Copy
The placenta: a key to schizophrenia?

Ron McKay, PhD, Chief Editor
Lieber Institute for Brain Development

Maltz Research Laboratories
Progress in understanding neuro-developmental disorders is accelerating by many measures. Even in the short time period covered by previous issues of neuroDEVELOPMENTS, the power of DNA and RNA sequencing methods has become increasingly apparent. The many bioinformatic tools employed to assemble sequence information allow us to decompose, reconstruct and project these data structures onto one another. In development—in our case, neural development—these approaches have generated a massive increase in the number of high-quality time series analyses of the genetic and cellular organization of the human brain. The elephant in the room that science has not yet addressed powerfully: We need a deeper understanding of how the genome and environment interact in human behavior. The two papers we focus on here and the comments from our board, all of which discuss the exciting finding that the placenta has something important to teach us about schizophrenia, give us a contemporary perspective on the task ahead.

--RM
neuroDEVELOPMENTS Editorial Board
 
Fred 'Rusty' Gage, PhD
President, The Salk Institute for Biological Studies

Daniel Geschwind, MD, PhD
Professor, UCLA School of Medicine

Elizabeth Grove, PhD
Professor, University of Chicago
 
Jürgen Knoblich, PhD
Director, Institute of Molecular Biotechnology, Austrian Academy of Sciences 

Arnold Kriegstein, MD
Professor, UCSF

Pat Levitt, PhD
Professor, Keck School of Medicine of USC

Mu-Ming Poo, PhD
Director, Institute of Neuroscience, Chinese Academy of Sciences

John Rubenstein, MD, PhD
Professor of Psychiatry, UCSF

Nenad Sestan, MD, PhD
Professor, Yale University 

Flora Vaccarino, MD
Professor, Yale University 

Chris Walsh, MD, PhD
Chief, Division of Genetics & Genomics, Boston Children's Hospital


______________________________________

Venkata S. Mattay, MD
Managing Editor

Michele Solis, PhD
Science Writer 
neuroDEVELOPMENTS 

A landscape of development


While most scientists agree that psychiatric disorders have their roots in brain development, pinpointing precisely what goes awry remains an elusive goal. During development, genetic and environmental factors can tilt the growing brain toward or away from a risk state, akin to navigating the metaphorical landscape described by Conrad Waddington: If development’s trajectory encounters a particular combination of hills and dales, it can be funneled into a different phenotypic valley.

These factors and their interactions are complicated, and seemingly limitless. Researchers have gravitated toward genetic explanations partly because the genome is finite compared to environmental factors. Most environmental factors remain uncharted. For this issue of neuroDEVELOPMENTS, we try to “get out of the DNA” in order to explore the role environment might have in elevating risk for neurodevelopmental disorders. Recent studies focused on the wellbeing of the pregnant mother have linked nutrition and infection with maternal stress during pregnancy and fetal outcome (e.g., Walsh et al., 2019). These results make the point that environmental factors active during the early stages of life in the womb offer rich terrain for exploration.

Here we discuss two recent studies with insights on how environmental influences impact early brain development. One suggests the placenta as a mediator of genetic risk for schizophrenia, which raises the provocative idea that targeting placenta health during pregnancy could help guide development onto healthy contours (Ursini et al., 2021). The second study brings us back into the nucleus with a focus on how the genome is modified to capture the early environmental signals that influence later psychiatric risk. (Boix et al., 2021)

 

“While the genetics and biology of this story is extremely complex, it should compel the field to look closely at all aspects of placental physiology when considering the etiology of neurodevelopmental disorders.” – John Rubenstein

[T]hese regulatory elements are where key epigenetic components operate, and can be influenced by the internal and external physiology of the cell, thus connecting the cell and organism with the environment.John Rubenstein

One interesting aspect is that the active marks are distinct across lineages, whereas repressive marks grouped samples across time/life stages. – Flora Vaccarino

Placental possibilities

When considering neurodevelopmental disorders like schizophrenia, it is automatic to think about how genetic risk factors impinge directly upon the brain. But researchers at the Lieber Institute for Brain Development (LIBD, and publisher of neuroDEVELOPMENTS) suggest an additional route that runs through the placenta. In 2018, they reported that the cumulative genetic risk for schizophrenia was five times higher in people who had experienced some early life complications, and found that much of this risk involved genes upregulated in the placenta (Ursini et al., 2018; Ursini et al., 2021). This suggests that some aspect of genetic risk acts through the placenta, which could in turn alter brain development or cause an early life complication that could then increase risk indirectly.

