Journal of Affective Disorders 191 (2016) 29–35

Contents lists available at ScienceDirect

Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad

Research report

Association between the serotonin transporter and cytokines: Implications for the pathophysiology of bipolar disorder Yuan-Hwa Chou a,n, Wen-Chi Hsieh a, Li-Chi Chen a, Jiing-Feng Lirng b, Shyh-Jen Wang c a

Departments of Psychiatry Taipei Veterans General Hospital and National Yang Ming University, Taipei, Taiwan Departments of Radiology Taipei Veterans General Hospital and National Yang Ming University, Taipei, Taiwan c Departments of Nuclear Medicine, Taipei Veterans General Hospital and National Yang Ming University, Taipei, Taiwan b

art ic l e i nf o

a b s t r a c t

Article history: Received 4 June 2015 Received in revised form 31 October 2015 Accepted 31 October 2015 Available online 10 November 2015

Background: Reduced brain serotonin transporter (SERT) has been demonstrated in bipolar disorder (BD). The aim of this study was to explore the potential role of cytokines on reduced SERT in BD. Methods: Twenty-eight BD type I patients and 28 age- and gender-matched healthy controls (HCs) were recruited. Single photon emission computed tomography with the radiotracer 123I ADAM was used for SERT imaging. Regions of interest included the midbrain, thalamus, putamen and caudate. Seven cytokines, including tumor necrosis factor-α (TNF-α), interferon-γ (IFN-γ), interleukin-1α (IL-1α), IL-1β, IL-4, IL-6 and IL-10, were measured using an enzyme linked immune-sorbent assay. Results: SERT availability in the midbrain and caudate was significantly lower in BD compared to HCs. IL-1β was significantly lower, whereas IL-10 was significantly higher in BD compared to HCs. Multiple linear regression analyses revealed that there were associations between cytokines, IL-1α, IL-1β, IL-6 and SERT availability in the midbrain but not in the thalamus, putamen and caudate. Furthermore, linear mixed effect analyses demonstrated that these associations were not different between HCs and BD. Conclusion: While many cytokines have been proposed to be important in the pathophysiology of BD, our results demonstrated that significant associations between cytokines and SERT availability may explain the role of cytokines in mood regulation. However, these associations were not different between HCs and BD, which imply the role of these cytokines is not specific for BD. & 2015 Elsevier B.V. All rights reserved.

Keywords: Bipolar disorder (BD) Serotonin transporter (SERT) Cytokines Single photon emission computed tomography (SPECT)

1. Introduction Serotonin is one of most extensively studied neurotransmitters in the brain. Serotonin affects our emotion and cognition (Elliott et al., 2011; Schmitt et al., 2006). Previous evidence has suggested the potential role of the serotonergic system in the etiology of mood disorders, including bipolar disorder (BD) and major depressive disorder (MDD) (Sobczak et al., 2002). The serotonin transporter (SERT) is a key regulator of central serotonergic activity, controlling the reuptake of serotonin and thereby terminating its action at the synapse. Importantly, brain imaging studies have demonstrated reduced SERT availability in the midbrain during depressive (Cannon et al., 2006; Oquendo et al., 2007) or euthymic state of BD (Chou et al., 2010). However, the underlying mechanism is unknown. n

Corresponding author. E-mail address: [email protected] (Y.-H. Chou).

http://dx.doi.org/10.1016/j.jad.2015.10.056 0165-0327/& 2015 Elsevier B.V. All rights reserved.

Cytokine abnormalities have been proposed to play a pivotal role in the pathophysiology of MDD (Maes, 1995). Until recently, a growing body of evidence, represented mainly by the finding of increased circulating levels of pro-inflammatory cytokines, suggests that immune-mediated mechanisms are related to the neurobiology of BD and its neuro-progression (Barbosa et al., 2014; Goldstein et al., 2009). One of the hypotheses was that pro-inflammatory cytokines stimulate the enzyme indoleamine 2,3-dioxygenase, which converts tryptophan into kynurenine (KYN), resulting in the reduction of the availability of tryptophan, the precursor for serotonin. Pro-inflammatory cytokines also enhance the activity kynurenine-3-monooxygenase, the enzyme that degrades KYN into 3-hydroxykynurenine, shifting the KYN pathway into the production of neurotoxic metabolites (Dantzer et al., 2011). Several meta-analyses have investigated immune function in BD with a particular focus on cytokine alterations (Modabbernia et al., 2013; Munkholm et al., 2013a, 2013b). Although these metaanalyses supported peripheral inflammatory alterations in BD,

30

Y.-H. Chou et al. / Journal of Affective Disorders 191 (2016) 29–35

differences in the samples examined make it difficult to generate definite conclusions. The interaction of cytokines and SERT availability has become the focus of recent studies. A number of pro-inflammatory cytokines, including TNF-α (Mossner et al., 1998a), interleukin-1β (Ramamoorthy et al., 1995), interferon-α, and interferon-γ (Morikawa et al., 1998), have been shown to up-regulate the SERT in cell models. In contrast, interleukin-4, an anti-inflammatory cytokine in the central nervous system, was reported to reduce the uptake of serotonin in a dose-dependent manner (Mossner et al., 2001). These findings suggest that a fine-tuned mechanism exists to communicate the state of the immune response in the central nervous system by differential modulation of the SERT via proinflammatory and anti-inflammatory cytokines (Baganz and Blakely, 2013). In the current study, we select the most frequently reported cytokines in literature and examine their interactions with SERT availability. Previous methods to image SERT in vivo had been hampered by the lack of a suitable radiotracer due to limitations in low signalto-noise ratio and low selectivity for SERT in different brain regions (Kuikka et al., 1995; Pirker et al., 2000). 123I-ADAM has been demonstrated to be a suitable radiotracer for imaging SERT in the brain regions of midbrain, thalamus, caudate and putamen but not for hippocampus (Chou et al., 2009b). By using 123I-ADAM, we have recently reported there was an association of IL-10 and thalamic SERT availability in euthymic BD (Hsu et al., 2014) but not in other brain regions, including the midbrain, putamen and caudate. This result suggested there might be regional effects of cytokines on SERT availability in different brain regions. Thus, using a within subject study design, this study aimed to examine the interaction between seven cytokines and SERT availability in different brain regions of BD. On the basis of our previous data, we hypothesized the interaction of cytokines and SERT availability is different across individual brain regions between BD and HCs.

