Elsevier

Journal of Psychiatric Research

Volume 93, October 2017, Pages 50-58
Journal of Psychiatric Research

BDNF Val66Met and childhood adversity on response to physical exercise and internet-based cognitive behavioural therapy in depressed Swedish adults

https://doi.org/10.1016/j.jpsychires.2017.05.007Get rights and content

Abstract

The genetic effect of Brain-derived neurotrophic factor (BDNF) on treatment response in depression is not consistent in the literature. Childhood adversity is a known risk factor for depression which has been reported to increase depression susceptibility by interacting with BDNF genetic variation. We aimed to explore whether the BDNF variation Val66Met and childhood adversity (CA) predicted the treatment response to a 12-week intervention with physical exercise (PE) or internet-based cognitive behavioural therapy (ICBT) when compared with treatment as usual (TAU). A prospective cohort study nested within a randomised control trial was conducted using data from 547 participants with mild to moderate depression. Depression severity at baseline and follow-up was measured using the Montgomery-Åsberg Depression Rating Scale. Met allele carriers without exposure to CA and current antidepressant use showed higher treatment response to PE than Val homozygotes. There was no evidence to support that BDNF Val66Met or CA alone predicted treatment response to PE and ICBT. The Met carriers had higher serum mature BDNF level. These data suggest that Met allele carriers benefit more from PE treatment but only if they are not exposed to early adversity.

Introduction

Depression is a leading cause of disability in both genders, imposing a significant burden of disease worldwide (DEPRESSION: A Global Crisis, 2012). Although evidence-based treatments exist, there are large inter individual differences in the therapeutic responses and the underlying explanations are sparse. Pharmacogenetic studies have suggested that the therapeutic response could be mediated by brain-derived neurotropic factor (BDNF) (Lee and Kim, 2010, Sakata, 2014). Higher serum BDNF levels were implicated in a positive response to treatment with antidepressants, and the treatment response was reported to be dependent on the single nucleotide polymorphism (SNP) in BDNF, rs6265, although there are inconsistencies between studies (Tsai et al., 2010).

BDNF is involved in regulation of neurogenesis, cell survival and synaptic plasticity (Lipsky and Marini, 2007, Zhang et al., 2003). It is secreted by neurons, but also by megakaryocytes and platelets which are the major source of BDNF in serum. BDNF exists in two forms, (i) a precursor BDNF (proBDNF) active via p75 neurotrophin receptor (P75NTR), and (ii) mature BDNF (mBDNF) active via tyrosin kinase receptor B (TRKB). ProBDNF is converted into mBDNF by extracellular proteases. Because of the differential receptor specificities mBDNF promotes neurogenesis and trophic plasticity effects whereas proBDNF supports neuronal migration and apopotosis (Barker, 2009, Colle et al., 2015a). Reduced levels of mBDNF have been reported to be involved in the pathophysiology of depression and mBDNF level restoration has been reported to have an antidepressive effect (supporting the neurotrophic hypothesis of depression) (Groves, 2007, Nestler et al., 2002). On the other hand, proBDNF levels were elevated in depressed patients (Yoshida et al., 2012, Zhou et al., 2013). The SNP rs6265 in BDNF is a G to A polymorphism encoding a Valine (Val) to Methionine (Met) substitution at codon 66 (Val66Met) and has been associated with depression risk and recovery (Mata et al., 2010, Sen et al., 2003, Verhagen et al., 2010). This Val to Met change impairs intracellular BDNF trafficking and activity dependent secretion of BDNF (Egan et al., 2003). Val66Met was reported to influence the response to antidepressants dependent on type of antidepressant; selective serotonin reuptake inhibitors (SSRIs) were reported to be more effective in Val/Val carriers whereas serotonin-norepinephrine reuptake inhibitors (SNRIs) or tricyclic antidepressants (TCAs) had a better effect in Met carriers (Colle et al., 2015b).

Physical exercise (PE) is known to influence serum/plasma BDNF levels and rodent hippocampal BDNF mRNA expression (Garza et al., 2004, Knaepen et al., 2010). In a cross-sectional study it was found that increased physical activity in adolescent girls (n = 82) with Met allele, but not Val/Val homozygosity, associated with reduced depressive symptoms (Mata et al., 2010). In contrast, a prospective study conducted on 1196 adolescents found no modifying effect of BDNF Val66Met on the association between physical activity and depression (Stavrakakis et al., 2013). Likewise, a cross-sectional study conducted on 1072 adults found no moderating effect of BDNF Val66Met on the association between self-reported physical activity and depression (Gujral et al., 2014). Accordingly, the relationship between antidepressive effect of PE and BDNF level seems consistent in animal models but inconsistent in humans (Russo-Neustadt and Chen, 2005, Toups et al., 2011). There are only few studies examining the relation between psychological therapy, the BDNF gene and protein levels in depression. However, BDNF Val66Met was found to predict treatment response to cognitive behavioural therapy (CBT) in posttraumatic stress disorder (Felmingham et al., 2013). In contrast, a recent study conducted on children with anxiety disorders observed no association between BDNF Val66Met and response to psychological therapy (Lester et al., 2012). Depression and its subtypes is considered to have polygenetic liabilities (Levine et al., 2014). A meta-analysis of 5 twin studies found that the additive genetic effect (heritability) is 37% (Sullivan et al., 2000) which can also explain why consistent genetic effects of anti-depressive effect of physical exercise and psychological therapy is hard to detect.

