Genomic Imprinting: Key for Neurodevelopment, with Christopher Gregg, PhD
"We're figuring out which circuits express imprinted genes and we are learning about critical stages of brain development during which maternally and/or paternally expressed genes function. We are also testing whether perturbations to imprinted genes, due to genetic or environmental effects, can alter specific aspects of brain development and function."
Christopher Gregg, Ph.D, is Assistant Professor of Neurobiology and Anatomy and Adjunct Assistant Professor of Human Genetics and University of Utah. Dr. Gregg's work explores epigenetic and genetic pathways that lay the foundation for adult health and behavior outcomes.
Much of his work looks at imprinted genes, a subset of genes that are expressed differently, depending on maternal and paternal origin, instead of expressing both parental alleles equally. His lab has uncovered a vast array of complex parental effects and indicate the existence of distinct maternal and paternal gene expression programs in the brain, including the discovery that 60% of imprinted genes expressed in the developing brain preferentially express the maternally inherited allele. However, in the adult brain, 70% of imprinted genes preferentially expressed the paternal allele. Thus, mothers and fathers appear to differentially influence developmental processes versus adult brain functions in offspring. His work also suggests the X chromosome is the maternal nexus of genetic control over the adult brain, which may provide a clue for why the X chromosome evolved an enrichment for genes that influence cognition. Interviewed by Jill Escher, November 2014 |
Links
Gregg Lab website: http://www.neuro.utah.edu/labs/gregg/Welcome.html University of Utah faculty page: http://www.neuro.utah.edu/people/faculty/gregg.html |
I’m delighted to have this opportunity to chat with you about your work in genomic imprinting, but first I want to find out more about your background, how you got into the science of epigenetics and neurodevelopment.
I started my career as a neural stem cell biologist and had a long training period focused on studying cells in culture, and then realized that I really wanted to get more insights into brain function and behavior. So I went to Harvard, to a lab that has expertise in understanding the genetic mechanisms that regulate social behaviors.
There I developed an expertise in genomics and doing behavior analysis, and became interested in studying genomic imprinting effects in the brain. in part because if you had imprinting effects on the X chromosome, then that might be a substrate that contributes to sexually dimorphic features and behaviors in the brain because imprinting on the X chromosome would only happen in females and not in males.
In the course of those studies, we came up with a method to profile gene expression from maternally versus paternally-inherited gene copies for any tissue type of the mouse at the time using new genomics technologies just emerging at the time. And this changed the landscape of the research entirely because now we could look across the entire genome in any tissue at any developmental stage and we could see which genes were showing a bias to express the mother's copy and which genes were showing a bias to express the father's copy.
Wow!
And we could compare those effects between brain regions and between males and females. And now we're studying these effects in other brain regions as well as between different tissues so different somatic tissues.
So those were the starting steps of my interest in genomic imprinting. And these studies opened up all kinds of possibilities because we could see these maternal and paternal allele expression effects and we've been working on resolving those at the levels of specific circuits in the brain and trying to figure out which neural circuits tend to express maternal alleles, which express paternal alleles, and where do neuropsychiatric risk factors fit into that architecture.
It's early stages but we think we're finding very interesting insights into parental origin of mutations, clearly there's a whole bunch of epigenetic machinery that may be causing these allele effects that we don't have a full understanding of at this point as well.
Can you explain a bit more about the difference between imprinted genes and non-imprinted genes?
Imprinted genes are very unique set of genes in the genome. We think at this stage that imprinting is a heritable epigenetic effect that occurs in mammals and flowering plants, but from what we can tell, most other species don't seem to have these effects.
There's something unique about mammals that has resulted in parents establishing marks on the genome in the germ cells. Mothers will establish different epigenetic marks than fathers at very specific locations in the genome. These locations are called imprinting control regions and when the sperm and egg unite in a fertilization event, those epigenetic marks, which are literally methyl groups that are covalently bound to specific cytosine residues in the genome sequence, are inherited by the offspring.
That means the chromosomes you get from your mother are not the same as the chromosomes you get from your father. And as a consequence for some genes—the numbers are still being debated, but we think that the number of genes that are affected by these marks are somewhere in the hundreds—those genes will show a bias to express only the mother's copy or only the father's copy.
We've learned that imprinting effects can be very tissue-specific. There are genes that will only express the father's copy in the brain but they'll express both the mother and the father's copy in the liver and the muscle for example. And there's other genes that will only express the mother's copy in the brain but express both copies in other tissues.