2. Materials and methods This study was approved by the Human Ethical Committee of Taipei Veterans General Hospital. All subjects were referred from the Department of Psychiatry. Single photon emission computed tomography (SPECT) was performed at the Department of Nuclear Medicine. 2.1. Subject selection Twenty-eight patients with euthymic BD and 28 age- and gender-matched healthy controls (HCs) were recruited. All subjects provided their informed consent prior to entering the study. Each HC was interviewed by a trained psychiatrist using the MiniInternational Neuropsychiatric Interview (M.I.N.I.) to exclude the possibility of co-morbidity with major psychiatric illnesses, or history of substance abuse. BD patients fulfilled the following inclusion criteria: (i) diagnosis of BD type I according to the DSM-IVTR, (ii) under stable treatment in the euthymic state, and (iii) received treatment with valproic acid only. Past use of antidepressants, if any, should have been discontinued for at least one year. Patients who had a past history of suicide attempt were also excluded. Euthymic state was defined as Montgomery–Åsberg Depression Rating Scale (MADRS) scores of less than 10 and Young Mania Rating Scale (YMRS) scores of less than 7 within an eightweek consecutive period. In addition, all participants who had the following conditions were excluded: demonstrated any infectious disease in the previous two weeks, as well as allergies, dermatitis, fibromyalgia, autoimmune disorders, neuro-inflammatory disorders or any medical illness that would require pharmacological

treatment with glucocorticoids. Pregnant or breast feeding females were also excluded. Only non-smokers were recruited. All participants were of Taiwanese origin. 2.2. Radiochemistry 123 I-ADAM was synthesized and prepared under good manufacturing practice standards by the Institute of Nuclear Energy Research, Taiwan. Preparation of 123I-ADAM has been published elsewhere (Oya et al., 2000). Briefly, 100 mg of a tin precursor of ADAM was reacted with approximately 5.55 GBq (150 mCi) of Na123I in the presence of hydrogen peroxide in dilute acetic acid. The reaction was quenched 5 min later with NaHSO3. Following neutralization, the reaction solution was loaded onto an octyl cartridge (Accubond, J&W Scientific, Folsom, CA, USA) and eluted. An injection solution was prepared in 50% (v/v) ethanol. Purified 123 I-ADAM was eluted from the cartridge using absolute ethanol and further diluted with a 0.9% saline solution to a specific activity of greater than 12,000 Ci/mmole. The radiochemical purity of 123 I-ADAM was normally more than 90%, as determined by highpressure liquid chromatography on a Hamilton PRP-1 column (4.1
250 mm2; Hamilton Co., Reno, NV, USA). Elution was performed via an isocratic acetonitrile/5 mM dimethyl glutaric acid (pH 7.0) 90:10 solution at a flow rate of 1 ml/min.

2.3. SPECT measurement Each subject received one SPECT measurement for SERT imaging and one magnetic resonance image (MRI) to exclude the possibility of an organic lesion in the brain and for co-registration of the brain anatomical location. Thirty minutes after an oral 180 mg KClO4 solution was performed for the protection of thyroid SPECT, a bolus intravenous injection of 5 mCi (185 MBq) 123I-ADAM was performed using a two-head gamma camera system (E-Cam variable angle, Siemens Medical Systems Inc.) equipped with high resolution fanbeam collimators. The system resolution was 7.3 mm. Each subject received a SPECT static measurement 240–270 min after 123I-ADAM injection. All scanning data were collected in step-and-shoot mode at 3° intervals over 360°, while 30-s projection views were obtained from each camera head. The radius of rotation was fixed at 13.5 cm. The image matrix size was 128
128, and the pixel size was 3.9 mm. All images were obtained through a filtered back projection reconstruction algorithm with a Metz filter using a Nyquist frequency cutoff at 0.55 and an order of 30. Photon attenuation correction was performed using Chang’s method (μ ¼0.12 cm  1) (Chang, 1978), and no scatter correction was employed. To minimize variability in SERT availability introduced by the effect of menstrual cycle (Maswood et al., 1999), all premenopausal female subjects were measured in the follicular phase of their menstrual cycle by oral report. 2.4. MRI acquisition Each subject was subjected to T1-weighted MRI to confirm the absence of organic lesions in the brain and to co-register with SPECT images for the delineation of anatomical locations. MRI were obtained using a 1.5 T GE scanner Excite-II system (TR/TE¼8.54 ms/ 1.836; FOV¼260
260
1.5; Matrix¼256
256
124; NEX¼1; TI¼400 ms; Flip angle¼ 15; BW¼15.63). 2.5. Regions of interest (ROIs) defined ROIs were drawn manually on an individual transaxial MRI image and co-registered with the SPECT image using a pixel-wise modeling tool PMOD version 3.0 software (PMOD Group, Zurich, Switzerland), implemented on a personal computer. The ROIs were selected a priori, which contained SERT-rich regions of the midbrain, thalamus,

Y.-H. Chou et al. / Journal of Affective Disorders 191 (2016) 29–35

putamen, caudate and SERT-poor regions of the cerebellum. The cortical region of the cerebellum served as a reference tissue due to its negligible density of SERT (Kish et al., 2005). 2.6. Calculation of specific uptake ratio (SUR) of SERT binding SUR was calculated as the primary measured outcome. Calculation of SUR has been previously described (Chou et al., 2007, 2009a), and its validity has been established in kinetic analysis studies (Acton et al., 2001). Briefly, the SPECT system measures the total counts in an ROI, CROI. The cortical region of the cerebellum, which is devoid of SERT, can be used as a reference region for the representation of free and nonspecific binding, Cn, in the brain (Acton et al., 2001; Erlandsson et al., 2005). Thus, specific binding of 123I-ADAM, Cb(t), in the ROIs was calculated by subtracting the mean counts per pixel in the cerebellum (Cn) from those in the ROIs (CROI). The equation is defined as the following: Cb(t) ¼CROI(t)  Cn(t).