Childhood adversities (CA) are known to increase the risk of depression later in life and to moderate antidepressant treatment effects. A recent meta-analysis showed that exposure to CA predicted worse treatment response over the 10 different clinical trials (Nanni et al., 2012). There is growing support that CA could have long lasting epigenetic effects which could influence the treatment response in depression (Menke and Binder, 2014, Szyf, 2009). Gene by environment (G × E) interaction has been proposed as a more promising approach, as genetic and environmental factors of antidepressant treatment response studies produced small main effect sizes (Keers, 2012). Until now, almost all G × E interaction studies of antidepressive treatment response aimed at exploring pharmacological therapies (Klengel and Binder, 2013). Keers and Uher (2012) emphasised that the use of early life stressors in G × E interaction studies aiming at exploring the response to non-pharmacological treatment would be worth investigating.

Previously, we have shown that the BDNF Val66Met × CA interaction is associated with depression (Lavebratt et al., 2010). In the present study we aimed at investigating (i) whether the BDNF Val66Met polymorphism predicted the treatment response to PE and Internet-based cognitive behavioural therapy (ICBT) in depressed adults, (ii) whether the interaction between the BDNF Val66Met and CA predicted treatment response to PE and ICBT, (iii) if there was any Val66Met dependent differences in serum BDNF (mBDNF and proBDNF) levels and (iv) if there was an association between serum BDNF levels and treatment response to PE. To our knowledge, this is the largest intervention study investigating the influence of BDNF genotype (Val66Met) and CA on the effects of PE and ICBT intervention.

Section snippets

Study design and participants

Data is derived from a randomized controlled trial, Regassa, aiming at assessing the relative effectiveness of PE, ICBT compared with treatment as usual (TAU) on mild to moderate depression. It was conducted from February 2011 to December 2012. Primary care patients from six county councils in Sweden aged 18 years or older and scoring 10 or higher on Patient Health Questionnaire (PHQ-9) were included. Exclusion criteria included severe somatic disorders, primary drug or alcohol use disorder,

Results

Among the 547 participants, 377 were Val/Val and 170 Met carriers. There were no significant differences in baseline characteristics between the genotype groups, see Table 1. Met carriers had lower odds ratios, although statistically non-significant, of treatment response when compared with Val/Val homozygotes see Table 2. The similar size improvement in treatment response by PE and ICBT compared to TAU in the Regassa study has been reported previously (Hallgren et al., 2015). No evidence of

Discussion

In the present study, of adult persons with mild to moderate depression, we found that Met allele carriers without exposure to adversity in childhood were more likely to be good responders to 12 weeks PE compared to Val/Val homozygotes. A statistically non-significant tendency for similar Val66Met × CA interaction was seen in the group treated with ICBT, but not for those treated as usual. The Met allele carrier group also had higher serum levels of free mBDNF, but these levels were not altered

Funding

This work was financially supported by the Vårdal Foundation, the County Councils in Stockholm, Skåne, Västra Götaland, Västmanland, Kronoberg and Blekinge, the Swedish Medical Research Council (www.vr.se; 2009–5546 YF, 2009–7053 SE, 2010–3631 CL, 2013–2838 SE), the Swedish Brain foundation: FO2016-0311, Karolinska Institutet and the regional agreement on medical training and clinical research between Stockholm County Council and Karolinska Institutet (www.forskningsstod.sll.se; SLL20110560 CL,

Author disclosure

All authors have approved the final version of the article.

Contributors

“MdSR, YF and CL designed the study. YF is the principal investigator (PI) of the Regassa study, CL organized all biological sample collections in regassa and is the PI of the molecular studies in Regassa. VM performed the genotyping and BDNF level measurements under the supervision of CL. MdSR and ZZ conducted statistical analysis under the supervision of YF and CL. MdSR drafted the manuscript. All authors contributed to the interpretation and discussion of the results. All authors critically

Conflict of interest

None.

References (54)

  • M. Szyf

    The early life environment and the epigenome

    Biochim. Biophys. Acta

    (2009)
  • M.S.P. Toups et al.

