Archive for the ‘Vinclozolin (Fungicide)’ Category

“The Toxins That Affected Your Great-Grandparents Could Be In Your Genes”

Biologist Michael Skinner has enraged the chemical community and shocked his peers with his breakthrough research

By Jeneen Interlandi for Smithsonian magazine, December 2013

Michael Skinner’s biggest discovery began, as often happens in science stories like this one, with a brilliant failure. Back in 2005, when he was still a traditional developmental biologist and the accolades and attacks were still in the future, a distraught research fellow went to his office to apologize for taking an experiment one step too far. In his laboratories at Washington State University, she and Skinner had exposed pregnant rats to an endocrine disruptor—a chemical known to interfere with fetal development—in the hope of disturbing (and thereby gaining more insight into) the process by which an unborn fetus becomes either male or female. But the chemical they used, an agricultural fungicide called vinclozolin, had not affected sexual differentiation after all. The scientists did find lower sperm counts and decreased fertility when the male offspring reached adulthood, but that was no surprise. The study seemed like a bust.

By accident, though, Skinner’s colleague had bred the grandchildren of those exposed rats, creating a fourth generation, or the great-grandchildren of the original subjects. “It’s OK,” Skinner told her. “You might as well analyze them.” If nothing else, he thought, the exercise might take her mind off her mistake. So she went ahead and studied the rats’ testes under a microscope.

What they found would not only change the direction of Skinner’s research but also challenge a bedrock principle of modern biology. And Skinner would become the forerunner of a new way of thinking about the possible long-term health consequences of exposure to environmental chemicals.

His discoveries touch on the basic question of how biological instructions are transmitted from one generation to the next. For half a century it has been common knowledge that the genetic material DNA controls this process; the “letters” in the DNA strand spell out messages that are passed from parent to offspring and so on. The messages come in the form of genes, the molecular equivalent of sentences, but they are not permanent. A change in a letter, a result of a random mutation, for example, can alter a gene’s message. The altered message can then be transmitted instead.

The strange thing about Skinner’s lab rats was that three generations after the pregnant mothers were exposed to the fungicide, the animals had abnormally low sperm counts—but not because of a change in their inherited DNA sequence. Puzzled, Skinner and his team repeated the experiments—once, twice, 15 times—and found the same sperm defects. So they bred more rats, and tested more chemicals, including substances that lead to diseases in the prostate, kidney, ovaries and immune system. Again and again, these diseases also showed up in the fourth- and fifth-generation offspring of mothers exposed to a chemical.

“In essence,” Skinner explains, “what your great-grandmother was exposed to could cause disease in you and your grandchildren.”

And, startlingly, whatever disease pathway a chemical was opening in the rats’ fur-covered bodies, it did not begin or end at a mutation in the genetic code. Skinner and his team found instead that as the toxins flooded in, they altered the pattern of simple molecules called methyl groups that latch onto DNA in the fetus’ germ-line cells, which would eventually become its eggs or sperm. Like burrs stuck to a knit sweater, these methyl molecules interfered with the functioning of the DNA and rode it down through future generations, opening each new one to the same diseases. These burrs, known to be involved in development, persisted for generations. The phenomenon was so unexpected that it has given rise to a new field, with Skinner an acknowledged leader, named transgenerational epigenetics, or the study of inherited changes that can’t be explained by traditional genetics.

A study by Skinner and colleagues published last year in the journal PLOS One has upped the ante considerably. The burrs were not just haphazardly attached, Skinner found. Instead, they fastened themselves in particular arrangements. When he bathed the insides of his pregnant rats in bug spray, jet fuel and BPA, the plastics component recently banned from baby bottles, each exposure left a distinct pattern of methyl group attachments that persisted in the great-grandchildren of exposed rats.

Not only is your great-grandmother’s environment affecting your health, Skinner concluded, but the chemicals she was exposed to may have left a fingerprint that scientists can actually trace.

The findings point to potentially new medical diagnostics. In the future, you may even go to your doctor’s office to have your methylation patterns screened. Exposure of lab rats to the chemical DDT can lead to obesity in subsequent generations—a link Skinner’s team reported in October. Hypothetically, a doctor might someday look at your methylation patterns early in life to determine your risk for obesity later. What’s more, toxicologists may need to reconsider how they study chemical exposures, especially those occurring during pregnancy. The work raises implications for monitoring the environment, for determining the safety of certain chemicals, perhaps even for establishing liability in legal cases involving health risks of chemical exposure.

