SEARCH

SEARCH BY CITATION

Keywords:

  • biomarker;
  • brain-derived neurotrophic factor;
  • gene;
  • mood disorders;
  • post-mortem brain

Abstract

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Mood disorders, such as major depressive disorder (MDD) and bipolar disorder (BPD), are the most prevalent psychiatric conditions, and are also among the most severe and debilitating. However, the precise neurobiology underlying these disorders is currently unknown. One way to combat these disorders is to discover novel biomarkers for them. The development of such biomarkers will aid both in the diagnosis of mood disorders and in the development of effective psychiatric medications to treat them. A number of preclinical studies have suggested that the brain-derived neurotrophic factor (BDNF) plays an important role in the pathophysiology of MDD. In 2003, we reported that serum levels of BDNF in antidepressant-naive patients with MDD were significantly lower than those of patients medicated with antidepressants and normal controls, and that serum BDNF levels were negatively correlated with the severity of depression. Additionally, we found that decreased serum levels of BDNF in antidepressant-naive patients recovered to normal levels associated with the recovery of depression after treatment with antidepressant medication. This review article will provide an historical overview of the role played by BDNF in the pathophysiology of mood disorders and in the mechanism of action of therapeutic agents. Particular focus will be given to the potential use of BDNF as a biomarker for mood disorders. BDNF is initially synthesized as a precursor protein proBDNF, and then proBDNF is proteolytically cleaved to the mature BDNF. Finally, future perspectives on the use of proBDNF as a novel biomarker for mood disorders will be discussed.

MOOD DISORDERS ARE among the most prevalent, recurrent, and disabling of mental illnesses. Major depressive disorder (MDD) is a serious disorder that affects approximately 17% of the population at some point in life, resulting in major social and economic consequences.1–3 According to the DSM-IV, MDD is a heterogeneous disorder that manifests with symptoms at the psychological, behavioral, and physiological levels. There is still very little known about the neurobiological alterations that underlie the pathophysiology or treatment of MDD. Several lines of evidence suggest that depression in most people is caused by interactions between a genetic predisposition and some environmental factors.3–7

Bipolar disorder (BPD), also known as manic-depressive illness, is a brain disorder that causes unusual shifts in a person's mood, energy, and ability to function. More than 2 million American adults, or about 1 percent of the population aged 18 and older in any given year, have BPD.2,8–10 BPD typically develops in late adolescence or early adulthood. However, some people have their first symptoms during childhood, and some develop them later in life. BPD is often not recognized as an illness, and people may suffer for years before it is properly diagnosed and treated. Like diabetes or heart disease, BPD is a long-term illness that must be carefully managed throughout a person's lifetime.2,4,5,9,10

The precise neurobiology underlying these mood disorders is currently unknown. One way to combat these disorders would be to discover novel biomarkers for them, which potentially could revolutionize their recognition and management.11–14 Identification of biomarkers would aid both in the diagnosis of these disorders, and in the development of effective psychiatric medications to treat them. In addition, biomarkers could provide the basis for early intervention and prevention efforts targeting at-risk individuals. Currently, there are no diagnostic biomarkers available for mood disorders, although easily accessible bodily fluids, including blood, urine, and cerebral spinal fluid (CSF), are potential sources for the identification of biomarkers.

This article provides a review of the recent findings on the role of brain-derived neurotrophic factor (BDNF) in the pathophysiology of mood disorders such as MDD and BPD, and discusses the potential use of BDNF as a biomarker for these disorders.

BDNF AND ITS PRECURSOR, proBDNF

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Mammalian neurotrophins are homodimeric proteins that include nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin (NT)-3, and NT-4/5. Most functions of neurotrophins are mediated by the tropomyosin receptor kinase (Trk) family of tyrosine kinase receptors. The interaction of neurotrophins with the Trk receptors is specific: NGF binds with TrkA; BDNF and NT-4 bind with TrkB; and NT-3 binds with TrkC.15–19 BDNF is found in the central nervous system and periphery. BDNF is a 27-kDa polypeptide that is recognized as playing an important role in the survival, differentiation, and outgrowth of select peripheral and central neurons during development and in adulthood.15,16 It is well known that BDNF participates in use-dependent plasticity mechanisms such as long-term potentiation, learning, and memory.15–17,20,21