(1)

SUR was defined as an integrated period from 240 to 270 min, which has been suggested as a suitable scanning duration for 123 I-ADAM binding (Chou et al., 2009a). The equation is defined as the following: 240

SUR =

∫270 Cb (t) 240

.

∫270 Cn (t)

(2)

This ratio corresponds to the binding potential, which reflects the ratio of Bmax (SERT density) over KD (affinity) (Farde et al., 1989). 2.7. Measurement of plasma levels of cytokines Blood samples were obtained using vacutainer tubes and kept on ice, collected between 1:00 and 2:00 pm when the subject received an injection of 123I-ADAM. Samples were centrifuged at 3000 rpm for 30 min at 4 °C. Plasma was carefully collected and subsequently snap-frozen on dry ice and stored at  80 °C, until use for further analyses. Seven cytokines were measured in this study, including the pro-inflammatory cytokine, tumor necrosis factor-α (TNF-α), interferon-γ (IFN-γ), interleukin-1α (IL-1α), IL-1β, IL-4, IL-6 and the anti-inflammatory cytokine, IL-10. Enzyme linked immunosorbent assay (eBioscience Ltd., Hatfield, UK) was used to quantify the concentrations. All assays were performed in duplicate by the same operator using the recommended buffers, diluents, and substrates. The concentrations of the samples in each plate were calculated according to each standard curve and dilution factor. Generally, the intra-assay coefficient of variation was 3.2–6.0% and the inter-assay coefficient of variation was 5.2–10.0%. 2.8. Statistics Student's t-test was used for comparison of continuous variables in demographic data and measured variables, such as SERT availability and cytokines between two groups. Because there was an interaction of group and brain regions, multiple linear regression analyses considering groups and seven cytokines were independently performed to predict SERT availability in different brain regions. A mixed effect model was used to examine the differentially predictive effect of cytokines, which was selected on the basis of the results of multiple linear regression analyses on SERT availability between groups. Groups and cytokines were a fixed effect. Statistics were performed using SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA) on a personal computer. Significance

31

Table 1 Demographic data in healthy controls (HCs) and bipolar disorder (BD).

Sample size Sex (M/F) Age (years) MADRS scores YMRS scores Dose of valproic acid (mg)

HCs

BD

t

p

28 (9/19) 36.6 77.9 – –

28 (9/19) 36.6 7 8.1 4.2 7 3.8 1.3 72.5 1153.5 7 466.3

– – –0.03 – –

– – 0.974 – –

MADRS: Montgomery–Åsberg Depression Rating Scale, YMRS: Young Mania Rating Scale, Data are the mean 7 SD.

was defined as p o0.05. Data were shown as the mean 7 S.D.

3. Results Twenty-eight BD and 28 age- and gender-matched HCs completed the protocol. Demographic data is shown in Table 1. None of the participants complained of side effects after injection of 123 I-ADAM. The gender ratio was 9:19 (females to males) in both groups. The average age was 36.67 7.9 years in HCs and 36.6 78.1 years in BD. The average scores of MADRS and YMRS were 4.2 73.8 and 1.37 2.5, respectively, which fitted the inclusion criteria. The average dose of valproic acid was 1153.5 7466.31 mg/ day The comparison of average SERT availability between HCs and BD was: midbrain, (2.9170.64 vs. 2.3770.62), thalamus (0.9370.43 vs. 0.8470.43), putamen (0.9370.36 vs. 0.7870.28) and caudate (0.8170.40 vs. 0.6170.27). The SERT availability was significantly lower in the midbrain (t¼3.231, p¼0.002) and caudate (t¼2.221, p¼0.031) in BD compared to HCs (Table 2 and Fig. 1). Among the seven cytokines measured in this study, IL-1β was significantly lower (3.8370.18 vs. 3.9570.19, t¼2.487, p¼0.016), whereas IL-10 was significant higher (3.7371.07 vs. 3.0770.43, t¼  2.985, p¼ 0.005) in BD compared to HCs. The remaining cytokines did not show statistically significant differences between BD and HCs, including TNF-α (7.83713.51 vs. 10.19725.08, t¼0.433, p¼0.667), IFN-γ (5.2778.09 vs. 1.8671.93, t¼  1.886, p¼0.073), IL-1α (0.7670.39 vs. 0.8170.36, t¼0.487, p¼0.629), IL-4 (14.93718.38 vs. 13.06716.42, t¼  0.402, p¼0.689), and IL-6 (1.6571.68 vs. 2.1872.39, t¼0.951, p¼0.346) (Table 2 and Fig. 2). Multiple linear regression analyses considering groups and cytokines on SERT availability in different brain regions found only Table 2 Comparison of serotonin transporter (SERT) and cytokines between healthy controls (HCs) and bipolar disorder (BD). HCs SERT Midbrain Thalamus Putamen Caudate

Cytokines TNF-α IFN-γ IL-1α IL-1β IL-4 IL-6 IL-10

BD

2.91 70.64 0.93 70.43 0.93 70.36 0.81 70.40

2.37 70.62 0.84 70.43 0.78 70.28 0.61 70.27

10.19 725.08 1.86 71.93 0.81 70.36 3.95 70.19 13.06 716.42 2.18 72.39 3.0770.43

7.83 713.51 5.27 78.09 0.76 70.39 3.83 70.18 14.93 718.38 1.65 71.68 3.73 71.07

t

p

3.231 0.794 1.741 2.221

0.002 0.430 0.087 0.031

0.433  1.886 0.487 2.487  0.402 0.951  2.985

0.667 0.073 0.629 0.016 0.689 0.346 0.005

TNF-α: Tumor necrosis factor alpha, IFN-γ: Interferon gamma, IL: interleukin. Data are the mean7 S.D.

