PRKCB methylation: a potential biomarker of MDD with childhood chronic stress, a cross-sectional study in drug-naive, first-episode adolescent MDD

ABSTRACT

The purpose of this study was to investigate the relationship between childhood chronic stress(CCS), Protein kinase C beta (PRKCB) methylation and adolescent major depressive disorder (MDD). After recruiting 100 adolescents with MDD and 50 healthy controls (HCs), we evaluated the severity of CCS. PRKCB methylation was assessed by pyrosequencing using whole blood-derived DNA. To explore the relationship between CCS, PRKCB and adolescent MDD, we conducted correlation analysis and regression analysis, and constructed multiplicative interaction models and generalized linear models. PRKCB methylation and CCS were both found to be associated with MDD, and CCS was associated with PRKCB methylation. No significant CCS-PRKCB methylation interactions were observed. However, we found the interaction of CCS and MDD on PRKCB methylation. Our results found that PRKCB methylation was influenced by CCS and the disease itself, and PRKCB methylation was significantly positively associated with MDD severity, suggesting that PRKCB methylation may be a potential biomarker for adolescent MDD. This study is a cross-sectional observational study, which cannot draw the conclusion of causality. Prospective cohort studies are needed to further examine the relationship between CCS, adolescent MDD, and PRKCB methylation.

Introduction

Major depressive disorder (MDD) is characterized by persistent depressed mood and cognitive impairment [1]. Studies have found that the prevalence of MDD increases significantly throughout adolescence [2]. Compared with adult MDD, atypical symptoms are more common in adolescents with MDD [3], leading to difficulties in diagnosis and treatment [4,5] and more serious impairment of social function [6]. Adolescents with MDD have more problems, such as suicide, academic difficulties, fighting and substance abuse [7–10], resulting in a significant social burden.

It is generally believed that the risk of MDD is determined by a combination of environmental and genetic factors [11]. Among environmental factors, the incidence of childhood chronic stress (CCS) is relatively high, with approximately 50.7% of adolescents reporting chronic stress experience in childhood [12], which is significantly correlated with the occurrence of MDD and suicide risk in adolescents [13–15]. Epigenetics refers to the influence on gene expression and translation without changing the DNA sequence [11]. With the rapid development of epigenetics, it is known that changes in gene activity established through epigenetic alterations may be the result of exposure to environmental factors such as social stress, and traumatic experiences [16]. Previous studies have found that there may be specific epigenetic variations associated with MDD, particularly DNA methylation, which helps distinguish MDD patients from healthy controls (HCs) [17].

Our group’s previous exploratory study found that Protein kinase C beta (PRKCB) was differentially methylated in MDD adolescents [18], and was enriched in many depression-related pathways, such as glutaminergic synapses, dopaminergic synapses, 5-serotonergic synapses and MAPK signaling pathways [19]. A previous genome-wide epigenetic association analysis of brain tissue also found a significant correlation between PRKCB methylation and MDD [20]. PRKCB is also involved in many important biological processes, such as the immune response, neuronal development, and signal transmission [21–23], which current studies have found to be associated with MDD and antidepressant treatment [24–26]. Studies have also found that single nucleotide polymorphisms and gene expression levels of PRKCB were associated with the risk of MDD and antidepressant treatment [27,28]. Elwood’s study found a positive association between PRKCB gene polymorphism and depression in the presence of stressful life events [29], which shows the complex relationship between environment factors, epigenetics and MDD. However, there are no studies on the relationship between PRKCB methylation, MDD and environmental stress.

Therefore, this study aimed to further validate the PRKCB differential methylation site found in our previous study and to explore the relationship between CCS, MDD adolescents and PRKCB methylation.

Methods

Subjects

We recruited 100 drug-naive, first-episode adolescents with MDD from the Mental Health Center of West China Hospital of Sichuan University. All enrolled patients were independently diagnosed by two senior psychiatrists using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition and the Kiddie-Schedule for Affective Disorders and Schizophrenia-Present and Lifetime Version, with an overall score of ≥ 14 on the Chinese version of Beck Depression Inventory-II(BDI) [30]. Over the same period, 50 adolescent HCs with no history of mental illness were recruited through advertising.

