Final Report Summary - SEEM (Assessing the effect of early social environment on epigenetic modification)
Among the molecular mechanisms that control gene expression, DNA methylation is a key factor. In this biochemical modification methyl groups are added to cytosines in CG dinucleotide sequences (so-called CG-sites) within the regulatory regions of genes. Due to the importance of DNA methylation in developmental processes, the majority of methylated gene regions are tissue-specific and don’t show high variability between individuals. However, an increasing number of studies do report changes in DNA methylation patterns following exposures to chemical, biological, or social environment; and small but reproducible changes could highlight important regulatory mechanisms of environmental effects. This project aimed to study the effect of early social environment, including prenatal maternal depression and postnatal care-giving quality, on the DNA methylation pattern of the offspring. For clarity, the results of the conducted experiments are presented according to the seven specific aims of the fellowship (points 1-4 at the outgoing phase and points 5-7 at the return phase). Whenever adequate, the knowledge transfer and the impact on the fellow’s carrier are highlighted.
1. Learning techniques to supplement genetic association analyses with epigenetic analyses:
In the host laboratory (headed by a pioneer in the field of epigenetics, Prof. Moshe Szyf), the fellow learned two different epigenetic techniques: 1. methylated DNA immunoprecipitation (MeDIP), based on a specific antibody's ability to bind methylated cytosine; 2. bisulfite conversion analyses, based on chemical conversion of non-methylated cytosine into uracil, which can be distinguished from methylated cytosine after an amplification step. The fellow used both types of techniques in human and animal studies resulting in research articles in the outgoing phase, and successfully applied them in the return phase, teaching her students and colleagues (see point 6).
2. Developing high-throughput, cost-effective methods for assessing DNA methylation levels:
Although the fellow tested various published methods using real-time PCR (a rapid and easily accessible method), and developed a one-step quantification method, their respective sensitivities were not satisfactory, especially when compared to that of the gold-standard method pyrosequencing. The lower accuracies encountered were possibly due to the very low levels of methylation changes (2-10%) often observed in psychological studies, which are far from the original target range of real-time PCR methods in cancer research where typically at least 2-fold differences are detected.
3. Assessing the usability of saliva and buccal cell samples in DNA methylation analyses:
Since non-invasive sampling procedures are optimal for large-scale studies and research involving children, mouth samples (saliva, mouthwash and buccal swabs) versus blood-derived samples were compared in a human pilot study involving 16 healthy individuals. Preliminary results showed that certain CG-sites had significantly lower DNA methylation levels when assessed from mouth as compared to blood samples, and were significantly affected by cell type ratios. Buccal swab samples showed the least cellular heterogeneity, whereas saliva and mouthwash samples showed the greatest variability in the ratio of epithelial and blood cells. This is of great importance because the different cell types present in these samples (buccal epithelial cells versus leukocytes) have markedly different epigenetic signatures. Interestingly, comparative analyses indicated that epithelial tissue might be a better surrogate for non-blood based diseases such as mental health problems, therefore, the fellow set-up a collection procedure to obtain cell samples with greater epithelial homogeneity. This resulted in two major technical advances: (1) An improved collection method leading to high epithelial : leukocyte ratios; (2) DNA methylation assays for a panel of tissue-specific CG-sites to obtain cell composition data of the original sample. This data can be used in order to control for cell heterogeneity in epigenetic association analyses. In addition, the fellow learned separation protocols of different cell types and various imaging techniques during the implementation of this project.
4. Assessing the validity of measuring DNA methylation changes in peripheral tissues in relation to brain-specific changes in an animal model:
From an on-going collaborative project, blood, saliva and buccal cell samples were collected from rhesus monkeys subjected to different rearing conditions (mother-reared versus surrogate-peer-reared). This model produces an array of physiological, behavioural, and neurobiological deficits (which parallel those identified in humans subjected to various forms of early life adversity), such as increased frequency of illnesses, higher mortality rate, aggressive behaviour or depressive and anxiety symptoms. The observed genome-wide DNA methylation differences of blood and buccal cells were found to be organized along functional gene pathways. In addition, the enriched pathways were connected to psychological disorders and to neuroactive ligand - receptor pairs, providing evidence that variations in DNA methylation - even when assessed in peripheral cells - constitute functionally organized patterns, which may underlie the observed phenotypic differences. The comparison of differently methylated genes in blood samples and previously obtained results from T lymphocyte and brain (prefrontal cortex) samples pointed to common regulators, such as stress-responsive elements.
5. Applying learned methodologies to national and international collaborative projects:
The fellow has successfully introduced DNA methylation analyses into previous collaborative projects conducted with developmental psychologists and obtained new collaborations with psychologists and psychiatrists working on depressive or borderline personality disorders in several countries (including Canada, Ireland, and Hungary). In this way, different patient cohorts are tested with the same methodology in order to identify biological mediator mechanisms involving DNA methylation changes in relation to environmental risk factors, such as childhood trauma, neglect, or low socioeconomic status. Importantly, the technical improvements described in point 3 were implemented in the on-going epigenetic association studies.
