Abnormal Serum BDNF and p-mTOR in MDD in Adolescents with Childhood Trauma

Introduction

Major depression is one of the leading causes of death and disease burden in adolescents aged 10–24.¹ Childhood trauma is an adverse environmental factor during the neurodevelopment and could affect the emotional circuit.² Childhood maltreatment could result in earlier onset, chronic and severe course of major depression and impaired cognitive function.^3,4 The patients who experienced emotional abuse in childhood had more serious negative cognition and stronger suicide; Both antidepressants and non-pharmacological treatments have poor efficacy.⁵ In addition, adolescents’ internal perception of emotions is often at odds with their parents’ external observation of their emotional behavior.⁶ This makes it difficult to seek early help to adolescents. It is a challenge to treat major depressive episodes in an adolescent who experienced a childhood trauma timely.

mTOR could integrate intracellular and extracellular signals and regulate nerve growth and protein synthesis. In a maternal separation (MS) model in mice, the mTOR signaling pathway is reported to be abnormal;^7,8 p-mTOR(p-mTOR) indicates the active mTOR. The amazing thing is the p-mTOR levels of mice separated at birth is significantly lower than that of chronic stress group. mTOR signaling pathway abnormalities have also been reported in MDD patients.^7,9 Increased mTOR activity was revealed in Prefrontal Cortex (PFC) / Brodmann Area 9 (BA9) in MDD patients who died by suicide, regardless of drug use.¹⁰ Interestingly, when ketamine rapidly improves treatment-resistant depressive disorder, the expressions of mTOR and brain-derived neurotrophic factor (BDNF) are up-regulated.¹¹ BDNF is the most prevalent neurotrophic factor in human, with widespread present throughout the nervous system. Its highest concentration is found in the hippocampus and cortex. The down-regulation of BDNF, which plays an important role in neuronal plasticity, has also been reported in many animal models of MDD.^12–14 Low levels of BDNF have been reported in peripheral blood and postmortem brain studies of suicidal individuals.^15,16 BDNF has also been suggested as a suicide risk predictor possibly independent of MDD^17,18.¹⁹ In previous study, BDNF could improve depressive symptoms by down-regulating the PI3K-AKT-mTOR pathway and up-regulating autophagy.²⁰ It is therefore clear that BDNF and the mTOR signaling pathway in major depressive episodes seem to be closely linked and complex.

 Whether BDNF and mTOR are involved in major depressive episodes in adolescents with childhood trauma and whether there is a regulatory relationship between BDNF and mTOR is worth exploring. Serum BDNF and p-mTOR levels have not been reported in adolescents with major depressive disorder associated with childhood trauma. We detected the levels of serum BDNF and p-mTOR in adolescents with major depressive episodes associated with childhood trauma to explore the specific markers in peripheral blood for major depressive episodes in adolescents with childhood trauma, and predict the change and outcome of the disease.

Materials and Methods

Subjects

Adolescents diagnosed with major depressive episode for the first time without medication were recruited from inpatient department from august 2022 to June 2023. Entry criteria: (1) Young people aged between 10 and 19 years (including 10 and 19 years); (2) Comply with the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-V) for major depressive episodes; (3) Two attending psychiatrists assessed the total CDRS-R score ≥40 points;(4) Adolescent patients with major depressive episode did not take any antidepressant drugs before enrollment;(6) Patients or family members sign informed consent. Exclusion criteria: (1) Patients with a history of mental retardation, epilepsy, encephalitis or other brain organic diseases, other neurological diseases. (2) Patients with diseases such as immune, or endocrine or metabolic disorders. (3) History of acute respiratory tract infection or chronic infection in the last half month. (4) People with a history of drug or psychoactive substance abuse. (5) Those who have experienced a major adverse life event in the last 6 months after age 16. (6) There is a clear risk of suicide, harm to oneself or others. Healthy controls aged 11–19 were recruited at physical examination center. Exclusion criteria are the same as above. All adolescent patients who met the aforementioned criteria and were hospitalized were recruited for the study. Children’s Depression Inventory (CDI) and CTQ scales were assessed on the same day, followed by fasting blood collection on the subsequent day. The physician responsible for diagnosing the disease was different from the one in charge of scale assessment. All subjects completed fasting blood collection the next day after admission. A total of 31 individuals participated in this study, with only two not meeting the childhood trauma criteria. We did not include 1 adolescent with depression who were not associated with childhood trauma, and only 30 MDD participants and 18 matched healthy controls were statistically analyzed.