All participants were between the ages of 12 and 17 years, right-handed, and had to complete at least primary education to ensure they understood the content of the scales. All adolescents underwent neurological examinations and detailed semi-structured interviews to ensure that they had no history of neurological disorders, psychotropic substance use, alcohol or drug abuse, or other Axis I/II psychiatric disorders. This study was approved by the Ethics Committee of the West China Hospital of Sichuan University. Written informed consent was provided by all participants and their guardians.

Childhood chronic stress

We used the Childhood Chronic Stress Questionnaire (CCSQ) to assess whether the participants had experienced chronic stress in childhood and the extent of its psychological effects [31]. The scale comprises three major dimensions, encompassing nine sub-dimensions: childhood peer bullying, which incorporates peer relationship bullying and physical bullying; childhood abuse and neglect, covering emotional and physical abuse, sexual abuse, emotional neglect, and physical neglect; and finally, adverse childhood experiences, including family, school and personal adverse experiences. This scale has good reliability and validity in China, cronbach’s α coefficient is 0.946 [31]. The stress level of the participants was divided into high and low stress levels with the mean CCSQ score + 1 SD of the control group as the standard [32].

PRKCB methylation

Peripheral venous blood was collected using EDTA anticoagulant tubes. Then, methylated pyrophosphate sequencing experiments were performed, DNA was extracted using the Blood Genomic DNA Extraction Kit (QIAGEN), and DNA was quantified using the Nanodrop instrument (Thermo). The methylation was then modified with sulfite using the EpiTect Bisulfite Kit (QIAGEN). PyroMark Assay Design 2.0 software was used to design the primers of PRKCB, and the primers were synthesized by the Beijing Genomics Institute. And they had the following sequences: 5’-TTTAGTTGAGGGGGAGGG-3,’ 5’-CTCCCCTAAACACATCTCCTA-3,’ and 5’-GTTGAGGGGGAGGGA-3.’

PCR amplification was performed as follows. The reaction system was a 50 μl mixture, including 10 μl 5X GC buffer (KAPA), 10 mm dNTP, 2 μl template DNA, 1 U/μl Taq polymerase, 50 pM upstream primer, 50 pM down-stream primer and 34.8 μl H₂O. The reaction started with 95°C for three minutes, followed by 40 cycles at 94°C for 30 s, 56°C for 30 s, and 72°C for 1 min, then 72°C for 7 min. The methylation status was analysed by Pyro Q-CpG software (PyroMark Q96 ID, QIAGEN). The percentage of methylation was calculated by determining the proportion of methylated cytosine of the total cytosines at the CpG site (methylated or unmethylated).

Statistical analysis

All statistical analyses were performed using SPSS 25.0, with significance level set at α = 0.05. Demographic and clinical characteristics of MDD adolescents and HCs were compared. Mann-Whitney U test was used for continuous variables and chi-square test was used for categorical variables. Using Spearman correlation to explore the relationship between CCS, PRKCB methylation and severity of depression in MDD patients. Binary logistic regression was utilized to explore the relationship between CCS, PRKCB methylation, and the diagnosis of MDD, while linear regression was applied to investigate the correlation between CCS and PRKCB methylation.

In order to explore the interactive effects of CCS and PRKCB methylation on MDD diagnosis, we constructed a multiplicative interaction model. This model used PRKCB methylation, CCS, and their centralized interaction terms as independent variables, and MDD diagnosis as the dependent variable in logistic regression analysis. Then, we used a two-way ANOVA to explore the interaction of CCS and diagnosis of MDD on the PRKCB methylation.

Results

Demographic and clinical characteristics

No significant differences were observed in age, sex, or BMI between MDD adolescents and HCs. However, in comparison to HCs, MDD patients exhibited significantly higher scores on the BDI scores, the total CCSQ score, as well as in all dimensions of the CCSQ (Table 1).

Table 1.

Demographic and clinical characteristics of MDD patients and HCs.

