C.-W. Tsao, PhD, Department of Nursing, Chung Hwa College of Medical Technology, No 89, Wen-Hwa 1st Street, Jen-Te Hsiang, Tainan County, Taiwan. E-mail: firstname.lastname@example.org
Clinical reports document that depression as a side effect is more prevalent in hepatic patients given interferon (IFN)-α therapy than in those given lamivudine. The mechanisms, however, are poorly understood. Serotonin transporter (5-HTT), via uptake of serotonin (5-HT) into presynaptic serotoninergic neurons, is an initial action site for antidepressants. Real-time polymerase chain reaction (PCR) was used to quantify 5-HTT mRNA expression in immune cells in order to evaluate whether 5-HTT acted as an indicator of depression. Results showed that the 5-HTT mRNA expression was much higher in T-cell and B-cell lines than that in a monocytic cell line. Treatment with either lamivudine or ribavirin reduced the 5-HTT mRNA expression, protein level and 5-HT uptake in T-cell line. Treatment with IFN-α, however, increased those levels in the same group. A similar effect was observed in peripheral blood mononuclear cells (PBMC). Mimicking clinical use by treating PBMC with a combination of IFN-α and ribavirin increased the 5-HTT mRNA expression level. Our study indicates that these therapeutic drugs regulate 5-HTT expression, which implies that 5-HTT might be a trait marker in IFN-α-induced depression after hepatic therapy.
Interferon (IFN)-α has antiviral, antiproliferative and immuno-modulating properties and has been available for clinical use in patients infected with chronic hepatitis B virus (HBV) [1–3]. An oral nucleoside analogue, lamivudine, which markedly inhibits HBV replication, is also used in HBV therapy [1–4]. IFN-α alone or in combination with ribavirin, an antiviral drug, is also effective for treating chronic hepatitis C virus (HCV) infection [5–10]. Several studies [11–14] indicate that IFN-α therapy caused side effects such as influenza-like syndrome, leukopenia and thrombocytopenia, and symptoms of neuropsychiatric disturbances including depression and suicide [11–14]. IFN-α plus ribavirin therapy for HCV infection has also caused depression . Adverse events derived from lamivudine treatment, such as diarrhoea, dizziness, nausea and vomiting, have been documented. However, only a few studies [4, 16] reported depression as a side effect of lamivudine treatment for hepatitis. Once these serious side effects occurred, immediate dose reduction or discontinuation was indicated. Some reports suggested that patients with IFN-α-induced depression might obtain symptom relief by treatment with antidepressants [17, 18].
The serotonin transporter (5-HTT) is critical for terminating serotoninergic neurotransmission by taking up serotonin (5-HT) into presynaptic neurons; it is therefore an initial action site for selective 5-HTT reuptake inhibitors . In addition to low levels of 5-HT receptor in the brain of depressed patients , 5-HTT was also thought to be a potential substrate for the pathophysiology of suicide and highly correlated with depressive symptoms . Depressed patients with extremely high levels of dysfunctional attitudes also had increased regional 5-HTT-binding potential in the brain . In an animal model, knockout mice without the 5-HTT reduced clearance of extracellular 5-HT, altered in 5-HT neuronal firing and receptor function, and increased anxiety-like behaviours .
5-HTT across the human platelet membrane has widely been used as a cell model for investigations of neuronal serotonin reuptake. Several studies have reported, however, that platelets are not suitable for depression verification due to contradictory results in radioligand-binding assays [24–29]. 5-HTT has been observed in the plasma membrane of serotonergic neurons, platelets and human placenta  as well as in lymphocytes (human immune cells) [31–33]. Lymphocytes have also been used as neural probes in studying psychiatric disorders . One study reported that 5-HTT numbers from peripheral blood mononuclear cells (PBMC) were significantly reduced in patients with major depression as compared with controls .
To investigate the physiological variations in gene expression, the quantification of relative expression ratio (a target gene versus a reference gene) seems adequate for most purposes . Porphobilinogen deaminase (PBGD) is used as a housekeeping gene for such relative quantification. To evaluate whether 5-HTT acts as an indicator of depression after hepatic therapy, we established in vitro cultures and used fluorescence real-time polymerase chain reaction (PCR) in a LightCycler to quantify the 5-HTT mRNA expression in various immune cell lines and PBMC, all of which had been treated with lamivudine, ribavirin or IFN-α, respectively. The present study demonstrated the drugs for hepatic therapy-regulated 5-HTT expression, which implicated that 5-HTT might act as a trait marker in IFN-α-induced depression after hepatic therapy.
