Exposure to environmental factors can induce the epigenetic transgenerational inheritance of disease. Alterations to the epigenome termed “epimutations” include “primary epimutations” which are epigenetic alterations in the absence of genetic change and “secondary epimutations” which form following an initial genetic change. To determine if secondary epimutations contribute to transgenerational transmission of disease following in utero exposure to the endocrine disruptor vinclozolin, we exposed pregnant female rats carrying the lacI mutation-reporter transgene to vinclozolin and assessed the frequency of mutations in kidney tissue and sperm recovered from F1 and F3 generation progeny. Our results confirm that vinclozolin induces primary epimutations rather than secondary epimutations, but also suggest that some primary epimutations can predispose a subsequent accelerated accumulation of genetic mutations in F3 generation descendants that have the potential to contribute to transgenerational phenotypes. We therefore propose the existence of “tertiary epimutations” which are initial primary epimutations that promote genome instability leading to an accelerated accumulation of genetic mutations.
Citation: McCarrey JR, Lehle JD, Raju SS, Wang Y, Nilsson EE, Skinner MK (2016) Tertiary Epimutations – A Novel Aspect of Epigenetic Transgenerational Inheritance Promoting Genome Instability. PLoS ONE 11(12): e0168038. https://doi.org/10.1371/journal.pone.0168038
Editor: Toshi Shioda, Massachusetts General Hospital, UNITED STATES
Received: July 22, 2016; Accepted: November 23, 2016; Published: December 19, 2016
Copyright: © 2016 McCarrey et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are presented in the primary manuscript and supplemental information.
Funding: This work was supported by Kleberg Foundation grant to JRM (http://www.klebergfoundation.org), NIH/NIEHS grant to MKS (http://www.niehs.nih.gov), and John Templeton Foundation grant to MKS and JRM (https://www.templeton.org).
Competing interests: The authors have declared that no competing interests exist.
Since the initial report of the induction of epigenetic transgenerational inheritance of adult-onset disease in F3 generation progeny from an F0 generation gestating female rat exposed to the endocrine disruptor vinclozolin , there have been numerous reports of similar effects caused by exposure of fetuses or adults to a variety of different environmental factors (see [2,3] for reviews), as well as reports of epigenetic transgenerational inheritance in numerous additional species [1,4–9]. These environmental factors include various toxicants, including many different endocrine disruptors, nutrition effects including famine, caloric restriction, high fat diets or folate deficiencies, and other stressors such as drought, smoking, alcohol or heat. Exposure of fetuses to these agents in utero, or of adults to these environmental factors, is associated with transgenerational increases in the incidence of adult-onset disease or abnormalities impacting the reproductive systems and fertility in both sexes, the incidence of cancer, the immune system, kidney function, the prostate, obesity, cardiovascular function, growth, insulin sensitivity, glucose tolerance, pulmonary function, neuronal function and social behavior, among other effects . These reports have revealed a novel etiology of adult-onset disease and a mechanism underlying the developmental origins of health and disease (DOHAD) . In addition, these observations suggest that the phenomenon of epigenetic transgenerational inheritance of disease is significantly more extensive than previously thought.
While it is not surprising that exposure to environmental factors or toxicants can alter the epigenome, especially during key developmental periods when epigenetic reprogramming is taking place , it is surprising that such alterations, known as epimutations, can persist over multiple generations in the absence of any further exposure to the disruptive agent [12–14]. Persistence of abnormalities for one or two generations following in utero exposure may simply reflect toxicity related effects induced by direct exposure of an F0 generation female, an F1 generation fetus, and/or the germ cells within that fetus that will give rise to the F2 generation. This phenomenon has been termed a “multigenerational exposure”  or more recently “intergenerational inheritance” , but, in fact, does not really represent true “inheritance” at all if it is limited to direct exposure toxicity effects. However, transmission of phenotypes through three or more generations following a single transient exposure to the causative agent requires some mechanism of transgenerational inheritance [1,13,17]. Epimutations induced by certain effects, for example the use of assisted reproductive technologies (ART), have been shown to be corrected by normal germline-specific epigenetic reprogramming such that they are detected in the F1 generation, but are not transmitted to subsequent generations . These are examples of intergenerational epimutations that are not transgenerational. This begs the question of how epimutations caused by exposure to certain environmental factors, such as endocrine disruptors, during critical windows of development are able to avoid subsequent correction by germline epigenetic reprogramming such that they persist over multiple generations via transgenerational inheritance.
One potential explanation for the observed transgenerational transmission of epimutations induced by a single exposure to vinclozolin or similar agents in utero is that this exposure could promote the formation of genetic mutations that impact epigenetic programming, such that the phenotype of these mutations is manifest as epimutations, but the underlying source of these defects is one or more genetic mutations. Indeed, two distinct types of epimutations have been previously described . “Primary epimutations” are epigenetic aberrations that occur in the absence of any genetic change and are propagated via mitosis to daughter cells or via meiosis to subsequent generations on the basis of epigenetic rather than genetic inheritance. “Secondary epimutations” are those that form as a consequence of, and thus subsequent to, an initial genetic change (mutation), and can therefore be transmitted on the basis of genetic transmission of the initial mutation or on the basis of epigenetic transmission of the subsequent epimutation, or both. Primary epimutations in the germ line would have the potential to be corrected by germline epigenetic reprogramming, whereas secondary epimutations would not be corrected so would be transmitted transgenerationally as genetic traits. Thus to understand the mechanism of epigenetic transgenerational inheritance, it is important to distinguish between primary and secondary epimutations.