	MDD patients (n = 100)	HCs (n = 50)	z/χ2	p
Age (years)	14 (13, 16)	14 (13, 15)	−1.821	0.069
Sex (Male/Female)	41/59	28/22	3.019a	0.082
BMI (kg/m [2])	19.64 (17.58, 22.45)	18.88 (17.08, 21.45)	−1.186	0.236
BDI score	33.00 (27.00, 40.00)	4.00 (0.00, 9.00)	−9.974	<0.001*
CCSQ	131.50 (108.25, 157.75)	75.00 (63.75, 90.75)	−8.15	<0.001*
Childhood peer bullying	25.00 (18.25, 35.50)	17.00 (15.00, 22.00)	−4.989	<0.001*
peer relationship bullying	14.00 (10.00, 20.00)	8.50 (7.00, 11.00)	−5.119	<0.001*
peer physical bullying	11.00 (8.00, 15.75)	8.00 (8.00, 10.00)	−3.465	0.001*
Childhood abuse and neglect	60.50 (48.25, 72.00)	32.50 (29.00, 41.00)	−7.731	<0.001*
emotional and physical abuse	21.00 (16.00, 30.00)	12.00 (11.00, 16.00)	−6.192	<0.001*
sexual abuse	7.00 (7.00, 7.00)	7.00 (7.00, 7.00)	−1.991	0.047*
emotional neglect	24.50 (17.00, 31.75)	8.00 (7.00, 13.00)	−7.850	<0.001*
physical neglect	6.00 (4.00, 9.00)	4.00 (4.00, 4.00)	−5.523	<0.001*
Adverse childhood experiences	42.00 (35.00, 50.00)	24.00 (18.00, 29.25)	−8.001	<0.001*
family adverse experiences	6.00 (4.00, 11.00)	4.00 (4.00, 6.00)	−3.968	<0.001*
school adverse experiences	17.00 (13.00, 22.00)	8.50 (6.00, 12.25)	−6.735	<0.001*
personal adverse experiences	17.00 (14.00, 21.00)	9.00 (6.75, 12.00)	−7.688	<0.001*
PRKCB methylation	22.83 (19.47, 25.00)	20.54 (18.31, 23.24)	−2.757	0.006*

Data were not normally distributed, expressed as the median (P25, P75); ^ameans Chi-square test, and the rest are Mann–Whitney U test.

Relationship between CCS and adolescent MDD

To investigate the association between CCS and severity of depression in MDD patients, we employed Spearman correlation analysis. The results showed that the CCSQ total score (r = 0.363, p < 0.001) and three major dimensions of childhood peer bullying (r = 0.223, p = 0.026), childhood abuse and neglect (r = 0.324, p = 0.001), and adverse childhood experiences (r = 0.352, p < 0.001) were positively correlated with BDI scores. Among the nine sub-dimensions, peer relationship bullying (r = 0.255, p = 0.011), emotional neglect (r = 0.361, p < 0.001), physical neglect (r = 0.346, p < 0.001) and personal adverse experiences (r = 0.425, p < 0.001) were also positively correlated with BDI scores.

To explore the relationship between CCS and MDD diagnosis, we performed binary logistic regression analysis using CCSQ scores as the independent variable, MDD diagnosis as the dependent variable, and age, sex, and BMI as covariates. The results showed that CCSQ total score and three major dimensions were associated with MDD diagnosis (p < 0.05). Of the nine sub-dimensions, all but sexual abuse (OR = 1.868, 95% CI: 0.841–4.149, p = 0.125) were associated with MDD diagnosis (Table 2).

Table 2.

Relationship between childhood chronic stress and MDD diagnosis.

	OR(95%CI)	p
CCSQ total score	1.080 (1.052-1.108)	<0.001*
Childhood peer bullying	1.150 (1.077-1.229)	<0.001*
peer relationship bullying	1.260 (1.140-1.393)	<0.001*
peer physical bullying	1.260 (1.106-1.434)	<0.001*
Childhood abuse and neglect	1.127 (1.081-1.176)	<0.001*
emotional and physical abuse	1.177 (1.095-1.265)	<0.001*
sexual abuse	1.868 (0.841-4.149)	0.125
emotional neglect	1.280 (1.172-1.397)	<0.001*
physical neglect	1.670 (1.281-2.177)	<0.001*
Adverse childhood experiences	1.212 (1.135-1.294)	<0.001*
family adverse experiences	1.259 (1.089-1.456)	0.002*
school adverse experiences	1.288 (1.171-1.417)	<0.001*
personal adverse experiences	1.437 (1.273-1.623)	<0.001*

Age, sex and BMI were included as covariates.