6. Establishing an epigenetic section at the molecular genetic laboratory in the return phase:
The newly acquired knowledge of the fellow enabled the return host laboratory to measure DNA methylation levels of selected candidate gene regions in association with social environmental effects using the sensitive pyrosequencing method. An active research project supported by national funds aims to characterize the effect of maternal care quality in a longitudinal study from the normative population. The results of this study could complement the findings obtained from patients with psychiatric disorders (see previous point). In another project of the return host department the fellow could use her knowledge of DNA immunoprecipitation methods to assess DNA methylation and hydroxymethylation patterns in a human fibroblast experiment. Understanding the importance of 5-hydroxymethylcytosine modification in human tissue development has been one of the hottest topics in epigenetics recently. Obtaining fundamental knowledge on its enzymatic processes (e.g. requirement for certain cofactors) would potentially impact molecular biological experimental designs in the future.
7. Assessing the usability of frozen, archived samples in DNA methylation analyses:
The fellow has tested DNA methylation analyses of archived DNA samples stored for more than 10 years in the biobank of the return host laboratory using the pyrosequencing method with the help of her undergraduate and graduate students. The quality of the final DNA methylation signal did not depend on storage conditions, but rather on the original quality and quantity of the DNA samples. Since the majority of archived DNA samples were isolated from buccal cells (originating from the mouth where a substantial amount of the DNA can be derived from bacteria), it was important to select samples with higher DNA amounts in order to achieve optimal amplification with human-specific primer sets. One of the present aims of the on-going longitudinal study mentioned previously (where the first samples have been collected from one-year old infants and the participants were recruited recently at the of age 16) is to assess the long-term stability of DNA methylation patterns in buccal swab samples.
CONCLUSIONS: Genome-wide analyses conducted at the host laboratory of the outgoing phase pointed to “system-wide” and “genome-wide” DNA methylation changes in a primate animal model of negative early social life experiences. These results support the hypothesis that DNA methylation patterns found in peripheral tissues can mirror the physiological effects of social experiences. The fellow also carried out and improved candidate gene methylation analyses by assessing cell heterogeneity of the samples which has to be corrected for in association studies due to the very different epigenetic landscapes of different cell types.
IMPACT OF THE PROJECT: Acquiring first-hand knowledge of epigenetic techniques has given the European return host laboratory a new and relevant dimension to add onto on-going collaborative projects in the field of developmental psychology and psychiatric genetics. The employment of non-invasively obtained samples will significantly impact epigenetic research involving infants as well as large-scale epidemiological studies involving adults. The technical achievements will potentially impact both sample collection at the starting point and accuracy of retrieved data at the end stage of epigenetic association studies. The impact of this fellowship on the fellow’s carrier was multi-layered: She could concentrate on research activities in the outgoing phase and establish her own work-group in the return phase.
1. Learning techniques to supplement genetic association analyses with epigenetic analyses:
In the host laboratory (headed by a pioneer in the field of epigenetics, Prof. Moshe Szyf), the fellow learned two different epigenetic techniques: 1. methylated DNA immunoprecipitation (MeDIP), based on a specific antibody's ability to bind methylated cytosine; 2. bisulfite conversion analyses, based on chemical conversion of non-methylated cytosine into uracil, which can be distinguished from methylated cytosine after an amplification step. The fellow used both types of techniques in human and animal studies resulting in research articles in the outgoing phase, and successfully applied them in the return phase, teaching her students and colleagues (see point 6).
2. Developing high-throughput, cost-effective methods for assessing DNA methylation levels:
Although the fellow tested various published methods using real-time PCR (a rapid and easily accessible method), and developed a one-step quantification method, their respective sensitivities were not satisfactory, especially when compared to that of the gold-standard method pyrosequencing. The lower accuracies encountered were possibly due to the very low levels of methylation changes (2-10%) often observed in psychological studies, which are far from the original target range of real-time PCR methods in cancer research where typically at least 2-fold differences are detected.
3. Assessing the usability of saliva and buccal cell samples in DNA methylation analyses:
Since non-invasive sampling procedures are optimal for large-scale studies and research involving children, mouth samples (saliva, mouthwash and buccal swabs) versus blood-derived samples were compared in a human pilot study involving 16 healthy individuals. Preliminary results showed that certain CG-sites had significantly lower DNA methylation levels when assessed from mouth as compared to blood samples, and were significantly affected by cell type ratios. Buccal swab samples showed the least cellular heterogeneity, whereas saliva and mouthwash samples showed the greatest variability in the ratio of epithelial and blood cells. This is of great importance because the different cell types present in these samples (buccal epithelial cells versus leukocytes) have markedly different epigenetic signatures. Interestingly, comparative analyses indicated that epithelial tissue might be a better surrogate for non-blood based diseases such as mental health problems, therefore, the fellow set-up a collection procedure to obtain cell samples with greater epithelial homogeneity. This resulted in two major technical advances: (1) An improved collection method leading to high epithelial : leukocyte ratios; (2) DNA methylation assays for a panel of tissue-specific CG-sites to obtain cell composition data of the original sample. This data can be used in order to control for cell heterogeneity in epigenetic association analyses. In addition, the fellow learned separation protocols of different cell types and various imaging techniques during the implementation of this project.