Mutation-reporter transgenes provide a convenient and sensitive approach to assess the frequency of genetic mutations. Such systems can facilitate testing of tens or hundreds of thousands of copies of a reporter transgene to accurately determine the frequency of point mutations in this gene without the need to sequence most of these copies. One of the most extensively studied mutation-reporter systems is the “Big Blue” rodent system originally developed by Stratagene (now part of Agilent) in the 1990s [20,21]. This approach provides a method to examine the frequency of mutations accumulated in a transgene that is integrated into the same genomic location in all cells, but is not expressed in any cell type of the animal so does not impact the function of those cells in any way. Specifically, the Big Blue system is based on the colorimetric detection of expression of the lacZ reporter gene as an indicator of mutations that have occurred in the lacI (repressor) gene. Copies of the transgene expressing the mutant phenotype are sequenced to confirm the presence and type of the relevant mutation in the lacI gene and to identify clonal mutations which are counted as a single mutagenic event regardless of how many times they are detected in the same sample.
To determine if in utero exposure to vinclozolin leads to the direct induction of genetic mutations, and, thus, potentially to the induction of secondary epimutations, we exposed pregnant female Big Blue Rats carrying the lacI mutation-reporter transgene to vinclozolin as previously described [1,10] and compared the frequency of mutations in the lacI mutation-reporter transgene in kidney tissue and sperm recovered from F1 generation offspring of vinclozolin-treated or control (vehicle-treated) dams. In addition, we bred F1 generation offspring to generate both vinclozolin- and control-lineage F2 generation and subsequently F3 generation descendants as described [3,22], and analyzed the frequency of spontaneous mutations in kidney and sperm cells from the F3 generation progeny as well. Our results reveal the unexpected finding of a third type of epimutation which we term “tertiary epimutations.” Tertiary epimutations are distinct from primary or secondary epimutations and have the potential to contribute to epigenetic transgenerational inheritance.
Materials and Methods
The lacI mutation-reporter transgene (‘Big Blue’) system has been used extensively for the assessment of relative mutation frequencies in different animals, cell types and/or samples for nearly 25 years [23,24]. Big Blue rats homozygous for 2–4 copies of the lacI mutation-reporter transgene per genome  were obtained from Stratagene (La Jolla, CA). In the Skinner lab the Big Blue rats at 70 to 100 days of age were fed ad lib with a standard rat diet and ad lib tap water. To obtain timed-pregnant females, female rats in proestrus were pair-mated with male rats. Sperm-positive (day 0) rats were monitored for diestrus and body weight. On days 8–14 of gestation, the females were administered daily intraperitoneal injections of vinclozolin (100 mg/kg BW/day) or dimethyl sulfoxide (vehicle) as a control as described . Generally sibling females were used for the control and vinclozolin lineage F0 generation females, respectively. In addition, sibling F0 generation males were used to generate control and exposure lineage offspring. A minimum of three distinct lineages were used for each exposure. Vinclozolin was obtained from Chem Service Inc. (West Chester, PA), and was injected in a 200 microliter DMSO/sesame oil vehicle as previously described .
The initial gestating female rats and the males to which they were mated were designated as the F0 generation. Offspring of F0 generation matings formed the F1 generation. Non-sibling females and males aged 70–90 days from F1 generation control or vinclozolin lineages were bred to generate F2 generation offspring. F2 generation rats were then bred to obtain F3 generation offspring. Only the F0 generation gestating female was directly treated transiently with vinclozolin. Different F0 generation females were used for each different experiment and one male per litter was selected as the individual animal to be analyzed for each specific experiment. Control and vinclozolin lineage animals were housed in the same room and racks with identical lighting and food + water supply as previously described [10,27,28]. Samples for analysis were recovered from F1 and F3 generation vinclozolin-lineage and control-lineage male descendants to distinguish between potential immediate mutagenic effects (F1 generation samples) and an indirect, subsequent increase in mutation frequencies (F3 generation samples). To collect tissue or cell samples, rats were placed into a chamber and exposed to CO2 until breathing ceased, followed by cervical dislocation as a secondary euthanasia method.
Kidney and sperm were selected as the cell types to be analyzed because pathological defects have previously been documented in each of these cell types recovered from both F1 and F3 generation male descendants of dams exposed to vinclozolin [1–3,10]. In addition, this provided one example of a somatic tissue (kidney) and one example of germline cells (sperm) among the cell types to be analyzed for mutation frequencies. To obtain sperm samples, the epididymis was dissected free of connective tissue and a small cut was made in the cauda portion, which was then placed in 5 ml of F12 culture medium containing 0.1% bovine serum albumin for 10 minutes at 37°C to allow sperm to emerge from the tissue and then kept at 4°C to immobilize the sperm. The epididymal tissue was minced and the released sperm centrifuged at 13, 000 x g and stored in fresh nucleus isolation medium buffer at -70°C as described [10,27,28]. Kidneys were also dissected and snap frozen at -70°C. Each tissue or cell sample was shipped to the McCarrey lab on dry ice for further analysis (see below). All experimental protocols for the procedures with rats were pre-approved by the Washington State University Institutional Animal Care and Use Committee (IACUC approval # 02568–039), and carried out in accordance with relevant guidelines and regulations.
