- Email: [email protected]
Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia
Biochemical and Biophysical Research Communications xxx (2018) 1e6
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia Takashi Ishii a, c, Kotaro Hattori a, b, Tomoko Miyakawa a, b, Kentaro Watanabe a, Shinsuke Hidese a, Daimei Sasayama a, g, Miho Ota a, Toshiya Teraishi a, Hiroaki Hori a, e, Sumiko Yoshida b, f, Akihiko Nunomura c, Kazuyuki Nakagome d, Hiroshi Kunugi a, * a
Department of Mental Disorder Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan Medical Genome Center, National Center of Neurology and Psychiatry, Tokyo, Japan Department of Neuropsychiatry, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Yamanashi, Japan d National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan e Department of Adult Mental Health, National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan f National Center of Neurology and Psychiatry Hospital, Tokyo, Japan g Department of Psychiatry, Shinshu University School of Medicine, Matsumoto, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 January 2018 Accepted 15 February 2018 Available online xxx
Inflammation has been implicated in a variety of psychiatric disorders. We aimed to determine whether levels of complement C5 protein in the cerebrospinal fluid (CSF), which may reflect activation of the complement system in the brain, are altered in patients with major psychiatric disorders. Additionally, we examined possible associations of CSF C5 levels with clinical variables. Subjects comprised 89 patients with major depressive disorder (MDD), 66 patients with bipolar disorder (BPD), 96 patients with schizophrenia, and 117 healthy controls, matched for age, sex, and ethnicity (Japanese). Diagnosis was made according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, criteria. CSF C5 levels were measured by enzyme-linked immunosorbent assay. CSF C5 levels were significantly increased in the patients with MDD (p < 0.001) and in the patients with schizophrenia (p ¼ 0.001), compared with the healthy controls. The rate of individuals with an “abnormally high C5 level” (i.e., above the 95th percentile value of the control subjects) was significantly increased in all psychiatric groups, relative to the control group (all p < 0.01). Older age, male sex, and greater body mass index tended to associate with higher C5 levels. There was a significantly positive correlation between C5 levels and chlorpromazine-equivalent dose in the patients with schizophrenia. Thus, we found, for the first time, elevated C5 levels in the CSF of patients with major psychiatric disorders. Our results suggest that the activated complement system may contribute to neurological pathogenesis in a portion of patients with major psychiatric disorders. © 2018 Elsevier Inc. All rights reserved.
Keywords: Cerebrospinal fluid Complement C5 Schizophrenia Bipolar disorder Major depressive disorder Body mass index
1. Introduction Current diagnostic systems for mental disorders rely upon the presentation of signs and symptoms, and thus do not adequately reflect the relevant neurobiological and behavioral systems [1]. There is an urgent need to elucidate molecular basis of mental disorders and to develop biomarkers that allow classification of psychiatric disorders based on the pathophysiology. In this context,
* Corresponding author. 4-1-1, Ogawahigashi, Kodaira, Tokyo, 187-8502, Japan. E-mail address: [email protected] (H. Kunugi).
investigation of the cerebrospinal fluid (CSF) might be one of the most promising approaches to detect molecules that contribute to the pathophysiology of psychiatric diseases within the brain, as CSF is in contact with the brain interstitial fluid over the surfaces of ventricles, brain, and spinal cord; therefore, molecules released from brain cells can directly diffuse into the CSF. Indeed, CSF biomarkers have already been established for some neurological diseases, such as tau, phosphorylated tau, and b-amyloid proteins, which diagnose Alzheimer's disease with a high (~90%) sensitivity and specificity [2]. There is a growing amount of evidence that supports increased levels of pro-inflammatory cytokines in the blood of patients with
https://doi.org/10.1016/j.bbrc.2018.02.131 0006-291X/© 2018 Elsevier Inc. All rights reserved.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
2
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
2. Material and methods Abbreviations 2.1. Participants BBB blood-brain barrier BMI body mass index BPD bipolar disorder CNS central nervous system CPeq chlorpromazine-equivalent dose CSF cerebrospinal fluid ELISA enzyme-linked immunosorbent assay GRID-HAMD-17 GRID-Hamilton Depression Rating Scale, 17item version IL-6 interleukin-6 IMIeq imipramine-equivalent dose MDD major depressive disorder MINI Mini International Neuropsychiatric Interview PANSS Positive and Negative Syndrome Scale YMRS Young Mania Rating Scale
schizophrenia and major depressive disorder (MDD) [3,4]. We previously reported increased interleukin-6 (IL-6) levels in the blood of patients with schizophrenia [4]. Contrary to the traditional view that the brain is an immunologically privileged site that is shielded behind the blood-brain barrier (BBB), studies in the past 20 years have uncovered complex interactions among the immune system, systemic inflammation, and brain functions, which can lead to changes in mood, cognition, and behavior [5]. As stated above, we have reported increased IL-6 in the CSF of patients with schizophrenia, as well as in patients with MDD, relative to healthy controls [6]; these findings were further supported by a recent meta-analysis [7]. Additionally, we have reported increased CSF fibrinogen levels in patients with MDD [2]. Fibrinogen is a coagulation factor which activates microglia, thereby inducing neuroinflammation and negatively affecting the brain [8]. A major aspect of the innate immune system is the complement system. In general, complement activation occurs via four pathways: the classical, alternative, mannan-binding lectin, and extrinsic protease pathways [9,10]. These pathways converge upon the cleavage of complement C5 into complement fragments C5a and C5b by C5 convertase, which is the final step of the complement activation. C5a is the most potent and stable of the anaphylatoxins, while C5b is an initial element in the formation of the membrane attack complex. Notably, the complement system also plays a role in synapse elimination/pruning, which is essential to the development of a precise neuronal network [11,12]. A recent study on adult Swedish twins enriched for schizophrenia reported that peripheral messenger RNA (mRNA) expression levels of two complement genes (C5, SERPING1) in peripheral blood mononuclear cells were associated with decreased superior frontal cortical thickness, which may be involved in the pathogenesis of schizophrenia [13]. A number of studies, including ours, have reported structural and functional changes in volume, gray matter, white matter, and functional activity, within the frontal lobe of schizophrenia patients [14,15]. Since C5 is involved in the final step of the activation of the complement system, we hypothesized that C5 levels would be altered in the brain of patients with schizophrenia. The aim of the present study was to determine whether CSF complement C5 levels were altered in patients with major psychiatric disorders, i.e., MDD, bipolar disorder (BPD), and schizophrenia, compared with healthy controls. We further examined the possible association of CSF C5 levels with clinical variables.
Subjects comprised 89 patients with MDD (age: 43.8 ± 10.4 years; 43 males), 66 patients with BPD (43.7 ± 12.3; 32 males), 96 patients with schizophrenia (40.1 ± 10.3; 58 males), and 117 healthy controls (42.5 ± 15.3; 66 males), matched for age, sex, and ethnicity (Japanese). Patients with BPD included 23 patients with bipolar I and 43 with bipolar II disorder. All participants were recruited at the National Center of Neurology and Psychiatry (NCNP), Tokyo, Japan, via advertising at the NCNP hospital, on our website, and in local free magazines. All participants underwent a structured interview using the Japanese version of the Mini International Neuropsychiatric Interview (MINI) [16], administered by a trained psychologist or board-certificated psychiatrist. Diagnosis was made according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition Criteria (American Psychiatric Association, 1994), on the basis of the MINI, additional unstructured interviews, and medical record (if available). Healthy controls were volunteers without a current or past history of contact with psychiatric services. Participants were excluded if they had a medical history of central nervous system (CNS) diseases, severe head injury, substance abuse, or mental retardation. The study protocol was approved by the ethics committee at the NCNP; the study was conducted according to the Declaration of Helsinki (World Medical Association, 2000). Written informed consent was obtained from every participant. 2.2. Clinical assessments The Positive and Negative Syndrome Scale (PANSS) was used to assess symptoms in patients with schizophrenia [17]. The GRIDHamilton Depression Rating Scale, 17-item version (GRID-HAMD17), was used to evaluate depressive symptoms in patients with MDD or BPD [18]. The Young Mania Rating Scale (YMRS) was used to evaluate manic symptoms in patients with BPD [19]. Daily doses of antipsychotics were converted into chlorpromazine-equivalent doses (CPeqs) and doses of antidepressants were converted into imipramine-equivalent doses (IMIeqs), using the published guidelines [20]. The status of medication use was recorded at the time of lumbar puncture. 2.3. Lumbar puncture After neurologic examinations, each participant underwent local anesthesia, followed by a lumbar puncture at the L3-4 or L4-5 interspace using an atraumatic pencil point needle (Universe 22G, 75 mm, Unisis Corp., Tokyo, Japan). CSF was collected using a low protein absorption tube (PROTEOSAVE SS 15 mL Conical tube, Sumitomo Bakelite Co., Tokyo, Japan) and immediately placed on ice. The CSF was centrifuged (4000 g for 10 min) at 4 C, then the supernatant was dispensed into 0.5 mL aliquots in low protein absorption tubes (PROTEOSAVE SS 1.5 mL Slim tube, Sumitomo Bakelite Co.) and stored in a deep freezer (80 C) until use. 2.4. Enzyme-linked immunosorbent assay (ELISA) We used a Human Complement C5 ELISA Kit (ab125963, Abcam Japan, Tokyo, Japan). The CSF samples were diluted 1:100, using the diluent attached to the kit. Prior to the experiment, the experimenter (T.I) validated the reproducibility of the ELISA at a level of 4% coefficient of variation, using the kit.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
2.5. Statistical analysis All statistical analyses were performed using the Statistical Package for the Social Sciences version 24.0 (IBM Japan, Ltd., Tokyo, Japan). Data are reported as mean ± standard deviation (SD). Continuous variables were compared among more than two groups of the patients and/or controls via analysis of variance. Categorical variables were compared between the patients and controls via the chi-squared test. Differences in C5 levels among the diagnostic groups were compared by analysis of covariance (ANCOVA), controlling for age and sex, and using the Bonferroni correction. The Pearson product-moment correlation was calculated to examine associations between CSF C5 levels and clinical variables within each diagnostic group. Three samples that showed a negative value, upon detection by optical density, were excluded from the analyses. Statistical significance was defined as a p-value of <0.05.