So for some reason, these epigenetic marks are controlling which parents copy is being expressed in which tissues. And we know that for some imprinted genes, if they have a mutation in the expressed copy the result is autism. So for example, UBE3A is a gene that is imprinted and if you get a mutation in the maternal copy of that gene [inherit a mutation in that gene from your mother] you will develop a syndrome that has autistic features called Angelman Syndrome.
Imprinting effects will tend to influence clusters of genes. So within the genome, you may find clusters of 4, 5, or 10 genes that all are regulated by this one mark that the mother or the father has established on the DNA. And as a consequence, that cluster of genes will show a bias to express the maternal or paternal copy.
What is the process by which those marks get established on the early sperm or early egg and when do those marks get established?
Those details are a major area of interest in the field and people have been trying to work out those mechanisms. And we don't know very much about it from humans though we know more from the mice, which are the classic research model organism for these sorts of problems.
Basically, a developing mouse has a set of cells, called primordial germ cells that are set aside to become the germ cells when it's older. By mid-gestation, those cells erase all the epigenetic marks that are on the DNA. And then as they get older the offspring establish new marks depending on whether they are male or female. If they're a son, they'll lay down marks that are appropriate for a male on the DNA of the sperm and if they're a female, then they'll lay down different marks in the eggs.
What are the mechanisms that determine those marks? Is it hormonal because it's male and female-determined or is it something else?
That's a great question. Researchers are working to discover those mechanisms and we have found some key enzymes that are important for making the marks.
Anyway, the epigenetic marks endure, they stick or they don't change even after the fusion of the egg and sperm to make the next generation, right?
That's right. When the egg and the sperm fuse for the next generation, those marks are protected. There is an erasing procedure that happens when the sperm and the egg meet. Most epigenetic marks in the sperm genome are stripped away when it enters the egg, but the imprinting marks are protected.
The egg genome has its own epigenetic marks and those are gradually diluted away as the embryo develops and cells divide. New epigenetic marks are established as undifferentiated cells start to commit to different cell lineages. Some cells are going to become brain cells, some are going to be endoderm, some are going to be mesoderm and they'll start to collect up the epigenetic marks that they need to become those different cell types. Throughout these processes, the parental imprints are maintained.
You were talking about the relationship between imprinted genes and specific neural circuits. Could you talk more broadly about how imprinted genes help control early neurodevelopment?
This is a very new area. We are still trying to get a handle on which genes are imprinted in the brain and how that compares to other tissues. Many researchers are studying the Angelman Syndrome gene, UBE3A. This gene plays a major role in synapse formation during development, regulating the formation of connections between neurons in the brain.
Other imprinted genes regulate the survival of neurons during postnatal development. For example, there is an imprinted gene called Paternally Expressed Gene 3 that influences the survival of neurons during the brain development when the brain is producing lots of cells and they have to be wired into functional circuits. The brain actually overproduces the number of cells that you need and many of them are pruned away and die. That “sculpting” process is regulated in part by imprinted genes.
What about more subtle impairments of the regulation of these imprinted genes? Is there any suggestion that short of an outright mutation that dysregulated imprinting could have an impairing effect on early neurodevelopment, or too soon to say?
We certainly know that mutations in some imprinted genes can cause neurodevelopmental disorders in humans. A challenge in the field is that we know a lot about imprinting in the mouse brain, but very little about the human brain. My lab is working on this problem.
We're figuring out which circuits express imprinted genes and we are learning about critical stages of brain development during which maternally and/or paternally expressed genes function. We are also testing whether perturbations to imprinted genes, due to genetic or environmental effects, can alter specific aspects of brain development and function. These are important questions and they're all of ahead of us to figure out at this stage.
Well, hopefully we'll get there.
For parents, it must seem incredibly slow.
I'm wondering if you could talk about the difference, as far as we know, in imprinting in humans as compared to model animals.
I would say that at this stage we have a sense that there are some differences and there are some similarities and we don't have a full sense of the landscape how different the two species are in terms of their imprinting.
I think the goal for the animal-based research is to gain insights into the mechanisms because the mechanisms will probably be the same. The proteins that have the job of establishing genomic imprints in the germ cells will be similar in mice and humans.
We can also use mice to determine whether there are particular environmental factors that impact the imprinting mechanisms. I think that's useful in terms of trying to understand what might cause loss of imprinting or other perturbations that could influence imprinting and gene expression in the brain.
One can also examine human patients to figure out if there is loss of an imprinting effect at a particular gene in the brain postmortem, for example, and determine whether the effect is statistically associated with autism. And then one can go back to the mouse model and try to work out the mechanisms involved and where in the sequence of events from the formation to the expression of imprinting can one disrupt things and cause neurodevelopmental changes that may be relevant to autism.