Epigenetic transgenerational phenotypes were not monitored in this study because the Big Blue rat model is maintained on the Fisher inbred rat genetic background which has previously been shown to repress the occurrence of such phenotypes , and because frequencies and spectra of mutations in the lacI reporter gene were assessed in animals at <1 year of age which is prior to the age at which transgenerational phenotypes typically become evident.
Analysis of mutation frequencies and spectra
High molecular weight DNA was prepared from each tissue or cell sample using the RecoverEase kit protocol from Stratagene with modifications as described [29,30,31]. Briefly, each sample was homogenized with a Dounce tissue homogenizer, filtered through a 100 μm filter (Millipore, Billerica, MA), subjected to digestion with ribonuclease (RNace-it, Stratagene) and proteinase K, then drop dialyzed against 1X TE using a 0.025 μm dialysis membrane (Millipore).
High molecular weight DNA recovered from each kidney or sperm sample was subjected to a phage packaging reaction using Transpack packaging extract (Stratagene) following the manufacturer’s protocol. Resulting infectious phage particles were plated onto a 25cm x 25cm lawn of SCS-8 E. coli cells to facilitate formation of phage plaques (up to 15,000 plaque forming units [pfu]/plate).
The lacI gene normally encodes a repressor of the lac operon that inhibits production of beta galactosidase by infected E. coli . However, point mutations incurred in the lacI gene while it was present as a transgene in the cells of the Big Blue rats can result in defective function of the lac repressor such that it fails to repress expression of the lac operon, resulting in production of beta galactosidase by E. coli infected with phage carrying a mutated lacI gene. The addition of X-gal and IPTG to the plating medium facilitated a blue/clear color selection method to distinguish rare mutant phage expressing beta galactosidase (blue plaques) from abundant wild type phage in which expression of beta galactosidase remained repressed (clear plaques) on each plate. Plaques displaying an apparent mutant (blue) phenotype were then picked and re-plated to confirm the mutant phenotype in ≥ 50% of the resulting phage. Plaques carrying an abundance of mutant phage were then prepared for Sanger sequencing of the lacI gene to confirm the presence of a mutation at the molecular level and to determine the position and type of mutation present. Sequencing was outsourced to the DNA Sequencing Facility at UT Austin.
DNA sequence data from each mutant plaque was used to determine the final frequency of spontaneous point mutations in each sample. Only confirmed independent (non-clonal) mutations were used to calculate the final frequency of mutations. Thus, if the same mutation in the same position within the lacI gene was detected more than once in the same DNA sample, it was assumed that this represented a single original mutation that was subsequently clonally amplified in the tissue or cell type, and such ‘clonal’ mutations were counted as a single mutation no matter how many times they were detected in a particular sample. Mutation frequencies were calculated by dividing the number of distinct, confirmed, non-clonal mutant plaques by the total number of pfu examined from each DNA sample. In general, sufficient total numbers of pfu were examined to identify at least five separate mutant plaques from each sample to facilitate statistical comparisons of the data from different samples. The DNA sequence data were also used to establish a mutation spectrum for each sample by cataloguing the relative prevalence of different types of point mutations detected in each sample, including base substitutions–transitions or transversions, small insertions, small deletions, or multiple base changes.
Numbers of mutations were analyzed by a Poisson model with parameter estimates obtained by the method of maximum likelihood . Because of the low expected frequencies, exact P-values were calculated by the exact conditional test for Poisson variables to compare differences between mutation frequencies, using the Exactci package implemented in R [34,35]. P ≤ 0.05 was considered statistically significant. Note that the extent of the standard error associated with each mutation frequency is dictated by the number of pfu examined and the number of independent mutations detected in those pfu.
Frequencies of point mutations detected in kidney and sperm cells from F1 generation offspring of dams exposed during pregnancy to either vehicle (DMSO) plus vinclozolin (vinclozolin-lineage animals) or vehicle alone (control-lineage animals) are shown in Table 1. No significant differences in frequencies of mutations were observed between vinclozolin- and control-lineage kidney or sperm samples, respectively, except for one vinclozolin-lineage sperm sample (marked with an asterisk in Fig 1A) in which we detected a mutation frequency that was significantly lower than the mean frequency of the F1 generation control-lineage sperm samples (p = 0.0352). Previous studies in the mouse revealed enhanced maintenance of genetic integrity in germline cells relative to that in differentiated somatic cells [31,36]. This distinction was not as robust between rat germ cells (sperm) and somatic cells (kidney). These data confirm previous reports that vinclozolin is not directly mutagenic , and suggest that neither the immediate phenotypic abnormalities observed in vinclozolin-lineage F1 generation offspring nor the transmission of those defects to the F2 generation can be ascribed to the direct induction of genetic mutations.