A new paper by Gianluca Ursini and colleagues at the LIBD pursues this idea further by tracking how placental genetic risk varies with early brain outcomes. Specifically, they isolated the genetic risk for schizophrenia involving genes upregulated in placental tissue, and asked whether it covaried with early brain outcomes measured in over 200 infants. They found inverse correlations, such that high placental-associated risk came with lower neonatal brain volumes, and with a worse measure of cognition and motor development at 1 year of age. The latter relationship was less pronounced at 2 years of age, which suggests that, with time, factors like socioeconomic status or education of parents or genetic effects unrelated to the placenta mask these features.

This connection between the placenta and early brain outcome was surprisingly specific to schizophrenia, at least in the context of this study, because it was not found for other disorders associated with early life complications, including attention deficit hyperactivity disorder and autism. Consistent with this, an inverse correlation between placental-associated genetic risk and brain volume could still be detected to a small degree in adults with schizophrenia, like a vestige from earlier life. This was not seen in normal subjects, who presumably “canalized” back towards more optimal developmental outcomes. These new findings stress how developmental trajectories initiated in the placenta may map onto a less healthy landscape of brain development.

Figure 1 Sex-related differences in the relationship between placental genomic risk for schizophrenia and early neurodevelopmental outcomes (A and B) Scatterplots of the relationship of neonatal ICV with placental genomic risk scores for schizophrenia (PlacGRS1, A; PlacGRS2, B) in males (n = 133; turquoise dots) and females (n = 109; violet dots). (C and D) Scatterplots of the relationship of MCS1 with placental genomic risk scores for schizophrenia (PlacGRS1, C; PlacGRS2, D) in male (n = 68; turquoise dots) and female singletons (n = 54; violet dots) (P values in the male sample are in turquoise; P values in the female sample are in violet).

Millions of openings


With this year bringing the 20th anniversary of the human genome’s publication, researchers are as busy as ever in trying to understand how its billions of base pairs work. This is key because how the genome’s different parts are activated or silenced alters gene transcription; a dense choreography of these activations eventually give rise to a person and keep that person humming. To identify what is turned on or off, several projects have worked to describe the epigenome—the pattern of chemical marks placed on top of the DNA sequence that determine how genes are turned on and off. These epigenome patterns differ between different tissues, which use different parts of the genome to enact their particular forms and functions.

Some of these chemical marks are also thought to be malleable by environmental factors, which shape gene regulation, and thus development. A recent paper from Manolis Kellis of the Massachusetts Institute of Technology and colleagues presents one of the most ambitious maps to date of this dashboard of genomic regulation (Boix et al., 2021). The researchers compiled epigenome data from 833 samples that came from different tissues, stages of life, and cell lines. Looking at patterns of chromatin marks, i.e. the diverse array of epigenetic signatures, they identified 18 different chromatin states, such as active or repressed states of gene expression. Finding overlaps between active states and spots on the genome where the DNA was actually accessible to transcription, they were able to discern 2.1 million active enhancer regions—those that contribute to enhancing expression—in different samples.

This pinpointed the controls involved in specifying a particular tissue. Shared patterns among these two million sites defined 300 different enhancer modules, most of which were specific to a tissue type. With these active enhancer sites revealed, the researchers also inferred their downstream genes and upstream regulators—forming connections that illuminate the genomic circuitry for a particular tissue.

These active enhancer regions are an important source of human variation. When the team examined the proximity of common genetic variants flagged by 803 genome wide association studies (GWAS) of human traits and diseases, they found intriguing connections between genetic risk signals and specific tissues and cells. Some involved only one type of tissue, like genes associated with educational attainment and the brain; others involved multiple tissues, like coronary heart disease signals found in enhancers for liver, heart, adipose tissue, and muscle. Interestingly, the detailed catalog of tissue enrichments for genes associated with schizophrenia found that after the brain, the placenta was the next organ most enriched for GWAS genes.