Materials and methods
Cell lines. Human T-leukaemia cell lines MOLT-4 (ATCC CRL-1582) and Jurkat T (ATCC TIB-152) and human monocytic cell line U937 (ATCC CRL-1593) were obtained from Dr B. C. Yang. Human B-cell line Raji (ATCC CCL-86) was a gift from Dr S. H. Chen (Department of Microbiology and Immunology, National Cheng Kung University Medical College, Tainan, Taiwan).
Chemicals. IFN-α-2b was obtained from Dr T. T. Chang (Department of Internal Medicine, National Cheng Kung University Medical College, Tainan, Taiwan). Lamivudine and ribavirin were gifts from GlaxoSmithKline (Taipei, Taiwan) and Schering-Plough (Madison, NJ, USA), respectively.
Preparation of PBMC. PBMC from healthy volunteers were obtained using sedimentation over a polymeric erythrocyte aggregating agents (Ficoll-Paque; Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA) and were centrifuged at 200 × g for 20 min at 20 °C. The PBMC-containing suspensions were then washed twice in RPMI-1640 medium (Gibco BRL, Grand Island, NY, USA) and centrifuged again at 125 × g for 10 min at 4 °C.
Cell cultures. MOLT-4, Jurkat T, Raji, U937 cells and PBMC were cultured in RPMI medium containing 10% fetal calf serum and 0.1% gentamicin in 24-well plates (106 cells/well). In some experiments, different doses of lamivudine, ribavirin or IFN-α were added to the cultures. Cells were harvested at various times, and the cell numbers were determined using a hemacytometer. The percentages of viability were determined using trypan blue dye exclusion.
RNA extraction. Isolation of RNA from various immune cell lines and PBMC (107) was performed using a reagent (trizol; Gibco BRL). After phenol/chloroform extraction and isopropanol precipitation, the RNA was washed in 75% ethanol, quantitated using a spectrophotometer (Gene Quant II; Amersham Pharmacia Biotech Inc., Cambridge, UK) at 260 nm and stored at −80 °C.
Real-time RT-PCR. Single-stranded cDNA were synthesized from 5 µg of total RNA using 25 units of Moloney murine leukaemia virus reverse transcriptase (Promega, Madison, WI, USA) and 6 µg/ml of oligo primers (Promega) at 37 °C for 2 h. PCR amplification was performed in a final 20 µl volume consisting of 5 µl of cDNA, 0.5 µm sense and antisense primers, 0.2 µm fluorescence probes, 4 mm MgCl2 and 2 µl FastStart Taq DNA polymerase (Roche Diagnostics GmbH, Roche Applied Science, Mannheim, Germany). Real-time PCR was monitored online using a thermocycler (LightCycler 2.0; Roche Molecular Systems, Alameda, CA, USA). Each sample (20 µl) was placed in a thermocycler glass capillary. Temperature cycling for the PCR run comprised denaturation at 95 °C for 10 min, the annealing conditions at 95 °C for 5 s, 55 °C for 10 s, an extension at 72 °C for 15 s (50 cycles) and, finally, a cooling step to 40 °C. The oligonucleotide primers and fluorescence resonance energy transfer probes of 5-HTT and PBGD were designed by TIB MOLBIOL (Berlin, Germany) (Table 1). PCR products were subjected to agarose gel electrophoresis and visualized by ultraviolet light in the presence of ethidium bromide.
Table 1. Oligonucleotide primers and fluorescence resonance energy transfer probes of serotonin transporter and porphobilinogen deaminase
Quantitative detection of 5-HTT mRNA was using the thermocycler, and all subsequent quantification steps were performed according to the manufacturer's instructions. The housekeeping gene, PBGD, was equally processed in separate tubes. The reaction product served as a reference to normalize each sample for relative quantification. A standard curve was generated by serial twofold dilutions of the template cDNA from the 6-h MOLT-4 medium control (copy numbers: 2.691 × 106−2.756 × 109 copies/µl). The cycle number was defined the position of each amplification curve at the crossing point (Cp). Determination of the Cp value of 5-HTT cDNA was derived from thermocycler analysis using the second derivative maximum method. A calibrator was used in each experiment. The graph of the linear regression and calculation of the regression coefficient r confirmed the accuracy and reproducibility of this approach (HTT: Slope = −3.480, Error = 0.075, r = −1.000; PBGD: Slope = −3.166, Error = 0.059, r = −1.000).