3
C5 levels than female subjects in each diagnostic group. BMI positively correlated with C5 levels in the patients with MDD, those with BPD, and the controls. We performed ANCOVA analysis on the subjects (N ¼ 307), in which C5 level was an dependent variable, sex and diagnosis were factors, and age and BMI were covariates. In this analysis, we found significant effects of age (F ¼ 11.8, df ¼ 1, p ¼ 0.001), sex (F ¼ 35.2, df ¼ 1, p < 0.001), BMI (F ¼ 8.37, df ¼ 1, p ¼ 0.004), and diagnosis (F ¼ 4.58, df ¼ 3, p ¼ 0.004) on C5 levels. Post-hoc analyses revealed that C5 levels were significantly increased in patients with MDD (p ¼ 0.001), and in patients with schizophrenia (p ¼ 0.003), but not in patients with BPD (p ¼ 0.18). As shown in Table 3, scores on HAMD-17, YMRS, or PANSS (including its subscales) showed no significant correlation with CSF C5 levels. There was a significantly positive correlation between CPeq and C5 levels in the patients with schizophrenia (p ¼ 0.002, Fig. 2). No significant correlation was detected between IMIeq and C5 levels in MDD or BPD patients.
3. Results Clinical features of the participants are shown in Table 1. There were no significant differences between each diagnostic group and the healthy control group in terms of age or sex distribution. The body mass index (BMI) in the patients with schizophrenia was significantly higher than the BMI in the controls. 3.1. C5 levels in diagnostic groups CSF C5 levels for the diagnostic groups are shown in Table 2 and Fig. 1. ANCOVA analysis, controlled for age and sex, revealed that CSF C5 levels were significantly increased in the patients with MDD (p < 0.001) and schizophrenia (p ¼ 0.001), but not in the patients with BPD (p ¼ 0.09), relative to the controls. When an “abnormally high C5 level” was defined as 424.0 ng/mL or higher, according to the 95th percentile value of the control group, individuals with this abnormally high C5 level were significantly more common in all of the included psychiatric groups, relative to the controls (all p < 0.01; Table 2). 3.2. Relationships of C5 levels with clinical variables Table 3 shows correlations between CSF complement C5 levels and multiple clinical variables. There was a significantly positive correlation between C5 levels and age in the patients with BPD, as well as in the controls. Male subjects tended to exhibit higher CSF
4. Discussion To our knowledge, this is the first report of increased CSF C5 levels in patients with major psychiatric disorders. C5 levels were significantly increased in the patients with MDD and in the patients with schizophrenia, compared with the healthy controls. The rate of individuals with an “abnormally high C5 level” was higher in all psychiatric groups than it was in the healthy controls. Older age, male sex, and greater BMI tended to associate with higher C5 levels. There was a significantly positive correlation between C5 levels and CPeq in the patients with schizophrenia. There are two possible mechanisms that may explain the increased CSF C5 levels in the patients with MDD and in the patients with schizophrenia. First, an increased permeability of the BBB may have permitted increased transition of the C5 molecule from the peripheral blood to the brain. There is some evidence that the albumin CSF-to-serum ratio, a marker for BBB permeability, was increased in the patients with MDD and in the patients with schizophrenia [21,22]. Another biomarker for BBB permeability is the serum S100B protein, which is produced mainly by astrocytes and oligodendrocytes. Several previous studies revealed that patients with MDD exhibit increased serum S100B levels [23,24]; however, conflicting reports stated that there was no significant association between serum S100B and depression severity [25,26]. Regarding schizophrenia, two meta-analyses determined that serum or plasma S100B levels were consistently increased in
Table 1 Clinical characteristics of the participants. Controls (n ¼ 117) Major depressive disorder (n ¼ 89)
Bipolar disorder (n ¼ 66)
Schizophrenia (n ¼ 96)
Mean ± SD
Mean ± SD
Mean ± SD
Age (years) 42.5 ± 15.3 Sexy, male (%)Sexy, male (%) 66 (56.4) BMI (kg/m2)BMI (kg/m2) 22.6 ± 3.4 CPeq (mg/day) IMIeq (mg/day) PANSS total positive negative general HAMD-17 S88 <88 YMRS
Mean ± SD
vs. Controls
43.8 ± 10.4 F ¼ 0.44, df ¼ 1, p ¼ 0.51 c2 ¼ 1.33, df ¼ 1, p ¼ 0.26 43 (48.3) 22.5 ± 3.3 F ¼ 0.07, df ¼ 1, p ¼ 0.79 39.7 ± 126.5 68.0 ± 119.3
vs. Controls
43.7 ± 12.3 F ¼ 0.28, df ¼ 1, p ¼ 0.60 32 (48.5) c2 ¼ 1.07, df ¼ 1, p ¼ 0.36 23.9 ± 4.8 F ¼ 3.87, df ¼ 1, p ¼ 0.05 114.2 ± 260.7 33.5 ± 83.5
vs. Controls
40.1 ± 10.3 F ¼ 1.72, df ¼ 1, p ¼ 0.19 58 (60.4) c2 ¼ 0.35, df ¼ 1, p ¼ 0.58 24.5 ± 5.5 F ¼ 7.87, df ¼ 1, p ¼ 0.006 533.3 ± 516.6 14.4 ± 43.4 60.9 ± 16.0 14.1 ± 5.1 16.2 ± 5.2 30.6 ± 8.8
16.6 ± 6.8 3.5 ± 2.5
14.9 ± 5.1 3.6 ± 2.5 6.3 ± 7.5
SD: standard deviation; BMI: body mass index; CPeq: chlorpromazine-equivalent dose; IMIeq: imipramine-equivalent dose; PANSS: Positive and Negative, Syndrome Scale; HAMD-17: 17-item version of the Hamilton Depression Rating Scale; YMRS: Young Mania Rating Scale; y: number of males.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
4
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
Table 2 Comparison of cerebrospinal fluid complement C5 levels and frequency of abnormally high C5 concentration between patients and controls. Cerebrospinal fluid C5 levels
Controls Major depressive disorder Bipolar disorder Schizophrenia
The number of patients with abnomally high C5 levels
n
Mean ± SD (ng/mL)
vs. Controls
vs. Controls
117 89 66 96
218.7 ± 111.1 317.7 ± 221.1 278.3 ± 190.7 306.6 ± 204.7
F ¼ 7.93, df ¼ 3, yp < 0.001 F ¼ 7.93, df ¼ 3, yp ¼ 0.09 F ¼ 7.93, df ¼ 3, yp ¼ 0.001
5 (4.3%) 22 (24.7%) 11 (16.7%) 20 (20.8%)
c2 ¼ 18.6, df ¼ 1, p < 0.001 c2 ¼ 8.1, df ¼ 1, p ¼ 0.006 c2 ¼ 14.0, df ¼ 1, p < 0.001
An “abnormally high C5 level” was defined as 424.0 ng/mL or higher, according to the 95th percentile value of the control group. y: ANCOVA controlling for age and sex.
ng/mL p = 0.001 p < 0.001
r=0.31 p=0.002
ng/mL
Fig. 2. Correlation between cerebrospinal fluid (CSF) C5 levels and chlorpromazineequivalent doses (CPeqs) in the patients with schizophrenia. In the patients with schizophrenia, there was a significant, but weakly positive, correlation between CSF C5 levels and CPeq.
n = 117
n = 89
n = 66
n = 96
Fig. 1. Cerebrospinal fluid (CSF) complement C5 levels in healthy controls (CONT), as well as in patients with major depressive disorder (MDD), bipolar disorder (BPD), and schizophrenia (SCH).
patients with the disorder [27]. Second, since human CNS cells produce components of the complement system under the appropriate stimulatory conditions [10], increased production of C5 in the brain may contribute to the observed increase in CSF C5 levels, both in the patients with MDD and in the patients with
schizophrenia. Although complement synthesis in the human brain is regarded as generally low or non-detectable under normal conditions [10], increased CSF C5a levels were reported in a study of patients with neuromyelitis optica; conversely, their serum C5a levels were unaltered [9]. We observed a positive correlation between C5 levels and antipsychotic medication doses in the patients with schizophrenia, which suggests that antipsychotics may increase C5 levels in the brain. Animal model and in vitro studies indicate that
Table 3 Correlations between cerebrospinal fluid complement C5 levels and multiple clinical variables.