32

Y.-H. Chou et al. / Journal of Affective Disorders 191 (2016) 29–35

4

*

HCs

SUR

BD

4. Discussion 2

The major findings in this study were: (1) SERT availability in the midbrain and caudate was significantly lower in BD; (2) IL-1β was significant lower, whereas IL-10 was significantly higher in BD, and (3) there were associations between cytokines IL-1α, IL1β, IL-6 and SERT availability in the midbrain. However, those associations were not different between BD and HCs.

*

0

Midbrain

Thalamus

Putamen

Caudate

Fig. 1. Comparison of serotonin transporter (SERT) availability in different brain region between HCs and BD. 40

pg / ml

groups in the predictive effect of IL-1α (df¼1.45, F¼ 0.08, p¼ 0.785), IL-1β (df¼1.45, F¼ 0.30, p¼ 0.588) and IL-6 (df ¼1.45, F¼0.14, p¼ 0.706) on SERT availability (supplement file).

20

*

*

4.1. Imaging SERT availability in BD Compared to our previous finding (Chou et al., 2010), the current study using a second patient group demonstrated a similar result of reduced SERT availability in the midbrain. Furthermore, SERT availability was measured in additional brain regions including the thalamus, putamen and caudate. We found reduced SERT availability not only in the midbrain but also in the caudate. Considering previous SERT imaging studies, a reduction in SERT availability has been consistently reported in the midbrain of patients in a depressive state of BD (Cannon et al., 2006, 2007; Oquendo et al., 2007), however, changes in SERT availability in other brain regions have not as congruent. The reduction of SERT availability in the midbrain may play a pivotal role in the pathophysiology of BD and requires further exploration. 4.2. Cytokines in euthymic BD

0

TNF-α

IFN-γ

IL-1α

IL-1β

IL-4

IL-6

IL-10

Fig. 2. Comparison of cytokines in plasma concentration between HCs and BD.

the model in the midbrain region was significant (Adjusted R2 ¼0.26, F¼ 3.11, p ¼0.008), but not in the thalamus (Adjusted R2 ¼0.12, F¼1.81, p ¼0.104), putamen (Adjusted R2 ¼0.08, F¼1.53, p ¼0.176) and caudate (Adjusted R2 ¼0.05, F¼ 1.31, p ¼0.265) (Table 3). Reduced SERT availability in the midbrain was significantly predicted by groups (β ¼  0.39, t¼  2.85, p ¼0.007), IL-1α (β ¼0.32, t¼2.44, p ¼0.019), IL-1β (β ¼  0.45, t¼  2.20, p ¼0.033) and IL-6 (β ¼0.68, t ¼2.51, p ¼0.016) but not TNF-α (β ¼ 0.01, t¼0.08, p¼ 0.937), IFN-γ (β ¼  1.49, t¼  1.78, p ¼0.083), IL-4 (β ¼1.26, t ¼1.64, p ¼0.108) and IL-10 (β ¼ 0.25, t ¼1.47, p ¼0.150) (Table 3). A mixed effect model examined the differentially predictive effect of cytokines on SERT availability in the midbrain between groups showed there was no significant difference between

The study of cytokines in BD is complicated due to the diversity and phase-dependent changes of cytokines. We recently reported the plasma concentration of TNF-α and IL-10 in the euthymic state of BD (17). Compared to our previous study, five additional cytokines, including IFN-γ, IL-1α, IL-1β, IL-4 and IL-6, were measured. The significant findings were that IL-1β was decreased, whereas IL-10 was increased in BD compared to HCs. The importance of IL-1β in the pathophysiology of BD has been reported in several studies. In an earlier study, Papiol et al. demonstrated that gray matter deficits in BD were associated with genetic variability in the IL-1β gene (2q13) (Papiol et al., 2008). Using postmortem brain tissues of BD, Rao et al. showed increased mRNA levels of IL-1β (Rao et al., 2010), further supporting the hypothesis of an activated IL-1 receptor cascade in BD. However, clinical studies have exhibited inconsistent results. Two studies measured IL-1β either in serum (Drexhage et al., 2011) or cerebrospinal fluid (Soderlund et al., 2011) found an increased

Table 3 Multiple linear regression analyses showed prediction of different cytokines on the serotonin transporter (SERT) in different brain regions.

Adjusted R2 F Sig. Group TNF-α IFN-γ IL-1α IL-1β IL-4 IL-6 IL-10

Midbrain

Thalamus

Putamen

Caudate

0.26 3.11 0.008 Beta  0.39 0.01  1.49 0.32  0.45 1.26 0.68 0.25

0.12 1.81 0.104 Beta  0.25  0.44  1.35 0.18  0.19 1.23 0.47 0.20

0.08 1.53 0.176 Beta  0.20  0.06  1.19 0.21  0.15 1.31 0.24 -0.05

0.05 1.31 0.265 Beta  0.36  0.20  0.90 0.08  0.31 0.93 0.24 0.12

t  2.85 0.08  1.78 2.44  2.20 1.64 2.51 1.47

Sig. 0.007 0.937 0.083 0.019 0.033 0.108 0.016 0.150

t  1.68  2.40  1.48 1.27  0.85 1.47 1.60 1.06

TNF-α: Tumor necrosis factor alpha, IFN-γ: Interferon gamma, IL: interleukin.

