Hostname: page-component-7c8c6479df-ph5wq Total loading time: 0 Render date: 2024-03-28T17:18:12.812Z Has data issue: false hasContentIssue false

Molecular differentiation of schizoaffective disorder from schizophrenia using BDNF haplotypes

Published online by Cambridge University Press:  02 January 2018

Todd Lencz*
Affiliation:
Center for Translational Psychiatry, Feinstein Institute for Medical Research and Division of Psychiatry Research, The Zucker Hillside Hospital, New York
Robert H. Lipsky
Affiliation:
Laboratory of Neurogenetics, National Institute on Alcohol Abuse and Alcoholism, Bethesda
Pamela DeRosse
Affiliation:
Division of Psychiatry Research, The Zucker Hillside Hospital
Katherine E. Burdick
Affiliation:
Center for Translational Psychiatry, Feinstein Institute for Medical Research and Division of Psychiatry Research, The Zucker Hillside Hospital, New York
John M. Kane
Affiliation:
Center for Translational Psychiatry, Feinstein Institute for Medical Research and Department of Psychiatry, The Zucker Hillside Hospital
Anil K. Malhotra
Affiliation:
Center for Translational Psychiatry, Feinstein Institute for Medical Research and Division of Psychiatry Research, The Zucker Hillside Hospital, New York, USA
*
Correspondence: Todd Lencz, The Zucker Hillside Hospital, Psychiatry Research, 75–59 263rd Street, Glen Oaks, New York 11004, USA. Email: lencz@lij.edu
Rights & Permissions [Opens in a new window]

Abstract

Background

Allelic variation in the gene encoding brain-derived neurotrophic factor (BDNF) has been associated with affective disorders, but generally not schizophrenia. Brain-derived neurotrophic factor variants may help clarify the status of schizoaffective disorder.

Aims

To test the hypothesis that BDNF haplotypes are associated with psychiatric illness marked by a prominent affective component.

Method

Frequencies of a 5-marker BDNF haplotype were examined in 600 White participants across four diagnostic categories and healthy controls.

Results

Individuals with schizoaffective disorder and other affective disorders were significantly more likely to carry two copies of the most common BDNF haplotype (containing the valine allele of the Val66Met polymorphism) compared with healthy volunteers. Moreover, when compared with people with schizophrenia, individuals with schizoaffective disorder were significantly more likely to carry two copies of the common haplotype.

Conclusions

To our knowledge, this is the first candidate gene study to demonstrate association with schizoaffective disorder but not schizophrenia. Variation in the BDNF gene may be associated with the clinical phenotype of affective dysregulation across several DSM–IV diagnostic categories.

Type
Papers
Copyright
Copyright © Royal College of Psychiatrists, 2009 

Since its introduction as a diagnostic entity, Reference Kasanin1 schizoaffective disorder has held an uncertain role in psychiatric nosology. Although Kraepelin himself had acknowledged the presence of such intermediate cases, Reference Marneros2 the so-called ‘Kraepelinian dichotomy’ (dementia praecox v. manic–depressive illness) has guided both clinical reasoning and pathophysiological research for a century. Consequently, clinicians and researchers often consider schizoaffective disorder to be a subtype of schizophrenia or bipolar disorder, each of which is seen as a discrete nosological entity. Reference Craddock, O'Donovan and Owen3 It is possible, however, that careful comparison of schizoaffective disorder with the other ‘major’ diagnostic categories may yield insights into the underlying nature of these disorders. Data from family studies Reference Cardno, Rijsdijk, Sham, Murray and McGuffin4 and population genetics Reference Laursen, Labouriau, Licht, Bertelsen, Munk-Olsen and Mortensen5 provide strong evidence that schizoaffective disorder shares a familial relationship with both schizophrenia and bipolar disorder. Moreover, molecular genetic studies are increasingly finding polymorphic sequence variants that increase susceptibility to mental illness across traditional diagnostic boundaries. Reference Funke, Malhotra, Finn, Plocik, Lake and Lencz6Reference Williams, Green, Macgregor, Dwyer, Norton and Williams8 Although it has been suggested that molecular genetics will introduce new nosological conceptualisations by undermining simple categorical constructs, Reference Kendler9 a genetic association approach has not yet been utilised to refine our understanding of schizoaffective disorder.

The present study was designed to examine genetic variation in the gene encoding brain-derived neurotrophic factor (BDNF) in people diagnosed with schizoaffective disorder, in comparison with people diagnosed with schizophrenia, bipolar disorder and unipolar depression, as well as healthy volunteers. Brain-derived neurotrophic factor was selected because it is highly expressed in critical brain regions such as the hippocampus, amygdala and striatum, where it is involved in key processes that may be implicated in psychopathology; such functions include associative learning and memory, fear conditioning and social defeat stress. Reference Rattiner, Davis and Ressler10Reference Egan, Kojima, Callicott, Goldberg, Kolachana and Bertolino12 In addition, the association of several psychiatric diagnoses with polymorphisms located within BDNF, especially the Val66Met amino acid substitution, has been examined in multiple studies. Several family studies have reported overtransmission of the Val66Met valine allele in bipolar disorder, Reference Sklar, Gabriel, McInnis, Bennett, Lim and Tsan13Reference Müller, de Luca, Sicard, King, Strauss and Kennedy16 and a few studies have also demonstrated an association of other BDNF polymorphisms with major depression. Reference Strauss, Barr, George, Devlin, Vetro and Kiss17,Reference Schumacher, Jamra, Becker, Ohlraun, Klopp and Binder18 By contrast, association studies of BDNF in schizophrenia have been largely negative, although one case–control study demonstrated a significant over-representation of the valine66 allele in people with schizophrenia or schizoaffective disorder. Reference Neves-Pereira, Cheung, Pasdar, Zhang, Breen and Yates19 Based on these data, BDNF variation has been hypothesised to be related to the affective components of psychiatric illness. Reference Craddock, O'Donovan and Owen3,Reference Duman and Monteggia20 The present study was designed to test the hypothesis that frequency of common variants within the BDNF gene would be associated with affective illnesses, including schizoaffective disorder (as well as bipolar disorder and major depressive disorder), but not schizophrenia.