Age (years) Sex BMI$ (kg/m2) HAMD-17 YMRS PANSS total positive negative general CPeq (mg/day) IMIeq (mg/day)
Controls (n ¼ 117)
MDD (n ¼ 89)
r
p
r
0.38 0.17 0.36
<0.001 0.07 <0.001
0.04 0.33 0.27 0.04
0.13
Bipolar disorder (n ¼ 66)
Schizophrenia (n ¼ 96)
p
r
p
r
p
0.71 0.002 0.02 0.72
0.41 0.27 0.32 0.14 0.19
0.001 0.03 0.01 0.29 0.16
0.12 0.31 0.11
0.24 0.002 0.33
0.08 0.09 0.06 0.12 0.31
0.47 0.40 0.55 0.24 0.002
0.22
0.02 0.15
0.87 0.23
MDD: major depressive disorder; BMI: body mass index; HAMD-17: 17-item version of the Hamilton, Depression Rating Scale; YMRS: Young Mania Rating Scale; PANSS: Positive and Negative Syndrome; CPeq:: chlorpromazine-equivalent dose; IMIeq: imipramine-equivalent dose; Significant pvalues are shown in bold cases. $: BMI data were missing for 61 subjects (16.6%).
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
antipsychotics might negatively affect the BBB [28,29]: antipsychotics may induce cytotoxic effects and apoptosis of BBB endothelial cells, with a concomitant impairment of barrier functionality [28]. In a manner consistent with this proposed medication effect, chronic antipsychotic treatment resulted in an increased density of microglia in the rat brain [29]. Increased CSF C5 levels may lead to microglial migration and activation, thereby altering immune conditions and inducing neuroinflammation in the CNS. Microglial migration is diminished in the complement C5a receptor knockout mouse [30], consistent with these findings, C5a-peptide vaccines reduced microglial activation and neuroinflammation [31]. From a developmental perspective, increased CSF C5 levels may contribute to complement-mediating synapse pruning, especially in adolescence. Peripheral mRNA expression levels of C5 were associated with superior frontal cortical thickness in patients with schizophrenia [13]. Moreover, structural alteration of C4, which is located upstream of C5, has been suggested to induce synapse pruning [11]. Interestingly, BMI positively correlated with C5 levels in patients with MDD, patients with BPD, and healthy controls. The mechanism of this correlation is unclear; however, several lines of evidence suggest that obesity may promote neuroinflammation and BBB hyperpermeability. Further, obesity appears to cause lowgrade chronic inflammation in peripheral tissues and blood circulation [32,33]. These previous studies did not evaluate neuroinflammation, but a separate report suggests that chronic inflammation correlates with BBB hyperpermeability in certain conditions related to the presence of CSF abnormalities [34]. Indeed, a high fat diet in mice induces neuroinflammation via BBB disruption at the hippocampus [35]. Those findings are especially interesting in the context of MDD pathophysiology, as obesity increases the risk of depression [36]; in a related manner, patients with MDD carry a significantly higher risk of developing obesity, relative to health individuals [37], and neuroinflammation has repeatedly been reported in MDD [2,38,39]. Future basic research should evaluate the effect of obesity on neuroinflammation and CSF markers, including in the context of medication. CSF C5 levels may reflect the pathophysiology of a subset of mental diseases. Although, in the present study, there was no association between CSF C5 levels and symptom scores (or scores of subscales) in each psychiatric group, a thorough investigation of the symptoms, brain imaging and physiological tests of high C5 patients is required in a future study. As novel treatment options for high C5 patients, C5, C5a, C5aR (induces microglia migration), and CD59 (inhibits membrane attack complex formation) may be promising targets to inhibit the complement cascade [40]. In the present study, the patients were biased towards milder forms of mental illness and the majority of patients were medicated. CSF was, therefore, collected in a “real-world setting”. Notably, the number of bipolar subjects was relatively small. CSF C5 levels were significantly increased in the patients with MDD and schizophrenia, compared with the healthy controls. The rate of individuals with an “abnormally high C5 level” was increased in all psychiatric groups, relative to the health controls. Our results suggest that the activated complement system may contribute to neurological pathogenesis in a portion of patients with major psychiatric disorders.
Conflicts-of-Interest The authors declare that they have no conflict of interest.
5
Funding This study was supported by Intramural Research Grants for Neurological and Psychiatric Disorders at NCNP [grant number 271] (H.K.), Strategic Research Program for Brain Sciences from Japan Agency for Medical Research and development (AMED) [grant number 16dm0107100h001] (H.K.), and Health and Labor Sciences Research Grants for Comprehensive Research on Persons with Disabilities from AMED [grant number 15dK0310061h001] (H.K.). These funding sources were involved only in the financial support. There was a significant increase in C5 levels in the patients with MDD (p < 0.001) and schizophrenia (p ¼ 0.001), compared with the healthy controls. Horizontal solid lines indicate the mean values of the diagnostic groups. The dotted line indicates the 95th percentile value of the control group (424.0 ng/mL). Acknowledgements We thank Mr. Ikki Ishida, Ms. Junko Matsuo, Ms. Moeko Hiraishi, Ms. Yuuki Yokota, Mr. Ryo Matsmura, Ms. Tomoko Kurashimo, Ms. Naoko Ishihara, Mr. Takahiro Tomizawa, and Ms. Chie Kimizuka for their assistance in recruitment and clinical assessments of participants. We also thank Ms. Misao Nakano and Mr. Shokichi Tajima for their sample management. References [1] B.N. Cuthbert, T.R. Insel, Toward the future of psychiatric diagnosis: the seven pillars of RDoC, BMC Med. 11 (2013) 126. [2] K. Hattori, M. Ota, D. Sasayama, S. Yoshida, R. Matsumura, T. Miyakawa, Y. Yokota, S. Yamaguchi, T. Noda, T. Teraishi, H. Hori, T. Higuchi, S. Kohsaka, Y. Goto, H. Kunugi, Increased cerebrospinal fluid fibrinogen in major depressive disorder, Sci. Rep. 5 (2015) 11412. [3] D.R. Goldsmith, M.H. Rapaport, B.J. Miller, A meta-analysis of blood cytokine network alterations in psychiatric patients: comparisons between schizophrenia, bipolar disorder and depression, Mol. Psychiatr. 21 (2016) 1696e1709. [4] D. Sasayama, C. Wakabayashi, H. Hori, T. Teraishi, K. Hattori, M. Ota, M. Ishikawa, K. Arima, T. Higuchi, N. Amano, H. Kunugi, Association of plasma IL-6 and soluble IL-6 receptor levels with the Asp358Ala polymorphism of the IL-6 receptor gene in schizophrenic patients, J. Psychiatr. Res. 45 (2011) 1439e1444. [5] G.M. Khandaker, L. Cousins, J. Deakin, B.R. Lennox, R. Yolken, P.B. Jones, Inflammation and immunity in schizophrenia: implications for pathophysiology and treatment, The Lancet. Psychiatry 2 (2015) 258e270. [6] D. Sasayama, K. Hattori, C. Wakabayashi, T. Teraishi, H. Hori, M. Ota, S. Yoshida, K. Arima, T. Higuchi, N. Amano, H. Kunugi, Increased cerebrospinal fluid interleukin-6 levels in patients with schizophrenia and those with major depressive disorder, J. Psychiatr. Res. 47 (2013) 401e406. [7] A.K. Wang, B.J. Miller, Meta-analysis of cerebrospinal fluid cytokine and tryptophan catabolite alterations in psychiatric patients: comparisons between schizophrenia, bipolar disorder, and depression, Schizophr. Bull. 44 (2018) 75e83. [8] D. Davalos, J.K. Ryu, M. Merlini, K.M. Baeten, N. Le Moan, M.A. Petersen, T.J. Deerinck, D.S. Smirnoff, C. Bedard, H. Hakozaki, S. Gonias Murray, J.B. Ling, H. Lassmann, J.L. Degen, M.H. Ellisman, K. Akassoglou, Fibrinogen-induced perivascular microglial clustering is required for the development of axonal damage in neuroinflammation, Nat. Commun. 3 (2012) 1227. [9] H. Kuroda, K. Fujihara, R. Takano, Y. Takai, T. Takahashi, T. Misu, I. Nakashima, S. Sato, Y. Itoyama, M. Aoki, Increase of complement fragment C5a in cerebrospinal fluid during exacerbation of neuromyelitis optica, J. Neuroimmunol. 254 (2013) 178e182. [10] T.M. Woodruff, R.R. Ager, A.J. Tenner, P.G. Noakes, S.M. Taylor, The role of the complement system and the activation fragment C5a in the central nervous system, NeuroMolecular Med. 12 (2010) 179e192. [11] A. Sekar, A.R. Bialas, H. de Rivera, A. Davis, T.R. Hammond, N. Kamitaki, K. Tooley, J. Presumey, M. Baum, V. Van Doren, G. Genovese, S.A. Rose, R.E. Handsaker, M.J. Daly, M.C. Carroll, B. Stevens, S.A. McCarroll, Schizophrenia risk from complex variation of complement component 4, Nature 530 (2016) 177e183. [12] W. Kakegawa, N. Mitakidis, E. Miura, M. Abe, K. Matsuda, Y.H. Takeo, K. Kohda, J. Motohashi, A. Takahashi, S. Nagao, S. Muramatsu, M. Watanabe, K. Sakimura, A.R. Aricescu, M. Yuzaki, Anterograde C1ql1 signaling is required in order to determine and maintain a single-winner climbing fiber in the mouse cerebellum, Neuron 85 (2015) 316e329. [13] D.M. Allswede, A.B. Zheutlin, Y. Chung, K. Anderson, C.M. Hultman, M. Ingvar,