(A) Mutation frequencies in kidney and sperm samples from F1 generation control- and vinclozolin-lineage animals. There were no statistically significant differences among the mutation frequencies detected in kidney or sperm samples from F1 generation control- and vinclozolin-lineage samples, except for one of the vinclozolin-lineage sperm samples (F1VL–marked with an asterisk) which showed a mutation frequency that was significantly lower than the mean of the F1 generation control-lineage samples (p = 0.00352). (B) Mutation frequencies in kidney and sperm samples from F3 generation control- and vinclozolin-lineage animals. A subset of both kidney and sperm samples from F3 vinclozolin-lineage descendants showed mutation frequencies that were not significantly different than the mean of the corresponding F3 generation control-lineage samples, although several of the F3 generation vinclozolin-lineage samples trended higher than the mean of the corresponding F3 generation control-lineage samples. However another subset of both kidney and sperm samples from F3 generation vinclozolin-lineage descendants showed mutation frequencies that were significantly higher than the mean of the corresponding F3 control-lineage samples. These mutation frequencies are marked with astrices, and include those found in the following samples: F3VLK6 (p = 0.00342), F3VLK7 (p = 0.00131), F3VLK8 (p = 0.0222), F3VLS1 (p = 0.00185), F3VLS2 (p = 0.03611)) and F3VLS6 (p = 0.00018). F1 = samples from F1 generation descendants, F3 = samples from F3 generation descendants, CL = samples from control-lineage descendants, VL = samples from vinclozolin-lineage descendants, K = kidney samples, S = sperm samples.
Frequencies of mutations detected in kidney and sperm cells from vinclozolin- or control-lineage F3 generation descendants of dams exposed during pregnancy to either vehicle plus vinclozolin or vehicle alone, respectively, are shown in Table 2. Interestingly, for both the control- and vinclozolin-lineage animals the frequencies of mutations detected in the F3 generation were consistently higher than those detected in F1 generation animals. The F1 and F3 generation samples were not generated or analyzed simultaneously, however, the control and vinclozolin-exposed samples within each generation were generated and analyzed simultaneously. Therefore, the relevant comparison in each case is between control and vinclozolin-exposed samples within the same generation.
When compared to mutation frequencies detected in kidney and sperm samples from F3 generation control-lineage animals, one subset of the kidney samples (7 of 10) and one subset of the sperm samples (5 of 8) from F3 generation vinclozolin-lineage animals showed mutation frequencies that were not statistically different from the control values, although the mutation frequencies detected in some of these F3 vinclozolin-lineage samples trended higher than those in the F3 control-lineage animals (Fig 1B). However, another subset of kidney samples (3 of 10) and sperm samples (3 of 8) from F3 generation vinclozolin-lineage animals showed mutation frequencies that were significantly higher than the mean frequency in the corresponding control-lineage samples (marked with astrices in Fig 1B; p < 0.05). Indeed, mutation frequencies detected in these divergent samples were 2–13 fold higher than the mean mutation frequency detected among the corresponding F3 generation control-lineage samples. This suggests that in utero exposure to vinclozolin predisposes a subsequent accumulation of mutations at an accelerated rate in some descendants. This effect appears to occur in a gradual and stochastic manner, such that it is not detectable in any vinclozolin-lineage F1 generation offspring but is manifest in a subset of vinclozolin-lineage F3 generation descendants.
The lacI gene from each mutant (blue) plaque recovered from each cell or tissue sample was sequenced to confirm that a mutation had occurred in each case and to determine the range of types of mutations associated with each cell sample. A comprehensive list of the position and type of each mutation detected in the lacI reporter gene recovered from each set of tissues is shown in S1 Table, and a summary of the spectrum of mutations detected in each set of samples is shown in Table 3.
The discovery of environmentally induced germline epimutations that are transmitted transgenerationally for three or more generations following exposure of only the F1 generation fetus suggests these epimutations are not corrected by the extensive epigenetic reprogramming that normally occurs in the developing germ line [11,13]. Many environmentally induced epimutations have been reported to be manifest as abnormalities in locus- or allele-specific patterns of DNA methylation [38–42], while others appear to reflect changes in patterns of histone modifications . In particular, vinclozolin has been shown to induce aberrations in DNA methylation patterns in F1 generation offspring [1,39]. Thus, studies of genome-wide DNA methylation patterns in both somatic and germ cells in descendants of pregnant dams treated with vinclozolin or vehicle only (controls) have revealed the occurrence of abundant alterations to DNA methylation patterns (epimutations) throughout the genomes of vinclozolin-lineage, but not control-lineage descendants . DNA methylation epimutations were found on all chromosomes except the Y in F1-F3 generation vinclozolin-lineage descendants, but occurred most frequently in CpG-poor regions of the genome (CpG deserts) . These DNA methylation epimutations were evident in F1 generation offspring of vinclozolin-treated dams and persisted until the F3 generation and beyond . Specific characteristics of vinclozolin-induced transgenerational differential DNA methylation regions (DMRs) in sperm included the nonrandom occurrence of unique consensus DNA sequences including zinc finger motifs and G-quadruplex sequences [44,45]. A recent report also documented alterations in small noncoding RNAs (sncRNAs) in the sperm from F3 generation control and vinclozolin lineage rats .
However, inherited patterns of both DNA methylation and histone modifications are largely erased in the developing germ line and then reset so as to provide proper epigenetic programming to support development of the subsequent generation during embryonic, fetal and postnatal stages [11,47]. This germline-specific epigenetic reprogramming has the potential to correct environmentally induced epimutations, thus preventing their transmission to subsequent generations, and such a correction process has been documented in some cases . Nevertheless, a growing number of cases of transgenerational transmission of environmentally induced epimutations have now been reported [2–10], indicating that exposure of individuals to environmental factors or stressful circumstances can induce epimutations that are transmitted transgenerationally and thus, somehow, escape normal germline reprogramming . These germline epimutations appear to be sustained in a manner similar to imprinted genes , and have the potential to promote a novel etiology of various types of disease or abnormal phenotypes in the exposed individuals and in their descendants, even in the absence of any ongoing exposure to the causative agent [2,10]. Our observation that there were no significant differences in the spectra of mutations detected in F1 or F3 generation vinclozolin-lineage or control-lineage samples, respectively (Table 3), suggests that a variety of different DNA repair and/or cell death pathways were impacted by exposure to vinclozolin in a stochastic manner, thus leading to a general increase in the frequency of all types of point mutations.