Methods

Participants

A total of 381 individuals were recruited from the in-patient and out-patient clinical services of the Zucker Hillside Hospital, a division of the North Shore–Long Island Jewish Health System. The Structured Clinical Interview for DSM–IV Axis I disorders (SCID, version 2.0) Reference First, Spitzer, Williams and Gibbon21 was administered by trained raters. All participants were screened, recruited and assessed by the same ratings team at a single site, the Zucker Hillside Hospital. On average, participants had spent 13.52 years (s.d.=8.91) in treatment at our institution (in-patient and out-patient), therefore allowing us to capture the majority of their illness course with our own medical records. All available family members (over age 18) were also interviewed to enhance diagnostic validity. Thus, the diagnostic process included the SCID, detailed medical record review, interviews with family members, consultation with the clinical treatment team and formulation of a detailed two to three page case summary of each participant that was discussed in a diagnostic consensus conference led by the project principal investigator (A.K.M.).

We have also operationalised DSM–IV Criterion C for schizoaffective disorder, which differentiates it from schizophrenia. 22 Criterion C requires that mood symptoms meeting criteria for a mood episode ‘must be present for a substantial portion of the entire period of illness’ for a diagnosis of schizoaffective disorder to be made. Our operationalised criteria took ‘a substantial portion’ to be greater than 20% of duration of illness. Following these procedures, all but three individuals (0.8%) were assigned a consensus diagnosis of schizophrenia, schizoaffective disorder, bipolar disorder or unipolar major depressive disorder. The consensus diagnosis and demographic information for these 378 individuals are presented in Table 1.

Table 1 Demographics of the five participant groups (total n=600)

n (% male) Mean age (s.d.)
Schizophrenia 211 (73.9) 38.3 (10.3)
Schizoaffective disorder 61 (52.5) 38.6 (12.2)
Bipolar I 77 (53.2) 36.6 (11.8)
Major depressive disorder 29 (44.8) 42.4 (9.4)
Healthy controls 222 (41.0) 47.6 (18.6)

Healthy controls (n=222) were recruited by use of local newspaper advertisements, flyers and community internet resources and underwent initial telephone screening to assess eligibility criteria. Participants who met eligibility criteria were administered the non-patient SCID (SCID–NP) to rule out the presence of an Axis I psychiatric disorder; a urine toxicology screen for drug use and an assessment of the participant's family history of psychiatric disorders were also performed. Reference First, Spitzer, Williams and Gibbon23 Exclusion criteria included (current or past) Axis I psychiatric disorder, psychotropic drug treatment, substance misuse, a first-degree family member with an Axis I psychiatric disorder and the inability to provide written informed consent. All participants were White by self-report and drawn from a single geographic location (surrounding Glen Oaks, New York, USA). After complete description of the study, written informed consent was obtained from all participants. As shown in Table 1, the groups significantly differed in age (F4,558=15.5, P<0.001). By design, somewhat older individuals were selectively recruited for the healthy control group so that they might be outside the window of maximal risk for onset of schizophrenia, schizoaffective disorder and bipolar disorder. As is typical in schizophrenia studies, males were over-represented among those with schizophrenia compared with the other groups (χ2=49.6, d.f.=1, P<0.001).

Genotyping

Genotyping was performed by 5′-exonuclease assay with allele-specific flourescence detection probes. As shown in online Fig. DS1, five single nucleotide polymorphisms (SNPs) encompassing the single protein-coding exon of BDNF were genotyped (listed in order of chromosomal position): rs4923463, rs6265 (the Val66Met polymorphism), rs11030104, rs2049045 and rs7103411. Single nucleotide polymorphisms were selected on the basis of three considerations:

  1. (a) availability of reliable TaqMan assays at the time of initiation of the study, which was prior to the completion of the HapMap;

  2. (b) coverage on both sides (3′ and 5′) of the single protein-coding exon; and

  3. (c) minor allele frequency of at least 10% in White populations, to permit adequate power.

Genotyping reactions (5 μl) were performed in 384-well plates containing 10 ng of genomic DNA, 0.5 μmol of primers, 0.2 μmol of probes, and 2.5 μl of Master Mix (Applied Biosystems Inc.). The reaction thermal cycle programme consisted of 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, 59°Cor 60°C for 1 min. End-point amplification genotypes were determined using an Applied Biosystems 7900 Sequence Detector with Sequence Detection System 2.0 software. Genotyping accuracy was verified by regenotyping at least 10% of the DNA samples, randomly selected. Genotyping accuracy was >99% and genotyping completion was >99%. All SNPs were in Hardy–Weinberg equilibrium in controls (all P>0.30).

