Introduction

Depression, a mood disorder, can be classified as unipolar or bipolar. Currently, depression greatly threatens the mental health of patients and accounts for the third largest share of the world’s burden of disease in terms of years lost to disability (GBD 2016 Disease and Injury Incidence and Prevalence Collaborators, 2017). In recent years, immunological mechanisms and inflammation are increasingly implicated in the pathogenesis of depressive symptoms (Bhattacharya and Drevets 2017). Some reports suggest that activation of the peripheral immune system has been consistently associated with depression (Chamberlain et al. 2019). However, the underlying pathobiology behind inflammation in depression is not fully understood, and further investigations are required to identify specific diagnostic biomarkers. As an important step towards immunotherapy for treatment of depression, it is essential to explore the peripheral immune biomarkers which are linked closely to the development of this disease.

The interplay of chronic mild stress, neuroinflammation, and altered immune reactivity has been shown to be important for pathophysiology of depressive disorders (Vasileva et al. 2019). There are an increasing number of studies suggesting that the nervous system is closely related to the immune system (Boorman et al. 2016; Duda et al. 2019). Depression can cause T/B lymphocytes imbalance and functional decline of natural killer T (NKT) cells, subsequently leading to low immune function, infection susceptibility, tumor formation, and other diseases (Boorman et al. 2016). In addition, stress exposure causes the dysfunction of peripheral T lymphocytes, and the lack of CD4+ T cells protects mice with xanthine metabolic disorder from stress-induced anxiety-like behavior, implying that CD4+ T cells trigger anxiety-like behavior (Fan et al. 2019). Regulatory immune cells can be divided into regulatory T cells and regulatory B cells, which play a negative regulatory role in immune regulation and can induce immune tolerance (Pierini et al. 2015; Correale et al. 2014). B lymphocytes also have been reported to play a significant role in suppressing inflammatory immune responses and preventing autoimmunity through the production of interleukin-10 (IL-10) (Mavropoulos et al. 2016).

Cell adhesion molecules (CAMs) are distributed to the extracellular environment for the adhesion of epithelial cells (Tang et al. 2019). CHL1 is a transmembrane CAM with unique functions in neural signal transduction pathway (Schlatter et al. 2008). Recent genome-wide expression profiling of human lymphoblastoid cell lines has identified CHL1 as a putative SSRI antidepressant response biomarker (Morag et al. 2011; Oved et al. 2012). CHL1 belongs to the immunoglobulin (Ig) superfamily cell adhesion molecules at 3p26.3, also known as L1 family cell adhesion molecules (Senchenko et al. 2011). These adhesion molecules consist of six Ig-like domain and five fibronectin type III extracellular domains (Kenwrick et al. 2000). CHL1 is expressed at the cell surface throughout the central and peripheral nervous system, mainly in areas of high plasticity (Probst-Schendzielorz et al. 2015). In the developing nervous system, CHL1 plays a pivotal role in the regulation of neuronal differentiation and survival, neurite outgrowth, and axon guidance (Barao et al. 2015; Schmalbach et al. 2015). In the mature nervous system, CHL1 is involved in the regulation of synaptic activity and plasticity, including learning and memory (Pratte and Jamon 2009; Kolata et al. 2008). Single-nucleotide polymorphisms, deletions, and duplications in the CHL1 gene were associated with a higher risk of developing various neurological disorders including Parkinson’s disease, schizophrenia, autism, epilepsy, and intellectual disability (Shoukier et al. 2013; Palumbo et al. 2015; Li et al. 2016; Guseva et al. 2018; Shaltout et al. 2013). At the present, little is known about the association of CHL1 expressed by lymphocytes with the prognosis of depression.

In the current study, we hypothesize that the level of CHL1 in blood lymphocytes of depressive patients is consistent with CHL1 in the brain using our well-established chronic unpredictable mild stress (CUMS) depression mouse model. We propose that the CHL1 levels in lymphocytes of depressive patients will mirror the CHL1 levels in the brain. In order to test this hypothesis and characterize lymphocyte subtypes which predominately express CHL1 during depression, we first assessed the mRNA and protein expression levels of CHL1 in the brain of mice under chronic stress regime, and then we examined the mRNA and protein expression levels of CHL1 in lymphocytes isolated from depressive patients, as well as their serum CHL1 level. In addition, we explored the CHL1 level in peripheral B cell subsets of depressive patients with and without antidepressant treatment compared with healthy controls.