Sig. 0.102 0.021 0.146 0.211 0.398 0.149 0.117 0.297

t  1.27  0.33  1.27 1.46  0.67 1.54 0.81 -0.26

Sig. 0.210 0.746 0.210 0.153 0.505 0.132 0.424 0.794

T  2.29  1.03  0.95 0.51  1.35 1.07 0.77 0.61

Sig. 0.027 0.311 0.349 0.613 0.186 0.291 0.446 0.549

Y.-H. Chou et al. / Journal of Affective Disorders 191 (2016) 29–35

concentration of IL-1β in BD. Two other studies showed either decreased (Knijff et al., 2007) or no difference (Remlinger-Molenda et al., 2012) of IL-1β in BD compared with HCs. The discrepancy of these clinical results may be attributed to the complexity of the phases of BD. Our results showed decreased IL-1β in euthymic BD. Because IL-1β is a pro-inflammatory cytokine, a straightforward explanation might be that the disease process becomes less inflammatory at this stage. However, it should be noted that Berk et al. recently proposed the nature of BD implies the presence of an active neuro-progression process that is considered to be at least partly mediated by inflammation (Berk et al., 2014). Therefore, an alternative explanation of our results may be the medication effect because all of the patients were taking valproic acid. If this is the case, our results can be supported by a recent animal study that demonstrated a reduction of IL-1β gene expression after valproic acid treatment (Zhang et al., 2012). Although it has been suggested to be relevant in depression, the role of IL-10 in BD is still ambiguous (Maes, 1993; Roque et al., 2009). Two previous studies have demonstrated that IL-10 did not differ between euthymic BD and HCs (Brietzke et al., 2009; Guloksuz et al., 2010). However, Kunz et al. (2011) recently reported that IL-10 was lower in HCs compared to euthymic BD, which is consistent with our finding. It has been reported that IL-10 suppresses various aspects of the immune response, including the synthesis of TNF-α, IL-1β, and IL-8, in an ischemic model (Cassatella et al., 1993; Hess et al., 1997). Considering the level of IL-1β in the euthymic state, higher levels of IL-10 in our study might indicate a counteracting effect of IL-10 and IL-1β. However, this idea is limited to the cross-sectional study design and warrants further study. 4.3. Interaction of cytokines on SERT availability in different brain regions The most prominent finding in this study was that three cytokines, IL-1α, IL-1β and IL-6, were associated with SERT availability in the midbrain of BD but not in other brain regions. Many previous studies have discussed the interaction of cytokines and serotonergic system in mood disorders, but most studies have focused on MDD (Myint and Kim, 2003). It is tempting to speculate that cytokine regulation of SERT availability may be via two pathways. First, cytokines activate the indoleamine-2,3-dioxygenase (IDO) system, which results in changes in the serotonin metabolism pathway. Subsequently, SERT availability might be altered due to changes in serotonin concentration. Second, it has been reported that cytokines could directly regulate SERT transcriptional activity. For example, IL-1β has been reported to elevate SERT activity via an increased production of SERT mRNA levels in the human choriocarcinoma cell line (Ramamoorthy et al., 1995). Although it is not possible to differentiate which pathway is more prominent regarding the interaction of IL-1β and SERT availability in this study, the negative association of IL-1β and SERT availability (β-value was 0.45) may represent an overall outcome between the interaction of IL-1β and SERT availability. To the best of our knowledge, this study is the first to describe the interaction of IL-1α and SERT availability in vivo. Unlike most cytokines which are rapidly up-regulated upon stimulation, IL-1α is present constitutively in resting cells under homeostatic conditions (Rider et al., 2013). Specifically, IL-1α is released and activates both innate and adaptive immunity when the cell is damaged (Dinarello, 2009). In an animal study, it has been demonstrated that IL-1α could induce serotonin release in the hippocampus (Broderick, 2002). The β-value of IL-1α was 0.33, which reflects a positive association between IL-1α and SERT availability. The opposing effects of IL-1α and IL-1β on SERT availability require further studies. Currently, studies approaching the interaction of IL-6 and SERT availability are limited. An earlier study demonstrated that SERT

33

function could be enhanced by TNF-α but not IL-6 (Mossner et al., 1998b). However, Yoshimura et al. recently showed that the plasma levels of IL-6 were significantly higher in serotonin transporter reuptake inhibitor responders compared to non-responders, which indicated a specific association of plasma IL-6 levels and SERT function (Yoshimura et al., 2013). It is worthy to mention that we recently published an article which demonstrated an association between thalamic SERT availability and IL-10 in euthymic BD (Hsu et al., 2014). Notably, we did not replicate a similar finding in this study. One of the possible explanations might be sample bias. There were ten overlapping cases between these two studies and the larger sample size in the new study was collected. Therefore, this result should still be viewed as preliminary. Although this study is the first to demonstrate the association between cytokines on SERT availability in vivo, it is not possible to explain the causal effect between cytokines and SERT availability because of the limitation of cross-sectional study design. Nevertheless, it seems odd that the association between cytokines and SERT availability was not different between HCs and BD. It is most likely that the association between cytokines and SERT availability was confounded by other factors such as stress hormone or neurotropic factors. Therefore, the underpinning of reduced SERT availability in BD needs to be further studied.

5. Limitations Several limitations in this study should be considered. First, cytokines could affect SERT expression directly activate the IDO system, which could indirectly result in changes in SERT. Thus, our results cannot clearly differentiae the role of each of the cytokines in BD. Second, in this study, the cytokine levels were measured in plasma, which may not reflect the levels in different brain regions. Srivastava et al. (Srivastava et al., 2012) recently found in a rat model of Japanese encephalitis that different cytokines may exist in different brain regions after infection, which may explain why the interaction of cytokines on SERT can only be found in the midbrain. Third, compared with previous studies using PET tracers, such as 11C-DASB, 123I-ADAM may not be as good as these tracers in measuring SERT availability in brain regions outside the midbrain. However, image analysis was performed in our study using individual MRI with PMOD software co-registration, which minimized bias (Hsu et al., 2011). Fourth, all patients in this study were on medication due to ethical considerations, and thus the effect of the medication on cytokines cannot be completely excluded. In particular, it has been reported that valproic acid could affect either brain structure (Brambilla et al., 2009) or plasma levels of cytokines (Himmerich et al., 2013). One of the strategies in our study design was within subjects and aimed to maintain the patients as homogenous as possible. Therefore, in this study all patients were only taking valproic acid. Fifth, a large inter-individual variability of cytokines was noted in this study. Some of them such as IL-1β and IL-10 even approached the inter-assay variability of approximate 5–6%. A recent published article that discussed the natural variability of circulating levels of cytokines in patients with congestive heart failure suggested that the sample size needed to show a statistically significant change in the circulating level of a given cytokine will vary depending on the specific cytokine that is being measured, as well as the time period over which that cytokine is being assayed (Dibbs et al., 1999) Therefore, although the statistic was significant in this study, a well-prepared sample with more specific targets is needed to replicate this result. Lastly, small sample size in this study was a weak point. However, most of previous PET or SPECT receptor image studies agreed that the suitable sample size was 20–30