The BDNF gene encodes a precursor peptide, proBDNF. BDNF is initially synthesized as a precursor protein (preproBDNF) in the endoplasmic reticulum. Following cleavage of the signal peptide, proBDNF is transported to the Golgi for sorting into either constitutive or regulated secretory vesicles. ProBDNF may be converted into mature BDNF intracellularly in the trans-Golgi by members of the subtilisin–kexin family of endoproteases, such as furin, or in the immature secretory granules by proprotein convertases (Fig. 1).17,22–25 ProBDNF could be converted to mature BDNF by extracellular proteases, such as plasmin and matrix mettaloproteinase-7 (Fig. 1).26,27 It has long been thought that only secreted mature BDNF is biologically active, and that proBDNF is exclusively localized intracellularly, serving as an inactive precursor. However, recent accumulating evidence suggests that proBDNF may also be biologically active.25,27,28 It has been reported that proBDNF induces neuronal apoptosis via activation of the p75NTR receptor.29 Taken together, these findings suggest that proBDNF and BDNF elicit opposite effects via the p75NTR and TrkB receptors, respectively (Fig. 1), and that both proBDNF and BDNF play an important role in several physiological functions.25,30–32

image

Figure 1. The synthesis of brain-derived neurotrophic factor (BDNF) from proBDNF. The BDNF gene produced precursor protein BDNF (preproBDNF) in the endoplasmic reticulum. ProBDNF binds to intracellular sortilin in the Golgi to facilitate proper folding of the mature domain. ProBDNF preferentially binds p75NTR receptors. ProBDNF is cleaved by proteases (e.g. plasmin, matrix metalloproteinase [MMP]-7) at the synapses and converted to mature BDNF. Mature BDNF preferentially binds tropomyosin receptor kinase (Trk) B receptor. This figure is a slight modification of that from the article by Hashimoto.31

Download figure to PowerPoint

PRECLINICAL STUDIES

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Antidepressant effects of BDNF

BDNF has been shown to have antidepressant effects in animal models of depression. Specifically, infusion of BDNF into the midbrain has an antidepressant-like influence on two animal models of depression, i.e. learned helplessness following exposure to inescapable shock, and learned helplessness following exposure to inescapable shock as well as in response to the forced-swim test.33 Furthermore, it has been reported that forced swimming decreased BDNF mRNA in particular regions (CA1, CA3, and the dentate gyrus) of the hippocampus, and that a combination of physical activity and antidepressant treatment increased the level of hippocampal BDNF mRNA to well above the baseline value as well as enhanced swimming time in an animal model,34 supporting the possibility that the upregulation of BDNF expression may be important in the clinical response to antidepressant treatment. Moreover, Shirayama et al.35 reported that a single bilateral infusion of BDNF into the dentate gyrus of the hippocampus produced an antidepressant effect in both the learned helplessness and forced-swim test paradigms. It should be noted that these effects were observed as early as 3 days after a single infusion of BDNF, and these effects lasted for at least 10 days.35 In addition, infusions of a broad-spectrum Trk tyrosine kinase inhibitor, K252a, or of a selective extracellular signal-regulated protein kinase (ERK) inhibitor, U0126, blocked the antidepressant effects of BDNF, suggesting that the TrkB/mitogen-activated protein (MAP) kinase cascade plays a role in the therapeutic action of antidepressants.35 Interestingly, it has been demonstrated that intra-ventral tegmental area (VTA) infusions of BDNF resulted in 57% shorter latency to immobility relative to control animals (a depression-like effect), and that rats given intra-nucleus accumbens (NAc) injections of a virus expressing the truncated version of the BDNF receptor had almost fivefold longer latency to immobility relative to rats that received a vehicle injection or a virus expressing the full-length version of the BDNF receptor (an antidepressant-like effect). These findings suggest that the action of BDNF in the VTA-NAc pathway might be related to development of a depressive-like phenotype.36