Material and Methods

CUMS Model

The study was carried out on 2-month-old male C57BL/6 J mice (23–27 g). All the animals were housed in groups of three to five per standard cage with food and water for at least 2 weeks before commencement of experimental tests. The animals were maintained under standard laboratory conditions (12 h light/dark cycle, 22 ± 1 °C, 52 ± 2% humidity). All experiments were performed in accordance with the Animal Management Rules of the Ministry of Health of the People’s Republic of China. All animal experiments were approved by the Animal Research Ethics Committee, Tianjin Medical University. All efforts were made to minimize the number of mice used and their suffering. For each treatment group in these series of experiments, 9–10 mice were used. The detailed CUMS procedures were described in our previous paper (Yang et al. 2017). The experimental schedules are depicted in Fig. 1a.

Fig. 1
figure 1

Effect of chronic treatment with antidepressants and CUMS on depressive and anxiety-like behavior. a Experimental protocol. b CUMS mice showed less sucrose consumption than the control mice. The treatment of FLU and CLO for 14 days significantly reversed the decrease of sucrose preference in CUMS mice. c, d A 4-week period of CUMS exposure resulted in a typical reduction in the total distance traveled and central square in mice, as compared with the controls in OFT. FLU and CLO treatment increased the total distance traveled and central square in the CUMS group. e, f FLU and CLO exposure significantly induced immobility time in FST and increased the number of climbs compared with VEH. g Treatment of antidepressant of CUMS mice significantly decreased anxiety-like behavior as displayed by increase in percentage of open/closed zone ratio in the EOM. CUMS, chronic unpredictable mild stress; VEH, vehicle; FLU, fluoxetine; CLO, clozapine; OFT, open-field test; FST, forced swim test; EOM, elevated zero maze. *p < 0.05, **p < 0.01, ***p < 0.001; #p < 0.05, ##p < 0.01, ###p < 0.001; all data are presented as mean ± SEM

Full size image

Experimental Design and Antidepressant Drug Treatment

Fluoxetine hydrochloride (FLU) was purchased from Lilly S. A. Spain. Clozapine (CLO) was made by the Jiangsu Nhwa Pharmaceutical Group. Both FLU and CLO were dissolved separately in sterile saline. The animals were randomly divided: (group I) unstressed mice with free access to standard food and water, used as normal control (n = 10); (group II) mice after CUMS exposure (n = 10) for 28 days; (group III) CUMS-induced depression models treated with daily intraperitoneal (I.P.) injections of sterile saline (0.01 ml/g) for 14 days, used as a vehicle (VEH) group (n = 10); (group IV) CUMS-induced depression models treated with daily I.P. injections of FLU (n = 10) (18 mg/kg; 0.01 ml/g) for 14 days; and (group V) CUMS-induced depression models treated with daily I.P. injections of CLO (n = 10) (20 mg/kg; 0.01 ml/g) for 14 days. Both groups IV and V were used as the positive controls for antidepressant action.

Behavioral Testing (SPT, OFT, FST, O Maze)

Sucrose preference (SPT), open-field (OFT), forced swimming (FST), and O maze tests were performed as previously described (Yang et al. 2017; Yang et al. 2020). Detailed procedures are described in supplementary material. During the behavioral experiments, there were some accidents resulted in data loss from some rodents. Specifically, two bottles were found leaking after SPT, one in VEH group and one in CLO group. During the O maze test, one mouse in FLU group always jumped to the open arm. The data from these animals were excluded. So control (n = 10), CUMS (n = 10), VEH (n = 9), FLU (n = 9), and CLO (n = 9) were used in the analysis of behavioral data. At the end point, animals were humanely killed by an overdose of carbon dioxide via inhalation, and brain tissues were isolated for total RNA and protein extraction and stored at − 80 °C.

Human Subjects

Forty-four patients, 18–60 years old, both male and female, were recruited from Tianjin Mental Health Center in this study. Major depression was diagnosed according to the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) criteria. The control population consisted of forty-two healthy age- and gender-matched volunteers at the Second Hospital of Tianjin Medical University. We declare that our study was carried out in accordance with the latest version of the Declaration of Helsinki. The Ethics Committee of Tianjin Mental Hospital approved all procedures (KY2018K015). Patients were enrolled to participate in the study after they had signed the informed consents for participation. The patients were all drug-naive. We have excluded the participants who had history of autoimmune diseases, chronic inflammatory diseases, or other immune-associated disorders in this study. After having obtained the permission, peripheral blood samples were collected from healthy volunteers and patients. The details of the blood collection for these studies were given in the supplementary material.

Enzyme-Linked Immunosorbent Assay of CHL1 (Isolation of Serum)

Serum was assayed for CHL1 in duplicate, using an ELISA assay kit (L1CAM, cat.#SEQ118Hu, China), according to the protocol provided by the manufacturer. Briefly, ELISA plates were coated with protein G-purified mouse anti-CHL1 monoclonal antibody (2 μg/ml in coating buffer), blocked with PBS-BSA, loaded with serum at 1:5 dilution, detected with detection antibodies, and incubated with TMB, and OD values were obtained at 450 nm.