These possibilities have not been lost on regulators, industries, scientists and others who have a stake in such matters. “There are two forces working against me,” Skinner says. “On one side, you have moneyed interests refusing to accept data that might force stronger regulations of their most profitable chemicals. On the other side, you have genetic determinists clinging to an old paradigm.”

Michael Skinner wears a gray Stetson with a tan strap, and leans back easily in his chair in his office on the Pullman campus. His fly-fishing rod stands in the corner, and a colossal northern pike is mounted on the wall. An avid fly fisherman, Skinner, age 57, was born and raised on the Umatilla Indian Reservation in eastern Oregon. The Skinners are not of Indian descent, but his parents owned a family farm there—“a good cultural experience,” he says. His father worked in insurance, and he and his four brothers grew up just as five generations of Skinners had before them—hunting and fishing and cowboying, learning a way of life that would sustain them into adulthood.

He loved the outdoors, and his fascination with how nature worked prompted a school guidance counselor’s suggestion that a career in science might be just the thing. He was about 12, and true to form he stuck with it. In high school and then at Reed College he wrestled competitively, and today his supporters and critics alike may detect a bit of his old grappling self in how he approaches a problem—head-on. “It probably taught me how to confront, rather than avoid challenges,” he says now. The sport also led him to his future wife, Roberta McMaster, or Bobbie, who served as his high-school wrestling team’s scorekeeper. “I was fascinated that someone so young knew exactly what he wanted to do with his life,” Bobbie recalls. He proposed marriage before heading for college, and the two have been together ever since and have two grown children.

He attended Washington State University for his PhD in biochemistry, and during that time he and Bobbie often lived on game that he’d hunted. It was not unheard of to find a freshly killed deer hanging in the carport of their student housing. “They were lean years,” Bobbie says. “But they were good ones.”

After positions at Vanderbilt and the University of California, San Francisco, Skinner returned to Washington State University. “I wanted a big research college in a rural town,” he says. He spent the next decade studying how genes turn on and off in ovaries and testes, and how those organs’ cells interact. He wasn’t aiming to take on the central idea in biology for much of the 20th century: genetic determinism, the belief that DNA is the sole blueprint for traits from hair and eye color to athletic ability, personality type and disease risk.

In some sense this interpretation of genetic determinism was always oversimplified. Scientists have long understood that environments shape us in mysterious ways, that nature and nurture are not opposing forces so much as collaborators in the great art of human-making. The environment, for example, can ramp up and pull back on gene activity through methyl groups, as well as a host of other molecules that modify and mark up a person’s full complement of DNA, called the genome. But only changes in the DNA sequence itself were normally passed to offspring.

So certain was everyone of this basic principle that President Bill Clinton praised the effort to complete the first full reading of the human genome, saying in June 2000 that this achievement would “revolutionize the diagnosis, prevention and treatment of most, if not all human diseases.” When stacked against such enthusiasm, Skinner’s findings have felt like heresy. And for a while, at least, he was criticized accordingly.


Critics of the Skinner-led research pointed out that the doses of vinclozolin in his rat studies were way too high to be relevant to human exposure, and injecting the rats as opposed to administering the toxins through their food exaggerated the effects. “What he’s doing doesn’t have any real obvious implications for the risk assessments on the chemical,” EPA toxicologist L. Earl Gray was quoted telling Pacific Standard magazine back in 2009. Until the results are replicated, “I’m not sure they even demonstrate basic science principles.”

Skinner responds to assaults on his data by saying that risk assessment, of the type that toxicologists do, has not been his goal. Rather, he’s interested in uncovering new biological mechanisms that control growth, development and inheritance. “My approach is basically to hit it with a hammer and see what kind of response we get,” he says. He remains calm, even when called on to defend that approach. “Conflicts with individuals solve very little,” he says. “The best way to handle these things is to let the science speak for itself.”