Relationship between PRKCB methylation and adolescent MDD

The methylation level of PRKCB was 22.83 (19.47, 25.00) in MDD group and 20.54 (18.31, 23.24) in control group, with significant differences between the two groups (z = −2.757, p = 0.006). Spearman correlation analysis showed a positive correlation between the PRKCB methylation level and BDI scores (r = 0.259, p = 0.009). A binary logistic regression analysis was conducted, taking the PRKCB methylation level as the independent variable, MDD diagnosis as the dependent variable, and the same covariates as before. The results showed a significant association between PRKCB methylation and MDD diagnosis (OR = 1.062, 95% CI: 1.001–1.126, p = 0.048).

Relationship between CCS and PRKCB methylation

Spearman correlation analysis showed that the CCSQ total score (r = 0.291, p = 0.003) and two major dimension scores, childhood peer bullying (r = 0.213, p = 0.033) and childhood abuse and neglect (r = 0.283, p = 0.004), were positively correlated with the PRKCB methylation level. Among the nine sub-dimensions, peer relationship bullying (r = 0.222, p = 0.026), emotional and physical abuse (r = 0.210, p = 0.036), emotional neglect (r = 0.237, p = 0.017), physical neglect (r = 0.208, p = 0.038), and school adverse experiences (r = 0. 261, p = 0.009) were also positively correlated with PRKCB methylation level.

A linear regression analysis was conducted, using CCSQ scores as the independent variables, PRKCB methylation as the dependent variable, and the same covariates as before. The results showed that the total CCSQ score (β = 0.291, p = 0.004) and the three major dimensions, childhood peer bullying (β = 0.218, p = 0.036), childhood abuse and neglect (β = 0.269, p = 0.009) and adverse childhood experiences (β = 0.211, p = 0.036), were correlated with PRKCB methylation level. Among the nine sub-dimensions, peer bullying (β = 0.231, p = 0.026), emotional and physical abuse (β = 0.220, p = 0.032), emotional neglect (β = 0.247, p = 0.017), physical neglect (β = 0.204, p = 0.041) and school adverse experiences (β = 0.323, p = 0.001) all exhibited associated with PRKCB methylation level.

Relationship between CCS, PRKCB methylation and MDD diagnosis

To explore the interaction of CCS and PRKCB methylation on MDD diagnosis, we constructed a multiplicative interaction model. The results showed that the interaction between CCS and PRKCB methylation had no effect on the diagnosis of MDD (Table 3).

Table 3.

Interaction effect between CCS and PRKCB methylation on MDD diagnosis.

	CCS		PRKCB		CCS*PRKCB methylation
	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
CCSQ total score	1.075 (1.049, 1.101)	<0.001*	1.040 (0.952, 1.137)	0.384	1.000 (0.996, 1.004)	0.929
Childhood peer bullying	1.134 (1.064, 1.208)	<0.001*	1.077 (1.001, 1.159)	0.047*	1.007 (0.996, 1.017)	0.200
peer relationship bullying	1.234 (1.118, 1.361)	<0.001*	1.047 (0.987, 1.111)	0.127	1.009 (0.996, 1.022)	0.188
peer physical bullying	1.229 (1.077, 1.402)	0.002*	1.033 (0.973, 1.098)	0.287	1.020 (0.991, 1.049)	0.177
Childhood abuse and neglect	1.119 (1.077, 1.163)	<0.001*	1.026 (0.933, 1.127)	0.600	1.000 (0.993, 1.006)	0.901
emotional and physical abuse	1.179 (1.097, 1.265)	<0.001*	1.049 (0.986, 1.116)	0.129	1.001 (0.988, 1.013)	0.916
sexual abuse	1.798 (0.823, 3.928)	0.141	1.070 (1.004, 1.139)	0.037*	0.978 (0.859, 1.114)	0.737
emotional neglect	1.258 (1.163, 1.360)	<0.001*	1.026 (0.944, 1.115)	0.550	0.998 (0.984, 1.011)	0.716
physical neglect	1.645 (1.250, 2.164)	0.001*	1.028 (0.957, 1.103)	0.452	1.004 (0.945, 1.068)	0.886
Adverse childhood experiences	1.212 (1.138, 1.291)	<0.001*	1.054 (0.972, 1.143)	0.204	0.995 (0.983, 1.007)	0.398
family adverse experiences	1.279 (1.104, 1.482)	0.001*	1.091 (1.006, 1.183)	0.034*	0.995 (0.977, 1.013)	0.600
school adverse experiences	1.287 (1.170, 1.415)	<0.001*	1.055 (0.981, 1.134)	0.151	1.008 (0.993, 1.024)	0.283
personal adverse experiences	1.456 (1.286, 1.648)	<0.001*	1.050 (0.969, 1.138)	0.235	0.989 (0.965, 1.014)	0.392

CCS*PRKCB methylation refers to the interaction of CCS and PRKCB methylation. CCS: childhood chronic stress.