Statistical analysis

Haplotype block structure was determined using Haploview version 3.32. Reference Barrett, Fry, Maller and Daly24 Haplotype blocks were defined in accordance with Gabriel method Reference Gabriel, Schaffner, Nguyen, Moore, Roy and Blumenstiel25 using default settings for all criteria. Using this method, a single haplotype block was identified with a very tight linkage disequilibrium structure (Fig. 1). A common haplotype (AGAGT), containing the valine-coding allele for rs6265, had a 74% frequency in the overall sample. The complementary haplotype (GAGCC) was the only other common haplotype (18% frequency), with several rare haplotypes also detected. PHASE 2.1.1 Reference Stephens and Donnelly26 was then used to determine phase of alleles and assign haplotypes for each participant individually. Seven participants' samples (three with schizophrenia, two healthy controls, one with schizoaffective disorder and one with bipolar disorder) could not be phased with greater than 95% confidence and were excluded from subsequent analyses.

Fig. 1 Linkage disequilibrium plot of the five single nucleotide polymorphisms (SNPs) genotyped in the present study; inter-SNP r 2 is displayed for each pair. D′ is at or near unity for all pairs of SNPs.

To reduce the number of multiple comparisons, analyses were conducted in a hierarchical fashion. Primary analyses were conducted comparing number of copies (zero, one or two) of the AGAGT haplotype, across members of each of the diagnostic groups. Pearson chi-squared tests permuted 500 000 times were used to empirically determine P-values. Groups were first examined individually (full matrix), and then certain groups were combined (e.g. all individuals with a major affective component to disorder) in order to test the hypothesis described above. Subsequent to significant effects of haplotype for a given comparison, individual genotypic effects of each constituent SNP were examined.

Results

As groups differed in ratio of males to females, gender effects on number of copies of the common haplotype were examined first. No differences in common haplotype frequency were observed between males and females either in the total sample (χ2=1.42, d.f.=2, P=0.49) or in any of the diagnostic subsamples (all P>0.43).

Frequencies of two, one or zero copies of the common AGAGT haplotype across the five diagnostic groups are displayed in Fig. 1. The frequency of individuals possessing two copies of the AGAGT haplotype was 54% across the entire sample; frequency of one-copy carriers was 41%, and only 5% of the total sample carried zero copies of AGAGT. As shown in Fig. 2, these frequencies differed across diagnostic groups. Slightly fewer than half of the healthy controls carried two copies of the common haplotype (48.6%) and slightly more than half of the individuals with schizophrenia (53.4%) were two-copy carriers (χ2=0.96, P=0.33). By contrast, at least 60% of people in each of the three affective groups (schizoaffective 60.0%; bipolar 61.8%; and major depressive disorder 69.0%) carried two copies of the common haplotype. Similarly, frequency of one-copy carriers can be seen to decline across the three affective groups relative to the non-affective groups (i.e. healthy controls and people with schizophrenia).

Fig. 2 Frequency of two, one and zero copies of the common AGAGT haplotype for members of the five diagnostic groups. HC, healthy control; Sz, schizophrenia; SzAf, schizoaffective disorder; Bip, bipolar disorder; MDD, major depressive disorder.

Statistical examination of these comparisons, designed to test the hypothesis that dosage of the BDNF common haplotype differs in individuals with an affective diagnosis (including schizoaffective disorder) relative to individuals who do not have an affective illness, are displayed in Table 2. First, the full matrix (five groups compared across three levels, with eight degrees of freedom) reveals a statistically significant difference across the entire sample (first row of Table 2). The second and third rows of Table 2 demonstrate that comparisons remain statistically significant when the affective illness groups are combined. The next two rows demonstrate that the combined affective group significantly differs from the healthy controls and the schizophrenia group, respectively. Finally, the last row of Table 2 reflects a specific comparison of people with schizoaffective disorder with people with schizophrenia, again demonstrating a significant difference. Because gender differences across groups were noted, all comparisons in Table 2 were re-analysed using multinomial logistic regression with haplotype and gender as predictors. For all comparisons, effects of gender were negligible, and all haplotype effects in Table 2 remained statistically significant in the regression analyses.

Table 2 Statistical tests of haplotype frequencies (two, one, or zero copies of the common haplotype) across groups.

Comparison n χ2 d.f. P Significant SNPsa
5 group: HC v. Sz v. SzAf v. Bip v. MDD 593 17.5 8 0.026 rs7103411
4 group: HC v. Sz v. SzAf v. (Bip+MDD) 593 15.5 6 0.017 rs7103411
3 group: HC v. Sz v. (SzAf+Bip+MDD) 593 13.2 4 0.010 rs4923463, rs7103411
2 group: HC v. (SzAf+Bip+MDD) 385 8.5 2 0.015 all
2 group: Sz v. (SzAf+Bip+MDD) 373 9.5 2 0.008 rs7103411
2 group: Sz v. SzAf 268 9.2 2 0.009 rs4923463, rs7103411

As demonstrated in the final column of Table 2, only one SNP (rs7103411) was consistently associated with diagnostic status in a manner parallel to the common haplotype. This was the strongest tag SNP for the haplotype, insofar as it had the highest minor allele frequency and was therefore most informative for identification of uncommon haplotypes. However, we also note that for each comparison, P-values were slightly lower (i.e. stronger) for the full haplotype, indicating that additional individuals with rare haplotypes were not fully captured by rs7103411. Specifically, four participants who were homozygous for the common allele at rs7103411 were carriers of rare haplotypes and three of these were healthy controls (the remaining one had bipolar disorder). Thus, use of the haplotypes results in a slightly more effective differentiation of cases and controls, resulting in an additional 1.4% reduction in common haplotype frequency in controls. By contrast, the Val66Met SNP (rs6265) examined in isolation was only significant in the comparison of healthy controls v. all individuals with schizoaffective or major affective disorder.