Autism is more common in males than females. Do you have any ideas about maybe why imprinting or X chromosome effects could be causing this gender difference?
I'm very interested in the idea that genetic and or epigenetic effects on the X chromosome may be playing a major role.
Currently, the mechanisms underlying this sex-based difference are unclear. The X chromosome is a good candidate since males have only one X chromosome and females have two, so they have a back-up copy. The X chromosome is enriched with genes that regulate brain function for reasons that are not clear. Males will be particularly susceptible to mutations, epigenetic or genetic, that influence X-linked genes because they just have the one copy and females won't. For example, in the case of Rett Syndrome, males do not survive the Rett Syndrome mutation, but females do, because of their backup copy.
I think the X chromosome is a really important place to look to solve this problem. That's all I can say at this point and we'll just see where things take us.
Thank you so much for your time today, this is a fascinating and important topic, and I thank you for your work and sharing your knowledge.
And thank you.
I started my career as a neural stem cell biologist and had a long training period focused on studying cells in culture, and then realized that I really wanted to get more insights into brain function and behavior. So I went to Harvard, to a lab that has expertise in understanding the genetic mechanisms that regulate social behaviors.
There I developed an expertise in genomics and doing behavior analysis, and became interested in studying genomic imprinting effects in the brain. in part because if you had imprinting effects on the X chromosome, then that might be a substrate that contributes to sexually dimorphic features and behaviors in the brain because imprinting on the X chromosome would only happen in females and not in males.
In the course of those studies, we came up with a method to profile gene expression from maternally versus paternally-inherited gene copies for any tissue type of the mouse at the time using new genomics technologies just emerging at the time. And this changed the landscape of the research entirely because now we could look across the entire genome in any tissue at any developmental stage and we could see which genes were showing a bias to express the mother's copy and which genes were showing a bias to express the father's copy.
Wow!
And we could compare those effects between brain regions and between males and females. And now we're studying these effects in other brain regions as well as between different tissues so different somatic tissues.
So those were the starting steps of my interest in genomic imprinting. And these studies opened up all kinds of possibilities because we could see these maternal and paternal allele expression effects and we've been working on resolving those at the levels of specific circuits in the brain and trying to figure out which neural circuits tend to express maternal alleles, which express paternal alleles, and where do neuropsychiatric risk factors fit into that architecture.
It's early stages but we think we're finding very interesting insights into parental origin of mutations, clearly there's a whole bunch of epigenetic machinery that may be causing these allele effects that we don't have a full understanding of at this point as well.
Can you explain a bit more about the difference between imprinted genes and non-imprinted genes?
Imprinted genes are very unique set of genes in the genome. We think at this stage that imprinting is a heritable epigenetic effect that occurs in mammals and flowering plants, but from what we can tell, most other species don't seem to have these effects.
There's something unique about mammals that has resulted in parents establishing marks on the genome in the germ cells. Mothers will establish different epigenetic marks than fathers at very specific locations in the genome. These locations are called imprinting control regions and when the sperm and egg unite in a fertilization event, those epigenetic marks, which are literally methyl groups that are covalently bound to specific cytosine residues in the genome sequence, are inherited by the offspring.
That means the chromosomes you get from your mother are not the same as the chromosomes you get from your father. And as a consequence for some genes—the numbers are still being debated, but we think that the number of genes that are affected by these marks are somewhere in the hundreds—those genes will show a bias to express only the mother's copy or only the father's copy.
We've learned that imprinting effects can be very tissue-specific. There are genes that will only express the father's copy in the brain but they'll express both the mother and the father's copy in the liver and the muscle for example. And there's other genes that will only express the mother's copy in the brain but express both copies in other tissues.
So for some reason, these epigenetic marks are controlling which parents copy is being expressed in which tissues. And we know that for some imprinted genes, if they have a mutation in the expressed copy the result is autism. So for example, UBE3A is a gene that is imprinted and if you get a mutation in the maternal copy of that gene [inherit a mutation in that gene from your mother] you will develop a syndrome that has autistic features called Angelman Syndrome.
Imprinting effects will tend to influence clusters of genes. So within the genome, you may find clusters of 4, 5, or 10 genes that all are regulated by this one mark that the mother or the father has established on the DNA. And as a consequence, that cluster of genes will show a bias to express the maternal or paternal copy.
What is the process by which those marks get established on the early sperm or early egg and when do those marks get established?
Those details are a major area of interest in the field and people have been trying to work out those mechanisms. And we don't know very much about it from humans though we know more from the mice, which are the classic research model organism for these sorts of problems.