Quantitative Real-Time PCR

Total RNA was extracted from the entire cortex, hippocampus tissues from the rodents, and peripheral blood mononuclear cells (PBMCs) from participant donors, which were extracted by TRIzol® reagent (Sigma), following the manufacturer’s recommendation. The RNA samples were dissolved in 30 μl of DEPC-H2O and subsequently stored at − 80 °C. The cDNA was obtained using the reverse transcription kit (Thermo-Fisher Scientific, CA, USA). Quantitative real-time PCR was performed with SYBR Green (Takara, Japan) on CFX96 Touch™ Deep Well Real-Time PCR Detection system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Q-PCR was performed as follows: 95 °C for 3 min, and 39 cycles of 95 °C for 10 s and 60 °C for 30 s. The experiment was repeated in triplicate. Gene expression level of CHL1 was detected by running quantitative real-time PCR. The gene expression of β-actin/GAPDH was detected as internal control for total RNA used. Data were processed using the 2-ΔΔCTmethod. The primer sequences used are shown in Table 1.

Table 1 The following primer sequences were used
Full size table

Western Blotting Analysis (Protein Isolation of Lymphocytes)

Protein was extracted from the entire cortex, hippocampus tissues from the rodents, and PBMCs from participant donors harvested at indicated times using radio immunoprecipitation assay (RIPA) lysis buffer supplemented with proteinase inhibitors. Protein concentrations were measured using the BCA Protein Assay Kit (Thermo Scientific, Waltham, MA). Equal amounts of protein were separated by 10% SDS/PAG, followed by immunoblotting with the primary antibodies (rabbit anti-CHL1 antibody 1:1000, cat.#ab106269, abcam, China; mouse β-actin 1:1000, cat.#CW0097, CoWin Biosciences, China; rabbit anti-GAPDH antibody, 1:5000, cat.#ab 9485, abcam, China). Membranes were then incubated with peroxidase-conjugated secondary antibody, and specific bands were detected with Image J software (NIH, USA).

Flow Cytometry

To measure the CHL1 expression in PBMCs from healthy volunteers and depressive patients, human PBMCs were isolated from peripheral blood samples of healthy donors (HD) and depressive patients by lymphoprep separation liquid (Stemcell Technology, Norway). Freshly isolated PBMCs were then counted manually with a hemocytometer, and cell viability was measured by trypan blue (> 96%). PBMCs were stained with fluorescently conjugated monoclonal antibodies (mAbs) at 4 °C for 30 min as follows, respectively: BV421 anti-human CD3 (cat.#ab562427, BD), ECD anti-human CD4 (cat.#ab6604727, BC), APC-CY7 anti-human CD8 (cat.#ab641400, BD), FITC anti-human CD20 (cat.#ab347673, BD), PC7 anti-human CD19 (cat.#abIM3628, BC), V500 anti-human CD45 (cat.#ab560777, BD), PE anti-human CD171 (CHL1, cat.#ab371604, Biolegend), and APC mouse IgG1(cat.#ab400119, Biolegend). After rinse, cells were read on the flow cytometer (Cystek, Fremont, CA, USA), and data were analyzed with FlowJo vX0.7 software. Single-stained cells or OneComp eBeads (eBioscience, San Diego, CA) were used for compensation calculations. Unstained cells, cells with 7-AAD, and fluorescence minus one (FMO) controls were used for cytometry and gating set up.

Statistical Analyses

Data were presented as mean ± SEM and analyzed by SPSS 19.0 version. All statistical analysis was performed with Prism software from GraphPad (San Diego, CA, USA). For difference comparison within groups, Student’s t test was used for the two groups, one-way analysis of variance (ANOVA) for more than two groups, and two-way ANOVA for two groups with two factors. The p < 0.05 was considered statistically significant.