Discussion

Results of the present study support the hypothesis that BDNF variation is associated with psychiatric disorders with a primary affective component. This association cuts across several traditional DSM–IV diagnostic categories, consistent with the oft-repeated complaint that the current categorical structure of the DSM system does not effectively describe the underlying biology of the disorders. Reference Van Os, Gilvarry, Bale, Van Horn, Tattan and White27 Further, the present study provides explicit support for the model proposed by Craddock et al, Reference Craddock, O'Donovan and Owen3 in which schizophrenia, schizoaffective disorder and bipolar disorder are seen as partially overlapping constructs against a background of genetic susceptibility to more specific traits such as psychosis or mood disturbance. Future studies incorporating dimensionalised ratings of psychopathology across multiple dimensions would be helpful in elaborating this model more fully.

Importantly, schizoaffective disorder haplotype frequencies were found to be similar to other affective disorders, and dissimilar from schizophrenia and healthy controls in the present study. To our knowledge, this is the first molecular genetic study to report such a distinction. In our prior study of DISC1, Reference Hodgkinson, Goldman, Jaeger, Persaud, Kane and Lipsky28 allelic heterogeneity was observed such that people with schizophrenia, schizoaffective disorder and bipolar disorder all significantly differed from controls, but at different SNP loci within the gene. Similarly, a recent linkage study of schizoaffective disorder (also the first of its kind), demonstrated a risk locus near DISC1. Reference Hamshere, Bennett, Williams, Segurado, Cardno and Norton29 In that study, no evidence of linkage was found on chromosome 11p near BDNF. That negative result is not inconsistent with the present study, insofar as the linkage study examined pedigrees derived roughly equally from studies of probands with schizophrenia and probands with bipolar disorder. Thus, the linkage study was designed to identify risk genes common to both disorders, such as DISC1. Another suggestive locus reported in the schizoaffective linkage study was at chromosome 22q11, and we have previously reported that a single COMT haplotype increases risk for all disorders in our sample. Reference Funke, Malhotra, Finn, Plocik, Lake and Lencz6 At the same time, other genetic variants predisposing to psychotic illness may be shared between schizophrenia and schizoaffective disorder, but not pure affective disorders. Reference Craddock, O'Donovan and Owen3

In the present study, the haplotype more commonly observed in people with affective diagnoses contained the valine allele at Val66Met. Thus, results are consistent with several prior studies identifying an over-transmission of the Val66Met valine allele to individuals with bipolar disorder, Reference Sklar, Gabriel, McInnis, Bennett, Lim and Tsan13Reference Müller, de Luca, Sicard, King, Strauss and Kennedy16 and no effect of this polymorphism on risk for schizophrenia per se. Reference Allen, Bagade, Tanzi and Bertram30 Notably, one study demonstrating an over-representation of the valine allele in a schizophrenia sample included an unspecified proportion of individuals with schizoaffective disorder. Reference Neves-Pereira, Cheung, Pasdar, Zhang, Breen and Yates19 Another identified a very rare BDNF haplotype that was associated with a depression subphenotype within a schizophrenia cohort. Reference Schumacher, Jamra, Becker, Ohlraun, Klopp and Binder18

The present study also extends prior genetic association findings on BDNF in primary affective disorder by demonstrating association with bipolar and unipolar affective disorders in a general population case–control sample. Most prior case–control association studies of BDNF Val66Met in bipolar disorder have been negative Reference Kanazawa, Glatt, Kia-Keating, Yoneda and Tsuang31 (except for one), Reference Lohoff, Sander, Ferraro, Dahl, Gallinat and Berrettini32 although two studies have reported association with the rapid cycling subtype. Reference Müller, de Luca, Sicard, King, Strauss and Kennedy16,Reference Green, Raybould, MacGregor, Hyde, Young and O'Donovan33 Although a few prior studies have reported increased Val66 in childhood-onset and geriatric-onset major depressive disorder, Reference Strauss, Barr, George, Devlin, Vetro and Kiss17,Reference Hwang, Tsai, Hong, Yang, Lirng and Yang34,Reference Borroni, Archetti, Costanzi, Grassi, Ferrari and Radeghieri35 to our knowledge, no studies have demonstrated association of Val66Met with major depression in a general White population.

It is perhaps counterintuitive that the more common haplotype was associated with increased risk for illness in the present study. One potential explanation of our findings is that the common haplotype may carry deleterious variants, not directly genotyped in our study, which are individually rare but have accumulated in the population over time. Reference Lencz, Lambert, DeRosse, Burdick, Morgan and Kane36 However, it is also notable that the genetics literature to date carries numerous examples of common alleles associated with disease. Approximately one-third of all significant illness-based genomewide association results have been with the common allele. Reference Iles37 In psychiatry, a recent study demonstrated that the most common haplotype of the PPP1R1B (DARPP-32) gene is associated with increased risk for schizophrenia; Reference Meyer-Lindenberg, Straub, Lipska, Verchinski, Goldberg and Callicott38 intriguingly, this common risk haplotype was associated with enhanced brain function. Similarly, the BDNF Val66 allele is associated with larger hippocampal volumes Reference Szeszko, Lipsky, Mentschel, Robinson, Gunduz-Bruce and Sevy39 but may lead to heightened risk for affective illness. It is important to note that these associations are by no means deterministic, and pleiotropy is likely to provide multiple positive and negative effects of a given polymorphism across a population. For example, a promoter region variant (the minor allele at rs4950928) in CHI3L1 has been associated with increased risk for schizophrenia Reference Zhao, Tang, Gao, Shi, Zhou and Guo40 but decreased risk for asthma. Reference Ober, Tan, Sun, Possick, Pan and Nicolae41