34

Y.-H. Chou et al. / Journal of Affective Disorders 191 (2016) 29–35

(Cannon et al., 2006, 2007; Hahn et al., 2014; Miller et al., 2013). Meanwhile, when we used either non-parametric statistics or logtransformation of cytokine the data showed the similar results (except the changes of IL-6). Therefore, although the sample size in our study was acceptable, a larger sample study is warranted. In conclusion, as many cytokines have been proposed to be important in the pathophysiology of BD, our results demonstrated a significant association between cytokines and SERT availability, which may explain the role of cytokines in mood regulation. However, because this association was not different between HCs and BD, it suggests an underlying mechanism in BD beyond these cytokines.

References Acton, P.D., Choi, S.R., Hou, C., Plossl, K., Kung, H.F., 2001. Quantification of serotonin transporters in nonhuman primates using [(123)I]ADAM and SPECT. J. Nucl. Med. 42, 1556–1562. Baganz, N.L., Blakely, R.D., 2013. A dialogue between the immune system and brain, spoken in the language of serotonin. ACS Chem. Neurosci. 4, 48–63. Barbosa, I.G., Bauer, M.E., Machado-Vieira, R., Teixeira, A.L., 2014. Cytokines in bipolar disorder: paving the way for neuroprogression. Neural Plast. 2014, 360481. Berk, M., Berk, L., Dodd, S., Cotton, S., Macneil, C., Daglas, R., Conus, P., Bechdolf, A., Moylan, S., Malhi, G.S., 2014. Stage managing bipolar disorder. Bipolar Disord. 16, 471–477. Brambilla, P., Bellani, M., Yeh, P.H., Soares, J.C., 2009. Myelination in bipolar patients and the effects of mood stabilizers on brain anatomy. Curr. Pharm. Des. 15, 2632–2636. Brietzke, E., Stertz, L., Fernandes, B.S., Kauer-Sant’anna, M., Mascarenhas, M., Escosteguy Vargas, A., Chies, J.A., Kapczinski, F., 2009. Comparison of cytokine levels in depressed, manic and euthymic patients with bipolar disorder. J. Affect. Disord. 116, 214–217. Broderick, P.A., 2002. Interleukin 1alpha alters hippocampal serotonin and norepinephrine release during open-field behavior in Sprague-Dawley animals: differences from the Fawn-Hooded animal model of depression. Prog. Neuropsychopharmacol. Biol. Psychiatry 26, 1355–1372. Cannon, D.M., Ichise, M., Fromm, S.J., Nugent, A.C., Rollis, D., Gandhi, S.K., Klaver, J. M., Charney, D.S., Manji, H.K., Drevets, W.C., 2006. Serotonin transporter binding in bipolar disorder assessed using [11C]DASB and positron emission tomography. Biol. Psychiatry 60, 207–217. Cannon, D.M., Ichise, M., Rollis, D., Klaver, J.M., Gandhi, S.K., Charney, D.S., Manji, H. K., Drevets, W.C., 2007. Elevated serotonin transporter binding in major depressive disorder assessed using positron emission tomography and [11C]DASB; comparison with bipolar disorder. Biol. Psychiatry 62, 870–877. Cassatella, M.A., Meda, L., Bonora, S., Ceska, M., Constantin, G., 1993. Interleukin 10 (IL-10) inhibits the release of proinflammatory cytokines from human polymorphonuclear leukocytes. Evidence for an autocrine role of tumor necrosis factor and IL-1 beta in mediating the production of IL-8 triggered by lipopolysaccharide. J. Exp. Med. 178, 2207–2211. Chang, L.T., 1978. A method for attenuation correction in radionuclide computed tomography. IEEE Trans. Nucl. Sci. 35, 638–643. Chou, Y.H., Huang, W.S., Su, T.P., Lu, R.B., Wan, F.J., Fu, Y.K., 2007. Dopamine transporters and cognitive function in methamphetamine abuser after a short abstinence: a SPECT study. Eur. Neuropsychopharmacol. 17, 46–52. Chou, Y.H., Wang, S.J., Lin, C.L., Mao, W.C., Lee, S.M., Liao, M.H., 2010. Decreased brain serotonin transporter binding in the euthymic state of bipolar I but not bipolar II disorder: a SPECT study. Bipolar Disord. 12, 312–318. Chou, Y.H., Yang, B.H., Chung, M.Y., Chen, S.P., Su, T.P., Chen, C.C., Wang, S.J., 2009a. Imaging the serotonin transporter using (123)I-ADAM in the human brain. Psychiatry Res. Neuroimage 172, 38–43. Chou, Y.H., Yang, B.H., Chung, M.Y., Chen, S.P., Su, T.P., Chen, C.C., Wang, S.J., 2009b. Imaging the serotonin transporter using (123)I-ADAM in the human brain. Psychiatry Res. 172, 38–43. Dantzer, R., O’Connor, J.C., Lawson, M.A., Kelley, K.W., 2011. Inflammation-associated depression: from serotonin to kynurenine. Psychoneuroendocrinology 36, 426–436. Dibbs, Z., Thornby, J., White, B.G., Mann, D.L., 1999. Natural variability of circulating levels of cytokines and cytokine receptors in patients with heart failure: implications for clinical trials. J. Am. Coll. Cardiol. 33, 1935–1942. Dinarello, C.A., 2009. Immunological and inflammatory functions of the interleukin-1 family. Annu. Rev. Immunol. 27, 519–550. Drexhage, R.C., Hoogenboezem, T.H., Versnel, M.A., Berghout, A., Nolen, W.A., Drexhage, H.A., 2011. The activation of monocyte and T cell networks in patients with bipolar disorder. Brain Behav. Immun. 25, 1206–1213. Elliott, R., Zahn, R., Deakin, J.F.W., Anderson, I.M., 2011. Affective cognition and its disruption in mood disorders. Neuropsychopharmacology 36, 153–182. Erlandsson, K., Sivananthan, T., Lui, D., Spezzi, A., Townsend, C.E., Mu, S., Lucas, R., Warrington, S., Ell, P.J., 2005. Measuring SSRI occupancy of SERT using the novel tracer [123I]ADAM: a SPECT validation study. Eur. J. Nucl. Med. Mol. Imaging