Western blotting. Cells (107) were centrifuged at 1500 rpm for 5 min at 4 °C, and the pellets were resuspended in 1 ml of ice-cold Dounce buffer containing 10 mm Tris-Cl (pH 7.5), 0.5 mm MgCl2, the protease inhibitors leupeptin (10 µg/ml), aprotinin (10 µg/ml), phenylmethylsulfonyl fluoride (1 mm) and iodoacetamide (1.8 mg/ml). The mixture was incubated on ice for 10 min, homogenized and then added to tonicity restoration buffer containing 10 mm Tris-Cl (pH 7.6), 0.5 mm MgCl2 and 0.6 m NaCl, plus protease inhibitors as in Dounce buffer. The mixture was centrifuged at 125 × g for 5 min at 4 °C. To the supernatant, 0.5 m EDTA was added to a final concentration of 5 mm, and the membrane suspensions were centrifuged at 160,000 × g for 45 min at 4 °C. The supernatant was removed, and the pellet was resuspended in Triton X-100 lysis buffer with protease inhibitors, continuously shaken in a vibrator at 4 °C for 45 min, and then centrifuged at 6000 × g for 15 min at 4 °C. Finally, the supernatant was transferred to a new Eppendorf tube and stored at −80 °C.
Cell membrane extract (15 µg in each sample) was separated using SDS-PAGE (100 V) for 60 min and transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA) (60 V) for 90 min. After blocking for 2 h, the membrane was washed five times with PBS containing 0.1% Tween 20, and blots were developed using 1–10 µg/ml rabbit anti-5-HTT polyclonal antibody (Chemicon International Inc., Temecula, CA, USA). Blots were then hybridized using anti-rabbit immunoglobulin (Ig)G conjugate peroxidase (1:10,000) for 1 h (Chemicon International Inc.) and developed using an AEC substrate kit (Zymed Laboratories Inc., San Francisco, CA, USA). A prestained protein ladder (Fermentas Inc., Hanover, MD, USA) was indicated as a marker. The optical densities of the bands were determined using bio-1d analysis software (Bio-Capt MW; Vilber Lourmat, Marne La Vallée, France). In addition, an equal amount (15 µg) of total cell lysate was analyzed as an internal control. Blots were hybridized with mouse anti-human β-actin (1:5000) (Sigma-Aldrich Inc., St Louis, MO, USA) followed by anti-mouse IgG conjugate peroxidase (1:10,000) (Sigma-Aldrich). Finally, the relative expressions of the 5-HTT levels were normalized to that of β-actin for each sample.
5-HT uptake. Jurkat T cells were cultured using different doses of lamivudine, ribavirin, IFN-α, or medium control in 24-well plates (106 cells/well), and harvested after 24 h. Cells (107) in each group were incubated with 0.1 µm 5-hydroxy[3H]tryptamine trifluoroacetate (Amersham Pharmacia Biotech Inc.) for 30 min at 25 °C with or without imipramine (0.1 mm). Cells were washed twice by centrifugation (125 × g, 10 min) in PBS containing 50 µm imipramine and solubilized with 1% SDS. The radioactivity of the cell extracts was measured using a liquid scintillation counter (LS6500; Beckman Coulter, Inc., Fullerton, CA, USA).
Statistical analysis. Student's t-test was used to examine the differences between the drug-treated group and the medium control group for 5-HTT genes as expressed in a sample versus reference gene PBGD at different periods of time. The results are presented as relative values. Student's t-test was also used to measure 5-HT uptake. Statistical significance was set at P < 0.05.
5-HTT mRNA expression in various immune cells
5-HTT mRNA expression was identified using fluorescence real-time RT-PCR in a LightCycler in various human immune cells for different periods of time. The cycle number, which defined the position of each amplification curve at the Cp of each sample from various cells, was also analyzed. The Cp ranges of 5-HTT were 35–36 cycles in MOLT-4 and Jurkat T cells, 34–35 cycles in Raji cells and >35 cycles in U937 cells within 24-h cultures. Using the equation derived from the standards, the Cp values of samples were converted to concentrations. Each sample from different cells was normalized using its PBGD content as a reference for relative quantification. The ratios of 5-HTT/PBGD mRNA expression at 0 h were higher in the MOLT-4 (1.001 ± 0.156) and Jurkat T (1.056 ± 0.357) T-cell lines and in the Raji (0.805 ± 0.030) B-cell line than in the U937 (0.025 ± 0.013) monocytic cell line (Fig. 1A).