Results

Effects of Antidepressant Drugs on Depressive and Anxiety-Like Behavior

During the analysis of behavioral data, there were VEH (n = 9), FLU (n = 9), and CLO (n = 9). The effect of antidepressant drugs on depressive and anxiety-like behavior is shown in Fig. 1. After 4 weeks, the percentage of sucrose consumption in the sucrose preference test was significantly decreased in the CUMS mice group (n = 10) compared with the control group (n = 10) (Student’s t test, ***p < 0.001, t = 10.541, Fig. 1b), and antidepressant (FLU) and antipsychotic drugs (CLO) treatment reversed the decrease (###p < 0.001, ###p < 0.001, F(2,24) = 27.968, Fig. 1b). In the OFT, a 4-week period of CUMS exposure resulted in a typical reduction in the total distance traveled (***p < 0.001, t = 4.482, Fig. 1c) and central square (**p = 0.002, t = 3.559, Fig. 1d) in mice, as compared with the controls. We have found that antidepressants significantly reversed these behavioral impairments in CUMS mice compared with the VEH group, respectively (##p = 0.006, ##p = 0.004, F(2,24) = 6.515, Fig. 1c; ###p < 0.001, ###p < 0.001, F(2,24) = 14.435, Fig. 1d). The results showed that antidepressant treatment impacted the locomotor activity and exploratory activity in the OFT. In addition, CUMS exposure caused the increased immobility time (***p < 0.001, t = −10.255, Fig. 1e) in the FST, and treatment with antidepressant drugs (FLU, CLO) significantly decreased the immobility time in these mice (##p = 0.002, ##p = 0.009, F(2,24) = 6.693, Fig. 1e). At the same time, the number of climbings (***p < 0.001, t = 5.18, Fig. 1f) in the FST was declined in CUMS mice vs the control group, and treatment with antidepressant drugs (FLU, CLO) increased the numbers in these mice (#p = 0.01, ##p = 0.004, F(2,24) = 5.894, Fig. 1f). Analogously, in the EOM, the percentage of time spent in the open arms/closed zone was significantly decreased following CUMS compared with the controls (**p = 0.002, t = 3.599, Fig. 1g). The results suggested that antidepressants impacted the anxiety activity behaviors compared with the VEH group (##p = 0.004, #p = 0.02, F(2,24) = 5.636, Fig. 1g).

Effects of Antidepressant and Antipsychotic Drugs on CHL1 in the Brain of CUMS Mice

Expression of CHL1 in the cortex and hippocampus was evaluated by real-time PCR (Fig. 2a, c) and Western blotting (Fig. 2b, d). The mRNA and protein level of CHL1 were lower in the CUMS group than that of the control group both in the cortex (*p = 0.029, t = 2.547, Fig. 2a; ***p < 0.001, t = 8.496, Fig. 2b) and hippocampus (*p = 0.022, t = 2.772, Fig. 2c; ***p < 0.001, t = 10.779, Fig. 2d). In order to assess whether antidepressant and antipsychotic drugs could regulate the expression of CHL1, we examined CHL1 expression in CUMS mice treated with FLU and CLO. The mRNA level of CHL1 in the cortex and hippocampus was significantly increased in the FLU group (###p < 0.001, F(2,24) = 10.965, Fig. 2a; #p = 0.027, F(2,24) = 11.074, Fig. 2c) and CLO group (##p = 0.003, F(2,24) = 10.965, Fig. 2a; ###p < 0.001, F(2,24) = 11.074, Fig. 2c) compared with the VEH group. Consistently, the protein level of CHL1 was also increased in the treatment with FLU (#p = 0.03, F(2,24) = 20.264, Fig. 2b; ##p = 0.009, F(2,24) = 48.982, Fig. 2d) and CLO (###p < 0.001, F(2,24) = 20.264, Fig. 2b; ###p < 0.001, F(2,24) = 48.982, Fig. 2d) compared with the VEH group in the cortex and hippocampus.

Fig. 2
figure 2

Effect of antidepressants on the expression of CHL1 mRNA and protein in the cortex and hippocampus. a, b The mRNA and protein expression of CHL1 are downregulated in the cortex. Treatment of CUMS mice with FLU and CLO reversed CHL1 mRNA and protein expression. c, d CHL1 mRNA and protein expression are downregulated in the hippocampus. Treatment of CUMS mice with FLU and CLO reversed CHL1 mRNA and protein expression. β-actin was detected as loading control. *p < 0.05, **p < 0.01, ***p < 0.001; #p < 0.05, ##p < 0.01, ###p < 0.001; all data are presented as mean ± SEM

Full size image

CHL1 Is Downregulated in the Serum and PBMCs in the Patients with Major Depressive Disorder

The demographic data of the enrolled depressive patients and normal donors was shown in Table 2. There were no significant differences of the average ages and WBC between normal group and patient group (p = 0.27, t = − 1.113; p = 0.318, t = − 1.136, Student’s t test). To examine whether CHL1 in the PBMCs of the enrolled depressive patients was downregulated, we performed real-time PCR on the cell lysates of PBMCs from patients and controls. The CHL1 mRNA level in PBMCs (***p < 0.001, t = 8.229, Fig. 3a) was reduced in the depressive group (n = 44) compared with the normal group (n = 42). Moreover, the protein expression of CHL1 in PBMCs (***p < 0.001, t = 4.583, Fig. 3b) was significantly lower in the depressive group (n = 44) than in the normal group (n = 42). Finally, antidepressant therapy (n = 18) induced high amount of CHL1 production not only at the mRNA level (###p < 0.001, F(2,101) = 43.208, Fig. 3a) but also at the protein level (###p < 0.001, F(2,101) = 181.581, Fig. 3b) compared with the depressive group (n = 44).