“It is also remarkable that overall, driver SNPs in 39% of GWAS loci mapped to tissue-enriched enhancers, suggesting that disease-relevant loci are frequently found in enhancer regions.” - Flora Vaccarino

“While the specificity is impressive, a note of caution is warranted, given that the field is in its infancy in attempts to gain a deep understanding of placental biology and its impact on near-term fetal and long-term postnatal brain circuit development that would be relevant for understanding disorder etiology, and possibly even early identification of high risk.” - Pat Levitt

Figure 2 Trait-trait network The network across 538 traits (by per-node false discovery rate correction) by similarity of epigenetic enrichments (cosine similarity ≥ 0.75), laid out using the Fruchterman–Reingold algorithm. Traits (nodes) are colored by the contributing groups (pie chart by the fraction of −log10P, and size by maximal −log10P) and interactions (edges) by the group with the maximal dot product of enrichments between two traits. The redundant node names indicate different GWAS. AD, Alzheimer disease; ADHD, attention-deficit/hyperactivity disorder; BMI, body mass index; CVD, cardiovascular disease; FEV1, forced expiratory volume in 1 s; T2D, type 2 diabetes; vWF, von Willebrand factor; WHR, waist-to-hip ratio

“[D]evelopmental processes that may underlie psychiatric disorders are likely to be impacted by multi-factorial, multi-organelle and multi-organ disruptions. This emphasizes the importance of incorporating collaborations that allow for the design of more holistic biological and computational studies, which is reflected in the papers highlighted in this issue.” - Pat Levitt

Radiating risk

As with Ursini et al., the findings indicate that genetic risk may act through more than one tissue, even unexpected ones. While the picture of genomic risk radiating out through multiple tissues may seem discouraging, it may also expose multiple potential therapeutic avenues. By approaching the problem in two directions—from outside the fetus and from deep inside its nuclei—researchers are mapping how the environment impacts the flowering of fetal development. The hope is for radically new and malleable therapeutic targets that might actually make prevention a reality.

###
The full commentary from our Editorial Board on the two papers highlighted in this issue of neuroDEVELOPMENTS.
 
1. From John Rubenstein, MD, PhD, University of California, San Francisco:
It is extremely important to search for the biological interfaces of where the environment impacts the genetic, molecular and cellular processes that regulate brain development and function. The paper by Walsh et al. provides mounting evidence that maternal stress increases the probability of a negative outcome on the baby. How this occurs is extremely complex, and likely involves a myriad of processes impacted by fetal and maternal physiology (e.g. placental, neuroendocrine, immunological, cardiovascular) and external variables (e.g. nutrition, medications, alcohol and drug consumption, infections). I think that every effort needs to be made to enact social changes that eliminate the risk to babies through helping mothers with proper nutrition, avoiding alcohol and drug consumption, and preventing and treating infections. Efforts need to be made to identify medications that put the baby at risk. Furthermore, addressing and treating the psychosocial causes of stress are critical.
 
The paper by Ursini et al. provides novel evidence that the genetic loci that predispose to schizophrenia may operate in part through the placenta. Early-life health complications to the fetus and postnatal infant increase the risk for schizophrenia; here they show that this risk is increased in individuals with genotypes having alleles implicated in schizophrenia. Furthermore, they find that placentas from complicated pregnancies show differential expression of RNAs encoded by schizophrenia risk-associated genes. This implicates defects in placental gene expression in the probability of developing schizophrenia. While the genetics and biology of this story is extremely complex, it should compel the field to look closely at all aspects of placental physiology when considering the etiology of neurodevelopmental disorders.
 
The paper by Boix et al. used H3K27c and ATAC-seq data to obtain a “compendium comprising 10,000 epigenomic maps across 800 samples, … to define chromatin states, high-resolution enhancers, enhancer modules, upstream regulators and downstream target genes.” This is an extremely useful resource to link human genetics with understanding tissue-specific gene regulation in health and disease. They provided evidence for existence of over 2 million enhancers and defined which of these candidate enhancers fall into tissue-specific gene expression modules, including in the brain. Their enhancer–gene links provide insight into the potential role of specific enhancer alleles in altering gene expression and therefore contributing to human disease.

Of course, functional demonstration of the biological impact of allelic versions of enhancers is complex and awaits deeper analyses. Nonetheless, this paper, and several others that have recently appeared in the literature, forms a road map for understanding what parts of the non-coding genome control gene expression and can contribute to disease risk. Furthermore, these regulatory elements are where key epigenetic components operate, and can be influenced by the internal and external physiology of the cell, thus connecting the cell and organism with the environment.

2. From Flora Vaccarino, MD, Yale University:
The most remarkable contribution of Boix et al. is the extensive map of enhancer elements, which allowed the discovery of common enhancer elements active across several tissues. Another strength is that this map relies on both chromatin state annotation based on multiple histone marks as well as chromatin accessibility, and is uniformly processed across samples. One interesting aspect is that the active marks are distinct across lineages, whereas repressive marks grouped samples across time/life stages. Across tissues, the authors identified the location of 2.1 million enhancer elements, perhaps the largest dataset identified so far and encompassing 13% of the human genome—more than six times the space encompassed by protein-coding regions. 
 
