Genesis of Schizophrenia May Lie in the Placenta

Pauline Anderson

May 30, 2018

An interaction between gene expression in the placenta and obstetric or neonatal complications may increase schizophrenia risk, new research suggests.

Dr Daniel Weinberger

Investigators found that the expression of schizophrenia risk genes is particularly enriched in the placenta from male offspring, which may explain the higher incidence of the disease in men.

The results underline the importance of the placenta in schizophrenia risk and perhaps in risk for other male-dominated disorders, such as autism and attention-deficit/hyperactivity disorder, Daniel R. Weinberger, MD, director and CEO, Lieber Institute for Brain Development, and professor of neurology, psychiatry, and neuroscience, Johns Hopkins University School of Medicine, Baltimore, Maryland, told Medscape Medical News.

"It's clear that the placenta, at a basic molecular level, mediates some genetic risk for developmental behavioral disorders," Weinberg said.

Yet the placenta remains "profoundly neglected. It's the only organ taken out of a human body that does not routinely get sent to the laboratory for examination but is typically thrown out," he said.

The study was published online May 28 in Nature Medicine.

Gene-Environment Interaction

Both genetic and environmental factors affect schizophrenia risk. But researchers have long wondered how these risks are related.

"The question is, are they independent risk factors, or are they risk factors that actually exaggerate each other's effects, which is the principle of the gene-environment interaction?" said Weinberger.

To examine the interaction of genomic and environmental risks, the researchers used a number of large samples. The "discovery" sample included 501 unrelated white US adults (234 with schizophrenia, and 267 healthy persons) who were participants in the Clinical Brain Disorders Branch Sibling Study of Schizophrenia at the National Institute of Mental Health.

The study also included more than 2000 unrelated adult individuals in four replication samples from Italy, Germany, and Japan. Two were composed of only patients with schizophrenia, and two were composed of schizophrenia patients and control persons.

Researchers calculated the cumulative polygenic risk score (PRS) for each participant. This is a measure of genomic risk, calculated as the weighted sum of risk alleles for schizophrenia from recent genome-wide association studies (GWASs).

Initially, they selected relevant genes mapping to the PRS1 loci. According to Weinberger, on the basis of GWAS significant alleles, PRS1 had a significance of P < 5 X 10-8.

The investigators examined the interaction between PRS and early-life complications (ELCs), which are conditions occurring during pregnancy, labor, delivery, and the neonatal period that are potentially harmful to the offspring.

Such complications include hypoxia, preeclampsia, intrauterine growth restriction, placenta previa, breach delivery, and premature birth.

The researchers assessed ELCs through medical records and personal interviews. They also used the well-validated McNeil-Sjöström questionnaire, in which each ELC is assigned a severity level of 1 to 6. The severity levels reflect the degree of inferred potential harm to the offspring's central nervous system.

ELCs with a severity weight ≥4 are considered potentially clearly harmful factors or factors that are relevant in fetal stress.

The researchers divided individuals into quintiles on the basis of their PRS1 levels and determined odds ratios associated with being in each quintile compared to the lowest quintile in the absence and in the presence of ELCs.

Investigators responsible for recruitment and clinical evaluation of control persons and patients were blinded to their PRSs and ELCs.

"Dramatic Finding"

In the discovery sample, results showed an interaction between PRS1 and ELCs in patients with schizophrenia. The risk, or "liability," of schizophrenia accounted for by PRS1 was more than five times greater in persons with ELCs compared to those without.

Weinberger said that without information on early-life complications, "you can predict about a 3% probability" of schizophrenia. "But if you know who didn't have an early-life complication and compare that to who did, there is a fivefold increase in prediction liability."

The researchers investigated whether the interaction between genomic risk and ELCs occurred in PRS levels 2-10, constructed from alleles showing association with schizophrenia, but at lesser statistical levels (ie, the associations were not GWAS significant).

They found that the ELC-dependent change in schizophrenia risk variance accounted for by PRS gradually decreased and that PRS2 — which, along with PRS1, showed the strongest association with schizophrenia in GWAS (P < 1 × 10–6) — was the only other PRS to interact with ELCs.