BDNF gene knock-out (or knock-in) mice

Based on the putative role of BDNF in depression, it is of interest to evaluate heterozygous BDNF knock-out (BDNF±) mice as a potential animal model of the disease. Two studies using mice with a genetic deletion of the BDNF gene have demonstrated that BDNF play a critical role in neuronal differentiation and survival.37,38 Heterozygous BDNF knock-out (BDNF±) mice that do not show abnormal mortality have been widely used because BDNF null mutant (BDNF−/−) mice die during the first few weeks after birth.37,38 Heterozygous BDNF knock-out (BDNF±) mice have forebrain BDNF mRNA and protein levels that are 50% of the levels in wild-type mice.39–41 In addition to abnormalities in serotonergic neurotransmission, these mice develop premature age-associated decrements in forebrain serotonin levels and fiber density, and exhibit enhanced intermale aggressiveness,42,43 as well as abnormal eating behaviors.43–45 However, MacQueen et al.46 have reported that heterozygous BDNF knock-out (BDNF±) mice and wild-type mice did not differ in measures of activity, exploration, or hedonic sensitivity, and also did not differ in response to the forced-swim test. Although heterozygous BDNF±mice were slower to escape after training than wild-type mice in the learned helplessness paradigm, this effect may have been due to a reduced sensitivity to centrally mediated pain.46 In the forced-swim test, the immobility time of imipramine-treated heterozygous BDNF±mice was not significantly different from that of the saline-treated (wild-type or heterozygous BDNF knock-out) mice.47 At the present time, therefore, it seems unlikely that heterozygous BDNF±mice will provide an effective model of genetic vulnerability to depression, although further detailed studies will be needed to confirm this.48

The development of a conditional targeting strategy for the BDNF gene will help to discriminate between the roles of BDNF during development and adulthood, as well as between central and peripheral contributions to impaired function in the absence of BDNF. It would also be of interest to evaluate conditional BDNF knock-out mice as an animal model of depression. Monteggia et al.49 showed that conditional BDNF knockout mice also display an increase in depression-like behavior in the forced-swim and sucrose preference tests, suggesting that low production of BDNF may precipitate depressive disorder. Subsequently, Adachi et al.50 used a viral-mediated gene transfer approach to assess the role of BDNF in subregions of the hippocampus. The loss of BDNF in either the CA1 or the dentate gyrus of the hippocampus did not alter locomotor activity, anxiety-like behavior, fear conditioning, or depression-related behaviors. However, the selective loss of BDNF in the dentate gyrus attenuated the actions of antidepressants (e.g. desipramine and citalopram) in the forced-swim test. These data suggest that BDNF in the DG might be essential in mediating the therapeutic effect of antidepressants.50

Chen et al.51 generated a variant proBDNF (66Met/Met) mouse that produces the phenotypic hallmarks in humans with a variant allele, and, when placed in stressed settings, proBDNF (66Met/Met) mice exhibited increased anxiety-related behaviors that were not attenuated by fluoxetine. This study suggests the impact of Met substitution of proBDNF on anxiety-related behaviors.31

Expression of BDNF by antidepressants

Several lines of evidence suggest that the expression of BDNF may be a downstream target of a variety of antidepressant treatments; BDNF might therefore be an important target for therapeutic recovery from depression, and it might also provide protection against stress-induced neuronal damage.6,7,52–58 The molecular elements known to regulate neuronal plasticity in models of learning and memory are also involved in the actions of drugs used for the treatment of MDD and BPD.6,7,52–58 Administration of antidepressants increased the expression of BDNF mRNA in limbic structures in response to chronic, but not acute, treatment; these results are consistent with the time course typically required for the therapeutic action of antidepressants to take effect.59–61 Subsequently, a number of studies reported an increase of BDNF protein levels in the brain after chronic treatment with antidepressants.57 In addition, the expression of BDNF has been shown to be decreased by exposure to stress,62–64 suggesting the possibility that BDNF might also be involved in the pathophysiology of stress-related mood disorders.

Electroconvulsive therapy

Electroconvulsive therapy (ECT) can be effectively used in the treatment of MDD.65,66 Electroconvulsive seizures increase the levels of both BDNF mRNA and its receptor TrkB mRNA in the rat hippocampus, and chronic administration of electroconvulsive seizures blocked the downregulation of BDNF mRNA in the hippocampus in response to restraint-induced stress.59,67 Electroconvulsive stimulus applied for 8 days increased the levels of BDNF protein in the hippocampus, striatum, and occipital cortex of rats.68 Furthermore, ten consecutive daily exposures to electroconvulsive seizures increased BDNF protein in the parietal cortex (219%), entorhinal cortex (153%), hippocampus (132%), frontal cortex (94%), neostriatum (67%), and septum (29%) of rats.39 Increases of BDNF protein peaked at 15 h after the last electroconvulsive seizures and lasted at least 3 days thereafter.39 Moreover, chronic treatment with electroconvulsive seizures induces sprouting of the granule cell mossy fiber pathway in the hippocampus, and electroconvulsive seizure-induced sprouting is significantly diminished in heterozygous BDNF±mice, suggesting that BDNF contributes to mossy fiber sprouting.69 Thus, the sprouting of the mossy fiber pathway would appear to oppose the mechanism of the action of stress, and could thereby contribute to the therapeutic effects of ECT; however, the functional consequences of ECT remain unclear at this time.