Table 2 Baseline data of the enrolled depressive patients and normal donors
Full size table
Fig. 3
figure 3

The CHL1 level in PBMCs and serum is reduced in depression. a The PBMCs level of CHL1 mRNA was significantly decreased in major depressive patients compared with health controls. There was a significant difference between major depressive patients and treatment group. b CHL1 protein was significantly increased in therapy patients when compared with major depressive patients. Antidepressant therapy induced significantly high amount of CHL1 production. c The ELISA result of CHL1 was compared among health controls, major depressive patients, and therapy group. d The correlation between CHL1 and HRSD scores. PBMCs, peripheral blood mononuclear cells; ELISA, enzyme-linked immunosorbent assay; HRSD, Hamilton Rating Scale for Depression. *p < 0.05, **p < 0.01, ***p < 0.001; #p < 0.05, ##p < 0.01, ###p < 0.001; all data are presented as mean ± SEM

Full size image

ELISA result showed that the CHL1 level in the depressive patients (n = 44) was significantly decreased in comparison with the normal donors (n = 42) (**p = 0.001, F(2,101) = 13.181, Fig. 3c). Before the release from hospital, the serum level of CHL1 in some enrolled depressive patients was significantly increased in the antidepressant treatment group (n = 18) (###p < 0.001, F(2,101) = 13.181, Fig. 3c) when compared with non-treated group (n = 44). The analysis of correlation between the relative CHL1 level in the serum and HRSD scores in patients was shown in Fig. 3d. There was significant negative correlation between CHL1 (n = 44, r = − 0.3879, p = 0.0093, Fig. 3d) and HRSD scores. The results indicated that the trend of CHL1 change in the PBMCs and serum was similar to that in the brain of CUMS mice (Fig. 2).

Downregulation of Blood Lymphocytes CHL1 on B Cells

To explore the potential contribution of the immune cell imbalance in the internal environment and any subpopulation of lymphocytes involved, we examined the distribution of immune cells and the expression of the CHL1 (CD171) on T/B cells from depressive patients. Peripheral blood lymphocytes from 23 normal donors, 26 depressive patients, and 16 treatment patients were phenotypically analyzed by flow cytometry for the expression of CD45, CD19, CD20, CD3, CD4, and CD8 surface markers (Fig. 4a). We next examined the frequency of the expression of CHL1 (CD171) on T/B cells from depressive patients (Fig. 5).

Fig. 4
figure 4figure 4

Representative density plots and frequency graphs of in immune subsets. a Representative images of flow cytometry for T and B lymphocytes in normal health control, depressive patients and therapy patients, gating of mononuclear cells isolated from the PBMCs. b Frequency of CD45+ lymphocytes was significantly decreased in the depressive group. These CD45+ lymphocytes from patients after treatment were increased compared with the depressive group. c, d Frequency of CD19+ lymphocytes and CD20+ lymphocytes was not statistically different from those in the depressive group compared with the normal group. They were slightly decreased in the treatment group compared with the depressive group. e No difference among the three groups in frequency of CD3+ lymphocytes. f Frequency of CD4+CD8−--CD3+ lymphocytes was significantly increased between the normal group and the depressive group. g Frequency of CD4−CD8+--CD3+ lymphocytes was decreased between the normal group and the depressive group. h Frequency of CD4+CD8+--CD3+ lymphocytes. i CD4/CD8 ratio was significantly increased in depressive patients compared with the normal group, but no significance was seen between the depressive group and the treatment group. *p < 0.05, **p < 0.01, ***p < 0.001; #p < 0.05, ##p < 0.01, ###p < 0.001; all data are presented as mean ± SEM

Full size image
Fig. 5
figure 5figure 5

Histograms and graphs show CHL1 expression on immune subset. CD19/CD20 B cells exhibit high expression of CHL1. a Frequency of CHL1+ cells in CD45+ lymphocytes was increased in the treatment group. And there was no significance from those in the depressive group compared with the normal group. b Frequency of CHL1+ cells in CD19+ B cells was significantly decreased between the normal group and the depressive group. It was dramatically increased in the treatment group. c Frequency of CHL1+ cells in CD20+ B cells was consistent to the frequency of CHL1+ cells in CD19+ B cells. d Frequency of CHL1+ cells in CD3+ T cells. e Frequency of CHL1+ cells in CD4+CD8− --CD3+ T cells. f Frequency of CHL1+ cells in CD4−CD8+ − -CD3+ T cells. g There was a significant difference between the normal group and the depressive group in the frequency of CHL1+ cells in CD4+CD8+ − -CD3+ T cells. *p < 0.05, **p < 0.01, ***p < 0.001; #p < 0.05, ##p < 0.01, ###p < 0.001; all data are presented as mean ± SEM