Then we constructed a two-way ANOVA to investigate the interaction of CCS and MDD diagnosis on PRKCB methylation. The results revealed a significant interaction between MDD diagnosis and two dimensions of CCSQ: childhood peer bullying (F = 4.235, p = 0.041, R² = 0.069) and school adverse experiences (F = 6.495, p = 0.012, R² = 0.079) on PRKCB methylation (Figure 1 and Table 4), and these findings were consistent when adjusted for age, sex and BMI (supplementary table S1). On this basis, we conducted a pairwise comparison analysis and discovered that in MDD adolescents, the PRKCB methylation level of those who endured high levels of childhood peer bullying (23.425 ± 0.860) was higher than that of individuals with low childhood peer bullying (21.039 ± 0.843). Specifically, among MDD adolescents, the average PRKCB methylation level of individuals with high childhood peer bullying was 2.386 (95%CI, 0.005–4.767) higher than those with low levels of bullying (p = 0.05). Moreover, within the group that endured high levels of childhood peer bullying, the PRKCB methylation level of MDD adolescents surpassed that of HCs by 6.138 (95%CI, 1.599–10.677), with the HCs’ level standing at 17.288 ± 2.129 (p = 0.008). For details, see supplementary tables S2 and S3.

Figure 1.

Interaction effect between CCS and MDD diagnosis on PRKCB methylation. (a, b). Interaction effect between childhood peer bullying and MDD diagnosis on PRKCB methylation (F = 4.235, p = 0.041, R² = 0.069). (c, d). Interaction effect between school adverse experiences and MDD diagnosis on PRKCB methylation (F = 6.495, p = 0.012, R² = 0.079). Estimated marginal means refers to the estimated marginal means of PRKCB methylation.

Table 4.

Interaction effect between CCS and MDD diagnosis on PRKCB methylation.

	CCS		MDD diagnosis		CCS*MDD diagnosis		R2
	F	p	F	p	F	p
CCSQ total score	0.138	0.711	3.856	0.051	0.296	0.587	0.036
Childhood peer bullying	0.055	0.815	6.929	0.009*	4.235	0.041*	0.069
peer relationship bullying	0.147	0.702	6.119	0.015*	2.593	0.109	0.053
peer physical bullying	0.002	0.962	6.091	0.015*	2.382	0.125	0.054
Childhood abuse and neglect	0.028	0.868	3.055	0.083	1.128	0.290	0.045
emotional and physical abuse	<0.001	0.983	5.144	0.025*	3.246	0.074	0.070
sexual abuse	1.525	0.219	0.487	0.486	0.511	0.476	0.044
emotional neglect	0.173	0.678	2.136	0.146	0.248	0.619	0.039
physical neglect	3.065	0.082	1.330	0.251	0.468	0.495	0.089
Adverse childhood experiences	0.574	0.450	4.655	0.033*	0.145	0.704	0.038
family adverse experiences	0.869	0.353	6.014	0.015*	1.119	0.292	0.042
school adverse experiences	0.557	0.457	6.772	0.010*	6.495	0.012*	0.079
personal adverse experiences	0.599	0.440	4.958	0.028*	0.160	0.690	0.038

CCS*MDD diagnosis refers to the interaction of CCS and MDD diagnosis. CCS: childhood chronic stress.

In the MDD group, the PRKCB methylation level in high school adverse experiences (23.004 ± 0.727) was 2.486 (95%CI, −0.052-5.024) higher than those with low school adverse experiences (20.518 ± 1.059), but it was not significant (p = 0.055). However, within the group that endured high levels of school adverse experiences, the PRKCB methylation level of MDD adolescents surpassed that of HCs by 7.105 (95%CI, 2.405–11.805), with the HCs’ level standing at 15.899 ± 2.264 (p = 0.003). For details, see supplementary tables S2 and S3.