Gersner et al.70 reported that repeated subconvulsive electrical stimulation of either the nucleus accumbens or the ventral but not the dorsal prelimbic cortex reversed the main behavioral deficit and the reduction of BDNF levels in the hippocampus that were induced by chronic mild stress. ECT was more effective because it also normalized a behavioral deficit associated with anxiety but produced a learning and memory impairment. This study suggests that local intermittent subconvulsive electrical stimulation can induce an antidepressant effect similar to that of ECT, without the cognitive impairment caused by the convulsive treatment.

Repetitive transcranial magnetic stimulation (rTMS)

Repetitive transcranial magnetic stimulation (rTMS) has been increasingly used as a therapeutic tool for the treatment of psychiatric diseases such as MDD.65,71–73 It has been reported that rTMS increased BDNF mRNA and BDNF protein in the hippocampus as well as in the parietal and piriform cortex.74 Such findings are similar to those observed after antidepressant drug treatment and electroconvulsive stimuli; it is therefore likely that rTMS and antidepressant treatment strategies share a common molecular mechanism of action.

Mood stabilizers

Lithium and valproate are widely prescribed mood stabilizers that effectively treat acute mood disorders, in addition to providing long-term prophylactic treatment of BPD. Although the mechanisms underlying the therapeutic effects of these drugs remain poorly understood, recent studies have implicated certain critical signal transduction pathways as being integral to the pathophysiology and treatment of BPD.9,54,55,75–77

Chronic administration of lithium or valproate has been shown to increase the expression of BDNF in the rat brain, suggesting that these mood stabilizers may produce a neurotrophic effect mediated by the upregulation of BDNF in the brain.78 Furthermore, pretreatment with either lithium or BDNF protected rat cortical neurons from glutamate excitotoxicity, and a BDNF-neutralizing antibody as well as K252a (an inhibitor of Trk tyrosine kinase) were both able to suppress the neuroprotective effects of lithium.79 Moreover, subchronic administration of another mood stabilizer, lamotrigine (30 mg/kg, twice-daily for 7 days), increased the expression of BDNF in the frontal cortex and hippocampus in both naive and stressed rats, and restored the stress-induced reduction of BDNF expression.80 All these findings suggest that the BDNF/TrkB pathway could play a role in mediating the neuroprotective effects of mood stabilizers.81

CLINICAL STUDIES

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES

BDNF gene polymorphisms and brain functions

The human BDNF gene maps to human chromosome 11p13 and is composed of six 5′ exons that are differentially spliced to a single 3′ terminal exon (exon 7) that encodes the entire sequence of mature BDNF.82 The BDNF 196G/A gene polymorphism converts a valine (Val) to methionine (Met) at codon 66 in the 5′ pro-region of the human BDNF gene (Fig. 2). Neurons transfected with GFP-labeled BDNF-66Met showed lower depolarization-induced secretion, while constitutive secretion was unchanged. Furthermore, GFP-labeled BDNF-66Met failed to localize to secretory granules or synapses.83 Interestingly, it has been demonstrated that the BDNF 196G/A gene polymorphism is associated with episodic memory, hippocampal activation (as assayed by functional magnetic resonance imaging [fMRI]), and decreased hippocampal N-acetyl aspartate (NAA) levels (as measured by magnetic resonance spectroscopy [MRS]), indicating that the BDNF gene plays a role in human memory and hippocampal function.83 In addition, a study using blood oxygenation level-dependent fMRI and a declarative memory task in healthy individuals has demonstrated an association between the BDNF gene 196G/A polymorphism and hippocampal activity during episodic memory processing.84

image

Figure 2. Structure of pro-brain-derived neurotrophic factor (BDNF) protein. Arrowheads indicate known protease cleavage sites involved in processing of the mature BDNF form, as well as of the secreted proBDNF. An arrow indicates the position of the single nucleotide polymorphism (SNP: 196G/A, Val66Met) of the BDNF gene. Met, methionine; Val, valine.

