Multigenerational Effects of Endocrine-Disrupting Molecules
"I would stress that our data show that in mice, social interactions, as a class of behaviors,
are disrupted by transgenerational exposure to 'human-like' levels of BPA."
Emilie Rissman, PhD, Professor of Biochemistry & Molecular Genetics, University of Virginia, studies mammalian behavioral genetics. Because many social behaviors are activated only when gonadal steroid hormones are present, the genes for steroid hormone receptors are currently under study in her lab. She has found that androgen receptor is required for normal social affiliative behavior in male mice. In addition, her work shows that several social behaviors are sexually dimorphic, in part, because of differences in sex chromosome genes. Her work in mice is relevant to several sexually dimorphic neurobehavioral diseases including autism spectrum disorder.
Interviewed by Jill Escher, April 2014
Dr. Rissman, I'm very pleased to be interviewing you because since you are one of the few researchers to explicitly examine multigenerational effects of endocrine disruption as it relates to abnormal behavior and neurodevelopment in offspring. How did you come to examine this question?
I have been interested how the environment and hormones regulate social behaviors for my whole career. Particularly I have studied social behaviors, and other behaviors that are sexually dimorphic. Autism and several other neurobehavioral developmental disorders are very dimorphic and typically males are more likely than females to develop these illnesses. Moreover, autistic children and adults have a number of issues, one of the major ones being unable to interact normally with other individuals. This is what made autism interesting to me. Given my background in hormones, which have powerful actions on developing organisms and adults, I was led to ask if endocrine disrupting compounds, present during embryonic development might have an impact on social behaviors.
Interviewed by Jill Escher, April 2014
Dr. Rissman, I'm very pleased to be interviewing you because since you are one of the few researchers to explicitly examine multigenerational effects of endocrine disruption as it relates to abnormal behavior and neurodevelopment in offspring. How did you come to examine this question?
I have been interested how the environment and hormones regulate social behaviors for my whole career. Particularly I have studied social behaviors, and other behaviors that are sexually dimorphic. Autism and several other neurobehavioral developmental disorders are very dimorphic and typically males are more likely than females to develop these illnesses. Moreover, autistic children and adults have a number of issues, one of the major ones being unable to interact normally with other individuals. This is what made autism interesting to me. Given my background in hormones, which have powerful actions on developing organisms and adults, I was led to ask if endocrine disrupting compounds, present during embryonic development might have an impact on social behaviors.
How did gestational BPA exposure differentially affect succeeding generations?
We have examined anxiety, activity, olfaction and social behaviors in male and female juvenile mice exposed either in utero (F1) or transgenerationally (the third, F3, or fourth, F4, generation after the initial exposure). We found that specifically social behaviors are influenced by BPA. We have published two papers on these data so far and have collected new data which also replicate these findings. The effects in the F3 and F4 generations are in fact more pronounced than in the directly exposed F1 mice. The mice in the BPA lineages (F3) are more interactive with adults and with other juveniles than are the control juveniles. Obviously the tasks we use cannot completely simulate how children interact, however we do think that the fact that BPA lineage mice display higher levels of interactions with other mice might be a form of perseveration. Perseveration is one of the classic characteristics of autism spectrum disorder. I would stress that our data show that in mice, social interactions, as a class of behaviors, are disrupted by transgenerational exposure to “human-like” levels of BPA.
What epigenetic mechanisms are at play with exposure to EDCs?
Several labs have shown that EDCs change either global DNA methylation or methylation at specific gene promoters. In addition, while less work has been done on chromatin, EDCs can change histone modifications, which in turn make DNA more or less accessible. This ultimately changes gene transcription and then levels of protein. We are excited about new funding from NIEHS we have received to track the epigenomic changes produced by BPA in a holistic manner. We will examine histone modifications, DNA methylation, and RNA expression. Together these data will point to the genes which are susceptible to BPA and responsible for the behavioral changes we have observed. We also have a grant pending that would extend this work to germ cells to determine which genes are epigenetically modified and passed on to the next generations.
What do we know about hormone and other receptors on and in germ cells? And somatic cells, if there's a difference.
Steroid hormone receptors are present in many cells, including neurons. Germ cells are influenced by hormones, which interact with receptors in somatic cells nearby (i.e. the granulosa cells in the ovary and the sertoli cells in the testes).
What role does the steroid hormone receptor play in epigenetic function?
There are several ways that steroid receptors might function in epigenetics. First the steroid receptors are transcription factors and might change the expression of various enzymes that are involved in the epigenetic machinery. Second steroid receptors can work in conjunction with other proteins to form protein complexes that in turn block or facilitate DNA methylation and/or chromatin modifications.
Why do hormone-disrupting molecules affect epigenetic activity?
The EDCs can work as steroid agonists in which case they might enhance the actions of the steroid receptors. They can also act as steroid antagonists which may block the normal actions of steroids. Either way the can promote or black steroid actions at inappropriate times in the organism.
Some of the scientists I've spoken with differentiated between "strong" exposures and weak ones in terms of likelihood of epigenetic impacts. Some strong ones mentioned included steroid hormones and their mimics, cigarette smoke and psychoactive drugs. Comparatively weak ones were thought to include variability in stress and nutrition. Do you agree with this general distinction?
I don’t actually. I think the strength of the exposure is going to depend on the stage in development as much as the type of compound. Some of the most interesting data on epigenetic inheritance comes from studies of nutrition, for example work on folate donors that can reverse the actions of BPA. The first suggestion of transgenerational epigenetic inheritance was from the “Dutch Famine” kids born during this time period were smaller than their siblings and later in life had more disease. These observations were essential for the development of the early origins of adult disease hypothesis. Work in animals on maternal separation stress has also shown changes in DNA methylation of genes in the hypothalamic pituitary adrenal axis.