The well documented alteration of DNA methylation patterns in cells of offspring exposed to vinclozolin noted above suggests that vinclozolin induces primary epimutations in the form of direct alterations of DNA methylation patterns that are subsequently propagated by epigenetic transgenerational inheritance. We tested the hypothesis that in utero exposure to vinclozolin might also induce genetic mutations that could contribute to secondary epimutations by which abnormalities in epigenetic programming result from alteration of one or more genetic functions normally required to establish such programming, because secondary epimutations would be expected to contribute to transgenerational transmission of the resulting phenotype on the basis of genetic inheritance. However, our data showing no difference in mutation frequencies between F1 generation vinclozolin- and control-lineage offspring corroborate previous reports that vinclozolin is not directly mutagenic . This argues against the likelihood that exposure to vinclozolin induces secondary epimutations, or any mutator effect , involving an initial genetic mutation that subsequently predisposes the accelerated accumulation of additional mutations.
We next investigated mutation frequencies in F3 generation vinclozolin- and control-lineage descendants of dams that were or were not exposed to vinclozolin, respectively. Surprisingly, we detected an elevated frequency of point mutations in a subset of F3 generation vinclozolin-lineage descendants that distinguished these animals from the F3 generation control-lineage animals and from the other F3 generation vinclozolin-lineage animals. Thus, our results suggest that in utero exposure to vinclozolin promotes the initial induction of primary epimutations, at least some of which have the potential to predispose an elevated accumulation of point mutations in subsequent generations, and that this could contribute to transmission of transgenerational phenotypes. We propose that this represents a third type of epimutation which we term a “tertiary epimutation” and define as an initial disruption of epigenetic programming that leads to a subsequent change in the frequency of genetic mutations.
The addition of tertiary epimutations to the list of possible scenarios by which heritable phenotypic alterations can be introduced and/or transmitted completes the options for potential combinations of genetic and epigenetic events involved in this process (Table 4). These include 1) disruption of genomic DNA sequence leading to genetic mutations transmittable by genetic inheritance, 2) disruption of epigenetic parameters leading directly to altered epigenetic programming (primary epimutations) transmittable by epigenetic inheritance, 3) disruption of genomic DNA sequence leading to subsequent aberrations in epigenetic programming (secondary epimutations) transmittable by either genetic or epigenetic inheritance, and 4) disruption of epigenetic programming leading to the subsequent accumulation of genetic mutations at an accelerated rate (tertiary epimutations) transmittable by either epigenetic or genetic inheritance.
Our finding that the frequency of point mutations is not increased in vinclozolin-lineage F1 generation offspring, but is increased in a subset of vinclozolin-lineage F3 descendants exemplifies the expected gradual inheritance pattern of a tertiary epimutation. These results are consistent with a recent report of an increase in copy number variations (CNV) in F3 generation, but not F1 generation descendants of vinclozolin-exposed dams , and corroborate previous reports suggesting that disruption of epigenetic programming can predispose multiple types of genetic aberrations [49–58].
Abnormalities in epigenetic programming could predispose genome instability through a variety of mechanisms. The most well documented example is predisposition of C-to-T transitions via deamination of methylated cytosines [31,49,50,57,59]. Thus, a change in the methylation status of a CpG dinucleotide, particularly the addition of DNA methylation where it normally does not occur, typically increases the likelihood that a C-to-T transition will change a G-C base pair into an A-T base pair. However, we did not observe a significant increase in C-to-T transitions in samples from F3 generation vinclozolin-lineage animals (Table 3), so this does not appear to be the primary source of the increased mutation frequencies we observed in these samples.
It is well established that epigenetic programming regulates chromatin structure [60,61] and that more or less condensed chromatin appears less or more susceptible, respectively, to either spontaneous or induced mutagenesis [62,63]. We have no data describing the extent of chromatin condensation associated with the lacI mutation-reporter transgene in kidney or sperm cells from F3 generation vinclozolin-lineage animals, so cannot comment on whether or not this source of accelerated mutagenesis may have contributed to the elevated mutation frequencies in a subset of F3 generation vinclozolin-lineage animals.
Finally, it has been shown that specific cellular states, such as those that distinguish pluripotent and differentiated cell types, or germ and somatic cell types are correlated with more or less enhanced maintenance of genetic integrity, and that this is at least partially controlled by epigenetic mechanisms that coordinate the function of different gene networks in cells that regulate pluripotency and genetic integrity [64,65]. To the extent that this type of coordinate regulation normally governs the relative level of expression of genetic integrity (DNA repair or cell death) genes, disruptions of epigenetic programming could alter maintenance of genetic integrity resulting in the subsequent accumulation of point mutations at an accelerated rate. This would only occur if regulation of a relevant gene network was impacted by the initial epigenetic disruption, which would normally occur stochastically, such that changes in mutation frequencies would be expected to occur in only a subset of descendants of animals exposed to the initial disruptor. Our observation that only a subset of F3 vinclozolin-lineage descendants displayed significant increases in mutation frequency is consistent with this notion. In addition, a multiple step process of this sort (initial disruption of the epigenome → dysregulation of a genetic integrity gene network → an increase in the rate at which spontaneous point mutations are accumulated) would likely lead to a gradual increase in mutation frequency. This is consistent with our observation of a detectable increase in mutation frequency in F3 generation, but not in F1 generation vinclozolin-lineage descendants.