32; , pp. 1329–1336. Farde, L., Eriksson, L., Blomquist, G., Halldin, C., 1989. Kinetic analysis of central [11C]raclopride binding to D2-dopamine receptors studied by PET–a comparison to the equilibrium analysis. J. Cereb. Blood Flow Metab. 9, 696–708. Goldstein, B.I., Kemp, D.E., Soczynska, J.K., McIntyre, R.S., 2009. Inflammation and the phenomenology, pathophysiology, comorbidity, and treatment of bipolar disorder: a systematic review of the literature. J. Clin. Psychiatry 70, 1078–1090. Guloksuz, S., Cetin, E.A., Cetin, T., Deniz, G., Oral, E.T., Nutt, D.J., 2010. Cytokine levels in euthymic bipolar patients. J. Affect. Disord. 126, 458–462. Hahn, A., Haeusler, D., Kraus, C., Hoflich, A.S., Kranz, G.S., Baldinger, P., Savli, M., Mitterhauser, M., Wadsak, W., Karanikas, G., Kasper, S., Lanzenberger, R., 2014. Attenuated serotonin transporter association between dorsal raphe and ventral striatum in major depression. Hum. Brain Mapp. 35, 3857–3866. Hess, P.J., Seeger, J.M., Huber, T.S., Welborn, M.B., Martin, T.D., Harward, T.R., Duschek, S., Edwards, P.D., Solorzano, C.C., Copeland, E.M., Moldawer, L.L., 1997. Exogenously administered interleukin-10 decreases pulmonary neutrophil infiltration in a tumor necrosis factor-dependent murine model of acute visceral ischemia. J. Vasc. Surg. 26, 113–118. Himmerich, H., Bartsch, S., Hamer, H., Mergl, R., Schonherr, J., Petersein, C., Munzer, A., Kirkby, K.C., Bauer, K., Sack, U., 2013. Impact of mood stabilizers and antiepileptic drugs on cytokine production in-vitro. J. Psychiatr. Res. 47, 1751–1759. Hsu, J.-W., Wang, S.-J., Lin, C.-L., Hsieh, W.-C., Lirng, J.-F., Shen, Y.-C., Liao, M.-H., Chou, Y.-H., 2011. Short term vs. long term test-retest reproducibility of 123I ADAM for the binding of serotonin transporters in the human brain. Psychiatry Res. 194, 224–229. Hsu, J.W., Lirng, J.F., Wang, S.J., Lin, C.L., Yang, K.C., Liao, M.H., Chou, Y.H., 2014. Association of thalamic serotonin transporter and interleukin-10 in bipolar I disorder: a SPECT study. Bipolar Disord. 16, 241–248. Kish, S.J., Furukawa, Y., Chang, L.J., Tong, J., Ginovart, N., Wilson, A., Houle, S., Meyer, J.H., 2005. Regional distribution of serotonin transporter protein in postmortem human brain: is the cerebellum a SERT-free brain region? Nucl. Med. Biol. 32, 123–128. Knijff, E.M., Breunis, M.N., Kupka, R.W., de Wit, H.J., Ruwhof, C., Akkerhuis, G.W., Nolen, W.A., Drexhage, H.A., 2007. An imbalance in the production of IL-1beta and IL-6 by monocytes of bipolar patients: restoration by lithium treatment. Bipolar Disord. 9, 743–753. Kuikka, J.T., Tiihonen, J., Bergstrom, K.A., Karhu, J., Hartikainen, P., Viinamaki, H., Lansimies, E., Lehtonen, J., Hakola, P., 1995. Imaging of serotonin and dopamine transporters in the living human brain. Eur. J. Nucl. Med. 22, 346–350. Kunz, M., Cereser, K.M., Goi, P.D., Fries, G.R., Teixeira, A.L., Fernandes, B.S., Belmontede-Abreu, P.S., Kauer-Sant’Anna, M., Kapczinski, F., Gama, C.S., 2011. Serum levels of IL-6, IL-10 and TNF-alpha in patients with bipolar disorder and schizophrenia: differences in pro- and anti-inflammatory balance. Rev. Bras. Psiquiatr. 33, 268–274. Maes, M., 1993. A review on the acute phase response in major depression. Rev. Neurosci. 4, 407–416. Maes, M., 1995. Evidence for an immune response in major depression: a review and hypothesis. Prog. Neuropsychopharmacol. Biol. Psychiatry 19, 11–38. Maswood, S., Truitt, W., Hotema, M., Caldarola-Pastuszka, M., Uphouse, L., 1999. Estrous cycle modulation of extracellular serotonin in mediobasal hypothalamus: role of the serotonin transporter and terminal autoreceptors. Brain Res. 831, 146–154. Miller, J.M., Hesselgrave, N., Ogden, R.T., Sullivan, G.M., Oquendo, M.A., Mann, J.J., Parsey, R.V., 2013. Positron emission tomography quantification of serotonin transporter in suicide attempters with major depressive disorder. Biol. Psychiatry 74, 287–295. Modabbernia, A., Taslimi, S., Brietzke, E., Ashrafi, M., 2013. Cytokine alterations in bipolar disorder: a meta-analysis of 30 studies. Biol. Psychiatry 74, 15–25. Morikawa, O., Sakai, N., Obara, H., Saito, N., 1998. Effects of interferon-alpha, interferon-gamma and cAMP on the transcriptional regulation of the serotonin transporter. Eur. J. Pharmacol. 349, 317–324. Mossner, R., Daniel, S., Schmitt, A., Albert, D., Lesch, K.P., 2001. Modulation of serotonin transporter function by interleukin-4. Life Sci. 68, 873–880. Mossner, R., Heils, A., Stober, G., Okladnova, O., Daniel, S., Lesch, K.P., 1998a. Enhancement of serotonin transporter function by tumor necrosis factor alpha but not by interleukin-6. Neurochem. Int. 33, 251–254. Mossner, R., Heils, A., Stober, G., Okladnova, O., Daniel, S., Lesch, K.P., 1998b. Enhancement of serotonin transporter function by tumor necrosis factor alpha but not by interleukin-6. Neurochem. Int. 33, 251–254. Munkholm, K., Brauner, J.V., Kessing, L.V., Vinberg, M., 2013a. Cytokines in bipolar disorder vs. healthy control subjects: a systematic review and meta-analysis. J. Psychiatr. Res. 47, 1119–1133. Munkholm, K., Vinberg, M., Vedel Kessing, L., 2013b. Cytokines in bipolar disorder: a systematic review and meta-analysis. J. Affect. Disord. 144, 16–27. Myint, A.M., Kim, Y.K., 2003. Cytokine-serotonin interaction through IDO: a neurodegeneration hypothesis of depression. Med. Hypotheses 61, 519–525. Oquendo, M.A., Hastings, R.S., Huang, Y.Y., Simpson, N., Ogden, R.T., Hu, X.Z., Goldman, D., Arango, V., Van Heertum, R.L., Mann, J.J., Parsey, R.V., 2007. Brain serotonin transporter binding in depressed patients with bipolar disorder using positron emission tomography. Arch. Gen. Psychiatry 64, 201–208. Oya, S., Choi, S.R., Hou, C., Mu, M., Kung, M.P., Acton, P.D., Siciliano, M., Kung, H.F., 2000. 2-((2-((dimethylamino)methyl)phenyl)thio)-5-iodophenylamine (ADAM): an improved serotonin transporter ligand. Nucl. Med. Biol. 27, 249–254. Papiol, S., Molina, V., Desco, M., Rosa, A., Reig, S., Sanz, J., Palomo, T., Fananas, L., 2008. Gray matter deficits in bipolar disorder are associated with genetic