Download figure to PowerPoint

Whereas this BDNF 196G/A gene polymorphism does not affect mature BDNF protein function, it has been shown to dramatically alter the intracellular trafficking and packaging of proBDNF.30,31,83 These findings suggest that the BDNF gene is involved in human episodic memory by virtue of its effects on hippocampal neuronal function, and that the BDNF gene might contribute to the pathogenesis of neuropsychiatric diseases such as mood disorders.

BDNF gene analysis in MDD

Mood disorders are common psychiatric disorders with complex causes that likely involve multiple genes in addition to non-genetic influences.85–87 Given the role of BDNF in the pathophysiology of MDD, it is of interest to examine the BDNF gene as potentially associated with the risk of developing MDD. However, in studies in a Chinese population, the genotype and/or allele frequencies of the BDNF 196G/A gene polymorphism were not significantly different among subjects in the MDD and control groups, suggesting that the BDNF 196G/A gene polymorphism plays no major role in the pathogenesis of MDD in the Chinese population.88,89 Furthermore, it has been reported that the BDNF 196G/A polymorphism was not related to the development of MDD but was related to the clinical features of MDD in a Japanese population90 and an Italian population.91 In contrast, an association has been demonstrated between the BDNF 196G/A polymorphism and geriatric depression in a Chinese population92 and depression associated with Alzheimer's disease in an Italian population.93

Schumacher et al.94 reported that haplotype analysis of the marker combination rs988748-(GT)n-rs6265 of the BDNF gene produced nominally significant associations for all investigated phenotypes (MDD, P = 0.00006; BPD, P = 0.0057). Association with MDD was the most robust finding and could be replicated in a second German sample of MDD patients and control subjects (P = 0.0092), suggesting that BDNF may be a susceptibility gene for MDD in the German population.94

Licinio et al.95 reported that six single-nucleotide polymorphisms on the BDNF gene (rs12273539, rs11030103, rs6265, rs28722151, rs41282918, and rs11030101) were associated with MDD and that two haplotypes in different blocks were significantly associated with MDD. Interestingly, a 5′ untranslated region polymorphism (rs61888800) was associated with antidepressant response after adjusting for age, sex, medication, and baseline score on the 21-item Hamilton Depression Rating Scale. However, there was no major impact of the BDNF gene on antidepressant treatment response.96 A recent meta-analysis study demonstrated an association between BDNF Val66Met polymorphism and the treatment response in patients with MDD, with Val66Met heterozygous patients showing a better response rate than the Val/Val homozygotes, especially in Asian populations.97 Further gene–gene and gene–environment interaction studies based on larger sample sizes and stratified by ethnicity will be necessary. Very recently, it was reported that a combination of several independent risk alleles within the TrkB (or NTRK2) gene were associated with suicide attempts among patients with MDD, suggesting that the BDNF-TrkB pathway plays a role in the pathophysiology of suicide.98

In vivo brain imaging studies have demonstrated reduced hippocampal volumes in patients with MDD.99–103 Interestingly, significantly smaller hippocampal volumes were observed for MDD patients and for controls carrying the BDNF-66Met allele compared with subjects homozygous for the BDNF-66Val allele in a German population,104 suggesting that BDNF-66Met allele carriers might be at risk of developing smaller hippocampal volumes and may be susceptible to MDD.104,105 We reported that the frequency of healthy individuals who carried the G/G (Val/Val) genotype was significantly lower in Japanese subjects than in Italians or in Americans (Table 1), suggesting an ethnic difference in the frequency of the BDNF 196G/A polymorphism.106 Given this possibility of an ethnic difference, further detailed studies using other ethnic samples will be needed to confirm the association between the BDNF Val66Met polymorphism and the reduction of hippocampal volume.

Table 1.  Allele and genotype frequencies of the brain-derived neurotrophic factor gene polymorphism at position 196
 JapanItalyUSA
  1. A slight modification of the table by Shimizu et al.106 reproduced with permission.