Weinberger stressed that it is the severity of the complications, not their frequency, that matters in the prediction.

The findings were essentially the same in the four replication samples.

The researchers also carried out analyses of RNA sequencing in placental tissue collected at the University of California, San Francisco. The samples came from complicated pregnancies as well as normal pregnancies.

They found that the genes in the regions of the genome that were shown in the clinical samples to interact with ELCs "are much more abundantly expressed" in placentae from complicated pregnancies compared to normal ones, said Weinberger.

They also compared gene expression in placental samples taken from male and female offspring. They found that schizophrenia-related genes that interact with ELCs are much more dramatically expressed in the placentae of male offspring than in those of female offspring.

"This is a very dramatic finding. We have known for a long time that all these developmental disorders are two to four times more common in males than females, and we have never understood how this could be," he said.

The Missing Link

At this point, researchers are not certain what accounts for the sex difference in placental gene expression, said Weinberger.

His team examined the potential role of stress response in the placenta. Many factors can put stress on the placenta, including exposure to a virus or smoking.

They found that the more that schizophrenia genes are expressed in the placenta, the more an immune response is generated.

"These effects in the placenta ultimately translate into risk for the fetus," said Weinberger.

Although researchers have long known that prenatal care is critical to early child development, "I think we have never before had an insight at a molecular and genetic level of how this plays out in the placenta," he said.

"Someone once said that the missing link between maternal risk and fetal brain development has been hiding in plain sight for a long time. Now we know it's the placenta," he said.

These new results generate a myriad of questions that, if answered, could provide insights into how to make the placenta healthier, said Weinberger.

"If we can reverse some of what looks like placental stress and genomic risk for stress, we might be able to prevent the emergence of some of these conditions in some individuals," he said.

To that end, the investigators are using stem cell technology to generate what he called "little placentas" from people at high risk for placental stress to try to determine what goes on in these cells very early in their development.

Weinberger predicts that monitoring the state of the placenta and determining what may be a risky threshold of gene expression in that organ will become "a new area of research."

"This is the first clue to a road map that could possibly lead to some type of prevention. This is preventing biology of risk long before this risk really gets established," he said.

The authors note that the frequency of ELCs in the study samples may not be representative of that in the general population.

Weinberger also pointed out that obstetric or neonatal complications occur in an estimated 15% to 20% of pregnancies and that in most cases, these do not lead to negative outcomes.

Paradigm Shift

Commenting on the study for Medscape Medical News, Christopher A. Ross, MD, PhD, professor of psychiatry and behavioral sciences, Johns Hopkins Medicine, Baltimore, Maryland, called it "groundbreaking" and "paradigm shifting."

"I think it's one of the most important studies in psychiatric genetics in the past few years," he said.

Researchers knew that genes contribute substantially to schizophrenia risk, but it was assumed that these were genes in the brain, because schizophrenia is a brain disorder, said Ross.

"But what this study puts together is the idea that part of that genetic contribution to schizophrenia actually comes from genes that are predominantly expressed in the placenta," he commented.

Researchers also knew that environmental factors — for example, malnutrition or infections — during pregnancy contribute to schizophrenia risk and that there seem to be subtle developmental abnormalities in people even before they have detectable schizophrenia symptoms.

"What this study basically does, in my view, is put all of those important pieces together and assembles them in an entirely unexpected way," said Ross. "It says that if a mother's placenta is not functioning optimally, that contributes to the risk for schizophrenia."

Although this is "a very first step," extrapolating these findings to treatment possibilities "is exciting," said Ross.

"This study opens up so many different new ways of thinking about schizophrenia, new avenues of investigation, and potentially new preventive interventions," he commented.

In light of these results, psychiatrists may want to "talk more" with their colleagues in obstetrics and pediatrics about avoiding obstetric complications, he said.

Authors' acknowledgements of financial and other support are listed in the original article. Dr Weinberger and Dr Ross have disclosed no relevant financial relationships.

Nat Med. Published online May 28, 2018. Abstract

For more Medscape Psychiatry news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.