Limitations

As described above, our primary findings represent novel extensions of the BDNF association literature. However, it is important to emphasise that the genetic association occurs at the group level, and is not diagnostic for any individual case. Additionally, it should be noted that the sample sizes of the individual affective disorders groups were relatively modest for genetic association studies, and larger samples will be needed to confirm our results. Nonetheless, sample size for the primary, full-group comparisons was robust (total n approaching 600), yielding statistically significant results. In this context, it is notable that significant differences were obtained in sub-comparisons with smaller numbers (Table 2), but not in the largest two-group comparison (schizophrenia group v. controls, total n>400). This contrast suggests that Type I error is not a likely explanation of our results, and that significant results for the affective disorders group were less likely to be a result of an atypical healthy control sample. Although our results are consistent with prior theory and literature as described above, lack of additional replication samples means that the possibility of false-positive findings cannot be excluded. In designing replication studies, it is important to acknowledge the trade-off between large multisite studies (with increased ascertainment, diagnostic and environmental heterogeneity) v. the ascertainment of smaller yet potentially more reliably ascertained cohorts of participants (patients and controls). Reference Brzustowicz42

Population stratification is a potential confound in any case–control study. In our recent genomewide association study of a case–control cohort collected at our institution, Reference Lencz, Morgan, Athanasiou, Dain, Reed and Kane43 we tested for stratification using 210 ancestry informative markers selected for maximal informativeness (minor allele frequencies >0.50 between ethnic groups) and did not observe any deviation from that predicted by chance, suggesting that undetected substructure is not present in our geographically homogeneous population. Moreover, none of the participants ascertained at the Zucker Hillside Hospital deviated from a single population as assessed by the structure program (http://pritch.bsd.uchicago.edu/software.html). We also note that, although undetected substructure in US White populations is a theoretical concern, empirical data collected to date suggests that self-identified race/ethnicity (SIRE) is sufficient with which to match study groups from the US population. Tang and colleagues Reference Tang, Quertermous, Rodriguez, Kardia, Zhu and Brown44 examined this question in 3636 people assessed with SIRE information, genotyped with 326 microsatellite markers and analysed with the structure program. These investigators reported ‘a nearly perfect correspondence between genetic cluster and SIRE for major ethnic groups living in the United States’ (p. 273), and suggested that the critical information needed to avoid confounding is SIRE, not necessarily additional genetic marker information.

Mechanism

It is worth noting that the risk haplotype in the present study contains the Val66Met valine allele. Therefore, results of the present study and others listed above could be described as demonstrating that the Met allele (or haplotypes that carry it) may be protective against affective disorders. The Val66Met substitution is a logical candidate for a causative variant, in one or more ways. First, this polymorphism may diminish the dendritic localisation and secretory activity of BDNF, thereby altering the amount of mature protein available for activation of the TrkB pathway. Reference Egan, Kojima, Callicott, Goldberg, Kolachana and Bertolino12 Second, it may affect the level and extracellular processing of proBDNF that acts through p75NTR, downregulating synaptic plasticity through long-term depression via a bidirectional pathway. Reference Woo, Teng, Siao, Chiaruttini, Pang and Milner45 Diminished synaptic strength among certain circuits, in particular excitatory glutamatergic networks, may contribute to a protective mechanism against development of affective disorders.

Although this mechanism is clearly speculative in terms of the pathophysiology of affective disorders, recent animal studies support this possibility. For example, BDNF and its receptor have been found in terminals critical to associative learning within the amygala, Reference Agassandian, Gedney and Cassell46 and changes in BDNF expression in the amygdala have been linked specifically to the acquisition of fear-conditioned responses. Reference Rattiner, Davis and Ressler10 Similarly, BDNF knockdown in the nucleus accumbens abolishes the learned avoidance response to aggression (social defeat stress); Reference Berton, McClung, Dileone, Krishnan, Renthal and Russo11 chronic, but not acute, administration of antidepressants had the same effect. Future studies in humans can extend these observations by examining the relationship of BDNF variation with neuroanatomic and functional measures of the amygdala and basal ganglia.

This line of research holds promise to enhance our understanding not only of the mechanisms underlying affective disorders, but also their treatment. Reference Duman and Monteggia20 Several recent studies have demonstrated that treatment with antidepressants or mood stabilisers upregulates BDNF expression. Reference Chen, Peng, Li, Yang, Wu and Wang47 Moreover, BDNF polymorphisms may also affect response to both mood stabilisers and antidepressants. Reference Choi, Kang, Lim, Oh and Lee48

Finally, although results of the present study are consistent with prior data concerning the functional effects of the Val66Met polymorphism, it should be noted that the haplotype contained several markers that have not been previously examined. Whereas all five markers, including Val66Met, significantly differentiated healthy controls from the combined sample of people with affective illnesses (fourth row of Table 2), only rs7103411 consistently differentiated the diagnostic groups as effectively as the entire haplotype. It is possible that the conflicting results of prior studies examining only Val66Met reflect effects of different haplotypic backgrounds.

It should be noted that our survey of genetic variation at the BDNF locus was not exhaustive: the Phase II HapMap indicates that additional haplotypes exist within the region examined in the present study, and potential splicing variants in the 5′ region of the gene were not examined in the present study. It also cannot be ruled out that prior associations of Val66Met to affective disorders reflect linkage disequilibrium with a distal variant.