  2. Met, methionine; Val, valine.

Allele   
 Allele A (Met)41.1 %29.7 %18.0 %
 Allele G (Val)58.9 %70.3 %82.0 %
Genotype   
 A/A (Met/Met)15.9 %8.1 %4.5 %
 G/A (Val/Met)50.3 %43.2 %27.1 %
 G/G (Val/Val)33.8 %48.7 %68.4 %

BDNF gene analysis in BPD

The family-based association studies have suggested that the BDNF gene is a potential risk locus for the development of BPD.107,108 The dinucleotide repeat (GT)n polymorphism at position –1040 bp on the BDNF gene showed that allele A3 was preferentially transmitted to affected individuals (P = 0.04), and that the BDNF 196G/A (val66met) polymorphism showed a significant association with BPD (P = 0.0006).107 Furthermore, an association study of the relationship between 76 candidate genes and BPD suggested that BDNF is a potential risk gene for the disease.108 It is of note that of the 76 candidate genes involving a variety of neurobiological systems, only the BDNF gene emerged as a potential risk locus after genotyping and haplotyping studies were carried out using the original trios and in a replication sample.108 Subsequent association studies replicated an association between the BDNF gene and BPD.109 However, no evidence has been found for an allelic or genotypic association of the two polymorphisms (-1360C/T and 196G/A) of the BDNF gene in Japanese (or Chinese) patients with BPD,89,110 suggesting that the BDNF gene is unlikely to confer susceptibility to BPD in the Asian population. As mentioned above, an ethnic difference in the frequency of the BDNF 196G/A polymorphism has been suggested.106 This phenomenon should be taken into consideration when verifying the role of certain variations in the BDNF gene as risk factors for BPD.

Post-mortem human brain sample studies

Several studies using post-mortem brain samples have suggested that BDNF plays a role in the pathophysiology of mood disorders. A post-mortem study of patients with MDD and BPD demonstrated evidence of the specific loss of neuronal and glial cells in mood disorders.111 Three patterns of morphometric cellular changes were noted: cell loss (subgenual prefrontal cortex), cell atrophy (dorsolateral prefrontal cortex and orbitofrontal cortex), and increased numbers of cells (hypothalamus, dorsal raphe nucleus). Thus, cellular changes in patients with mood disorders are suggested to play a role in stress and prolonged prefrontal cortex development, and neurotrophic/neuroprotective factors are also suggested to be involved in these disorders. The precise anatomic localization of dysfunctional neurons and glia in mood disorders may reveal novel therapeutic targets for mood disorders, including MDD and BPD.58

Chen et al.112 observed increased levels of BDNF immunoreactivity in post-mortem hippocampal tissue obtained from subjects who were being treated with antidepressant medications at the time of death, compared with the BDNF levels observed in samples from antidepressant-untreated subjects. Interestingly, Dunham et al.113 reported that reductions in proBDNF were seen in all layers of the right but not the left hippocampus, with no changes in the dentate gyrus of the brains of MDD patients from the Stanley consortium. The pattern was similar but less marked for BPD. In addition, BPD, but not MDD patients, had bilateral reductions in p75NTR in hippocampal layers, but not in the dentate gyrus. These findings suggest that both MDD and BPD may be associated with impairment in proBDNF expression,113 although further detailed studies on proBDNF will be needed to confirm this.

Suicide is a major public health problem that is at least partly related to mood disorders.114,115 Dwivedi et al.116 reported that the mRNA levels of BDNF and its receptor TrkB were significantly reduced in both the prefrontal cortex and hippocampus of suicide subjects, as compared with those in control subjects. These reductions were associated with significant decreases in the protein levels of BDNF and of full-length TrkB. These findings suggest that the BDNF-TrkB pathway may play an important role in the pathophysiological aspects of suicidal behavior.116 Furthermore, Karege et al.117 reported a significant decrease in BDNF and NT-3 levels in the hippocampus and prefrontal cortex (only BDNF), but not in the entorhinal cortex, of suicide victims who were drug-free compared with non-suicide controls. In drug-treated suicide victims, neurotrophin levels were not significantly different from those in non-suicide controls.117 Moreover, BDNF protein was significantly decreased in the prefrontal cortex, but not the hippocampus, of teenage suicide victims compared with normal control subjects.118 In addition, a decrease of the T1 splice variant of TrkB has been detected in the frontal cortex of suicide completers.119

Very recently, it has been reported that post-mortem brain samples from suicide subjects showed a statistically significant increase in DNA methylation at specific CpG sites in the BDNF promoter/exon IV compared with non-suicide control subjects, suggesting a novel link between epigenetic alteration in the brain and suicidal behavior.120 All these findings support a role of the BDNF-TrkB pathway in the suicidal behavior associated with pathogenesis of mood disorders.