Full size image

As shown in Fig. 4, in contrast to the normal group, the frequency of CD45+ cells was significantly decreased in the depressive group (*p = 0.037, F(2,62) = 3.145, Fig. 4b). Moreover, the percentage of CD45+ leukocytes from patients after treatment was increased compared with the depressive group (#p = 0.045, F(2,62) = 3.145, Fig. 4b). We observed that CD19+ B cells (p = 0.44, F(2,62) = 1.141, Fig.4c) and CD20+ B cells (p = 0.681, F(2,62) = 0.578, Fig. 4d) were not statistically different from those in the depressive group compared with the normal group. Notably, they were slightly decreased in the treatment group compared with the depressive group (p = 0.411, F(2,62) = 1.141, Fig. 4c; p = 0.478, F(2,62) = 0.578, Fig. 4d). We examined the expression of CD3+ T cells and showed no difference among the three group (p = 0.839, F(2,62) = 0.034; p = 0.816, F(2,62) = 0.034, Fig. 4e). Interestingly, CD4+CD8−-CD3+ T cells from the depressive group (**p = 0.005, F(2,62) = 4.848, Fig. 4f) and the treatment group (p = 0.542, F(2,62) = 4.848, Fig. 4f) showed higher frequency than the normal group. On the opposite, CD4−CD8+-CD3+ T cells from the depressive group (*p = 0.013, F(2,62) = 3.663, Fig. 4g) and the treatment group (p = 0.746, F(2,62) = 3.663, Fig. 4g) showed significantly lower frequency than the normal group. However, cells expressing CD4+CD8+-CD3+ T cells seemed to be increased in the depressive group compared with the normal group (p = 0.109, F(2,62) = 1.316, Fig. 4h), but no significant difference was seen compared with the treatment group (p = 0.447, F(2,62) = 1.316, Fig. 4h). As the result, the ratio of CD4+/CD8+ was increased in peripheral blood of depressive patients compared with the normal group (**p = 0.009, F(2,62) = 3.931, Fig. 4i), but no significance was seen between the depressive group and the treatment group (p = 0.659, F(2,62) = 3.931, Fig. 4i). These results reveal that the change of CD45+ leukocytes, CD4+CD8−-CD3+ T cells, and CD4−CD8+-CD3+ T cells perhaps be associated with depression.

CHL1 (CD171) expression was examined by flow cytometry. As shown in Fig. 5, the expression of CHL1 (CD171) in CD45+ leukocytes was increased in the treatment group (###p < 0.001, F(2,62) = 16.698, Fig. 5a). And there was no significance from those in the depressive group compared with the normal group (p = 0.573, F(2,62) = 16.698, Fig. 5a). The analysis performed by flow cytometry showed the abundance expression of CHL1 (CD171) in CD19+/CD20+ B cells. There were around 17% CHL1 (CD171)+ in CD19+ B cells in the normal group. Percentage of CHL1 (CD171)+ in CD19+ B cells was decreased to 4% in the depressive group (***p < 0.001, F(2,62) = 44.588, Fig. 5b). In contrast, the percentage of CHL1 (CD171)+ in CD19+ B cells was dramatically increased in the treatment group (###p < 0.001, F(2,62) = 44.588, Fig. 5b). Consistently, the trend of CHL1 (CD171)+ in CD19+ B cells and CD20+ B cells (***p < 0.001, F(2,62) = 33.207; ###p < 0.001, F(2,62) = 33.207, Fig. 5c) was extremely similar. The expression of CHL1 (CD171) in the CD3+ T cells (p = 0.146, F(2,62) = 1.131, Fig. 5d), CD4+ T cells (p = 0.605, F(2,62) = 0.766, Fig. 5e), and CD8+ T cells (p = 0.229, F(2,62) = 0.994, Fig. 5f) was not different between the depressive group and the normal group. The expression of CHL1 (CD171) after treatment administration did not change in the CD3+ T cells (p = 0.737, F(2,62) = 1.131, Fig. 5d), CD4+ T cells (p = 0.221, F(2,62) = 0.766, Fig. 5e), and CD8+ T cells (p = 0.893, F(2,62) = 0.994, Fig. 5f). However, the CHL1 (CD171) expression in the CD4+CD8+ T cells of the depressive group was significantly higher than the normal group (***p < 0.001, F(2,62) = 9.119, Fig. 5g). No significant change of CHL1 in CD4+CD8+ T cells was seen between the depressive group and the treatment group (p = 0.858, F(2,62) = 9.119, Fig. 5g). Taken together, the results showed that CHL1 (CD171) was downregulated in the CD19+ and CD20+ B cell subpopulation, which could be recovered after treatment.