Nevertheless, results of the present study suggest that negative conclusions of recent meta-analyses Reference Kanazawa, Glatt, Kia-Keating, Yoneda and Tsuang31,Reference Chen, Lawlor, Lewis, Yuan, Abdollahi and Timpson49 of BDNF Val66Met may be premature; notably, these meta-analytic reports combined studies across multiple ethnicities with very different baseline allele frequencies and excluded several positive reports from family-based samples. Moreover, the combined sample sizes in the psychiatric genetics literature may be inadequate at this point to reach definitive conclusions. A prominent recent example from the diabetes literature illustrates a seeming paradox in the relationship between sample size and detection of genetic associations. Specifically, the association of the PPARG Pro12Ala polymorphism with type 2 diabetes was initially identified in a sample of 91 people. Reference Deeb, Fajas, Nemoto, Pihlajamäki, Mykkänen and Kuusisto50 Although several subsequent studies with hundreds or even thousands of participants failed to replicate the association, the most recent analysis of tens of thousands of participants has definitively confirmed the association. Reference Zeggini, Scott, Saxena, Voight, Marchini and Hu51 Thus, further research on BDNF as a promising candidate for affective psychopathology is warranted, especially in samples that may be informative for future nosological refinements.

Funding

This work was supported by the following grants from the National Institute of Mental Health: K01 MH65580 (T.L.), P30 MH60575 (J.M.K.), K23 MH01760 (A.K.M.), and a General Clinical Research Center (M01 RR18535) to the North Shore – Long Island Jewish Health System. This work was also supported by the National Alliance for Research on Schizophrenia and Depression (T.L. and A.K.M.).

Footnotes

Declaration of interest

None.