That science has received a lot of attention (the vinclozolin study has been cited in the scientific literature more than 800 times). Recently, the journal Nature Reviews Genetics asked five leading researchers to share their views on the importance of epigenetic inheritance. A “mixture of excitement and caution,” is how the editors described the responses, with one researcher arguing that the phenomenon was “the best candidate” for explaining at least some transgenerational effects, and another noting that it might, if fully documented, have “profound implications for how we consider inheritance, for mechanisms underlying diseases and for phenotypes that are regulated by gene-environment interactions.”

Though most of Skinner’s critics have been reassured by new data from his lab and others, he says he still feels embattled. “I really try to be a scientist first and foremost,” he says. “I’m not a toxicologist, or even an environmentalist. I didn’t come to this as an advocate for or against any particular chemical or policy. I found something in the data, and I pursued it along a logical path, the way any basic researcher would.”

Read more: http://www.smithsonianmag.com/ideas-innovations/The-Toxins-That-Affected-Your-Great-Grandparents-Could-Be-In-Your-Genes-231152741.html#ixzz2mIaKLsRH
Follow us: @SmithsonianMag on Twitter

Read Full Post »

Read Full Post »

Chronic Dietary Exposure to a Low-Dose Mixture of Genistein and Vinclozolin Modifies the Reproductive Axis, Testis Transcriptome, and Fertility

Florence Eustache,1* Françoise Mondon,2,3,4 Marie Chantal Canivenc-Lavier,5 Corinne Lesaffre,2,3,4 Yvonne Fulla,6 Raymond Berges,5 Jean Pierre Cravedi,7 Daniel Vaiman,2,3,4,8 and Jacques Auger1

Environ Health Perspect. 2009 August; 117(8): 1272–1279.
Published online 2009 April 1. doi: 10.1289/ehp.0800158.



The reproductive consequences and mechanisms of action of chronic exposure to low-dose endocrine disruptors are poorly understood.


We assessed the effects of a continuous, low-dose exposure to a phytoestrogen (genistein) and/or an antiandrogenic food contaminant (vinclozolin) on the male reproductive tract and fertility.


Male rats were exposed by gavage to genistein and vinclozolin from conception to adulthood, alone or in combination, at low doses (1 mg/kg/day) or higher doses (10 and 30 mg/kg/day). We studied a number of standard reproductive toxicology end points and also assessed testicular mRNA expression profiles using long-oligonucleotide microarrays.


The low-dose mixture and high-dose vinclozolin produced the most significant alterations in adults: decreased sperm counts, reduced sperm motion parameters, decreased litter sizes, and increased post implantation loss. Testicular mRNA expression profiles for these exposure conditions were strongly correlated. Functional clustering indicated that many of the genes induced belong to the “neuroactive ligand-receptor interactions” family encompassing several hormonally related actors (e.g., follicle-stimulating hormone and its receptor). All exposure conditions decreased the levels of mRNAs involved in ribosome function, indicating probable decreased protein production.


Our study shows that chronic exposure to a mixture of a dose of a phytoestrogen equivalent to that in the human diet and a low dose—albeit not environmental—of a common anti-androgenic food contaminant may seriously affect the male reproductive tract and fertility.