Basically, a developing mouse has a set of cells, called primordial germ cells that are set aside to become the germ cells when it's older. By mid-gestation, those cells erase all the epigenetic marks that are on the DNA. And then as they get older the offspring establish new marks depending on whether they are male or female. If they're a son, they'll lay down marks that are appropriate for a male on the DNA of the sperm and if they're a female, then they'll lay down different marks in the eggs.
What are the mechanisms that determine those marks? Is it hormonal because it's male and female-determined or is it something else?
That's a great question. Researchers are working to discover those mechanisms and we have found some key enzymes that are important for making the marks.
Anyway, the epigenetic marks endure, they stick or they don't change even after the fusion of the egg and sperm to make the next generation, right?
That's right. When the egg and the sperm fuse for the next generation, those marks are protected. There is an erasing procedure that happens when the sperm and the egg meet. Most epigenetic marks in the sperm genome are stripped away when it enters the egg, but the imprinting marks are protected.
The egg genome has its own epigenetic marks and those are gradually diluted away as the embryo develops and cells divide. New epigenetic marks are established as undifferentiated cells start to commit to different cell lineages. Some cells are going to become brain cells, some are going to be endoderm, some are going to be mesoderm and they'll start to collect up the epigenetic marks that they need to become those different cell types. Throughout these processes, the parental imprints are maintained.
You were talking about the relationship between imprinted genes and specific neural circuits. Could you talk more broadly about how imprinted genes help control early neurodevelopment?
This is a very new area. We are still trying to get a handle on which genes are imprinted in the brain and how that compares to other tissues. Many researchers are studying the Angelman Syndrome gene, UBE3A. This gene plays a major role in synapse formation during development, regulating the formation of connections between neurons in the brain.
Other imprinted genes regulate the survival of neurons during postnatal development. For example, there is an imprinted gene called Paternally Expressed Gene 3 that influences the survival of neurons during the brain development when the brain is producing lots of cells and they have to be wired into functional circuits. The brain actually overproduces the number of cells that you need and many of them are pruned away and die. That “sculpting” process is regulated in part by imprinted genes.
What about more subtle impairments of the regulation of these imprinted genes? Is there any suggestion that short of an outright mutation that dysregulated imprinting could have an impairing effect on early neurodevelopment, or too soon to say?
We certainly know that mutations in some imprinted genes can cause neurodevelopmental disorders in humans. A challenge in the field is that we know a lot about imprinting in the mouse brain, but very little about the human brain. My lab is working on this problem.
We're figuring out which circuits express imprinted genes and we are learning about critical stages of brain development during which maternally and/or paternally expressed genes function. We are also testing whether perturbations to imprinted genes, due to genetic or environmental effects, can alter specific aspects of brain development and function. These are important questions and they're all of ahead of us to figure out at this stage.
Well, hopefully we'll get there.
For parents, it must seem incredibly slow.
I'm wondering if you could talk about the difference, as far as we know, in imprinting in humans as compared to model animals.
I would say that at this stage we have a sense that there are some differences and there are some similarities and we don't have a full sense of the landscape how different the two species are in terms of their imprinting.
I think the goal for the animal-based research is to gain insights into the mechanisms because the mechanisms will probably be the same. The proteins that have the job of establishing genomic imprints in the germ cells will be similar in mice and humans.
We can also use mice to determine whether there are particular environmental factors that impact the imprinting mechanisms. I think that's useful in terms of trying to understand what might cause loss of imprinting or other perturbations that could influence imprinting and gene expression in the brain.
One can also examine human patients to figure out if there is loss of an imprinting effect at a particular gene in the brain postmortem, for example, and determine whether the effect is statistically associated with autism. And then one can go back to the mouse model and try to work out the mechanisms involved and where in the sequence of events from the formation to the expression of imprinting can one disrupt things and cause neurodevelopmental changes that may be relevant to autism.
Autism is more common in males than females. Do you have any ideas about maybe why imprinting or X chromosome effects could be causing this gender difference?
I'm very interested in the idea that genetic and or epigenetic effects on the X chromosome may be playing a major role.
Currently, the mechanisms underlying this sex-based difference are unclear. The X chromosome is a good candidate since males have only one X chromosome and females have two, so they have a back-up copy. The X chromosome is enriched with genes that regulate brain function for reasons that are not clear. Males will be particularly susceptible to mutations, epigenetic or genetic, that influence X-linked genes because they just have the one copy and females won't. For example, in the case of Rett Syndrome, males do not survive the Rett Syndrome mutation, but females do, because of their backup copy.
I think the X chromosome is a really important place to look to solve this problem. That's all I can say at this point and we'll just see where things take us.
Thank you so much for your time today, this is a fascinating and important topic, and I thank you for your work and sharing your knowledge.
And thank you.