Scale Evaluation

The Childhood Depression Scale (CDI) is specifically designed to evaluate the level of depression in children and adolescents aged 7–17. Participants were requested to select the most appropriate way to articulate their emotions experienced within the past two weeks. It comprises a total of 27 items categorized into five dimensions: anhedonia, negative affect, negative self-perception, inefficacy, and interpersonal difficulties. Responses are rated on a 54-point scale ranging from 0 to 2, representing “infrequently”, “frequently”, and “consistently” respectively denoting frequency.²¹

The Childhood Trauma Questionnaire (CTQ), developed by Bernstein in 1998, is globally recognized as one of the most widely utilized assessment tools for measuring childhood abuse. The Chinese version of the Childhood Abuse Questionnaire was revised in 2004 by Xing-Fu Zhao, Ya-Lin Zhang, and Long-Fei Li. It consists of a total of 28 items categorized into five subscales: emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect. Scoring is based on a 5-level scale where higher scores indicate greater levels of abuse.²²

Sample Collection

About 5 mL of fasting venous blood was taken with a coagulant tube on the day of admission or in the morning of the next day. The samples were left at room temperature for several hours, centrifuged at 4000 rpm for 10 min, serum was obtained, and stored at −80 °C until testing. The control group was enrolled with the same method of blood collection and storage.

BDNF, p-mTOR Concentration Measurement

The serum concentrations of BDNF and p-mTOR were determined by Human BDNF Enzyme-Linked Immunosorbent Assay (ELISA) KIT (mlbio, ml98839-J) and Human p-mTOR ELISA KIT(ST-H10642). Samples were analyzed in duplicate, and the results were averaged. One extreme value of serum BDNF and p-mTOR were excluded (they are more than 2.5 times the mean). Serum BDNF and p-mTOR were quantified by 752 Ultraviolet Spectrophotometer at 450nm.

Statistical Analysis

SPSS 21.0 was used for statistical analysis. To get better statistical power, the serum BDNF and p-mTOR levels were transformed by square root in the case and the control group. The comparison of transformed BDNF and p-mTOR between the case and control group was performed by t test, including comparison of major depressive adolescents with suicide or self-injury or without suicide or self-injury with the control group. Education represents the years of study. χ² test was used for sex and t test was used for age and education. According to whether they are continuous variable, whether they are normal distribution and homoscedasticity, Pearson correlation analysis was used to determine the relationship between serum BDNF and p-mTOR, and their correlation with emotional abuse, emotional neglect respectively in the patient group. The relationship between serum BDNF, p-mTOR and somatic abuse, sexual abuse, somatic neglect was analyzed by Spearman Correlation Analysis. A receiver operating characteristic curve was performed to confirm the diagnostic value of BDNF, p-mTOR. The test level α=0.05 was used for bilateral test.

Results

Characteristics of the Case Group and the Control Group

There were significant differences in education (F=0.109, p=0.01) between the MDD and healthy control group. There was no significant difference in age and gender between the two groups. See Table 1.

Table 1 Demographic Data for the Cases and Healthy Controls

Biochemical Parameters

Serum level of BDNF were significantly lower (Transformed, F=0.416, p < 0.001) and p-mTOR level (Transformed, F=0.521, p =0.013) were significantly higher in adolescents with MDD compared to healthy controls [Figure 1A and Table 2]. Of the 31 adolescents with major depressive episodes, 17 had suicide or self-injury behaviors. Of the 31 enrolled patients, 13 had attempted suicide. There were 13 individuals who had self-inflicted injuries. Seven had both suicide attempts and self-harm. Compared with healthy people, there were more lower levels in serum BDNF levels (F=1.452, p<0.001), and there were more higher levels in serum p-mTOR levels (F=0.072, p=0.010) in patients with suicide or self-injury. See Figure 1B and Table 3. While compared with healthy people, those with no suicide or self-injury only showed significant difference in serum p-mTOR level (F=0.997, p=0.028). See Figure 1C and Table 4. There were no significant differences in serum BDNF or p-mTOR levels between adolescents with suicide or self-injury and those with no suicide or self-injury. See Figure 1D.