    Effects of serum Brain Derived Neurotrophic Factor on exercise augmentation treatment of depression

    J. Psychiatr. Res.

    (2011)
  • A.-L. Undén et al.

    Development of a social support instrument for use in population surveys

    Soc. Sci. Med.

    (1989)
  • J. Zhang et al.

    Neurotrophins regulate proliferation and survival of two microglial cell lines in vitro

    Exp. Neurol.

    (2003)
  • L. Zhou et al.

    Upregulation of blood proBDNF and its receptors in major depression

    J. Affect. Disord.

    (2013)
  • Y.-F. Zou et al.

    Meta-analysis of BDNF Val66Met polymorphism association with treatment response in patients with major depressive disorder

    Eur. Neuropsychopharmacol.

    (2010)
  • P.A. Barker

    Whither proBDNF?

    Nat. Neurosci.

    (2009)
  • S. Bonaccorso et al.

    The brain-derived neurotrophic factor (BDNF) Val66Met polymorphism is associated with increased body mass index and insulin resistance measures in bipolar disorder and schizophrenia

    Bipolar Disord.

    (2015)
  • R. Colle et al.

    BDNF/TRKB/P75NTR polymorphisms and their consequences on antidepressant efficacy in depressed patients

    Pharmacogenomics

    (2015)
  • DEPRESSION: A Global Crisis,...
  • J.O. Groves

    Is it time to reassess the BDNF hypothesis of depression?

    Mol. Psychiatry

    (2007)
  • M. Hallgren et al.

    Physical exercise and internet-based cognitive-behavioural therapy in the treatment of depression: randomised controlled trial

    Br. J. Psychiatry

    (2015)
  • K.-H. Joe et al.

    Decreased plasma brain-derived neurotrophic factor levels in patients with alcohol dependence

    Alcohol. Clin. Exp. Res.

    (2007)
  • J. Kaufman et al.

    Social supports and serotonin transporter gene moderate depression in maltreated children

    Proc. Natl. Acad. Sci. U. S. A.

    (2004)
  • R. Keers

    Will gene–environment interactions explain differential antidepressant response?

    Per. Med.

    (2012)
  • R. Keers et al.

    Gene-environment interaction in major depression and antidepressant treatment response

    Curr. Psychiatry Rep.

    (2012)
  • T. Klengel et al.

    Gene × environment interactions in the prediction of response to antidepressant treatment

    Int. J. Neuropsychopharmacol.

    (2013)
  • Cited by (12)

    • Biological markers evaluated in randomized trials of psychological treatments for depression: a systematic review and meta-analysis

      2019, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      A serotonin transporter (Kohen et al., 2011) trial found no interaction effect between treatment and predictor, but in exploratory post-hoc comparisons 3 of the 6 markers tested were associated with differential response to psychotherapy. Another trial on the brain-derived neurotrophic factor Val/Met (Rahman et al., 2017) polymorphism reported the Met/Met genotype did not predict treatment response across intervention arms, nor did it differentially predict response compared to Val genotypes or TAU. The authors did find that the interaction of Met genotypes and childhood adversity predicted treatment response across all arms.

    • Neuropeptide Y, stressful life events and personality trait conscientiousness: Preliminary associations from a Swedish longitudinal study

      2018, Psychiatry Research
      Citation Excerpt :

      With regard to childhood SLEs, the following stressful incidents were considered: (a) the participant lost one or both parents, (b) the participant's parents divorced, (c) there were long periods with economic difficulties in the family, or (d) there were long periods of family friction at home. A summed up continuous score (i.e. 0 − 4) was used in the analyses, similar to previous genetic epidemiological studies (Liu et al., 2015; Rahman et al., 2017). Second, we scored adult adversities, i.e. SLEs that occurred within 12 months preceding the completion of the PART questionnaire at wave 1 and, separately, at wave 2.

    • The serotonin transporter promoter variant (5-HTTLPR) and childhood adversity are associated with the personality trait openness to experience

      2017, Psychiatry Research
      Citation Excerpt :

      In addition, using the PART cohort in previous studies we were able to show childhood adversity effects on social interaction and risk for depression in adulthood (Lavebratt et al., 2010; Liu et al., 2015). Moreover, using the same childhood adversity questionnaire we recently found that childhood adversity influenced the Brain Derived Neurotrophic Factor (BDNF) Val66Met effect on antidepressive treatment response to physical exercise (Rahman et al., 2017). Finally, the genetic single nucleotide polymorphism rs25531, previously reported to render approximately 6% of 5-HTTLPR LL-allele carriers into a medium 5-HTT expression type (Wendland et al., 2006), was not analyzed in our study.

    View all citing articles on Scopus
    View full text