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
6
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
[14] [15]
[16]
[17] [18]
[19] [20] [21]
[22]
[23]
[24]
[25]
[26]
T.D. Cannon, Complement gene expression correlates with superior frontal cortical thickness in humans, Neuropsychopharmacology 43 (2018) 525e533. A. Mubarik, H. Tohid, Frontal lobe alterations in schizophrenia: a review, Trends Psychiatr. Psychother. 38 (2016) 198e206. S. Hidese, M. Ota, J. Matsuo, I. Ishida, M. Hiraishi, T. Teraishi, K. Hattori, H. Kunugi, Association between the scores of the Japanese version of the Brief Assessment of Cognition in Schizophrenia and whole-brain structure in patients with chronic schizophrenia: a voxel-based morphometry and diffusion tensor imaging study, Psychiatr. Clin. Neurosci. 71 (2017) 826e835. D.V. Sheehan, Y. Lecrubier, K.H. Sheehan, P. Amorim, J. Janavs, E. Weiller, T. Hergueta, R. Baker, G.C. Dunbar, The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10, J. Clin. Psychiatr. 59 (Suppl 20) (1998) 22e33 quiz 34e57. S.R. Kay, A. Fiszbein, L.A. Opler, The positive and negative syndrome scale (PANSS) for schizophrenia, Schizophr. Bull. 13 (1987) 261e276. J.B. Williams, K.A. Kobak, P. Bech, N. Engelhardt, K. Evans, J. Lipsitz, J. Olin, J. Pearson, A. Kalali, The GRID-HAMD: standardization of the Hamilton depression rating scale, Int. Clin. Psychopharmacol. 23 (2008) 120e129. R.C. Young, J.T. Biggs, V.E. Ziegler, D.A. Meyer, A rating scale for mania: reliability, validity and sensitivity, Br. J. Psychiatry 133 (1978) 429e435. T. Inada, A. Inagaki, Psychotropic dose equivalence in Japan, Psychiatr. Clin. Neurosci. 69 (2015) 440e447. P. Gudmundsson, I. Skoog, M. Waern, K. Blennow, S. Palsson, L. Rosengren, D. Gustafson, The relationship between cerebrospinal fluid biomarkers and depression in elderly women, Am. J. Geriatr. Psychiatr. 15 (2007) 832e838. K. Bechter, H. Reiber, S. Herzog, D. Fuchs, H. Tumani, H.G. Maxeiner, Cerebrospinal fluid analysis in affective and schizophrenic spectrum disorders: identification of subgroups with immune responses and blood-CSF barrier dysfunction, J. Psychiatr. Res. 44 (2010) 321e330. M. Rothermundt, V. Arolt, M. Wiesmann, U. Missler, M. Peters, S. Rudolf, H. Kirchner, S-100B is increased in melancholic but not in non-melancholic major depression, J. Affect. Disord. 66 (2001) 89e93. M.L. Schroeter, H. Abdul-Khaliq, A. Diefenbacher, I.E. Blasig, S100B is increased in mood disorders and may be reduced by antidepressive treatment, Neuroreport 13 (2002) 1675e1678. B.S. Jang, H. Kim, S.W. Lim, K.W. Jang, D.K. Kim, Serum S100B levels and major depressive disorder: its characteristics and role in antidepressant response, Psychiatry Investig. 5 (2008) 193e198. O. Ambree, V. Bergink, L. Grosse, J. Alferink, H.A. Drexhage, M. Rothermundt, V. Arolt, T.K. Birkenhager, S100B serum levels predict treatment response in patients with melancholic depression, Int. J. Neuropsychopharmacol. 19 (2015) pyv103.
[27] T.A. Pollak, S. Drndarski, J.M. Stone, A.S. David, P. McGuire, N.J. Abbott, The blood-brain barrier in psychosis, The lancet. Psychiatry 5 (2018) 79e92. [28] E. Elmorsy, L.M. Elzalabany, H.M. Elsheikha, P.A. Smith, Adverse effects of antipsychotics on micro-vascular endothelial cells of the human blood-brain barrier, Brain Res. 1583 (2014) 255e268. [29] M.C. Cotel, E.M. Lenartowicz, S. Natesan, M.M. Modo, J.D. Cooper, S.C. Williams, S. Kapur, A.C. Vernon, Microglial activation in the rat brain following chronic antipsychotic treatment at clinically relevant doses, Eur. Neuropsychopharmacol 25 (2015) 2098e2107. [30] D. Song, M.E. Sulewski Jr., C. Wang, J. Song, R. Bhuyan, J. Sterling, E. Clark, W.C. Song, J.L. Dunaief, Complement C5a receptor knockout has diminished light-induced microglia/macrophage retinal migration, Mol. Vis. 23 (2017) 210e218. [31] C. Landlinger, L. Oberleitner, P. Gruber, B. Noiges, K. Yatsyk, R. Santic, M. Mandler, G. Staffler, Active immunization against complement factor C5a: a new therapeutic approach for Alzheimer's disease, J. Neuroinflammation 12 (2015) 150. [32] G.S. Hotamisligil, Inflammation and metabolic disorders, Nature 444 (2006) 860e867. [33] M.F. Gregor, G.S. Hotamisligil, Inflammatory mechanisms in obesity, Annu. Rev. Immunol. 29 (2011) 415e445. [34] E. Elwood, Z. Lim, H. Naveed, I. Galea, The effect of systemic inflammation on human brain barrier function, Brain Behav. Immun. 62 (2017) 35e40. [35] Z. Tucsek, P. Toth, D. Sosnowska, T. Gautam, M. Mitschelen, A. Koller, G. Szalai, W.E. Sonntag, Z. Ungvari, A. Csiszar, Obesity in aging exacerbates blood-brain barrier disruption, neuroinflammation, and oxidative stress in the mouse hippocampus: effects on expression of genes involved in beta-amyloid generation and Alzheimer's disease, J. Gerontol. A Biol. Sci. Med. Sci 69 (2014) 1212e1226. [36] F.S. Luppino, L.M. de Wit, P.F. Bouvy, T. Stijnen, P. Cuijpers, B.W. Penninx, F.G. Zitman, Overweight, obesity, and depression: a systematic review and meta-analysis of longitudinal studies, Arch. Gen. Psychiatr. 67 (2010) 220e229. [37] R.B. Mansur, E. Brietzke, R.S. McIntyre, Is there a “metabolic-mood syndrome”? A review of the relationship between obesity and mood disorders, Neurosci. Biobehav. Rev. 52 (2015) 89e104. [38] H. Kunugi, H. Hori, S. Ogawa, Biochemical markers subtyping major depressive disorder, Psychiatr. Clin. Neurosci. 69 (2015) 597e608. [39] M.L. Monje, H. Toda, T.D. Palmer, Inflammatory blockade restores adult hippocampal neurogenesis, Science 302 (2003) 1760e1765. [40] D. Ricklin, J.D. Lambris, Complement-targeted therapeutics, Nat. Biotechnol. 25 (2007) 1265e1275.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia Takashi Ishii a, c, Kotaro Hattori a, b, Tomoko Miyakawa a, b, Kentaro Watanabe a, Shinsuke Hidese a, Daimei Sasayama a, g, Miho Ota a, Toshiya Teraishi a, Hiroaki Hori a, e, Sumiko Yoshida b, f, Akihiko Nunomura c, Kazuyuki Nakagome d, Hiroshi Kunugi a, * a
Department of Mental Disorder Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan Medical Genome Center, National Center of Neurology and Psychiatry, Tokyo, Japan Department of Neuropsychiatry, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Yamanashi, Japan d National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan e Department of Adult Mental Health, National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan f National Center of Neurology and Psychiatry Hospital, Tokyo, Japan g Department of Psychiatry, Shinshu University School of Medicine, Matsumoto, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 January 2018 Accepted 15 February 2018 Available online xxx
Inflammation has been implicated in a variety of psychiatric disorders. We aimed to determine whether levels of complement C5 protein in the cerebrospinal fluid (CSF), which may reflect activation of the complement system in the brain, are altered in patients with major psychiatric disorders. Additionally, we examined possible associations of CSF C5 levels with clinical variables. Subjects comprised 89 patients with major depressive disorder (MDD), 66 patients with bipolar disorder (BPD), 96 patients with schizophrenia, and 117 healthy controls, matched for age, sex, and ethnicity (Japanese). Diagnosis was made according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, criteria. CSF C5 levels were measured by enzyme-linked immunosorbent assay. CSF C5 levels were significantly increased in the patients with MDD (p < 0.001) and in the patients with schizophrenia (p ¼ 0.001), compared with the healthy controls. The rate of individuals with an “abnormally high C5 level” (i.e., above the 95th percentile value of the control subjects) was significantly increased in all psychiatric groups, relative to the control group (all p < 0.01). Older age, male sex, and greater body mass index tended to associate with higher C5 levels. There was a significantly positive correlation between C5 levels and chlorpromazine-equivalent dose in the patients with schizophrenia. Thus, we found, for the first time, elevated C5 levels in the CSF of patients with major psychiatric disorders. Our results suggest that the activated complement system may contribute to neurological pathogenesis in a portion of patients with major psychiatric disorders. © 2018 Elsevier Inc. All rights reserved.