Y.-H. Chou et al. / Journal of Affective Disorders 191 (2016) 29–35

variability at interleukin-1 beta gene (2q13). Genes Brain Behav. 7, 796–801. Pirker, W., Asenbaum, S., Hauk, M., Kandlhofer, S., Tauscher, J., Willeit, M., Neumeister, A., Praschak-Rieder, N., Angelberger, P., Brucke, T., 2000. Imaging serotonin and dopamine transporters with 123I-beta-CIT SPECT: binding kinetics and effects of normal aging. J. Nucl. Med. 41, 36–44. Ramamoorthy, S., Ramamoorthy, J.D., Prasad, P.D., Bhat, G.K., Mahesh, V.B., Leibach, F.H., Ganapathy, V., 1995. Regulation of the human serotonin transporter by interleukin-1 beta. Biochem. Biophys. Res. Commun. 216, 560–567. Rao, J.S., Harry, G.J., Rapoport, S.I., Kim, H.W., 2010. Increased excitotoxicity and neuroinflammatory markers in postmortem frontal cortex from bipolar disorder patients. Mol. Psychiatry 15, 384–392. Remlinger-Molenda, A., Wojciak, P., Michalak, M., Karczewski, J., Rybakowski, J.K., 2012. Selected cytokine profiles during remission in bipolar patients. Neuropsychobiology 66, 193–198. Rider, P., Carmi, Y., Voronov, E., Apte, R.N., 2013. Interleukin-1alpha. Semin. Immunol. 25, 430–438. Roque, S., Correia-Neves, M., Mesquita, A.R., Palha, J.A., Sousa, N., 2009. Interleukin10: a key cytokine in depression? Cardiovasc. Psychiatry Neurol. 2009, 187894.

35

Schmitt, J.A.J., Wingen, M., Ramaekers, J.G., Evers, E.A.T., Riedel, W.J., 2006. Serotonin and human cognitive performance. Curr. Pharm. Des. 12, 2473–2486. Sobczak, S., Honig, A., van Duinen, M.A., Riedel, W.J., 2002. Serotonergic dysregulation in bipolar disorders: a literature review of serotonergic challenge studies. Bipolar Disord. 4, 347–356. Soderlund, J., Olsson, S.K., Samuelsson, M., Walther-Jallow, L., Johansson, C., Erhardt, S., Landen, M., Engberg, G., 2011. Elevation of cerebrospinal fluid interleukin1ss in bipolar disorder. J. Psychiatry Neurosci.: Jpn. 36, 114–118. Srivastava, R., Kalita, J., Khan, M.Y., Misra, U.K., 2012. Status of proinflammatory and anti-inflammatory cytokines in different brain regions of a rat model of Japanese encephalitis. Inflamm. Res. 61, 381–389. Yoshimura, R., Hori, H., Ikenouchi-Sugita, A., Umene-Nakano, W., Katsuki, A., Atake, K., Nakamura, J., 2013. Plasma levels of interleukin-6 and selective serotonin reuptake inhibitor response in patients with major depressive disorder. Hum. Psychopharmacol. 28, 466–470. Zhang, Z., Zhang, Z.Y., Wu, Y., Schluesener, H.J., 2012. Valproic acid ameliorates inflammation in experimental autoimmune encephalomyelitis rats. Neuroscience 221, 140–150.