References

1 Kasanin, J. The acute schizoaffective psychoses. Am J Psychiatry 1933; 90: 97126.Google Scholar
2 Marneros, A. Beyond the Kraepelinian dichotomy: acute and transient psychotic disorders and the necessity for clinical differentiation. Br J Psychiatry 2006; 189: 12.CrossRefGoogle ScholarPubMed
3 Craddock, N, O'Donovan, MC, Owen, MJ. Genes for schizophrenia and bipolar disorder? Implications for psychiatric nosology. Schizophr Bull 2006; 32: 916.CrossRefGoogle ScholarPubMed
4 Cardno, AG, Rijsdijk, FV, Sham, PC, Murray, RM, McGuffin, P. A twin study of genetic relationships between psychotic symptoms. Am J Psychiatry 2002; 159: 539–45.CrossRefGoogle ScholarPubMed
5 Laursen, TM, Labouriau, R, Licht, RW, Bertelsen, A, Munk-Olsen, T, Mortensen, TB. Family history of psychiatric illness as a risk factor for schizoaffective disorder: a Danish register-based cohort study. Arch Gen Psychiatry 2005; 62: 841–8.Google Scholar
6 Funke, B, Malhotra, AK, Finn, CT, Plocik, AM, Lake, SL, Lencz, T, et al. COMT genetic variation confers risk for psychotic and affective disorders: a case control study. Behav Brain Funct 2005; 1: 19.Google Scholar
7 Green, EK, Raybould, R, Macgregor, S, Gordon-Smith, K, Heron, J, Hyde, S, et al. Operation of the schizophrenia susceptibility gene, neuregulin 1, across traditional diagnostic boundaries to increase risk for bipolar disorder. Arch Gen Psychiatry 2005; 62: 642–8.CrossRefGoogle ScholarPubMed
8 Williams, NM, Green, EK, Macgregor, S, Dwyer, S, Norton, N, Williams, H, et al. Variation at the DAOA/G30 locus influences susceptibility to major mood episodes but not psychosis in schizophrenia and bipolar disorder. Arch Gen Psychiatry 2006; 63: 366–73.CrossRefGoogle Scholar
9 Kendler, KS. Reflections on the relationship between psychiatric genetics and psychiatric nosology. Am J Psychiatry 2006; 163: 138–46.CrossRefGoogle ScholarPubMed
10 Rattiner, LM, Davis, M, Ressler, KJ. Differential regulation of brain-derived neurotrophic factor transcripts during the consolidation of fear learning. Learn Mem 2004; 11: 727–31.Google Scholar
11 Berton, O, McClung, CA, Dileone, RJ, Krishnan, V, Renthal, W, Russo, SJ, et al. Essential role of BDNF in the mesolimbic dopamine pathway in social defeat stress. Science 2006; 311: 864–8.Google Scholar
12 Egan, MF, Kojima, M, Callicott, JH, Goldberg, TE, Kolachana, BS, Bertolino, A, et al. The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell 2003; 112: 257–69.Google Scholar
13 Sklar, P, Gabriel, SB, McInnis, MG, Bennett, P, Lim, YM, Tsan, G, et al. Family-based association study of 76 candidate genes in bipolar disorder: BDNF is a potential risk locus. Mol Psychiatry 2002; 7: 579–93.CrossRefGoogle ScholarPubMed
14 Neves-Pereira, M, Mundo, E, Muglia, P, King, N, Macciardi, F, Kennedy, JL. The brain-derived neurotrophic factor gene confers susceptibility to bipolar disorder. Evidence from a family-based association study. Am J Hum Genet 2002; 71: 651–5.Google Scholar
15 Geller, B, Badner, JA, Tillman, R, Christian, SL, Bolhofner, K, Cook, EH Jr. Linkage disequilibrium of the brain-derived neurotrophic factor Val66Met polymorphism in children with a prepubertal and early adolescent bipolar disorder phenotype. Am J Psychiatry 2004; 161: 1698–700.CrossRefGoogle ScholarPubMed
16 Müller, DJ, de Luca, V, Sicard, T, King, N, Strauss, J, Kennedy, JL. Brain-derived neurotrophic factor (BDNF) gene and rapid-cycling bipolar disorder: family-based association study. Br J Psychiatry 2006; 189: 317–23.CrossRefGoogle ScholarPubMed
17 Strauss, J, Barr, CL, George, CJ, Devlin, B, Vetro, A, Kiss, E, et al. Brain-derived neurotrophic factor variants are associated with childhood-onset mood disorder: confirmation in a Hungarian sample. Mol Psychiatry 2005; 10: 861–7.Google Scholar
18 Schumacher, J, Jamra, RA, Becker, T, Ohlraun, S, Klopp, N, Binder, EB, et al. Evidence for a relationship between genetic variants at the brain-derived neurotrophic factor (BDNF) locus and major depression. Biol Psychiatry 2005; 58: 307–14.CrossRefGoogle ScholarPubMed
19 Neves-Pereira, M, Cheung, JK, Pasdar, A, Zhang, F, Breen, G, Yates, P, et al. BDNF gene is a risk factor for schizophrenia in a Scottish population. Mol Psychiatry 2005; 10: 208–12.CrossRefGoogle Scholar
20 Duman, RS, Monteggia, LM. A neurotrophic model for stress-related mood disorders. Biol Psychiatry 2006; 59: 1116–27.CrossRefGoogle ScholarPubMed
21 First, MB, Spitzer, R, Williams, JBW, Gibbon, M. Structured Clinical Interview for Axis I DSM–IV Disorders – Patient Edition (SCIDI/P, Version 2.0). Biometric Research Department, 1998.Google Scholar
22 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorder, (4th edn) (DSM–IV). APA, 1994.Google Scholar
23 First, MB, Spitzer, R, Williams, JBW, Gibbon, M. Structured Clinical Interview for Axis I DSM–IV Disorders – Non-Patient Edition (SCID–NP, Version 2.0). Biometric Research Department, 1998.Google Scholar
24 Barrett, JC, Fry, B, Maller, J, Daly, MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 21: 263–5.Google Scholar
25 Gabriel, SB, Schaffner, SF, Nguyen, H, Moore, JM, Roy, J, Blumenstiel, B, et al. The structure of haplotype blocks in the human genome. Science 2002; 296: 2225–9.Google Scholar
26 Stephens, M, Donnelly, P. A comparison of bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet 2003; 73: 1162–9.CrossRefGoogle ScholarPubMed
27 Van Os, J, Gilvarry, C, Bale, R, Van Horn, E, Tattan, T, White, I, et al. A comparison of the utility of dimensional and categorical representations of psychosis. UK700 Group. Psychol Med 1999; 29: 595606.CrossRefGoogle ScholarPubMed
28 Hodgkinson, CA, Goldman, D, Jaeger, J, Persaud, S, Kane, JM, Lipsky, RH, et al. Disrupted in schizophrenia1(DISC1): association with schizophrenia, schizoaffective disorder, and bipolar disorder. Am J Hum Genet 2004; 75: 862–72.Google Scholar
29 Hamshere, ML, Bennett, P, Williams, N, Segurado, R, Cardno, A, Norton, N, et al. Genomewide linkage scan in schizoaffective disorder: significant evidence for linkage at 1q42 close to DISC1, and suggestive evidence at 22q11 and 19p13. Arch Gen Psychiatry 2005; 62: 1081–8.CrossRefGoogle ScholarPubMed