Discussion

The overall goal of this study was to investigate the relationship between CCS, adolescent MDD, and PRKCB methylation. Our primary analyses sought to test whether CCS have a potentially interacting association with adolescent MDD and PRKCB methylation. For adolescent MDD, we found no interaction between CCS and PRKCB methylation. However, the interaction between MDD diagnosis and two dimensions of CCS (childhood peer bullying and school adverse experiences) had an impact on PRKCB methylation.

Previous studies have found that CCS is an important potential risk factor for MDD [33]. This study found that CCS was correlated with the severity of depression in MDD patients, indicating that adolescents with MDD have experienced more CCS, and the higher the level of chronic stress experienced in childhood, especially abuse and neglect, the higher the severity of depression in MDD patients may be. In addition to sexual abuse, all dimensions of CCS were associated with MDD diagnosis, which is consistent with previous findings [34,35]. However, this study did not find a relationship between sexual abuse and MDD diagnosis, which is inconsistent with previous research findings [35]. In China, the sex education system is still not perfect, adolescents have little knowledge of sexual abuse, and sexual stigma is still very serious, so participants may conceal it [36]. All of the above factors may have an impact on the results, which may be the reason for the inconsistency with previous research results. Childhood sexual abuse may have a lasting psychological impact on victims and is more likely to result in suicidal behavior and substance abuse [37]. Therefore, it is necessary to improve education on sexual knowledge and sexual abuse and the treatment process after sexual abuse to better protect this special group.

Current studies suggest that epigenetics may play an important role in MDD [38]. Many methylation site variations have been found in genome-wide methylation studies of MDD, and these differential methylation sites are thought to be associated with increased vulnerability to MDD and may be potential biomarkers of MDD [39,40].

Our group’s previous exploratory study found that PRKCB was differentially methylated in MDD adolescents, which was consistent with the results of previous genome-wide DNA methylation studies [20], and previous studies also supported the involvement of PRKCB in important biological processes related to MDD [41]. Therefore, this study expanded the sample size to further investigate the relationship between PRKCB methylation and adolescent MDD. We found that the methylation level of PRKCB in MDD patients was significantly higher than that in HCs and was correlated with adolescent MDD. For every unit increased in PRKCB methylation levels, the risk of adolescent suffering from MDD increased by 1.062 times. This suggests that PRKCB methylation may be a potential biomarker of adolescent MDD, and further validation of the results by expanding the sample is needed.

Previous studies have found that CCS can lead to changes in DNA methylation levels [42]. For example, a genome-wide methylation study of the human hippocampus found 362 sites was differentially methylated in individuals who experienced severe childhood abuse [43], and similar methylation variations were observed in blood studies [44–46]. This study tried to preliminarily explore the relationship between CCS and PRKCB methylation, and found that CCS, especially childhood abuse and neglect, is related to PRKCB methylation. According to previous studies, PRKCB is involved in many pathways and biological processes related to depression [41]. This implies that CCS may be involved in important pathways and biological processes associated with MDD by affecting the methylation level of PRKCB, thereby leading to susceptibility to the disease. Therefore, relevant mechanisms can be studied in the future.

Current longitudinal studies have shown that antidepressant treatment can reverse stress-induced changes in DNA methylation in specific genes [47,48]. Additionally, researches have revealed a strong association between DNA methyltransferases (DNMT) and depressive states [49,50], suggesting that DNA methylation may play a crucial role in the response to antidepressants. Furthermore, DNMT inhibitors have been found to induce antidepressant-like effects and can alter gene methylation levels [51,52]. For instance, Zhong et al. observed increased expression of DNMT1 in prenatally stressed mice, and DNMT1 inhibitor may alleviate the reduction of nerve cells and depression-like behaviors in prenatally stressed mice by improving the GABAergic system [53]. Li et al. observed increased binding of DNMT1 to the dopamine 2 receptor promoter and hypermethylation of the dopamine 2 receptor promoter in prenatal stress mice. They also found that DNMT1 inhibitors restored striatal synaptic plasticity and alleviate depression-like behaviors through dopamine 2 receptor-mediated dopamine signaling [54]. These finding underscores the significant potential of DNMT inhibitors in antidepressant therapy. Therefore, additional longitudinal studies are necessary to investigate whether PRKCB methylation levels change after antidepressant treatment and to determine whether DNMT inhibitors impact PRKCB methylation levels. This will further elucidate the role of PRKCB methylation in the onset and development of MDD.

