a specific 5-HT transporter (SERT), which mediates the intracellular reuptake of 5-HT and can be specifically blocked by selective 5-HT reuptake inhibitors (SSRIs) such as paroxetine and fluoxetine. SERT is widely expressed in intestinal epithelial cells, in central or peripheral serotonergic neurons and in platelets; it shares common molecular and physiological features in these locations [13][14][15]. Clinical studies have demonstrated the efficacy of SSRI in FM [16], but the data are not unequivocal. Nociception refers to the physiological process of transmitting a painful stimulus from the periphery through afferent neurons to the cerebral cortex. It has been postulated that serotonergic neurotransmission has a significant function in nociception [17,18]; alterations in 5-HT metabolism and transmission might therefore be important in the pathogenesis of FM. These findings support the proposal that aberrant pain perception in FM also results from an instability of the 5-HT system in FM. There is also evidence that changes in the expression of SERT are due to a polymorphism in the transcription region in patients with FM [19].
In the present study, both the expression and functionality of SERT were determined in platelets collected from patients with FM, with the following aims: first, to perform a comparison with the pharmacological profile of platelet SERT in healthy volunteers, and second, to examine putative correlations of SERT characteristics with the severity of symptoms such as tiredness, Fibromyalgia Impact Questionnaire (FIQ) score or tender point index (TPi).

Subjects
Forty patients (all female) affected by primary FM, aged 53 ± 13 years (mean ± SD) took part in the study, and 38 healthy females age-matched to the patients (50 ± 12 years) were used as a control group. The American College of Rheumatology criteria for FM [1] were used to make the diagnosis of FM. The inclusion criteria for the study groups comprised a negative history for psychoactive drug treatment and other neurological disorders. None of the subjects had comorbid psychiatric disorders or had received treatment with antidepressant drugs. No patient was under pharmacological treat-ment. All patients underwent a wash-out period of 2 months before the study. They were enrolled at the University Division of Rheumatology, Santa Chiara Hospital, Pisa. Written consent was obtained from all patients and controls after a full explanation of the procedure. The study was approved by the local Ethics Committee.

Evaluation of clinical parameters
For each patient and control, tenderness at tender points was evaluated by means of the Fischer dolorimeter [20]. The total fibromyalgic tender point score (right plus left) was used for statistical analysis. Each positive tender point had a pain score between 0 (no pain) and 3 (severe pain). We also calculated the TPi as the sum of the scores of all tender points divided by the total number of tender points.
To estimate the impact of FM on the quality of life, all patients and controls received an FIQ [21]. The total score was the sum of the values of the 10 FIQ items, which reflected the impact of FM and ranged from 0 (no impact) to 100 (maximum impact).
Tiredness was evaluated as measured by the FIQ tiredness item, which consisted of a visual analogic scale numbered between 0 and 10.
To exclude major psychiatric disorders, all patients were evaluated by means of a diagnostic interview consisting of the administration of the Structured Clinical Interview for DSM-IV axis-I disorder (SCID-I/P) [22]. This assessment was conducted by psychiatrists who were trained and certified in the use of the study instruments at our department.

Separation of platelets
Venous blood (30 ml) was collected from each subject and gently mixed with 1 ml of anticoagulant (0.15 M EDTA). Platelet-rich plasma was obtained by low-speed centrifugation (200g for 20 minutes at 22°C). Platelets were counted automatically with a flux cytometer (Cell-dyn 3500 system; Abbott, Milano, Italy).
For measurement of [ 3 H]serotonin reuptake, platelets were used immediately, whereas for [ 3 H]paroxetine binding, platelets were precipitated by centrifugation at 10,000g for 10 min- Protein concentration was determined with the method of Lowry and colleagues [24] using bovine serum albumin as a standard.

Data analysis
Equilibrium-saturation binding data, the maximum binding capacity (B max , fmol/mg of protein) and the dissociation constant (K d , nM) were analysed by means of the iterative curvefitting computer programs EBDA and LIGAND (Kell for Windows, v. 6.0).
The maximal uptake rate of SERT (V max , pmol/10 9 cells per minute) and the Michaelis constant (K m , nM) were obtained by direct weighted nonlinear regression of uptake rate against [ 3 H]serotonin concentration with GraphPad PRISM software (GraphPad, San Diego, CA, USA).
Statistical analysis was performed with Student's t test. The relationship between variables was checked with a two-tailed Spearman analysis.

Results
[ 3 H]Paroxetine binding sites on the SERT of platelet membranes were used as a first marker of serotonergic function.  Table 3). A comparison of these parameters with those in healthy volunteers indicated significant differences for V max values but not for K m values: whereas the mean V max in the patients with FM was significantly lower (90 ± 4 versus 114 ± 6 pmol/10 9 platelets per minute for controls; p < 0.001), the mean K m in the patients with FM was not significantly different (96 ± 15 versus 114 ± 15 nM for controls; Table 2).
The severity of the symptoms was related to V max and K m in that the statistical analysis showed a negative correlation between the symptom scores and the respective V max values (p < 0.0001 for tiredness; p < 0.05 for TPi and p < 0.001 for FIQ). We also found a positive correlation between K m values and tiredness (p < 0.001), TPi (p < 0.001) and FIQ (p < 0.001).
The covariate age was shown not to have an effect on the variability of the V max or K m values in either patients with FM or healthy volunteers.