Blood levels of BDNF in patients with MDD

BDNF is also present in human blood, although it is more highly concentrated in brain tissue.121,122 Previously, it was reported that BDNF could cross the blood–brain barrier,123 and that BDNF levels in the brain and serum underwent similar changes during the maturation and aging processes in rats,124 suggesting that serum BDNF levels may reflect BDNF levels in the brain. Karege et al.125 reported that serum BDNF levels were significantly decreased in antidepressant-free patients with MDD, and that serum BDNF levels were negatively correlated with the severity of depression. We reported that serum BDNF levels in antidepressant-naive patients with MDD were significantly lower than those of patients medicated with antidepressants and normal controls; we also demonstrated that serum BDNF levels were negatively correlated with the severity of depression.126 Interestingly, we reported a preliminary finding showing that decreased serum BDNF levels in antidepressant-naive patients recovered to normal levels associated with lower Hamilton Depression Rating Scale scores after treatment with antidepressant medication.126 Subsequently, a number of reports have shown decreased BDNF levels in patients with MDD and increased BDNF levels in patients treated with antidepressants.127–136 Two meta-analysis studies on blood levels in MDD have been reported. One meta-analysis study demonstrated strong evidence that BDNF levels were lower in depressed subjects than healthy control subjects (P < 6.8 × 10−8), and that BDNF levels were significantly (P = 0.003) increased after antidepressant treatment.137 As shown in Figure 3, serum BDNF levels in patients with MDD were lower than those of normal controls in all reports.125–136 The other meta-analysis similarly showed that BDNF levels increased significantly after antidepressant treatment (effect size: 0.62), and that there was a significant (P = 0.02) correlation between changes in BDNF level and depression score changes.138 Recently, it has been reported that plasma BDNF and tissue-type plasminogen activator (tPA) levels were significantly lower in patients with late-onset geriatric depression, suggesting the role of BDNF and tPA in late-onset geriatric depression.139 Taken together, these findings suggest that serum BDNF levels are likely to be a biomarker for MDD, and support the notion that amelioration of the disease might be associated with the neuroplastic changes achieved by antidepressant treatment.

image

Figure 3. Serum brain-derived neurotrophic factor (BDNF) levels in (inline image) healthy control subjects and (inline image) patients with major depressive disorder. The data are the mean of each study. The data show the first author name and the reference number for each article. This figure is a slight modification of that from the article by Hashimoto.136

Download figure to PowerPoint

It has been reported that ECT increased serum (or plasma) BDNF levels in treatment-resistant patients with MDD,140–143 although some reports did not replicate these findings.144,145 In addition, acute prefrontal cortex TMS was reported not to affect the serum BDNF levels in healthy control subjects.146 Further detailed studies of the effects of ECT (or TMS) on blood levels of BDNF will be needed.

Blood levels of BDNF in patients with BPD

Accumulating evidence suggests a role of BDNF in the pathogenesis of BPD. It has been reported that blood levels of BDNF were decreased in BPD patients during manic,147–150 depressed,147,150,151 and even euthymic states.135 But these findings were not replicated in other reports.151–154 A recent meta-analysis study demonstrated that patients with BPD had lower levels of BDNF than healthy controls (P = 1 × 10−4).155 Based on different affective status, however, the results showed statistically significant differences in BDNF only between the patients in a manic (P = 0.0008) or depressed (P = 0.02) state and the controls, and not between the patients in a euthymic state and the controls (P = 0.25). In addition, BDNF levels were significantly (P = 0.01) increased after the pharmacological treatment of the manic state. These findings indicate that BDNF levels are abnormally reduced in manic and depressed states of BPD, and that the reduced level in the manic state increases after pharmacological treatment. This suggests a potential role of the blood BDNF level as a state-dependent biomarker of BPD.

Blood levels of BDNF in patients with other psychiatric diseases

Depression is a comorbidity of several psychiatric diseases, such as eating disorders,156 panic disorder,157 and schizophrenia.158 It has been reported that serum levels of BDNF in patients with eating disorders (anorexia nervosa and bulimia nervosa) were lower than those of normal controls,159–161 suggesting a role of BDNF in the pathophysiology of eating disorders.162 Furthermore, Kobayashi et al.163 reported that serum BDNF levels of patients with a poor response to group cognitive behavioral therapy (CBT) were significantly lower than those of patients with a good response to group CBT, suggesting that BDNF might contribute to the therapeutic response of panic disorder. Moreover, Strohle et al.164 reported that serum BDNF levels in patients with panic disorder were lower than those of healthy control subjects, and that acute exercise significantly increased serum BDNF levels in these patients. This study suggests that acute exercise ameliorates reduced BDNF levels in patients with panic disorder, although it is uncertain whether long-term exercise training also increases serum BDNF levels. In addition, it is also reported that serum BDNF levels in patients with MDD or conversion disorder are significantly lower than those of normal controls, suggesting a similar role of BDNF in the pathophysiology of MDD and conversion disorder.165

There are many articles on decreased serum BDNF levels in chronic patients with schizophrenia,166 although no change of serum BDNF in drug-naive patients was reported by our group.167 A further, detailed meta-analysis study on serum BDNF levels in schizophrenia will be necessary to examine the possibility of using BDNF as a biomarker for schizophrenia.