30 Allen, NC, Bagade, S, Tanzi, R, Bertram, L. The SchizophreniaGene Database. Schizophrenia Research Forum (http://www.schizophreniaforum.org/res/sczgene/default.asp).Google Scholar
31 Kanazawa, T, Glatt, SJ, Kia-Keating, B, Yoneda, H, Tsuang, MT. Meta-analysis reveals no association of the Val66Met polymorphism of brain-derived neurotrophic factor with either schizophrenia or bipolar disorder. Psychiatr Genet 2007; 17: 165–70.Google Scholar
32 Lohoff, FW, Sander, T, Ferraro, TN, Dahl, JP, Gallinat, J, Berrettini, WH. Confirmation of association between the Val66Met polymorphism in the brain-derived neurotrophic factor (BDNF) gene and bipolar I disorder. Am J Med Genet B Neuropsychiatr Genet 2005; 139: 51–3.Google Scholar
33 Green, EK, Raybould, R, MacGregor, S, Hyde, S, Young, AH, O'Donovan, MC, et al. Genetic variation of brain-derived neurotrophic factor (BDNF) in bipolar disorder. Case–control study of over 3000 individuals from the UK. Br J Psychiatry 2006; 188: 21–5.Google Scholar
34 Hwang, JP, Tsai, SJ, Hong, CJ, Yang, CH, Lirng, JF, Yang, YM. The Val66Met polymorphism of the brain-derived neurotrophic-factor gene is associated with geriatric depression. Neurobiol Aging 2006; 27: 1834–7.Google Scholar
35 Borroni, B, Archetti, S, Costanzi, C, Grassi, M, Ferrari, M, Radeghieri, A, et al. Role of BDNF Val66Met functional polymorphism in Alzheimer's disease-related depression. Neurobiol Aging 2008 Jan 5 (Epub ahead of print).Google Scholar
36 Lencz, T, Lambert, C, DeRosse, P, Burdick, KE, Morgan, TV, Kane, JM, et al. Runs of homozygosity reveal highly penetrant recessive loci in schizophrenia. Proc Natl Acad Sci USA 2007; 104: 19942–7.Google Scholar
37 Iles, MM. What can genome-wide association studies tell us about the genetics of common disease? PLoS Genet 2008; 4: e33.Google Scholar
38 Meyer-Lindenberg, A, Straub, RE, Lipska, BK, Verchinski, BA, Goldberg, T, Callicott, JH, et al. Genetic evidence implicating DARPP-32 in human frontostriatal structure, function, and cognition. J Clin Invest 2007; 117: 672–82.CrossRefGoogle ScholarPubMed
39 Szeszko, PR, Lipsky, R, Mentschel, C, Robinson, D, Gunduz-Bruce, H, Sevy, S, et al. Brain-derived neurotrophic factor val66met polymorphism and volume of the hippocampal formation. Mol Psychiatry 2005; 10: 631–6.Google Scholar
40 Zhao, X, Tang, R, Gao, B, Shi, Y, Zhou, J, Guo, S, et al. Functional variants in the promoter region of Chitinase 3-like 1 (CHI3L1) and susceptibility to schizophrenia. Am J Hum Genet 2007; 80: 12–8.Google Scholar
41 Ober, C, Tan, Z, Sun, Y, Possick, JD, Pan, L, Nicolae, R, et al. Effect of variation in CHI3L1 on serum YKL-40 level, risk of asthma, and lung function. N Engl J Med 2008; 358: 1682–91.CrossRefGoogle ScholarPubMed
42 Brzustowicz, L. Size matters: the unexpected challenge of detecting linkage in large cohorts. Am J Psychiatry 2007; 164: 192–4.Google Scholar
43 Lencz, T, Morgan, TV, Athanasiou, M, Dain, B, Reed, CR, Kane, JM, et al. Converging evidence for a pseudoautosomal cytokine receptor gene locus in schizophrenia. Mol Psychiatry 2007; 12: 572–80.Google Scholar
44 Tang, H, Quertermous, T, Rodriguez, B, Kardia, SL, Zhu, X, Brown, A, et al. Genetic structure, self-identified race/ethnicity, and confounding in casecontrol association studies. Am J Hum Genet 2005; 76: 268–75.CrossRefGoogle Scholar
45 Woo, NH, Teng, HK, Siao, CJ, Chiaruttini, C, Pang, PT, Milner, TA, et al. Activation of p75NTR by proBDNF facilitates hippocampal long-term depression. Nat Neurosci 2005; 8: 1069–77.Google Scholar
46 Agassandian, K, Gedney, M, Cassell, MD. Neurotrophic factors in the central nucleus of amygdala may be organized to provide substrates for associative learning. Brain Res 2006; 1076: 7886.Google Scholar
47 Chen, PS, Peng, GS, Li, G, Yang, S, Wu, X, Wang, CC, et al. Valproate protects dopaminergic neurons in midbrain neuron/glia cultures by stimulating the release of neurotrophic factors from astrocytes. Mol Psychiatry 2006; 11: 1116–25.Google Scholar
48 Choi, MJA, Kang, RH, Lim, SW, Oh, KS, Lee, MS. Brain-derived neurotrophic factor gene polymorphism (Val66Met) and citalopram response in major depressive disorder. Brain Res 2006; 1118: 176–82.Google Scholar
49 Chen, L, Lawlor, DA, Lewis, SJ, Yuan, W, Abdollahi, MR, Timpson, NJ, et al. Genetic association study of BDNF in depression. Finding from two cohort studies and a meta-analysis. Am J Med Genet B Neuropsychiatr Genet 2008; 147B: 814–21.Google Scholar
50 Deeb, SS, Fajas, L, Nemoto, M, Pihlajamäki, J, Mykkänen, L, Kuusisto, J, et al. A Pro12Ala substitution in PPARgamma2 associated with decreased receptor activity, lower body mass index and improved insulin sensitivity. Nat Genet 1998; 20: 284–7.CrossRefGoogle ScholarPubMed
51 Zeggini, E, Scott, LJ, Saxena, R, Voight, BF, Marchini, JL, Hu, T, et al. Metaanalysis of genome-wide association data and large-scale replication identifies additional susceptibility loci for type 2 diabetes. Nat Genet 2008; 40: 638–45.Google Scholar
Figure 0

Table 1 Demographics of the five participant groups (total n=600)

Figure 1

Fig. 1 Linkage disequilibrium plot of the five single nucleotide polymorphisms (SNPs) genotyped in the present study; inter-SNP r2 is displayed for each pair. D′ is at or near unity for all pairs of SNPs.

Figure 2

Fig. 2 Frequency of two, one and zero copies of the common AGAGT haplotype for members of the five diagnostic groups. HC, healthy control; Sz, schizophrenia; SzAf, schizoaffective disorder; Bip, bipolar disorder; MDD, major depressive disorder.

Figure 3

Table 2 Statistical tests of haplotype frequencies (two, one, or zero copies of the common haplotype) across groups.

Supplementary material: PDF

Lencz et al. supplementary material

Supplementary Figure S1

Download Lencz et al. supplementary material(PDF)
PDF 222.8 KB
Submit a response

eLetters

No eLetters have been published for this article.