Taken together, our results support a novel mechanism by which initial epimutations can predispose a subsequent accelerated accumulation of genetic mutations, and such tertiary epimutations could contribute to transgenerational transmission of defective phenotypes on the basis of either epigenetic or genetic inheritance, or both. As summarized in Table 4, tertiary epimutations complete the list of possible combinations of epigenetic and genetic effects that can contribute to the initial occurrence and subsequent transgenerational transmission of heritable phenotypic alterations. It is likely that various combinations of genetic and epigenetic mechanisms of transmission are involved in the propagation of many epimutations. Therefore future studies should include comprehensive genome-wide analyses of both mechanisms.
JRM is the Kleberg Distinguished University Chair in Cellular & Molecular Biology at UTSA. MKS is the Eastlick Distinguished Professor at WSU. We acknowledge the assistance of Ms. Heather Johnson in preparation of the manuscript. The research was supported by a grant from the Kleberg Foundation to JRM, an NIH grant to MKS and a John Templeton Foundation grant to MKS and JRM.
- Conceptualization: JRM MKS.
- Formal analysis: YW.
- Funding acquisition: JRM MKS.
- Investigation: JDL SSR.
- Methodology: JRM MKS.
- Project administration: JRM MKS.
- Resources: EEN MKS.
- Supervision: JRM MKS.
- Visualization: JRM.
- Writing – original draft: JRM.
- Writing – review & editing: JRM MKS.
- 1. Anway MD, Cupp AS, Uzumcu M, Skinner MK. Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science. 2005; 308:1466–1469. pmid:15933200
- 2. Nilsson EE, Skinner MK. Environmentally Induced Epigenetic Transgenerational Inheritance of Reproductive Disease. Biology of Reproduction. 2015; 93:145. pmid:26510870
- 3. Skinner MK. Endocrine disruptor induction of epigenetic transgenerational inheritance of disease. Molecular and Cellular Endocrinology. 2014; 398: 4–12. pmid:25088466
- 4. Xing Y, Shi S, Le L, Lee CA, Silver-Morse L, Li WX. Evidence for transgenerational transmission of epigenetic tumor susceptibility in Drosophila. 2007; 3:1598–1606. pmid:17845077
- 5. Kelly WG. Transgenerational epigenetics in the germline cycle of Caenorhabditis elegans. Epigenetics Chromatin. 2014; 7:6. pmid:24678826
- 6. Bhandari RK, Deem SL, Holliday DK, Jandegian CM, Kassotis CD, Nagel SC, et al. Effects of the environmental estrogenic contaminants bisphenol A and 17α-ethinyl estradiol on sexual development and adult behaviors in aquatic wildlife species. Gen Comp Endocrinol. 2015; 214:195–219. pmid:25277515
- 7. Brun JM, Bernadet MD, Cornuez A, Leroux S, Bodin L, Basso B, et al. Influence of grand-mother diet on offspring performances through the male line in Muscovy duck. BMC Genet. 2015; 16:145. pmid:26690963
- 8. Braunschweig M, Jagannathan V, Gutzwiller A, Bee G. Investigations on transgenerational epigenetic response down the male line in F2 pigs. Plos One. 2012; 7:e30583. pmid:22359544
- 9. Kaati G, Bygren LO, Pembrey M, Sjostrom M. Transgenerational response to nutrition, early life circumstances and longevity. Eur J Hum Genet. 2007; 15:784–790. pmid:17457370
- 10. Skinner MK, Manikkam M, Guerrero-Bosagna C. Epigenetic transgenerational actions of environmental factors in disease etiology. Trends Endocrinol Metab. 2010; 21: 214–222. pmid:20074974
- 11. Lee HJ, Hore TA, Reik, W. Reprogramming the methylome: erasing memory and creating diversity. Cell Stem Cell. 2014; 14:710–719. pmid:24905162
- 12. McCarrey JR. The epigenome as a target for heritable environmental disruptions of cellular function. Molecular and Cellular Endocrinology. 2012; 354:9–15. pmid:21970811
- 13. McCarrey JR. Distinctions between transgenerational and non-transgenerational epimutations. Molecular and Cellular Endocrinology. 2014; 398:13–23. pmid:25079508
- 14. McCarrey JR. The epigenome—a family affair. Science. 2015; 350:634–635. pmid:26542557
- 15. Skinner MK. What is an epigenetic transgenerational phenotype? F3 or F2. Reproductive Toxicology. 2008; 25:2–6. pmid:17949945