It is also remarkable that overall, driver SNPs in 39% of GWAS loci mapped to tissue-enriched enhancers, suggesting that disease-relevant loci are frequently found in enhancer regions. Here, the extended dataset revealed the complexities of disease genetic pathophysiology, where some disease-related variants mapped to multiple enhancer elements, sometimes functionally converging onto a single gene. Finally, trait co-enrichment in tissues distinguished uni-factorial and multi-factorial traits, enriched in one or several tissues respectively (where each tissue contributed a distinct set of enhancers to the trait). The diversity of trait-enhancer associations revealed multiple components of disease pathophysiology. This relies on their careful establishment of functional links between enhancer elements and target genes.

3. From Arnold Kriegstein, MD, University of California, San Francisco:
In this issue of neuroDEVELOPMENTS, two papers are highlighted that link the expression of schizophrenia risk genes with the placenta. These observations support the critical role that maternal-fetal interactions can have on diseases that manifest later in life and underscores the sensitivity of the developing brain to environmental signals. It is particularly intriguing that among neurodevelopmental diseases, schizophrenia is most strongly linked to placental or maternal stress.

The concept that one’s risk of developing schizophrenia can be influenced by an adverse fetal environment has been supported by observations connecting prenatal maternal stress to schizophrenia in offspring. The Dutch 'hunger winter' of 1944-45, provoked by food shortages, as well as famines in China in the 1960s both led to a doubling in the likelihood of babies who later developed schizophrenia.

Stresses that can influence development in utero are not always physical. In this context, the findings of Walsh et al. are particularly relevant. They studied 187 pregnancies to determine the kinds of stress that have the greatest impact on fetal development, preterm birth, and birth outcomes, though schizophrenia per se was not considered. They concluded that a low index of social support—which they defined as having people with whom to talk, to spend time, and on whom to rely for material help—emerged as a strong indicator of prenatal maternal stress. Yet another reason to vaccinate the world population and reduce the social isolation imposed by the current pandemic. The realization that placental and maternal stress can significantly impact brain development is of concern, but the data also suggests that the negative effects can be reduced by environmental factors and experiences in infancy and childhood. A better understanding of the mechanisms conferring schizophrenia risk through placental dysfunction should help to design mitigation strategies.

4. From Pat Levitt, PhD, University of Southern California:
If anyone still had questions regarding major drivers of phenotypic heterogeneity in typical and atypical development, and in disease states, the Ursini et al. and Walsh et al. papers will quench their thirst for answers. First, continued accumulation of experimental studies of environmental manipulations, and the genetic studies highlighted in this issue of neuroDEVELOPMENTS, support the active role of the placenta in mediating the maternal-fetal relationship that influences the developmental landscape. Measures of correlation of genomic risk scores (GRS) with developmental milestones, such as the Mullen Scales of Early Development used by Ursini et al., are challenging because of the influence of non-heritable factors on cognition. This is recognized by the authors, who turn to demonstrating the specificity of GRS enrichment of placental genes that are expressed at schizophrenia risk loci, but not identified loci for several other neurodevelopmental disorders. While the specificity is impressive, a note of caution is warranted, given that the field is in its infancy in attempts to gain a deep understanding of placental biology and its impact on near-term fetal and long-term postnatal brain circuit development that would be relevant for understanding disorder etiology, and possibly even early identification of high risk.

Second, even as a non-geneticist, I can appreciate the Boix et al. paper as a remarkable tour de force in not only the deep cataloging of genome-wide regulatory elements, but in establishing relations between enhancers and, as noted by others in this issue, their putative target genes. This allows for the authors to demonstrate, computationally, pleiotropy at a level of complexity that has not been realized previously. In some ways, this work should encourage experimental efforts to deploy methods for introducing relevant and likely high impact driver variants that can be studied for their developmental influences at a cellular and circuit level in the brain. The study also highlights the importance of recognizing that the brain is connected to the rest of the body (and placenta prenatally!), and that in reflecting on the mission of the LIBD, developmental processes that may underlie psychiatric disorders are likely to be impacted by multi-factorial, multi-organelle and multi-organ disruptions. This emphasizes the importance of incorporating collaborations that allow for the design of more holistic biological and computational studies, which is reflected in the papers highlighted in this issue.
Share Share
Tweet Tweet
Forward Forward
Copyright © 2021 Lieber Institute, All rights reserved.


Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Email Marketing Powered by Mailchimp