Keywords: Cerebrospinal fluid Complement C5 Schizophrenia Bipolar disorder Major depressive disorder Body mass index
1. Introduction Current diagnostic systems for mental disorders rely upon the presentation of signs and symptoms, and thus do not adequately reflect the relevant neurobiological and behavioral systems [1]. There is an urgent need to elucidate molecular basis of mental disorders and to develop biomarkers that allow classification of psychiatric disorders based on the pathophysiology. In this context,
* Corresponding author. 4-1-1, Ogawahigashi, Kodaira, Tokyo, 187-8502, Japan. E-mail address: [email protected] (H. Kunugi).
investigation of the cerebrospinal fluid (CSF) might be one of the most promising approaches to detect molecules that contribute to the pathophysiology of psychiatric diseases within the brain, as CSF is in contact with the brain interstitial fluid over the surfaces of ventricles, brain, and spinal cord; therefore, molecules released from brain cells can directly diffuse into the CSF. Indeed, CSF biomarkers have already been established for some neurological diseases, such as tau, phosphorylated tau, and b-amyloid proteins, which diagnose Alzheimer's disease with a high (~90%) sensitivity and specificity [2]. There is a growing amount of evidence that supports increased levels of pro-inflammatory cytokines in the blood of patients with
https://doi.org/10.1016/j.bbrc.2018.02.131 0006-291X/© 2018 Elsevier Inc. All rights reserved.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
2
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
2. Material and methods Abbreviations 2.1. Participants BBB blood-brain barrier BMI body mass index BPD bipolar disorder CNS central nervous system CPeq chlorpromazine-equivalent dose CSF cerebrospinal fluid ELISA enzyme-linked immunosorbent assay GRID-HAMD-17 GRID-Hamilton Depression Rating Scale, 17item version IL-6 interleukin-6 IMIeq imipramine-equivalent dose MDD major depressive disorder MINI Mini International Neuropsychiatric Interview PANSS Positive and Negative Syndrome Scale YMRS Young Mania Rating Scale
schizophrenia and major depressive disorder (MDD) [3,4]. We previously reported increased interleukin-6 (IL-6) levels in the blood of patients with schizophrenia [4]. Contrary to the traditional view that the brain is an immunologically privileged site that is shielded behind the blood-brain barrier (BBB), studies in the past 20 years have uncovered complex interactions among the immune system, systemic inflammation, and brain functions, which can lead to changes in mood, cognition, and behavior [5]. As stated above, we have reported increased IL-6 in the CSF of patients with schizophrenia, as well as in patients with MDD, relative to healthy controls [6]; these findings were further supported by a recent meta-analysis [7]. Additionally, we have reported increased CSF fibrinogen levels in patients with MDD [2]. Fibrinogen is a coagulation factor which activates microglia, thereby inducing neuroinflammation and negatively affecting the brain [8]. A major aspect of the innate immune system is the complement system. In general, complement activation occurs via four pathways: the classical, alternative, mannan-binding lectin, and extrinsic protease pathways [9,10]. These pathways converge upon the cleavage of complement C5 into complement fragments C5a and C5b by C5 convertase, which is the final step of the complement activation. C5a is the most potent and stable of the anaphylatoxins, while C5b is an initial element in the formation of the membrane attack complex. Notably, the complement system also plays a role in synapse elimination/pruning, which is essential to the development of a precise neuronal network [11,12]. A recent study on adult Swedish twins enriched for schizophrenia reported that peripheral messenger RNA (mRNA) expression levels of two complement genes (C5, SERPING1) in peripheral blood mononuclear cells were associated with decreased superior frontal cortical thickness, which may be involved in the pathogenesis of schizophrenia [13]. A number of studies, including ours, have reported structural and functional changes in volume, gray matter, white matter, and functional activity, within the frontal lobe of schizophrenia patients [14,15]. Since C5 is involved in the final step of the activation of the complement system, we hypothesized that C5 levels would be altered in the brain of patients with schizophrenia. The aim of the present study was to determine whether CSF complement C5 levels were altered in patients with major psychiatric disorders, i.e., MDD, bipolar disorder (BPD), and schizophrenia, compared with healthy controls. We further examined the possible association of CSF C5 levels with clinical variables.
Subjects comprised 89 patients with MDD (age: 43.8 ± 10.4 years; 43 males), 66 patients with BPD (43.7 ± 12.3; 32 males), 96 patients with schizophrenia (40.1 ± 10.3; 58 males), and 117 healthy controls (42.5 ± 15.3; 66 males), matched for age, sex, and ethnicity (Japanese). Patients with BPD included 23 patients with bipolar I and 43 with bipolar II disorder. All participants were recruited at the National Center of Neurology and Psychiatry (NCNP), Tokyo, Japan, via advertising at the NCNP hospital, on our website, and in local free magazines. All participants underwent a structured interview using the Japanese version of the Mini International Neuropsychiatric Interview (MINI) [16], administered by a trained psychologist or board-certificated psychiatrist. Diagnosis was made according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition Criteria (American Psychiatric Association, 1994), on the basis of the MINI, additional unstructured interviews, and medical record (if available). Healthy controls were volunteers without a current or past history of contact with psychiatric services. Participants were excluded if they had a medical history of central nervous system (CNS) diseases, severe head injury, substance abuse, or mental retardation. The study protocol was approved by the ethics committee at the NCNP; the study was conducted according to the Declaration of Helsinki (World Medical Association, 2000). Written informed consent was obtained from every participant. 2.2. Clinical assessments The Positive and Negative Syndrome Scale (PANSS) was used to assess symptoms in patients with schizophrenia [17]. The GRIDHamilton Depression Rating Scale, 17-item version (GRID-HAMD17), was used to evaluate depressive symptoms in patients with MDD or BPD [18]. The Young Mania Rating Scale (YMRS) was used to evaluate manic symptoms in patients with BPD [19]. Daily doses of antipsychotics were converted into chlorpromazine-equivalent doses (CPeqs) and doses of antidepressants were converted into imipramine-equivalent doses (IMIeqs), using the published guidelines [20]. The status of medication use was recorded at the time of lumbar puncture. 2.3. Lumbar puncture After neurologic examinations, each participant underwent local anesthesia, followed by a lumbar puncture at the L3-4 or L4-5 interspace using an atraumatic pencil point needle (Universe 22G, 75 mm, Unisis Corp., Tokyo, Japan). CSF was collected using a low protein absorption tube (PROTEOSAVE SS 15 mL Conical tube, Sumitomo Bakelite Co., Tokyo, Japan) and immediately placed on ice. The CSF was centrifuged (4000 g for 10 min) at 4 C, then the supernatant was dispensed into 0.5 mL aliquots in low protein absorption tubes (PROTEOSAVE SS 1.5 mL Slim tube, Sumitomo Bakelite Co.) and stored in a deep freezer (80 C) until use. 2.4. Enzyme-linked immunosorbent assay (ELISA) We used a Human Complement C5 ELISA Kit (ab125963, Abcam Japan, Tokyo, Japan). The CSF samples were diluted 1:100, using the diluent attached to the kit. Prior to the experiment, the experimenter (T.I) validated the reproducibility of the ELISA at a level of 4% coefficient of variation, using the kit.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
2.5. Statistical analysis All statistical analyses were performed using the Statistical Package for the Social Sciences version 24.0 (IBM Japan, Ltd., Tokyo, Japan). Data are reported as mean ± standard deviation (SD). Continuous variables were compared among more than two groups of the patients and/or controls via analysis of variance. Categorical variables were compared between the patients and controls via the chi-squared test. Differences in C5 levels among the diagnostic groups were compared by analysis of covariance (ANCOVA), controlling for age and sex, and using the Bonferroni correction. The Pearson product-moment correlation was calculated to examine associations between CSF C5 levels and clinical variables within each diagnostic group. Three samples that showed a negative value, upon detection by optical density, were excluded from the analyses. Statistical significance was defined as a p-value of <0.05.