We also found 5-HTT mRNA expression in PBMC (range of Cp = 32–34 cycles from 7 healthy volunteers) (Fig. 1A). Moreover, the levels of the PBGD mRNA expression in PBMC were 10–100 times lower than those of the cell lines, because the PBGD of the PBMC was observed at a later stage (Cp ≥ 29 cycles in PBMC versus Cp ≥ 21 cycles in cell lines) (data not shown). Therefore, the ratios of 5-HTT/PBGD appeared much higher than in the other cell lines, e.g. 4.290 ± 0.083 in MOLT-4, 5.689 ± 2.382 in Jurkat T, 1.454 ± 0.417 in Raji and 0.045 ± 0.064 in U937 versus 348.892 ± 65.003 in PBMC after 3 h of culturing (Figs. 1A,2; Table 2). Data are the averages of ratios of 5-HTT/PBGD mRNA expression obtained from four experiments.
Table 2. Effects of interferon-α, lamivudine and ribavirin on serotonin transporter (5-HTT) mRNA expression in various immune cells
P < 0.05.
P < 0.01 compared with medium control.
MOLT-4, Jurkat T, Raji and U937 cells (106 cells) were cultured in medium alone or with lamivudine, ribavirin or interferon-α. Cells were harvested at 3 h, and then RNA was isolated using Trizol reagent. Single-stranded cDNA was synthesized as described in Materials and methods. Real-time polymerase chain reaction was done using a LightCycler. Data are expressed as mean ± SD of the averages of ratios of 5-HTT genes displayed by a sample versus reference gene porphobilinogen deaminase obtained from four experiments.
In addition to PBGD, human RNA polymerase II (RPII) (GenBank accession No. NM_000937) was also used as another reference for relative quantification. The Cp ranges of RPII were 27–29 cycles in MOLT-4 cells, Jurkat T cells and PBMC, 23–26 cycles in Raji cells and 26–28 cycles in U937 cells within 24-h cultures. The ratios of 5-HTT/RPII mRNA expression at 0 h, similar to 5-HTT/PBGD expression, were higher in MOLT-4 (0.575 ± 0.206) and Jurkat T (0.645 ± 0.118) T-cell lines, the Raji (1.058 ± 0.782) B-cell line and PBMC (0.780 ± 0.584) than in the U937 (0.025 ± 0.001) monocytic cell line.
5-HTT mRNA expression during cell growth
Continued cell culture indicated that the levels of 5-HTT mRNA expression were highest at 3 h and decreased at 6 h in MOLT-4, Jurkat T and Raji cells (Fig. 1A). There was no significant change in cell numbers in the various cell lines during the first 6 h of culturing (Fig. 1B). After 24 h of culturing, however, the number of cells increased by about 50%, and 5-HTT mRNA expression returned to the level of 0 h. Similar effects were also observed in PBMC: after 24 h of culturing, 5-HTT mRNA expression in PBMC decreased to below the level of 0 h. 5-HTT mRNA expression was low in U937 cells during the 24-h period (Fig. 1). Similar results were also observed in 5-HTT/RPII mRNA expression using RPII as the endogenous control (data not shown).
Effects of lamivudine, ribavirin and IFN-α on 5-HTT mRNA expression in various immune cells
A dose-dependent decrease in the levels of 5-HTT mRNA expressions in MOLT-4, Jurkat T and Raji cells was found after lamivudine (10−10−10−8m) and ribavirin (10−7−10−6m) treatment for 3 h (Table 2). The decrease continued for the 24 h of cell culture (data not shown). The levels in these groups increased, however, when treated with optimal doses (50 and 100 units) of IFN-α (Table 2). Similar results were also observed in the 5-HTT/RPII mRNA expression (data not shown). Compared with the medium control cells, no significant differences in cell numbers or percentages of viability were detected with various doses of lamivudine, ribavirin or IFN-α treatment (data not shown).
