Investigators identified a mechanism on the dopamine receptor, known as the autoreceptor, which regulates how much dopamine is released from the presynaptic neuron. Impairment of this autoreceptor leads to poorly controlled dopamine release and excessive dopamine flow.
The researchers found decreased expression of this autoreceptor accounts for the genetic evidence of schizophrenia risk, and, using a suite of statistical routines, they showed that this relationship is probably causative.
“Our research confirms the scientific hypothesis that too much dopamine plays a likely causative role in psychosis and precisely how this is based on genetic factors,” study investigator Daniel Weinberger, MD, director and CEO of the Lieber Institute for Brain Development, told Medscape Medical News.
“Drugs that treat psychosis symptoms by simply blocking dopamine receptors have harsh side effects…Theoretically, scientists could now develop therapies that target these malfunctioning autoreceptors to treat this devastating condition with fewer side effects,” he said.
The study was published online November 1 in Nature Neuroscience.
“Large international genetic studies known as genome-wide association studies (GWAS) have identified hundreds of regions of the human genome housing potential risk genes for schizophrenia,” Weinberger said.
“However, these regions are still poorly resolved in terms of specific genes, and treatments and diagnostic techniques are far from what they should be.” Moreover, “treatments for schizophrenia address the symptoms of psychosis but not the cause,” he said.
“For more than 70 years, neuroscientists have suspected that dopamine plays a key role in schizophrenia, but what kind of role, exactly, has remained a mystery,” Weinberger noted. “It occupied a privileged spot in the principal hypothesis about schizophrenia for over 60 years — the so-called ‘dopamine hypothesis’.”
Antipsychotic drugs that reduce dopamine “are the principal medical treatments but they cause serious side effects, including an inability to experience pleasure and joy — a sad reality for patients and their families,” he continued.
The current study “set out to understand how dopamine acts in schizophrenia” using “analysis of the genetic and transcriptional landscape” of the post-mortem caudate nucleus from 443 donors (245 neurotypical, 154 with schizophrenia, and 44 with bipolar disorder).
Brain samples were from individuals of diverse ancestry (210 were of African ancestry [AA] and 2233 were of European ancestry [EA]).
New Treatment Target?
The researchers performed an analysis of transancestry expression quantitative trait loci (eQTL), genetic variants that explain variations in gene expression levels, which express in the caudate, annotating “hundreds of caudate-specific cis-eQTLs.”
Then they integrated this analysis with gene expression that emerged from the latest genome-wide association study (GWAS) and transcriptome-wide association study, identifying hundreds of genes that “showed a potential causal association with schizophrenia risk in the caudate nucleus,” including a specific isoform of the dopamine D2 receptor, which is upregulated in the caudate nucleus of those with schizophrenia.
“If autoreceptors don’t function properly the flow of dopamine in the brain is poorly controlled and too much dopamine flows for too long,” said Weinberger.
In particular, they observed “extensive differential gene expression” for schizophrenia in 2701 genes in those with schizophrenia (compared to those without): glial cell-derived neutrophic factor antisense RNA (GDNF-AS1) was a “top-up” gene and tyrosine hydroxylase (TH), which is a rate-limiting enzyme in dopamine synthesis, was a “downregulated” gene. Dopamine receptors DRD2 and DRD3 were differentially expressed.
Having done this, they looked at the effects of antipsychotic medications that target D2 regions on gene expression in the caudate by testing for differences between individuals with schizophrenia who were taking antipsychotics at the time of death, those not taking antipsychotics at the time of death (n = 104 and 49, respectively), and neurotypical individuals (n = 239).
There were 2692 differentially expressed genes (DEGs) between individuals taking antipsychotics versus neurotypical individuals (false discovery rate [FDR] < .05). By contrast, there were only 665 DEGs (FDR < .05) between those not taking antipsychotics and neurotypical individuals.
“We found that antipsychotic medication has an extensive influence on caudate gene expression,” the investigators note.
They then developed a new approach to “infer gene networks from expression data.” This method is based on deep neural networks, obtaining a “low-dimensional representation of each gene’s expression across individuals.” The representation is then used to build a “gene neighborhood graph and assign genes to modules.”
This method identified “several modules enriched for genes associated with schizophrenia risk.” The expression representations captured in this approach placed genes in “biologically meaningful neighborhoods, which can provide insight into potential interactions if these genes are targeted for therapeutic intervention,” the authors summarize.
“Now that our new research has identified the specific mechanism by which dopamine plays a causative role in schizophrenia, we hope we have opened the door for more targeted drugs or diagnostic tests that could make life better for patients and their families,” Weinberger said.
No Causal Link?
Commenting for Medscape Medical News, Rifaat El-Mallakh, MD, director of the Mood Disorders Research Program, Department of Psychiatry and Behavioral Sciences, University of Louisville School of Medicine, Kentucky, called it an “excellent study performed by an excellent research group” that “fills an important lacuna in our research database.”
However, El-Mallakh, who was not involved in the research, disagreed that the findings show causality. “The data that can be gleaned from this study is limited and the design has significant limitations. As with all genetic studies, this is an association study. It tells us nothing about the cause–effect relationship between the genes and the illness.”
“We do not know why genes are associated with the illness. Genetic overrepresentation can have multiple causes, and more so when the data is a convenience sample. As noted by the authors, much of what they observed was probably related to medication effect. I don’t think this study specifically tells us anything clinically,” he added.
The study was supported by the LIBD, the BrainSeq Consortium, an NIH fellowship to two of the authors, and a NARSAD Young Investigator Grant from the Brain & Behavior Research Foundation to one of the authors. Weinberger has reported no relevant financial relationships. Disclosures for the other authors are listed with the article. El-Mallakh has declared no specific financial relationships relevant to the study but has reported being a speaker for several companies that manufacture antipsychotics.
Batya Swift Yasgur MA, LSW, is a freelance writer with a counseling practice in Teaneck, NJ. She is a regular contributor to numerous medical publications, including Medscape and WebMD, and is the author of several consumer-oriented health books as well as Behind the Burqa: Our Lives in Afghanistan and How We Escaped to Freedom (the memoir of two brave Afghan sisters who told her their story).