- 16. Rando OJ. Intergenerational Transfer of Epigenetic Information in Sperm. Cold Spring Harb Perspect Med. 2016;
- 17. Skinner MK. Endocrine disruptors in 2015: Epigenetic transgenerational inheritance. Nat Rev Endocrinol. 2016; 12, 68–70. pmid:26585656
- 18. de Waal E, Yamazaki Y, Ingale P, Bartolomei MS, Yanagimachi R, McCarrey JR. Primary epimutations introduced during intracytoplasmic sperm injection (ICSI) are corrected by germline-specific epigenetic reprogramming. Proceedings of the National Academy of Sciences of the United States of America. 2012; 109, 4163–4168. pmid:22371603
- 19. Whitelaw NC, Whitelaw E. Transgenerational epigenetic inheritance in health and disease. Curr Opin Genet Dev. 2008; 18:273–279. pmid:18662779
- 20. Kohler SW, Provost GS, Fieck A, Kretz PL, Bullock WO, Putman DL, et al. Analysis of spontaneous and induced mutations in transgenic mice using a lambda ZAP/lacI shuttle vector. Environmental and Molecular Mutagenesis 1991; 18, 316–321. pmid:1836179
- 21. Kohler SW, Provost GS, Fieck A, Kretz PL, Bullock WO, Sorge JA, et al. Spectra of spontaneous and mutagen-induced mutations in the lacI gene in transgenic mice. Proceedings of the National Academy of Sciences of the United States of America 88, 7958–7962 (1991). pmid:1832771
- 22. Guerrero-Bosagna C, Covert TR, Haque MM, Settles M, Nilsson EE, Anway MD, et al. Epigenetic Transgenerational Inheritance of Vinclozolin Induced Mouse Adult Onset Disease and Associated Sperm Epigenome Biomarkers. Reproductive Toxicology. 2012; 34:694–707. pmid:23041264
- 23. Provost GS, Kretz PL, Hammer RT, Matthews CD, Rogers BJ, Lundberg KS, et al. Transgenic systems for in vivo mutation analysis. Mutat Res 1993; 288:133–149 pmid:7686257
- 24. Valentine CR, Delongchamp RR, Pearce MG, Rainey HF, Dobrovolsky VN, Malling HV, et al. In vivo mutation analysis using the phiX174 transgenic mouse and comparisons with other transgenes and endogenous genes. Mutat Res; 2010 705:205–216. pmid:20637298
- 25. Dycaico MJ, Provost GS, Kretz PL, Ransom SL, Moores JC, Short JM. The use of shuttle vectors for mutation analysis in transgenic mice and rats. Mutat Res 1994; 307:461–478. pmid:7514720
- 26. Nilsson E. E., Anway M. D., Stanfield J. & Skinner M. K. Transgenerational epigenetic effects of the endocrine disruptor vinclozolin on pregnancies and female adult onset disease. Reproduction. 2008; 135:713–721. pmid:18304984
- 27. Manikkam M, Guerrero-Bosagna C, Tracey R, Haque MM, Skinner MK. Transgenerational actions of environmental compounds on reproductive disease and identification of epigenetic biomarkers of ancestral exposures. PLoS One. 2012; 7:e31901. pmid:22389676
- 28. Anway MD, Leathers C, Skinner MK. Endocrine disruptor vinclozolin induced epigenetic transgenerational adult-onset disease. Endocrinology. 2006; 147:5515–5523. pmid:16973726
- 29. Caperton L, Murphey P, Yamazaki Y, McMahan CA, Walter CA, Yanagimachi R, et al. Assisted reproductive technologies do not alter mutation frequency or spectrum. Proceedings of the National Academy of Sciences of the United States of America. 2007; 104:5085–5090. pmid:17360354
- 30. Murphey P, Yamazaki Y, McMahan CA, Walter CA, Yanagimachi R, McCarrey JR. Epigenetic regulation of genetic integrity is reprogrammed during cloning. Proceedings of the National Academy of Sciences of the United States of America. 2009; 106:4731–4735. pmid:19255429
- 31. Murphey P, McLean DJ, McMahan CA, Walter CA, McCarrey JR. Enhanced genetic integrity in mouse germ cells. Biology of Reproduction. 2013; 88:6 pmid:23153565
- 32. Muller-Hill B. Lac repressor and lac operator. Progress in Biophysics and Molecular Biology. 1975; 30:227–252. pmid:792953
- 33. Agresti A. Categorical Data Analysis. 2002; Wiley, New York.
- 34. Fay M. Two-sided Exact Tests and Matching Confidence Intervals for Discrete Data. R Journal. 2010;2/1, 53–58.
- 35. Hirji K. Exact Analysis of Discrete Data. 2006; Chapman and Hall/CRC. New York. ISBN: 142003619X, 9781420036190.