3
C5 levels than female subjects in each diagnostic group. BMI positively correlated with C5 levels in the patients with MDD, those with BPD, and the controls. We performed ANCOVA analysis on the subjects (N ¼ 307), in which C5 level was an dependent variable, sex and diagnosis were factors, and age and BMI were covariates. In this analysis, we found significant effects of age (F ¼ 11.8, df ¼ 1, p ¼ 0.001), sex (F ¼ 35.2, df ¼ 1, p < 0.001), BMI (F ¼ 8.37, df ¼ 1, p ¼ 0.004), and diagnosis (F ¼ 4.58, df ¼ 3, p ¼ 0.004) on C5 levels. Post-hoc analyses revealed that C5 levels were significantly increased in patients with MDD (p ¼ 0.001), and in patients with schizophrenia (p ¼ 0.003), but not in patients with BPD (p ¼ 0.18). As shown in Table 3, scores on HAMD-17, YMRS, or PANSS (including its subscales) showed no significant correlation with CSF C5 levels. There was a significantly positive correlation between CPeq and C5 levels in the patients with schizophrenia (p ¼ 0.002, Fig. 2). No significant correlation was detected between IMIeq and C5 levels in MDD or BPD patients.
3. Results Clinical features of the participants are shown in Table 1. There were no significant differences between each diagnostic group and the healthy control group in terms of age or sex distribution. The body mass index (BMI) in the patients with schizophrenia was significantly higher than the BMI in the controls. 3.1. C5 levels in diagnostic groups CSF C5 levels for the diagnostic groups are shown in Table 2 and Fig. 1. ANCOVA analysis, controlled for age and sex, revealed that CSF C5 levels were significantly increased in the patients with MDD (p < 0.001) and schizophrenia (p ¼ 0.001), but not in the patients with BPD (p ¼ 0.09), relative to the controls. When an “abnormally high C5 level” was defined as 424.0 ng/mL or higher, according to the 95th percentile value of the control group, individuals with this abnormally high C5 level were significantly more common in all of the included psychiatric groups, relative to the controls (all p < 0.01; Table 2). 3.2. Relationships of C5 levels with clinical variables Table 3 shows correlations between CSF complement C5 levels and multiple clinical variables. There was a significantly positive correlation between C5 levels and age in the patients with BPD, as well as in the controls. Male subjects tended to exhibit higher CSF
4. Discussion To our knowledge, this is the first report of increased CSF C5 levels in patients with major psychiatric disorders. C5 levels were significantly increased in the patients with MDD and in the patients with schizophrenia, compared with the healthy controls. The rate of individuals with an “abnormally high C5 level” was higher in all psychiatric groups than it was in the healthy controls. Older age, male sex, and greater BMI tended to associate with higher C5 levels. There was a significantly positive correlation between C5 levels and CPeq in the patients with schizophrenia. There are two possible mechanisms that may explain the increased CSF C5 levels in the patients with MDD and in the patients with schizophrenia. First, an increased permeability of the BBB may have permitted increased transition of the C5 molecule from the peripheral blood to the brain. There is some evidence that the albumin CSF-to-serum ratio, a marker for BBB permeability, was increased in the patients with MDD and in the patients with schizophrenia [21,22]. Another biomarker for BBB permeability is the serum S100B protein, which is produced mainly by astrocytes and oligodendrocytes. Several previous studies revealed that patients with MDD exhibit increased serum S100B levels [23,24]; however, conflicting reports stated that there was no significant association between serum S100B and depression severity [25,26]. Regarding schizophrenia, two meta-analyses determined that serum or plasma S100B levels were consistently increased in
Table 1 Clinical characteristics of the participants. Controls (n ¼ 117) Major depressive disorder (n ¼ 89)
Bipolar disorder (n ¼ 66)
Schizophrenia (n ¼ 96)
Mean ± SD
Mean ± SD
Mean ± SD
Age (years) 42.5 ± 15.3 Sexy, male (%)Sexy, male (%) 66 (56.4) BMI (kg/m2)BMI (kg/m2) 22.6 ± 3.4 CPeq (mg/day) IMIeq (mg/day) PANSS total positive negative general HAMD-17 S88 <88 YMRS
Mean ± SD
vs. Controls
43.8 ± 10.4 F ¼ 0.44, df ¼ 1, p ¼ 0.51 c2 ¼ 1.33, df ¼ 1, p ¼ 0.26 43 (48.3) 22.5 ± 3.3 F ¼ 0.07, df ¼ 1, p ¼ 0.79 39.7 ± 126.5 68.0 ± 119.3
vs. Controls
43.7 ± 12.3 F ¼ 0.28, df ¼ 1, p ¼ 0.60 32 (48.5) c2 ¼ 1.07, df ¼ 1, p ¼ 0.36 23.9 ± 4.8 F ¼ 3.87, df ¼ 1, p ¼ 0.05 114.2 ± 260.7 33.5 ± 83.5
vs. Controls
40.1 ± 10.3 F ¼ 1.72, df ¼ 1, p ¼ 0.19 58 (60.4) c2 ¼ 0.35, df ¼ 1, p ¼ 0.58 24.5 ± 5.5 F ¼ 7.87, df ¼ 1, p ¼ 0.006 533.3 ± 516.6 14.4 ± 43.4 60.9 ± 16.0 14.1 ± 5.1 16.2 ± 5.2 30.6 ± 8.8
16.6 ± 6.8 3.5 ± 2.5
14.9 ± 5.1 3.6 ± 2.5 6.3 ± 7.5
SD: standard deviation; BMI: body mass index; CPeq: chlorpromazine-equivalent dose; IMIeq: imipramine-equivalent dose; PANSS: Positive and Negative, Syndrome Scale; HAMD-17: 17-item version of the Hamilton Depression Rating Scale; YMRS: Young Mania Rating Scale; y: number of males.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
4
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
Table 2 Comparison of cerebrospinal fluid complement C5 levels and frequency of abnormally high C5 concentration between patients and controls. Cerebrospinal fluid C5 levels
Controls Major depressive disorder Bipolar disorder Schizophrenia
The number of patients with abnomally high C5 levels
n
Mean ± SD (ng/mL)
vs. Controls
vs. Controls
117 89 66 96
218.7 ± 111.1 317.7 ± 221.1 278.3 ± 190.7 306.6 ± 204.7
F ¼ 7.93, df ¼ 3, yp < 0.001 F ¼ 7.93, df ¼ 3, yp ¼ 0.09 F ¼ 7.93, df ¼ 3, yp ¼ 0.001
5 (4.3%) 22 (24.7%) 11 (16.7%) 20 (20.8%)
c2 ¼ 18.6, df ¼ 1, p < 0.001 c2 ¼ 8.1, df ¼ 1, p ¼ 0.006 c2 ¼ 14.0, df ¼ 1, p < 0.001
An “abnormally high C5 level” was defined as 424.0 ng/mL or higher, according to the 95th percentile value of the control group. y: ANCOVA controlling for age and sex.
ng/mL p = 0.001 p < 0.001
r=0.31 p=0.002
ng/mL
Fig. 2. Correlation between cerebrospinal fluid (CSF) C5 levels and chlorpromazineequivalent doses (CPeqs) in the patients with schizophrenia. In the patients with schizophrenia, there was a significant, but weakly positive, correlation between CSF C5 levels and CPeq.
n = 117
n = 89
n = 66
n = 96
Fig. 1. Cerebrospinal fluid (CSF) complement C5 levels in healthy controls (CONT), as well as in patients with major depressive disorder (MDD), bipolar disorder (BPD), and schizophrenia (SCH).
patients with the disorder [27]. Second, since human CNS cells produce components of the complement system under the appropriate stimulatory conditions [10], increased production of C5 in the brain may contribute to the observed increase in CSF C5 levels, both in the patients with MDD and in the patients with
schizophrenia. Although complement synthesis in the human brain is regarded as generally low or non-detectable under normal conditions [10], increased CSF C5a levels were reported in a study of patients with neuromyelitis optica; conversely, their serum C5a levels were unaltered [9]. We observed a positive correlation between C5 levels and antipsychotic medication doses in the patients with schizophrenia, which suggests that antipsychotics may increase C5 levels in the brain. Animal model and in vitro studies indicate that
Table 3 Correlations between cerebrospinal fluid complement C5 levels and multiple clinical variables.