Discussion

The present study demonstrated that the protein and mRNA levels of CHL1 were downregulated in the brain of mice with chronic stress. Consistent with the animal experimental studies, the expression of CHL1 in the serum and PBMCs was also decreased in the depressive patients examined. CHL1 was preferentially expressed in CD19+/CD20+ B cells in the peripheral blood of normal subjects but was decreased in depressive patients. The treatment with antidepressants significantly reversed the biomarker in both rodents and patients. No significant reduction in the CHL1 expression in CD3+, CD4+, and CD8+ T cell subpopulations was found after different treatment. Our study suggests that CHL1 may contribute to pathogenesis of major depression and the dysfunction of immune cells and can be a tentative biomarker for depression.

The Level of CHL1 Is Decreased in the Brain of Mice with Chronic Stress and in PBMCs from Depressive Patients

Over 30% of depressed patients fail to reach sufficient remission after 2–4 weeks of SSRI treatment (Gelenberg 2010). Moreover, it usually takes several weeks to know if antidepressant drug treatment works in a patient. Using genome-wide expression profiling of human LCLs, CHL1 has been reported to be as a leading tentative SSRI sensitivity biomarker (Oved et al. 2012). In our experimental mice models, OFT and EOM tests showed both depressive and anxiety-like behavior, as reflected by a decreased locomotor activity, travel distance in the central square of the OFT, and a decreased time spent in the open arms. However, depressive behavior can be distinguished from anxiety-like behavior through SPT and FST examinations in which mice with depression-like phenotype have less sucrose consumption and a longer immobility time (Zhong et al. 2018). We confirmed that in our CUMS model, there were both depressive and anxiety-like behaviors. In this study, we found that the mRNA and protein level of CHL1 were reduced in the cortex and hippocampus after modeling of mild chronic stress, which is consistent with previous studies (Desarnaud et al. 2008; Laifenfeld et al. 2005a). These results indicate that chronic stress not only elicits depression and anxiety-like phenotypes but also influences expression of CHL1.

Consistent with the animal study, the expression of CHL1 was also decreased in the serum and PBMCs of the depressive patients, suggesting the reduction in CHL1 is a systemic reaction in the depressed patients. In contrast, the expression of CHL1 mRNA and protein was reported to be increased in the prefrontal cortex but decreased in the ventral parieto-occipital cortex, in the depressed patients (Laifenfeld et al. 2005b). In addition, another study reported reciprocal alteration in the expression of NCAM-140, and CHL1 mRNAs could be state dependent and associated with the pathophysiology of bipolar disorder (Wakabayashi et al. 2008). In this study, we collected peripheral blood samples from unipolar and bipolar patients. Our main finding was that the level of CHL1 in the serum was significantly reduced in the depressed patients. In this group, the level of CHL1 had a negative correlation with HRSD-21 scores for major depression severity. These results implicated that a reduction in CHL1 expression is associated with depression severity and may play important roles in mood disorders (Wakabayashi et al. 2008) and multiple neurobiological functions (Sandi 2004; Hochstrasser et al. 2010).

Antidepressant (FLU) and Antipsychotic (CLO) Drugs Restore the Level of CHL1

To further confirm the link between CHL1 and mood regulation, we measured whether antidepressant (FLU) and antipsychotic (CLO) drugs can change the CHL1 level in the brain. Our data showed that the first-line antidepressant drug SSRI (FLU) (Wong et al. 1974) reversed the CUMS-induced depression and anxiety-like behaviors and normalized the CHL1 expression in the cortex and hippocampus of mice brains. These results are consistent with previous studies reporting that chronic antidepressant treatments including FLU increased the MAP-2 (Yang et al. 2015) and BDNF transcriptional levels in rodent brains (Yang et al. 2020). Moreover, CLO reversed depressive and anxiety-like behavior resulted in an increased sucrose preference, total distance traveled, central square, and a decreased immobility time in the FST. CLO is a drug which is used in schizophrenia and other psychotic problems. It mainly blocks receptors in the brain for several neurotransmitter substances including serotonin, D2, D4, and NMDA receptors (Yang et al. 2015). Our study provides the first evidence supporting a novel mechanism underlying the efficacy of FLU and CLO in the treatment of depression, that is, to increase the CHL1 level in the brain of depressive subjects. The result also supports the link between CHL1 and serotoninergic systems in mood regulation.