4. Assessing the validity of measuring DNA methylation changes in peripheral tissues in relation to brain-specific changes in an animal model:
From an on-going collaborative project, blood, saliva and buccal cell samples were collected from rhesus monkeys subjected to different rearing conditions (mother-reared versus surrogate-peer-reared). This model produces an array of physiological, behavioural, and neurobiological deficits (which parallel those identified in humans subjected to various forms of early life adversity), such as increased frequency of illnesses, higher mortality rate, aggressive behaviour or depressive and anxiety symptoms. The observed genome-wide DNA methylation differences of blood and buccal cells were found to be organized along functional gene pathways. In addition, the enriched pathways were connected to psychological disorders and to neuroactive ligand - receptor pairs, providing evidence that variations in DNA methylation - even when assessed in peripheral cells - constitute functionally organized patterns, which may underlie the observed phenotypic differences. The comparison of differently methylated genes in blood samples and previously obtained results from T lymphocyte and brain (prefrontal cortex) samples pointed to common regulators, such as stress-responsive elements.
5. Applying learned methodologies to national and international collaborative projects:
The fellow has successfully introduced DNA methylation analyses into previous collaborative projects conducted with developmental psychologists and obtained new collaborations with psychologists and psychiatrists working on depressive or borderline personality disorders in several countries (including Canada, Ireland, and Hungary). In this way, different patient cohorts are tested with the same methodology in order to identify biological mediator mechanisms involving DNA methylation changes in relation to environmental risk factors, such as childhood trauma, neglect, or low socioeconomic status. Importantly, the technical improvements described in point 3 were implemented in the on-going epigenetic association studies.
6. Establishing an epigenetic section at the molecular genetic laboratory in the return phase:
The newly acquired knowledge of the fellow enabled the return host laboratory to measure DNA methylation levels of selected candidate gene regions in association with social environmental effects using the sensitive pyrosequencing method. An active research project supported by national funds aims to characterize the effect of maternal care quality in a longitudinal study from the normative population. The results of this study could complement the findings obtained from patients with psychiatric disorders (see previous point). In another project of the return host department the fellow could use her knowledge of DNA immunoprecipitation methods to assess DNA methylation and hydroxymethylation patterns in a human fibroblast experiment. Understanding the importance of 5-hydroxymethylcytosine modification in human tissue development has been one of the hottest topics in epigenetics recently. Obtaining fundamental knowledge on its enzymatic processes (e.g. requirement for certain cofactors) would potentially impact molecular biological experimental designs in the future.
7. Assessing the usability of frozen, archived samples in DNA methylation analyses:
The fellow has tested DNA methylation analyses of archived DNA samples stored for more than 10 years in the biobank of the return host laboratory using the pyrosequencing method with the help of her undergraduate and graduate students. The quality of the final DNA methylation signal did not depend on storage conditions, but rather on the original quality and quantity of the DNA samples. Since the majority of archived DNA samples were isolated from buccal cells (originating from the mouth where a substantial amount of the DNA can be derived from bacteria), it was important to select samples with higher DNA amounts in order to achieve optimal amplification with human-specific primer sets. One of the present aims of the on-going longitudinal study mentioned previously (where the first samples have been collected from one-year old infants and the participants were recruited recently at the of age 16) is to assess the long-term stability of DNA methylation patterns in buccal swab samples.
CONCLUSIONS: Genome-wide analyses conducted at the host laboratory of the outgoing phase pointed to “system-wide” and “genome-wide” DNA methylation changes in a primate animal model of negative early social life experiences. These results support the hypothesis that DNA methylation patterns found in peripheral tissues can mirror the physiological effects of social experiences. The fellow also carried out and improved candidate gene methylation analyses by assessing cell heterogeneity of the samples which has to be corrected for in association studies due to the very different epigenetic landscapes of different cell types.
IMPACT OF THE PROJECT: Acquiring first-hand knowledge of epigenetic techniques has given the European return host laboratory a new and relevant dimension to add onto on-going collaborative projects in the field of developmental psychology and psychiatric genetics. The employment of non-invasively obtained samples will significantly impact epigenetic research involving infants as well as large-scale epidemiological studies involving adults. The technical achievements will potentially impact both sample collection at the starting point and accuracy of retrieved data at the end stage of epigenetic association studies. The impact of this fellowship on the fellow’s carrier was multi-layered: She could concentrate on research activities in the outgoing phase and establish her own work-group in the return phase.