Age (years) Sex BMI$ (kg/m2) HAMD-17 YMRS PANSS total positive negative general CPeq (mg/day) IMIeq (mg/day)
Controls (n ¼ 117)
MDD (n ¼ 89)
r
p
r
0.38 0.17 0.36
<0.001 0.07 <0.001
0.04 0.33 0.27 0.04
0.13
Bipolar disorder (n ¼ 66)
Schizophrenia (n ¼ 96)
p
r
p
r
p
0.71 0.002 0.02 0.72
0.41 0.27 0.32 0.14 0.19
0.001 0.03 0.01 0.29 0.16
0.12 0.31 0.11
0.24 0.002 0.33
0.08 0.09 0.06 0.12 0.31
0.47 0.40 0.55 0.24 0.002
0.22
0.02 0.15
0.87 0.23
MDD: major depressive disorder; BMI: body mass index; HAMD-17: 17-item version of the Hamilton, Depression Rating Scale; YMRS: Young Mania Rating Scale; PANSS: Positive and Negative Syndrome; CPeq:: chlorpromazine-equivalent dose; IMIeq: imipramine-equivalent dose; Significant pvalues are shown in bold cases. $: BMI data were missing for 61 subjects (16.6%).
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
antipsychotics might negatively affect the BBB [28,29]: antipsychotics may induce cytotoxic effects and apoptosis of BBB endothelial cells, with a concomitant impairment of barrier functionality [28]. In a manner consistent with this proposed medication effect, chronic antipsychotic treatment resulted in an increased density of microglia in the rat brain [29]. Increased CSF C5 levels may lead to microglial migration and activation, thereby altering immune conditions and inducing neuroinflammation in the CNS. Microglial migration is diminished in the complement C5a receptor knockout mouse [30], consistent with these findings, C5a-peptide vaccines reduced microglial activation and neuroinflammation [31]. From a developmental perspective, increased CSF C5 levels may contribute to complement-mediating synapse pruning, especially in adolescence. Peripheral mRNA expression levels of C5 were associated with superior frontal cortical thickness in patients with schizophrenia [13]. Moreover, structural alteration of C4, which is located upstream of C5, has been suggested to induce synapse pruning [11]. Interestingly, BMI positively correlated with C5 levels in patients with MDD, patients with BPD, and healthy controls. The mechanism of this correlation is unclear; however, several lines of evidence suggest that obesity may promote neuroinflammation and BBB hyperpermeability. Further, obesity appears to cause lowgrade chronic inflammation in peripheral tissues and blood circulation [32,33]. These previous studies did not evaluate neuroinflammation, but a separate report suggests that chronic inflammation correlates with BBB hyperpermeability in certain conditions related to the presence of CSF abnormalities [34]. Indeed, a high fat diet in mice induces neuroinflammation via BBB disruption at the hippocampus [35]. Those findings are especially interesting in the context of MDD pathophysiology, as obesity increases the risk of depression [36]; in a related manner, patients with MDD carry a significantly higher risk of developing obesity, relative to health individuals [37], and neuroinflammation has repeatedly been reported in MDD [2,38,39]. Future basic research should evaluate the effect of obesity on neuroinflammation and CSF markers, including in the context of medication. CSF C5 levels may reflect the pathophysiology of a subset of mental diseases. Although, in the present study, there was no association between CSF C5 levels and symptom scores (or scores of subscales) in each psychiatric group, a thorough investigation of the symptoms, brain imaging and physiological tests of high C5 patients is required in a future study. As novel treatment options for high C5 patients, C5, C5a, C5aR (induces microglia migration), and CD59 (inhibits membrane attack complex formation) may be promising targets to inhibit the complement cascade [40]. In the present study, the patients were biased towards milder forms of mental illness and the majority of patients were medicated. CSF was, therefore, collected in a “real-world setting”. Notably, the number of bipolar subjects was relatively small. CSF C5 levels were significantly increased in the patients with MDD and schizophrenia, compared with the healthy controls. The rate of individuals with an “abnormally high C5 level” was increased in all psychiatric groups, relative to the health controls. Our results suggest that the activated complement system may contribute to neurological pathogenesis in a portion of patients with major psychiatric disorders.
Conflicts-of-Interest The authors declare that they have no conflict of interest.
5
Funding This study was supported by Intramural Research Grants for Neurological and Psychiatric Disorders at NCNP [grant number 271] (H.K.), Strategic Research Program for Brain Sciences from Japan Agency for Medical Research and development (AMED) [grant number 16dm0107100h001] (H.K.), and Health and Labor Sciences Research Grants for Comprehensive Research on Persons with Disabilities from AMED [grant number 15dK0310061h001] (H.K.). These funding sources were involved only in the financial support. There was a significant increase in C5 levels in the patients with MDD (p < 0.001) and schizophrenia (p ¼ 0.001), compared with the healthy controls. Horizontal solid lines indicate the mean values of the diagnostic groups. The dotted line indicates the 95th percentile value of the control group (424.0 ng/mL). Acknowledgements We thank Mr. Ikki Ishida, Ms. Junko Matsuo, Ms. Moeko Hiraishi, Ms. Yuuki Yokota, Mr. Ryo Matsmura, Ms. Tomoko Kurashimo, Ms. Naoko Ishihara, Mr. Takahiro Tomizawa, and Ms. Chie Kimizuka for their assistance in recruitment and clinical assessments of participants. We also thank Ms. Misao Nakano and Mr. Shokichi Tajima for their sample management. References [1] B.N. Cuthbert, T.R. Insel, Toward the future of psychiatric diagnosis: the seven pillars of RDoC, BMC Med. 11 (2013) 126. [2] K. Hattori, M. Ota, D. Sasayama, S. Yoshida, R. Matsumura, T. Miyakawa, Y. Yokota, S. Yamaguchi, T. Noda, T. Teraishi, H. Hori, T. Higuchi, S. Kohsaka, Y. Goto, H. Kunugi, Increased cerebrospinal fluid fibrinogen in major depressive disorder, Sci. Rep. 5 (2015) 11412. [3] D.R. Goldsmith, M.H. Rapaport, B.J. Miller, A meta-analysis of blood cytokine network alterations in psychiatric patients: comparisons between schizophrenia, bipolar disorder and depression, Mol. Psychiatr. 21 (2016) 1696e1709. [4] D. Sasayama, C. Wakabayashi, H. Hori, T. Teraishi, K. Hattori, M. Ota, M. Ishikawa, K. Arima, T. Higuchi, N. Amano, H. Kunugi, Association of plasma IL-6 and soluble IL-6 receptor levels with the Asp358Ala polymorphism of the IL-6 receptor gene in schizophrenic patients, J. Psychiatr. Res. 45 (2011) 1439e1444. [5] G.M. Khandaker, L. Cousins, J. Deakin, B.R. Lennox, R. Yolken, P.B. Jones, Inflammation and immunity in schizophrenia: implications for pathophysiology and treatment, The Lancet. Psychiatry 2 (2015) 258e270. [6] D. Sasayama, K. Hattori, C. Wakabayashi, T. Teraishi, H. Hori, M. Ota, S. Yoshida, K. Arima, T. Higuchi, N. Amano, H. Kunugi, Increased cerebrospinal fluid interleukin-6 levels in patients with schizophrenia and those with major depressive disorder, J. Psychiatr. Res. 47 (2013) 401e406. [7] A.K. Wang, B.J. Miller, Meta-analysis of cerebrospinal fluid cytokine and tryptophan catabolite alterations in psychiatric patients: comparisons between schizophrenia, bipolar disorder, and depression, Schizophr. Bull. 44 (2018) 75e83. [8] D. Davalos, J.K. Ryu, M. Merlini, K.M. Baeten, N. Le Moan, M.A. Petersen, T.J. Deerinck, D.S. Smirnoff, C. Bedard, H. Hakozaki, S. Gonias Murray, J.B. Ling, H. Lassmann, J.L. Degen, M.H. Ellisman, K. Akassoglou, Fibrinogen-induced perivascular microglial clustering is required for the development of axonal damage in neuroinflammation, Nat. Commun. 