Several reports demonstrated that the CHL1 level might predict responsiveness to SSRIs (Probst-Schendzielorz et al. 2015; Oved et al. 2012). They found that CHL1 was produced at lower levels in the most responsive cell lines and at higher levels in the least responsive cell lines from 58 depressed patients (Probst-Schendzielorz et al. 2015). They thought that paroxetine caused increased production of the ITGB3gene—whose protein product is thought to interact with CHL1 to promote the development of new neurons and synapses (Probst-Schendzielorz et al. 2015). However, a series of studies have shown that chronic stress can markedly affect hippocampal cell adhesion molecules (CAMs) expression (Bisaz et al. 2009). A chronic restraint procedure was found to diminish NCAM but increase expression of another neuronal CAM L1, in the hippocampus (Sandi 2004). According to the theories proposed by Blanpied et al., if the adhesion molecules are not in the correct position at the synapse, the synaptic structure will be destroyed, and the neurotransmitters will not be completed (Tang et al. 2016; Biederer et al. 2017). In the present study, we found that the serum level of CHL1 was significantly increased after antidepressant treatment of depressive patients. Furthermore, we also found that antidepressant therapy induced high amount of CHL1 mRNA and protein production in the PBMCs, raising the possibility of using their levels as biomarkers for major depression.

The Relationship Between Major Depression and the CHL1 Expression in the B/T Cells

Many studies reported that there was a close link between depression and immune disorders of the body and brain (Bullmore 2018; Lynall et al. 2019). Although depression is often associated with a reduction in cellular immune responses, other types of emotional disturbance and psychopathology can activate certain aspects of immunity (Lemieux et al. 2008). In the present study, depressed patients exhibit significantly lower percentage of CD45+ lymphocytes. The change of cells expressing CHL1 on CD45+ lymphocytes was consistent with total blood CD45+ lymphocytes. Reduced blood lymphocyte counts in depression could be reversed after antidepressant treatment, suggesting that CHL1 may influence lymphocytes function via an autocrine/paracrine pathway. Many researchers have reported that activation markers on T cells, in particular, have been found to be elevated in depression and anxiety-like behavior (Laumet et al. 2018; Fan et al. 2019). However, emerging evidence implicates that B cells homeostasis is likely to be involved in the depressive behavior (Ahmetspahic et al. 2018). In our cases, there were no differences in the percentage of CD19+, CD20+, and CD3+ cells among the study samples, whereas a significantly lower percentage of CHL1 expressed in CD19+/CD20+ B cells was found in depressed patients. The treatment with antidepressants obviously reversed the biomarker of CHL1+CD19+/CD20+ B cells in depressed patients. We are the first to have defined the specific changes in CHL1 expression in B cells but not in T cells. Furthermore, we found that a significantly increased CD4+CD8−-CD3+ cells and decreased percentage of CD4−CD8+-CD3+ cells in major depression, and the CD4+/CD8+ ratio was significantly increased. Therefore, we speculate that CHL1 expression could be an important diagnostic biomarker for depression and the response indicator to therapeutic interventions. The function of CHL1 in lymphocyte and immune function is not known, our finding that CHL1 expression in B cells warrants further investigation to elucidate its role in the immune regulation, such as in proliferation and differentiation.

Potential Relationship Between CHL1 and Synaptogenesis

Synapses are the functionally connected parts of neurons, and they are also the structural basis for information transmission in the nervous system (Biederer et al. 2017). Some studies have reported that synapses in the prefrontal cortex and hippocampus are reduced in animal experiments of depression, consistent with evidence of lower number of synapses, and corresponding lower expression of synaptic-function-related genes in the brain of depressed patients in postmortem (Kang et al. 2012; Duric et al. 2013). A recent study also shows that lower synaptic density is associated with depression severity by positron emission tomography (PET) in depressed patients (Holmes et al. 2019) and synaptic regeneration can maintain ketamine’s antidepressant effect in animal trials (Moda-Sava et al. 2019). In the mature nervous system, CHL1 was involved in regulation of synaptic activity and plasticity (Morellini et al. 2007; Andreyeva et al. 2010). Blanpied’s group also concerned about the activity of adhesion molecules, which may be an important part of transcellular nanoalignment of synaptic function (Biederer et al. 2017). These studies all point to the view that the relationship between CHL1 and synaptogenesis is critical for the pathogenesis of depression and likely the drug target of antidepressants. Our current investigation adds additional evidence that reduced CHL1 is associated with depression severity, consistent with the change in synaptic density. Furthermore, antidepressant treatment can increase the expression of CHL1 in the brain of mice with chronic stress and in PBMCs from depressive patients. Certainly, findings from this study provide further incentive to evaluate interventions that restore CHL1 and synaptic connections to treat depression. Further studies are important to elucidate whether CHL1 is directly involved in the pathogenesis of depression and the underlying pathogenic mechanisms by loss of function and/or gain of function experiments.

Conclusion

In summary, the present study modeled the depression in rodents and demonstrated that the downregulation of CHL1 strongly correlated with severity of depression. The downregulation of blood CHL1 in B cells but not T cells may play an important role in pathological immune microenvironments in depressed patient. In the future, CHL1 should be further explored in animal trials in order to confirm whether overexpression of CHL1 could relieve depressive and anxiety-like behaviors or vice versa downregulation/silencing of CHL1 expression could result in development of depression.