FUTURE RESEARCH DIRECTIONS

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES

As described above, BDNF plays an important role in the pathophysiology of mood disorders such as MDD and BPD, and it is likely that the serum BDNF level is a biomarker of these disorders. Given the opposing biological effects of proBDNF and BDNF, it would be of great interest to study the precise mechanisms controlling the cleavage of proBDNF to BDNF. Considering the role of BDNF in the pathophysiology of mood disorders, it is likely that its precursor, proBDNF, also plays an important role in the pathophysiology of these disorders. The commercially available enzyme-linked immunosorbent assays (ELISA) for BDNF cannot address differences in the relative amounts of each form of BDNF, because the antibodies can recognize only the region of mature BDNF. Currently, there are two reports showing the presence of proBDNF in human serum168 and human saliva169 using an immunoblot method. Therefore, it will be important to determine the blood levels of both proBDNF and BDNF as novel biomarkers for mood disorders. In this regard, there is an urgent need to develop a highly sensitive ELISA system that can distinguish between proBDNF and BDNF.

The Val66Met gene variant of the BDNF gene is thought to affect intracellular trafficking and BDNF secretion, and has been associated with hippocampal volume and episodic memory. Interestingly, it has been demonstrated that BDNF levels in the amniotic fluids of subjects with proBDNF-66Met carriers (Met/Met and Met/Val) were significantly lower than those of non-carriers (Val/Val).170 It is also anticipated that both proBDNF-66Met and proBDNF-66Val could exist in the human bodily fluids of subjects with the BDNF 196G/A genotype, since the frequency of the BDNF 196G/A genotype is highest in the Japanese population (Table 1).106 It is therefore possible that the conversion process of the two proBDNF to mature BDNF may be different in subjects with the BDNF 196G/A genotype, suggesting that the levels of proBDNF-66Met and proBDNF-66Val in human bodily fluids may differ in the subjects with the BDNF 196G/A genotype. If a highly sensitive ELISA system could be developed to distinguish between proBDNF-66Met and proBDNF-66Val, it would be possible to quantify the levels of these two proBDNF, as well as the mature BDNF levels in the bodily fluids of healthy subjects and patients with mood disorders (Fig. 4). Furthermore, it would be of great interest to study the possibility of using these two proBDNF proteins (proBDNF-66Val and proBDNF-66Met) as novel biomarkers of mood disorders (Fig. 4).

image

Figure 4. Future perspectives of novel biomarkers for mood disorders. Both pro-brain-derived neurotrophic factor (BDNF)-66methionine(Met) and proBDNF-66valine(Val) could exist in the human bodily fluids of subjects with the BDNF 196G/A genotype. Therefore, these two proBDNF proteins (proBDNF-66Val and proBDNF-66Met) would be novel biomarkers of mood disorders. CSF, cerebral spinal fluid; ELISA, enzyme-linked immunosorbent assays.

Download figure to PowerPoint

Accumulating evidence suggests the important role of BDNF in glutamatergic neurotransmission;171,172 glutamate has also been implicated in the pathophysiology of mood disorders such as MDD.58,173–176 Therefore, it will also be interesting to study the correlations between proBDNF (or BDNF) and glutamate-related compounds as biomarkers of mood disorders. Finally, further detailed studies on the correlations between proBDNF (or BDNF) and glutamate would contribute to the development of novel therapeutic drugs for mood disorders.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES

This study was supported partly by a grant from the Ministry of Health, Labor and Welfare, Japan.

REFERENCES

  1. Top of page
  2. Abstract
  3. BDNF AND ITS PRECURSOR, proBDNF
  4. PRECLINICAL STUDIES
  5. CLINICAL STUDIES
  6. FUTURE RESEARCH DIRECTIONS
  7. ACKNOWLEDGMENTS
  8. REFERENCES