- 36. Walter CA, Intano GW, McCarrey JR, McMahan CA, Walter RB. Mutation frequency declines during spermatogenesis in young mice but increases in old mice. Proceedings of the National Academy of Sciences of the United States of America. 1998; 95:10015–10019. pmid:9707592
- 37. Hrelia P, Fimognari C, Maffei F, Vigagni F, Mesira R, Pozzetti L, et al. The genetic and non-genetic toxicity of the fungicide Vinclozolin. Mutagenesis. 1996; 11:445–453. pmid:8921505
- 38. Jirtle RL, Skinner MK. Environmental epigenomics and disease susceptibility. Nature Reviews Genetics. 2007; 8:253–262. pmid:17363974
- 39. Skinner MK, Guerrero-Bosagna C. Role of CpG Deserts in the Epigenetic Transgenerational Inheritance of Differential DNA Methylation Regions. BMC Genomics. 2014; 15,
- 40. Youngson NA, Lecomte V, Maloney CA, Leung P, Liu J, Hesson LB, et al. Obesity-induced sperm DNA methylation changes at satellite repeats are reprogrammed in rat offspring. Asian J Androl. 2015;
- 41. Ma J, Chen X, Liu Y, Xie Q, Sun Y, Chen J, et al. Ancestral TCDD exposure promotes epigenetic transgenerational inheritance of imprinted gene Igf2: Methylation status and DNMTs. Toxicology and Applied Pharmacology. 2015; 289:193–202. pmid:26455773
- 42. Blaze J, Roth TL. Evidence from clinical and animal model studies of the long-term and transgenerational impact of stress on DNA methylation. Semin Cell Dev Biol. 2015; 43:76–84. pmid:25917771
- 43. Siklenka K, Erkek S, Godmann M, Lambrot R, McGraw S, Lafleur C, et al. Disruption of histone methylation in developing sperm impairs offspring health transgenerationally. Science. 2015; 350:
- 44. Haque MM, Holder LB, Skinner MK. Genome-wide locations of potential epimutations associated with environmentally induced epigenetic transgenerational inheritance of disease using a sequential machine learning prediction approach. PLoS One. 2015;
- 45. Guerrero-Bosagna C, Weeks S, Skinner MK. Identification of genomic geatures in environmentally induced epigenetic transgenerational inherited sperm epimutations. PLoS One. 2014;
- 46. Schuster A, Skinner MK, Yan W. Ancestral vinclozolin exposure alters the epigenetic transgenerational inheritance of sperm small noncoding RNAs. Env. Epigenetics. 2016;
- 47. Peters AH, Schubeler D. Methylation of histones: playing memory with DNA. Current Opinion in Cell Biology. 2005; 17, 230–238. pmid:15780602
- 48. Skinner MK, Guerrero-Bosagna C, Haque MM. Environmentally Induced Epigenetic Transgenerational Inheritance of Sperm Epimutations Promote Genetic Mutations. Epigenetics. 2015; 10:762–771. pmid:26237076
- 49. Sved J, Bird A. The expected equilibrium of the CpG dinucleotide in vertebrate genomes under a mutation model. Proceedings of the National Academy of Sciences of the United States of America. 1990; 87:4692–4696. pmid:2352943
- 50. Lutsenko E, Bhagwat AS. Principal causes of hot spots for cytosine to thymine mutations at sites of cytosine methylation in growing cells. A model, its experimental support and implications. Mutat Res. 1999; 437:11–20. pmid:10425387
- 51. Becker C, Hagmann J, Muller J, Koenig D, Stegle O, Borgwardt K, et al. Spontaneous epigenetic variation in the Arabidopsis thaliana methylome. Nature. 2011; 480:245–249. pmid:22057020
- 52. Tang MH, Varadan V, Kamalakaran S, Zhang MQ, Dimitrova N, Hicks J. Major chromosomal breakpoint intervals in breast cancer co-localize with differentially methylated regions. Front Oncol. 2012; 2:197. pmid:23293768
- 53. Macia A, Munoz-Lopez M, Cortes JL, Hastings RK, Morell S, Lucena-Aguilar G, et al. Epigenetic control of retrotransposon expression in human embryonic stem cells. Molecular and Cellular Biology. 2011; 31:300–316. pmid:21041477
- 54. Feinberg AP. The epigenetics of cancer etiology. Semin Cancer Biol. 2004; 14:427–432. pmid:15489135
- 55. You JS, Jones PA. Cancer genetics and epigenetics: two sides of the same coin? Cancer Cell. 2012; 22, 9–20. pmid:22789535
- 56. Almal SH, Padh H. Implications of gene copy-number variation in health and diseases. J Hum Genet. 2012; 57, 6–13. pmid:21956041
- 57. Wu J, Grzeda KR, Stewart C, Grubert F, Urban AE, Snyder , et al. Copy Number Variation detection from 1000 Genomes Project exon capture sequencing data. BMC Bioinformatics. 2012; 13:
- 58. Miyake K, Yang C, Minakuchi Y, Ohori K, Soutome M, Hirasawa T, et al. Comparison of Genomic and Epigenomic Expression in Monozygotic Twins Discordant for Rett Syndrome. PloS One. 2013; 8:e66729, pmid:23805272
- 59. Collins DW, Jukes TH. Rates of transition and transversion in coding sequences since the human-rodent divergence. Genomics. 1994; 20:386–396. pmid:8034311
- 60. Rea S, Eisenhaber F, O'Carroll D, Strahl BD, Sun ZW, Schmid M, et al. Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature. 2000; 406:593–599. pmid:10949293
- 61. Hashimshony T, Zhang J, Keshet I, Bustin M, Cedar H. The role of DNA methylation in setting up chromatin structure during development. Nat Genet. 34, 2003:187–192. pmid:12740577
- 62. Smerdon MJ. DNA repair and the role of chromatin structure. Current Opinion in Cell Biology. 1991; 3:422–428. pmid:1892653
- 63. Friedberg EC, Walker GC, Siede W, Schultz RA. DNA Repair and Mutagenesis (2nd ed.). 2006; ASM Press, Washington DC. ISBN 978-1-55581-319-2.
- 64. Cooper DJ, Walter CA, McCarrey JR. Co-regulation of pluripotency and genetic integrity at the genomic level. Stem Cell Res. 2014; 13:508–519. pmid:25451711
- 65. Chen I-C, Hernandez C, Xu X, Cooney A, Wang Y, McCarrey JR. Dynamic variations in genetic integrity accompany changes in cell fate. Stem Cells Dev. 2016; 25:1698–1708. pmid:27627671