Effects of lamivudine, ribavirin and IFN-α on 5-HTT mRNA expression in PBMC
PBMC from seven healthy individuals were cultured for 24 h. Compared with the medium control cells, no significant differences in the cell numbers or percentages of viability were detected through various doses for numerous trials of lamivudine, ribavirin or IFN-α treatment (data not shown). At 3 h, however, there was a marked dose-dependent decrease in the levels of 5-HTT mRNA expressions after treatment with lamivudine (10−10−10−8m) and ribavirin (10−7−10−6m) compared with medium control (Fig. 2A); 5-HTT mRNA expression levels significantly increased in response to treatment with IFN-α at the optimal dose of 50 units in this group (Fig. 2B). Furthermore, adding 50 units of IFN-α and 10−6 M of ribavirin to PBMC, mimicking clinical use, also increased 5-HTT expression levels. These increased levels were significantly lower than those treated with IFN-α alone but higher than those treated with ribavirin (Fig. 2B). The results suggested that hepatic therapeutic drugs might regulate 5-HTT mRNA expression. Similar results were also observed in 5-HTT/RPII mRNA expression (data not shown).
Effects of lamivudine, ribavirin and IFN-α on 5-HTT protein expression in various immune cells
The deduced amino acid sequences of mammalian 5-HTT cDNA clones predict a polypeptide of approximately 68 kDa [30, 37, 38]. Several forms of human 5-HTT were distinguished: a small band (approximately 60 kDa) that is present only in the intracellular fraction, a broad band (70–100 kDa) that is typical of mature 5-HTT [37, 38] and a larger band (approximately 220 kDa) in both surface and intracellular fractions that may represent an SDS-resistant 5-HTT protein complex. We found two obvious bands close to 72 kDa (Fig. 3A). The expression of 5-HTT protein was much higher in Jurkat T cells, Raji cells and PBMC than in U937 cells at 0-h culture. The relative densities of the mature surface 5-HTT (approximately 72 kDa) bands were averaged and normalized to those of β-actin for different periods of time (Fig. 3B,C). The protein levels of 5-HTT increased in Jurkat T and Raji cells after 3–6 h of cell culture and were sustained to 24 h (Fig. 3B). Furthermore, a decrease in the levels of 5-HTT was detected after treating Jurkat T cells (at 3 h) and Raji cells (at 24 h) with lamivudine (10−10−10−8m) and after treating them both (at 24 h) with ribavirin (10−7−10−6m) (Fig. 3C). However, Jurkat T cells exposed to 50 units of IFN-α showed a slight (18.1%) increase in 5-HTT levels at 3 h (data not shown) and a significantly larger (69.4%) increment at 24 h (Fig. 3C). Raji cells also showed (at 3 h) an increase in 5-HTT levels after treatment with IFN-α at 50 and 100 units. In growing U937 cells, we found no changeable 5-HTT proteins after treatment with these three drugs (data not shown).
Effects of lamivudine, ribavirin and IFN-α on 5-HT uptake in Jurkat T cells
Because we found that three hepatic drugs affected 5-HTT mRNA levels, we chose to examine 5-HT uptake in Jurkat T cells, which presented higher 5-HTT mRNA expression. Compared with medium control, Jurkat T cells treated for 24 h with lamivudine (10−10−10−8m) and ribavirin (10−7−10−6m) showed a reduction in 5-HT uptake from 96.1 to 68.2% and from 85.1 to 74.7%, respectively. On the contrary, when Jurkat T cells were treated with IFN-α, 5-HT uptake increased to 119.4% at 50 units and 105.5% at 100 units (Fig. 4).
Several studies [6,8–10] reported that screening and monitoring depression in patients as they undergo IFN-α treatment for chronic hepatitis is important. Our study showed that IFN-α treatment upregulated 5-HTT mRNA expression in MOLT-4, Jurkat T and Raji cells. An increase in 5-HTT protein level and 5-HT uptake was also detected when Jurkat T cells were exposed to IFN-α. One report  on IFN-α regulation of 5-HTT expression at the transcriptional level in human placental choriocarcinoma (BeWo) cells and the midbrain and adrenal glands of mice corresponded with this observation. Moreover, this upregulation caused a significant increase in the uptake of 5-HT, which was abolished by actinomycin D, indicating that the IFN-α-induced increase in 5-HTT on the plasma membrane was elicited by the upregulation of 5-HTT transcription.