Discussion
The present study is the first examination of the relationships between SERT expression and kinetic parameters of 5-HT uptake. We have found statistically significant differences in B max and V max values between patients with FM and controls. Moreover, we found a negative correlation between symptom scores and the respective B max and V max values (B max and V max versus tiredness, TPi and FIQ score). The platelet is considered to be a peripheral model of neuronal activity with respect to 5-HT function. In fact, previous studies have demonstrated that the same SERT is expressed in the central nervous system and platelets [14]. Moreover, the identity between the two structures, as confirmed by sequence homologies through cloning studies [15], has provoked a surge of different studies in neuropsychiatric disorders, given the possibility of exploring peripherally a mechanism of the central nervous system [26].
Recently it has been proposed that altered serotonergic neuronal function might be related to the pathophysiology of FM [5,8,27]. These findings prompted us to investigate the characteristics of SERT in the platelets of patients with FM. B max and K d values of [ 3 H]Paroxetine binding were assumed to represent SERT density and ligand binding affinity, respectively, whereas V max and K m values of [ 3 H]Serotonin uptake were taken as estimates of SERT rate and affinity, respectively. A very interesting observation was that both binding and uptake parameters differed significantly from those of healthy volunteers.
The patients with FM have fewer SERTs expressed on the cellular membrane than healthy subjects (a decreased B max , perhaps because the SERTs are less transcribed). Besides having fewer SERTs, patients with FM have a deficit in functionality (demonstrated by a decrease in transport rate).
Such combined changes in B max and V max values allow the inference that the efficiency of 5-HT uptake by platelet SERT is altered. Our previous studies demonstrated an alteration of SERT density and of the uptake rate of SERT in psychiatric patients [28,29]. Consistent with this suggestion was the correlation analysis in the present study: the lower the density and rate of SERT on platelet membranes, the higher the severity of FM symptoms. Moreover, the K m values were also positively correlated with tiredness, TPi and FIQ.
A reduced density and rate of SERT are consistent with previous observations indicating that levels of 5-HT are altered in patients with FM [30,31]. The biophysiological mechanism of FM has been proposed to be similar to that in depression, and it has been suggested that this is likely to result from a neuroendocrine/neurotransmitter dysregulation [32]. However, we suppose that the alterations in B max and V max values are not related to the pathophysiology of FM but are a consequence of FM. Our hypothesis is that a decrease in B max and V max of SERT is due to a pain stimulus [33,34].
It has been shown that the decreased pain perception threshold during depression is likely to result from a dysfunction in several neurotransmitter systems, especially the serotoninergic one, which is also involved in the pathophysiology of depression [35]. In addition, an excessive stimulation of peripheral 5-HT receptors would account for pain and might explain why the clinical use of 5-HT 3 receptor antagonists such as tropisetron or granisetron can promote the relief of disturbance associated with FM [11,36].
SERT has been investigated previously in patients with FM, with discordant results. Russell and colleagues [37] found a higher B max in patients with FM than in healthy controls, whereas other authors found normal B max values [28,35] using [ 3 H]Paroxetine or [ 3 H]Imipramine [38]. In our experiments we used [ 3 H]Paroxetine, which binds with high affinity to a specific population of binding sites located on human platelets and neuronal membranes, associated with 5-HT uptake mechanisms [39]. The present results indicate a decrease in the density and rate of platelet SERT in patients with FM, and allow us to propose a specific role for SERT in the pathogenesis of FM. In fact, we avoided the inclusion of patients with FM who had psychiatric components because it is known that in psychiatric disorders such as depression, the expression of SERT is altered [40][41][42] and it is very difficult to identify the role of the two components in patients with FM who have comorbid psychiatric disorders. Thus, the changes in B max and V max demonstrated in our study may be due to FM only.
There is also a possible contribution from 5-HT to the aetiology of FM because of the efficacy of SSRIs in the management of chronic pain in idiopathic pain disorders [43]. Thus, in view of the decreased B max and V max values found in our subset of Results are means ± SD. FIQ, Fibromyalgia Impact Questionnaire; FM, fibromyalgia; TPi, tender point index. Results are shown as means ± SEM. FM, fibromyalgia; K m , Michaelis constant; V max , maximal uptake rate of specific serotonin transporter. a p < 0.001. mentally healthy patients, who were SSRI free, we propose that there is a compensatory mechanism in the central nervous system to relieve the pain. This may clarify the improvement in the therapeutic effectiveness of SSRI in the patients with FM.

Conclusion
This is to our knowledge the first observation that, apart from a decrease in expression, there is also an alteration in the rate of SERT which seems to depend not only on the SERT number. In fact, B max is not correlated with V max (data not shown).
In the patients with FM, the decrease in 5-HT levels, which had already been observed, together with the impaired SERT functionality, might contribute to the pathogenesis of the disease, both in quantity and rate. In fact, these two factors are important because they are correlated with the level of disease severity.
Thus, the results of the present study are in agreement with the hypothesis that a deficit in the 5-HT transporter site could be of pathogenetic significance in FM syndrome.