Table 2 Comparison of p-mTOR, BDNF in Peripheral Blood Between the Controls and the Cases

Table 3 Comparison of p-mTOR, BDNF in Peripheral Blood Between the Control Group and the Cases with Suicide or Self-Injury

Table 4 Comparison of p-mTOR, BDNF in Peripheral Blood Between the Control Group and the Cases with No Suicide or Self-Injury

Figure 1 Comparison of serum BDNF and p-mTOR levels among all groups (A) Comparison of serum BDNF and p-mTOR levels between the case group and control group; (B) Comparison of serum BDNF and p-mTOR levels in patients with suicide or self-injury and the control group; (C)Comparison of serum BDNF and p-mTOR levels in patients with no suicide or self-injury and the control group. (D)Comparison of serum BDNF and p-mTOR levels in patients with no suicide or self-injury and the patients with suicide or self-injury. Notes: *Indicated a significant difference of P≤0.05. ***Indicated a significant difference of P≤0.001.

Correlation Analysis

We analyzed the correlation between the levels of serum BDNF, p-mTOR and CDI, CDRS-R, CTQ and sub-scores respectively. By Pearson Correlation Analysis, p-mTOR was positively correlated with CDI (r=0.364, P=0.048); BDNF was negatively correlated with CDRS-R (r=−0.427, p=0.006). According to Bonferroni correction, p< 0.0025. There was no statistical correlation between serum BDNF, p-mTOR levels and CDI or CDRS-R scores; there was no statistical correlation between serum BDNF, p-mTOR levels and the sub-scores of physical abuse, emotional abuse, somatic neglect, and emotional neglect of CTQ-SF. See Figure 2.

Figure 2 Correlation analysis of serum BDNF, p-mTOR with CDI, CDRS-R (A)Serum p-mTOR levels were positively correlated with CDI scores in major depressive episodes in adolescents with childhood trauma; (B)Serum BDNF levels were negatively correlated with CDRS-R scores. But they were corrected by Bonferroni, P< 0.0025.

Receiver Operating Characteristic (ROC) Curve Analysis

ROC curve analysis was performed to detect whether serum BDNF and p-mTOR levels could be effective as biomarkers for distinguishing major depressive episodes in adolescents with childhood trauma from healthy controls. As shown in Figure 3, the AUC of BDNF was 0.883 (95% CI, 0.777–0.989; P = 0.000; Figure 3A), the best cutoff value was 0.354(specificity was 0.889; sensitivity,0.857). The AUC of p-mTOR was 0.773 (95% CI, 0.620–0.925; P =0.002; Figure 3B), the best cutoff value was 0.392(specificity was 0.882; sensitivity,0.800). The combined AUC clearly distinguished patients from healthy controls, with an AUC of 0.887 (95% CI, 0.783–0.993; P = 0.000; Figure 3C), the best cut-off value was 0.512(specificity was 0.765; sensitivity,0.926).

Figure 3 Diagnostic evaluation of serum BDNF, p-mTOR, and combine of them ROC curves of (A) BDNF and (B) p-mTOR, and (C) their combined roc curve. ROC, receiver operating characteristic; AUC, area under the curve.

Discussion

We detected significantly lower serum BDNF and higher serum p-mTOR levels in treatment naive adolescents during a major depressive episode and with a history of childhood trauma compared to healthy controls. In the ROC curve analysis, serum BDNF and p-mTOR levels could be effective as biomarkers for distinguishing major depressive episodes in adolescents with childhood trauma from healthy controls. These could reveal an association between BDNF and p-mTOR in adolescents with major depression associated with childhood trauma. Although BDNF, mTOR have been respectively revealed in MDD, as a key signaling pathway, mTOR has never been validated in adolescents with major depressive episodes associated with childhood trauma.