A number of studies about 5-HTT regulation [40–44] have reported various signalling pathways involving cAMP, cGMP, PKC, Ca2+, calmodulin and oxidants. In in vitro culture systems, upregulation of 5-HTT mRNA by cytokines, including interleukin-1β, tumour necrosis factor-α, IFN-α, and IFN-γ, has also been a focus of interest [39,45,46]. Our study further demonstrated that lamivudine and ribavirin downregulated the levels of 5-HTT mRNA and protein and 5-HT uptake.
IFN-α combined with ribavirin induces neuropsychiatric effects [6,8–10]. Ribavirin at higher doses (>800 mg/day) has been suggested to be one important factor in IFN-α-induced neuropsychiatric effects [47, 48]; in fact, depression and anxiety scores in patients with HCV were significantly higher in the subgroup treated with both IFN-α and ribavirin than those treated with IFN-α monotherapy [49, 50]. Currently, however, there is little evidence for the potential contribution of ribavirin to IFN-α-induced adverse psychiatric events, especially depression. Our study demonstrated that ribavirin at a nontoxic dosage (10−7−10−6m) reduced 5-HTT mRNA expression in immune cells for short periods of time (Table 2; Fig. 2). Although the addition of IFN-α and ribavirin to PBMC increased the levels of 5-HTT expression, they were significantly lower than those after adding IFN-α alone (Fig. 2). These results suggested that if 5-HTT were a trait marker of depression, ribavirin might not be a contributor to IFN-α-induced neuropsychiatric effects. Nevertheless, nonexclusive possibility occurred when a larger dose of ribavirin caused cell damage, led to subsiding 5-HT levels and hastened the onset of depressive symptoms. In addition, HBV has been found in extrahepatic sites, including PBMC, and the frequencies of HBV-positive cells are 50–500-fold higher in chronic than in acute hepatitis B . More clinical studies with large patient populations infected with HBV and treated with IFN-α alone or in combination with ribavirin are needed to clarify this issue.
Our study is the first to show that both T and B cells expressed high 5-HTT mRNA levels and that the highest expression levels occurred at 3 h. The protein levels of 5-HTT increased in T and B cells after 6 h of cell culturing and remained at that level until the 24th hour. These results might be related to cell growth, and they seem consistent with previous studies [52, 53], both of which concluded that 5-HTT expression was associated with the hypoxia-induced proliferation of pulmonary vascular smooth muscle cells and the proliferation of foetal heart cells exposed to exogenous serotonin at physiological doses. In our 24-h culturing of U937 monocytes, however, we found much lower 5-HTT mRNA expression and no significantly changeable 5-HTT proteins. Moreover, 5-HTT inhibitor fluoxetine (10−11−10−7m) caused a delay in the proliferation of Jurkat T cells but not in U937 cells during 24 h of culturing (data not shown). It appears that 5-HTT might play a role in cell proliferation depending on immune cell types and their levels of 5-HTT expression.
Lamivudine at 10−9−10−8m, which inhibited supernatant HBV DNA levels by about 50% , noticeably downregulated 5-HTT mRNA expression in immune cells. A reduction in 5-HT uptake in Jurkat T cells after treatment with lamivudine was also detected. Clinical studies  have also indicated that the occurrence of depressive disorders was two to eight times more common in patients treated with IFN-α alone than in patients treated with lamivudine. Although a number of hypotheses involving changes in central adrenergic, serotonergic, opioid and neuroendocrine pathways have been proposed , the mechanisms of IFN-α-induced neuropsychiatric effects are still poorly understood. Recently, several studies [54–56] demonstrated that cytokines downregulated 5-HT synthesis by lowering the availability of its precursor tryptophan through activation of the tryptophan-metabolizing enzyme indoleamine-2,3-dioxygenase. Tryptophan depletion was associated with triggering depression in patients undergoing IFN-α therapy for HCV infection . The present study showed differential regulation of 5-HTT mRNA expression, protein level and 5-HT uptake in vitro by lamivudine and IFN-α. This supports the hypothesis that an increase in 5-HTT in circulating blood lymphocytes after IFN-α therapy may decrease levels of circulating 5-HT, thereby causing depression.
This work was supported by grant NSC90–2314B-273–002 from the National Science Council, Taiwan and by Dr Huang's Dental Clinic, Taiwan. We also thank Bill Franke for editorial assistance.