3 (2012) 1227. [9] H. Kuroda, K. Fujihara, R. Takano, Y. Takai, T. Takahashi, T. Misu, I. Nakashima, S. Sato, Y. Itoyama, M. Aoki, Increase of complement fragment C5a in cerebrospinal fluid during exacerbation of neuromyelitis optica, J. Neuroimmunol. 254 (2013) 178e182. [10] T.M. Woodruff, R.R. Ager, A.J. Tenner, P.G. Noakes, S.M. Taylor, The role of the complement system and the activation fragment C5a in the central nervous system, NeuroMolecular Med. 12 (2010) 179e192. [11] A. Sekar, A.R. Bialas, H. de Rivera, A. Davis, T.R. Hammond, N. Kamitaki, K. Tooley, J. Presumey, M. Baum, V. Van Doren, G. Genovese, S.A. Rose, R.E. Handsaker, M.J. Daly, M.C. Carroll, B. Stevens, S.A. McCarroll, Schizophrenia risk from complex variation of complement component 4, Nature 530 (2016) 177e183. [12] W. Kakegawa, N. Mitakidis, E. Miura, M. Abe, K. Matsuda, Y.H. Takeo, K. Kohda, J. Motohashi, A. Takahashi, S. Nagao, S. Muramatsu, M. Watanabe, K. Sakimura, A.R. Aricescu, M. Yuzaki, Anterograde C1ql1 signaling is required in order to determine and maintain a single-winner climbing fiber in the mouse cerebellum, Neuron 85 (2015) 316e329. [13] D.M. Allswede, A.B. Zheutlin, Y. Chung, K. Anderson, C.M. Hultman, M. Ingvar,
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131
6
T. Ishii et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e6
[14] [15]
[16]
[17] [18]
[19] [20] [21]
[22]
[23]
[24]
[25]
[26]
T.D. Cannon, Complement gene expression correlates with superior frontal cortical thickness in humans, Neuropsychopharmacology 43 (2018) 525e533. A. Mubarik, H. Tohid, Frontal lobe alterations in schizophrenia: a review, Trends Psychiatr. Psychother. 38 (2016) 198e206. S. Hidese, M. Ota, J. Matsuo, I. Ishida, M. Hiraishi, T. Teraishi, K. Hattori, H. Kunugi, Association between the scores of the Japanese version of the Brief Assessment of Cognition in Schizophrenia and whole-brain structure in patients with chronic schizophrenia: a voxel-based morphometry and diffusion tensor imaging study, Psychiatr. Clin. Neurosci. 71 (2017) 826e835. D.V. Sheehan, Y. Lecrubier, K.H. Sheehan, P. Amorim, J. Janavs, E. Weiller, T. Hergueta, R. Baker, G.C. Dunbar, The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10, J. Clin. Psychiatr. 59 (Suppl 20) (1998) 22e33 quiz 34e57. S.R. Kay, A. Fiszbein, L.A. Opler, The positive and negative syndrome scale (PANSS) for schizophrenia, Schizophr. Bull. 13 (1987) 261e276. J.B. Williams, K.A. Kobak, P. Bech, N. Engelhardt, K. Evans, J. Lipsitz, J. Olin, J. Pearson, A. Kalali, The GRID-HAMD: standardization of the Hamilton depression rating scale, Int. Clin. Psychopharmacol. 23 (2008) 120e129. R.C. Young, J.T. Biggs, V.E. Ziegler, D.A. Meyer, A rating scale for mania: reliability, validity and sensitivity, Br. J. Psychiatry 133 (1978) 429e435. T. Inada, A. Inagaki, Psychotropic dose equivalence in Japan, Psychiatr. Clin. Neurosci. 69 (2015) 440e447. P. Gudmundsson, I. Skoog, M. Waern, K. Blennow, S. Palsson, L. Rosengren, D. Gustafson, The relationship between cerebrospinal fluid biomarkers and depression in elderly women, Am. J. Geriatr. Psychiatr. 15 (2007) 832e838. K. Bechter, H. Reiber, S. Herzog, D. Fuchs, H. Tumani, H.G. Maxeiner, Cerebrospinal fluid analysis in affective and schizophrenic spectrum disorders: identification of subgroups with immune responses and blood-CSF barrier dysfunction, J. Psychiatr. Res. 44 (2010) 321e330. M. Rothermundt, V. Arolt, M. Wiesmann, U. Missler, M. Peters, S. Rudolf, H. Kirchner, S-100B is increased in melancholic but not in non-melancholic major depression, J. Affect. Disord. 66 (2001) 89e93. M.L. Schroeter, H. Abdul-Khaliq, A. Diefenbacher, I.E. Blasig, S100B is increased in mood disorders and may be reduced by antidepressive treatment, Neuroreport 13 (2002) 1675e1678. B.S. Jang, H. Kim, S.W. Lim, K.W. Jang, D.K. Kim, Serum S100B levels and major depressive disorder: its characteristics and role in antidepressant response, Psychiatry Investig. 5 (2008) 193e198. O. Ambree, V. Bergink, L. Grosse, J. Alferink, H.A. Drexhage, M. Rothermundt, V. Arolt, T.K. Birkenhager, S100B serum levels predict treatment response in patients with melancholic depression, Int. J. Neuropsychopharmacol. 19 (2015) pyv103.
[27] T.A. Pollak, S. Drndarski, J.M. Stone, A.S. David, P. McGuire, N.J. Abbott, The blood-brain barrier in psychosis, The lancet. Psychiatry 5 (2018) 79e92. [28] E. Elmorsy, L.M. Elzalabany, H.M. Elsheikha, P.A. Smith, Adverse effects of antipsychotics on micro-vascular endothelial cells of the human blood-brain barrier, Brain Res. 1583 (2014) 255e268. [29] M.C. Cotel, E.M. Lenartowicz, S. Natesan, M.M. Modo, J.D. Cooper, S.C. Williams, S. Kapur, A.C. Vernon, Microglial activation in the rat brain following chronic antipsychotic treatment at clinically relevant doses, Eur. Neuropsychopharmacol 25 (2015) 2098e2107. [30] D. Song, M.E. Sulewski Jr., C. Wang, J. Song, R. Bhuyan, J. Sterling, E. Clark, W.C. Song, J.L. Dunaief, Complement C5a receptor knockout has diminished light-induced microglia/macrophage retinal migration, Mol. Vis. 23 (2017) 210e218. [31] C. Landlinger, L. Oberleitner, P. Gruber, B. Noiges, K. Yatsyk, R. Santic, M. Mandler, G. Staffler, Active immunization against complement factor C5a: a new therapeutic approach for Alzheimer's disease, J. Neuroinflammation 12 (2015) 150. [32] G.S. Hotamisligil, Inflammation and metabolic disorders, Nature 444 (2006) 860e867. [33] M.F. Gregor, G.S. Hotamisligil, Inflammatory mechanisms in obesity, Annu. Rev. Immunol. 29 (2011) 415e445. [34] E. Elwood, Z. Lim, H. Naveed, I. Galea, The effect of systemic inflammation on human brain barrier function, Brain Behav. Immun. 62 (2017) 35e40. [35] Z. Tucsek, P. Toth, D. Sosnowska, T. Gautam, M. Mitschelen, A. Koller, G. Szalai, W.E. Sonntag, Z. Ungvari, A. Csiszar, Obesity in aging exacerbates blood-brain barrier disruption, neuroinflammation, and oxidative stress in the mouse hippocampus: effects on expression of genes involved in beta-amyloid generation and Alzheimer's disease, J. Gerontol. A Biol. Sci. Med. Sci 69 (2014) 1212e1226. [36] F.S. Luppino, L.M. de Wit, P.F. Bouvy, T. Stijnen, P. Cuijpers, B.W. Penninx, F.G. Zitman, Overweight, obesity, and depression: a systematic review and meta-analysis of longitudinal studies, Arch. Gen. Psychiatr. 67 (2010) 220e229. [37] R.B. Mansur, E. Brietzke, R.S. McIntyre, Is there a “metabolic-mood syndrome”? A review of the relationship between obesity and mood disorders, Neurosci. Biobehav. Rev. 52 (2015) 89e104. [38] H. Kunugi, H. Hori, S. Ogawa, Biochemical markers subtyping major depressive disorder, Psychiatr. Clin. Neurosci. 69 (2015) 597e608. [39] M.L. Monje, H. Toda, T.D. Palmer, Inflammatory blockade restores adult hippocampal neurogenesis, Science 302 (2003) 1760e1765. [40] D. Ricklin, J.D. Lambris, Complement-targeted therapeutics, Nat. Biotechnol. 25 (2007) 1265e1275.
Please cite this article in press as: T. Ishii, et al., Increased cerebrospinal fluid complement C5 levels in major depressive disorder and schizophrenia, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.02.131