Previous studies have found that the interaction between environmental factors and epigenetics has an impact on affective disorders [55,56]. Klengel et al. found the interaction between DNA methylation and childhood trauma on depression [57]. This study analyzed the interaction between CCS and PRKCB methylation on adolescent MDD diagnosis, and the results showed no significant effect. Previous studies on whether the interaction between DNA methylation and childhood trauma has an impact on depression have inconsistent results [58,59], which may be related to differences in sample size, diagnostic criteria, ethnicity, and methylation sites. These results indicate the complex relationship between childhood stress and MDD, and multiple factors may affect the occurrence of MDD after exposure to childhood stress [60].

However, our study found that the interaction of MDD diagnosis and CCS (childhood peer bullying and school adverse experiences) has an impact on PRKCB methylation. This suggested that PRKCB methylation might be influenced by the interaction between school-related stress events and MDD itself. Moreover, PRKCB methylation was positively associated with the severity of depression in MDD patients, suggesting that PRKCB methylation may be a potential biological indicator for adolescent MDD. Peer relationships are important during adolescence, and peer bullying is a recognized mental health risk affecting adolescents [61]. Bullying has been a growing problem in schools across China [62]. A multicentre study conducted in China found that 42.9% of Chinese adolescents surveyed had been bullied by their peers, with the school environment being the most common place for peer bullying (34.7%) [63]. Bullying can lead to emotional problems such as depression and anxiety, and is also associated with crime, suicide, and substance abuse [64]. Therefore, it is important to take anti-bullying measures to reduce the occurrence of bullying, and provide a supportive school environment and timely help the bullied adolescents. At the same time, the long-term prospects should also be improved by addressing the common causes of victimization to reduce the risk of further bullying of vulnerable groups [62,64].

There are some limitations in this study. First, the CCSQ used in our study was a retrospective self-reported measure, and therefore may induce recall bias. Second, the sample size of this study is small. Third, this study is a cross-sectional observational study, which cannot draw the conclusion of causality. Therefore, follow-up longitudinal studies are needed to further examine the relationship between CCS, adolescent MDD, and PRKCB methylation.

In conclusion, we found that CCS and PRKCB methylation were associated with adolescent MDD. No significant CCS-PRKCB methylation interactions were observed. However, our study found that the interaction of MDD diagnosis and CCS has an impact on PRKCB methylation. This suggested that PRKCB methylation might be influenced by the interaction of CCS and MDD itself, indicating that the relationship between CCS, DNA methylation and adolescent MDD is complex. Changes in DNA methylation after CCS and MDD exposure may be influenced by a variety of factors. The relevant mechanism research still needs to be continued.

Abbreviations

CCS

childhood chronic stress

PRKCB

Protein kinase C beta

MDD

major depressive disorder

HCs

healthy controls

BDI

Beck Depression Inventory-II

CCSQ

Childhood Chronic Stress Questionnaire

DNMT

DNA methyltransferase

Funding Statement

This work was supported by the National key R&D program of Chinese Science & Technology Department[2023YFE0118600], National Natural Science Foundation of China [81801357], Science & Technology Department of Sichuan Province [2022YFS0351 & 2020JDKP0013], Health Commission of Sichuan province program [21PJ020], Science and Technology Department of Chengdu [2019-YF05-00284-SN]. These funding agencies were not involved in any of the research processes.

Authors’ contributions

Methodology and original draft preparation: Li Yin and Yuanmei Tao. Data collection: Meijiang Jin, Yuanmei Tao, Hang Zhang, Hanmei Xu, Shoukang Zou, Fang Deng, Lijuan Huang, Hong Zhang, Xiaolan Wang, Xiaohong Ma, Huijin Hou, Shufang Liang and Yanping Wang. Project administration and funding acquisition: Li Yin. All authors contributed to the article and approved the submitted version.

Availability of data and materials

All data generated or analysed during this study are included in this published article and its supplementary information files.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the West China Hospital of Sichuan University. Written informed consent was provided by all participants and their guardians.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15592294.2024.2408159