Before, low BDNF expression in MDD peripheral blood has been repeatedly shown, which even was expected to predict the recurrence of MDD.^23,24 In addition, single nucleotide polymorphism of BDNF is associated with treatment-resistant depression (TRD).²⁵ So, it is clear that BDNF is involved in the pathogenesis of MDD. Intriguingly, BDNF Val66Met polymorphism is also thought as an independent risk factor for MDD suicide attempts.^15,16 Early Cognitive behavioral therapy (CBT) intervention is revealed to reduce suicidal thoughts and self-injury in young people;^26,27 Serum BDNF levels increased after long-term CBT intervention.²⁸ The effect of CBT was thought to be depend on the BDNF Val66Met polymorphism.²⁹ It is worth noting that improvements in BDNF levels and working memory may be identified as markers of ketamine’s anti-suicide effects.^30,31 Increased mTOR activity has also been reported in PFC/BA9 in MDD deaths from suicide.¹⁰ In animal models, rapamycin, an inhibitor of the mTOR pathway, has been reported to have antidepressant activity.³² Our study also confirmed high p-mTOR and low BDNF levels, especially in adolescents with suicidal behavior or self-injury.

Trauma exposure is a key risk factor for recurrent MDD.³³ In multiracial American adults, suicidal ideation is associated with interpersonal violence and childhood abuse.³⁴ Childhood trauma may lead to significant clinical heterogeneity of MDD.³⁵ Childhood trauma and cortisol levels may be mediators of antidepressant response at different treatment stages.³⁶ Astonishingly, the mTOR signaling pathway has also been shown to be involved in stress-related disorders. Low p-mTOR level was reported in animal model of maternal separation; while RNA and Assay for Transposase Accessible Chromatin (ATAC) sequencing revealed that monocytes from stressed mice and humans showed activation of mTOR.^7,37 BDNF was also associated with stress-related disorders. Under long-term stress, increased glucocorticoids enter the hippocampus and activate the Mitogen-Activated Protein Kinase (MAPK) pathway through Tropomyosin receptor kinase B (TrkB) phosphorylation, enhancing the formation of negative memories.^38,39 Glucocorticoid stress hormone acts on Glucocorticoid Receptor (GR) and interacts with BDNF Val66Met polymorphism to determine hippocampal gated fear and spatial memory.⁴⁰ Previous studies have shown that childhood trauma is more likely to lead to negative attention bias.⁴¹ This suggests that BDNF may be involved in negative attention bias caused by trauma in childhood. While the mTOR inhibitor rapamycin, acting alone or in combination with the cannabinoid CB1/2 receptor, may diminish reactivated emotional memories, which have been thought a potential therapeutic strategy for traumatic memory-related disorders.⁴²

In addition, as the most prominent neurotrophic factor, BDNF could activate mTORC1 and enhance protein synthesis involving in dendrite formation, axon extension during brain development.⁴³ Interestingly, Ketamine could directly and rapidly activate mammalian target protein complex of rapamycin 1 (mTORC1), enhance brain-derived neurotrophic factor (BDNF) -mTOR pathway, and promote the rapid release of BDNF in the brain.^44,45 In previous studies, BDNF could improve depressive symptoms by down-regulating the PI3K-AKT-mTOR pathway and up-regulating autophagy.²⁰ BDNF could mediate autophagy through the PI3K-Akt-mTOR pathway.²⁰ Pharmacological enhancement of autophagy may have a stabilizing effect on mood.⁴⁶ In the study, the expression of serum BDNF was significantly lower, while p-mTOR was significantly increased, the correlation between BDNF, p-mTOR and autophagy cannot be excluded.

This study only detected the levels of serum BDNF and p-mTOR, which not fully explain the central system, and we only focus on serum BDNF and p-mTOR, which could not further elucidate the regulatory relationship and the signal pathways involved. Secondly, this study only included adolescents with MDD who experienced childhood trauma, and healthy controls, we did not include adolescents with MDD without childhood trauma. What’ more, we have not completed the follow-up of the MDD group, then the changes of serum BDNF and mTOR after treatment can be detected to better explain the relationship between the two. In the future, whether BDNF, p-mTOR may be key molecule in autophagy involving in adolescent MDD, as well as the latter in traumatic memory, should be further explore. Lastly, due to illness, these adolescents are unable to learn normally, resulting in different years of education than healthy controls. Obviously, our sample size is small. Next, we would continue to complete the enrollment of the study and healthy group, the therapeutic follow-up of the MDD group, and the retest of BDNF and p-mTOR levels after treatment.

Serum BDNF and p-mTOR may be involved in major depressive disorder in adolescents with childhood trauma, especially those with suicide and self-injury.