We have examined anxiety, activity, olfaction and social behaviors in male and female juvenile mice exposed either in utero (F1) or transgenerationally (the third, F3, or fourth, F4, generation after the initial exposure). We found that specifically social behaviors are influenced by BPA. We have published two papers on these data so far and have collected new data which also replicate these findings. The effects in the F3 and F4 generations are in fact more pronounced than in the directly exposed F1 mice. The mice in the BPA lineages (F3) are more interactive with adults and with other juveniles than are the control juveniles. Obviously the tasks we use cannot completely simulate how children interact, however we do think that the fact that BPA lineage mice display higher levels of interactions with other mice might be a form of perseveration. Perseveration is one of the classic characteristics of autism spectrum disorder. I would stress that our data show that in mice, social interactions, as a class of behaviors, are disrupted by transgenerational exposure to “human-like” levels of BPA.
What epigenetic mechanisms are at play with exposure to EDCs?
Several labs have shown that EDCs change either global DNA methylation or methylation at specific gene promoters. In addition, while less work has been done on chromatin, EDCs can change histone modifications, which in turn make DNA more or less accessible. This ultimately changes gene transcription and then levels of protein. We are excited about new funding from NIEHS we have received to track the epigenomic changes produced by BPA in a holistic manner. We will examine histone modifications, DNA methylation, and RNA expression. Together these data will point to the genes which are susceptible to BPA and responsible for the behavioral changes we have observed. We also have a grant pending that would extend this work to germ cells to determine which genes are epigenetically modified and passed on to the next generations.
What do we know about hormone and other receptors on and in germ cells? And somatic cells, if there's a difference.
Steroid hormone receptors are present in many cells, including neurons. Germ cells are influenced by hormones, which interact with receptors in somatic cells nearby (i.e. the granulosa cells in the ovary and the sertoli cells in the testes).
What role does the steroid hormone receptor play in epigenetic function?
There are several ways that steroid receptors might function in epigenetics. First the steroid receptors are transcription factors and might change the expression of various enzymes that are involved in the epigenetic machinery. Second steroid receptors can work in conjunction with other proteins to form protein complexes that in turn block or facilitate DNA methylation and/or chromatin modifications.
Why do hormone-disrupting molecules affect epigenetic activity?
The EDCs can work as steroid agonists in which case they might enhance the actions of the steroid receptors. They can also act as steroid antagonists which may block the normal actions of steroids. Either way the can promote or black steroid actions at inappropriate times in the organism.
Some of the scientists I've spoken with differentiated between "strong" exposures and weak ones in terms of likelihood of epigenetic impacts. Some strong ones mentioned included steroid hormones and their mimics, cigarette smoke and psychoactive drugs. Comparatively weak ones were thought to include variability in stress and nutrition. Do you agree with this general distinction?
I don’t actually. I think the strength of the exposure is going to depend on the stage in development as much as the type of compound. Some of the most interesting data on epigenetic inheritance comes from studies of nutrition, for example work on folate donors that can reverse the actions of BPA. The first suggestion of transgenerational epigenetic inheritance was from the “Dutch Famine” kids born during this time period were smaller than their siblings and later in life had more disease. These observations were essential for the development of the early origins of adult disease hypothesis. Work in animals on maternal separation stress has also shown changes in DNA methylation of genes in the hypothalamic pituitary adrenal axis.
It seems we forget there's a long and complicated molecular phase of the human life cycle. For example, those who say BPA or pesticides are safe cite studies on adults, children or fetuses. But isn't the most vulnerable phase gametes and precursor cells? What do you consider to be important windows of vulnerability?
The classical windows during which EDCs might affect inherited traits are periods of time when the germ cells are undergoing epigenetic reorganization. This occurs at fertilization and again in mid-gestation when the germ cells move into the gonads. On the other hand, the presence of EDCs during critical periods of brain development, which in humans extend from the fetus to the young child, may modify development of neural circuits, migration and cell survival.
While nearly all scientists find endocrine disruption a concern, some say "there's just not enough evidence yet" to be concerned about ambient exposures. How do you respond to that? What would be "enough" evidence?
Other countries have banned several of the EDCs that we are still debating in the US. I think the research on EDCs is getting better all the time and that we are honing in on important aspects of their actions. For example we know that low doses may have different effects than high doses, and as discussed here that critical period exposures are more partially disruptive than at other times. I am not a regulator so I don’t know how much data is sufficient. I would have thought that when the FDA suggested a ban on BPA exposure to infants that they would have also recommended a ban during gestation. It has been well documented that gestation is likewise a critical period, and science should underlie policy decisions.
The classical windows during which EDCs might affect inherited traits are periods of time when the germ cells are undergoing epigenetic reorganization. This occurs at fertilization and again in mid-gestation when the germ cells move into the gonads. On the other hand, the presence of EDCs during critical periods of brain development, which in humans extend from the fetus to the young child, may modify development of neural circuits, migration and cell survival.
While nearly all scientists find endocrine disruption a concern, some say "there's just not enough evidence yet" to be concerned about ambient exposures. How do you respond to that? What would be "enough" evidence?
Other countries have banned several of the EDCs that we are still debating in the US. I think the research on EDCs is getting better all the time and that we are honing in on important aspects of their actions. For example we know that low doses may have different effects than high doses, and as discussed here that critical period exposures are more partially disruptive than at other times. I am not a regulator so I don’t know how much data is sufficient. I would have thought that when the FDA suggested a ban on BPA exposure to infants that they would have also recommended a ban during gestation. It has been well documented that gestation is likewise a critical period, and science should underlie policy decisions.