Estrogenic and antiandrogenic endocrine-disrupting compounds (EDCs) cause a wide spectrum of developmental and fertility detrimental effects [for review, see Sharpe (2003) and Gray et al. (2001), respectively]. Most studies have used high doses of a single compound and short exposure periods, generally during the critical uterine or neonatal period. However, humans and wild animals are exposed simultaneously to various environmental and food EDCs, generally at low levels, throughout their lives. Therefore, it would be valuable to determine the effects of a chronic exposure to low doses of EDCs on the reproductive axis and to identify the mechanisms involved. In particular, it is not known whether lifetime exposures to low (environmental) doses of estrogenic “feminizing” and antiandrogenic “demasculinizing” EDCs can have adverse effects on male reproductive function of the same magnitude as those of acute exposure to high nonenvironmental doses of these compounds in isolation.
In this study, we used a rat model of prolonged exposure to EDCs by gavage to determine the effects on male reproduction of two EDCs that may be associated in the human diet: the phytoestrogen genistein and the antiandrogenic fungicide vinclozolin. Genistein is an estrogenic isoflavonoid found in leguminous plants (Breinholt et al. 2000). It is particularly abundant in diets containing soya or soya-derived products, leading to a dietary exposure of up to 2 mg/kg body weight/day; for infants fed milk formulas containing soya, dietary exposure can reach 1 mg/kg body weight/day (Setchell et al. 1998). In previous studies involving transient gestational, lactational, or adult intakes (Santti et al. 1998), rodents have been exposed to various doses of genistein but the findings are equivocal: some report male reproductive anomalies (Wisniewski et al. 2003), whereas other do not (Faqi et al. 2004; Jung et al. 2004). Human exposure to genistein may affect the responsiveness and sensitivity to other xenobiotics, particularly environmental estrogenic chemicals (You et al. 2002) and other endocrine-active dietary contaminants. Vinclozolin, a dicarboximide fungicide extensively used on fruit and vegetables, is recognized as a human diet contaminant acting—essentially through its two main metabolites M1 and M2—as an androgen-receptor (AR) binding antagonist (Kelce et al. 1994, 1997; Nellemann et al. 2003). A recent French study reported that 20% of 139 meal samples from work canteens contained measurable levels of vinclozolin (Leblanc et al. 2000). In addition, assays of metabolites in urine revealed that > 80% of a population in central Italy was exposed to noticeable levels of vinclozolin and similar pesticides (Turci et al. 2006). Vinclozolin administered to experimental animals in vivo at various doses, by various routes, and for exposure periods (gestation, lactation, puberty, adulthood) produces a wide spectrum of reproductive defects: reduced anogenital distance (AGD); persistent nipples; cleft phallus; hypospadias; cryptorchidism; reduced weights of the ventral prostate, seminal vesicles, and epididymis; and reduced sperm counts (Gray et al. 1999; Monosson et al. 1999; Yu et al. 2004). It is highly plausible that vinclozolin can induce such anomalies of the reproductive tract in humans (Kavlock and Cummings 2005). However, most studies used doses 100 times the U.S. Environmental Protection Agency (EPA) no observed adverse effect level (NOAEL) of 1.2 mg/kg body weight/day based on a combination of chronic toxicity, carcinogenicity, and reproductive toxicity in rats (U.S. EPA 2003).
To our knowledge, only one recent study has investigated the reproductive consequences (the frequency of hypospadias) of in utero exposure to both genistein and vinclozolin (Vilela et al. 2007). Using a lifelong exposure scheme, we found significant alterations of reproductive development and impairment of several fertility end points by these compounds, the most severe effect resulting from combined exposure to a dietary level of genistein and a level of vinclozolin lower than the U.S. EPA-proposed NOAEL. In addition, we found that mRNA expression profiles in the adult testis are notably and differentially modified according to the exposure protocol. We also describe functional clustering of the genes affected into ontologic families.



Genistein with a purity of 99% was synthesized at the Laboratoire de Chimie Organique et Organométallique (Université Bordeaux 1, Talence, France). We extracted vinclozolin from the commercial formulation Ronilan (BASF France, Levallois-Perret, France) according to Bursztyka et al. (2008). The extract was dried under vacuum and then recrystallized from methanol. Vinclozolin has a melting point of 108–109°C and its purity, as verified by HPLC-diode-array detection (from 192 to 400 nm) and gas chromatography/mass spectrometry analyses, was > 96% (data not shown). In addition, we tested the absence of the degradation products M1 and M2 by liquid chromatography/mass spectrometry as previously described (Bursztyka et al. 2008).

Doses used

The exposure scheme consisted of a “high” and a “low” dose for each compound, and the corresponding combinations. The high doses we used were higher than the reported NOAEL of vinclozolin and the plausible levels of genistein in the human diet; we chose these doses to be sufficiently low to maintain normal growth, as well as food and water intake. The genistein high dose, 10 mg/kg body weight (G10), was greater than the genistein levels found in human diets in Southeast Asia (Tanaka et al. 2008) and was several times lower than the doses used in some reproductive studies; the low dose, 1 mg/kg body weight (G1), was similar to that in soya-based diets (Tanaka et al. 2008). The vinclozolin high dose, 30 mg/kg body weight (V30), was substantially greater than real-life exposure levels but was 3–10 times lower than the doses used in several male reproductive studies (Gray et al. 2001). The acceptable daily intake (ADI) of vinclozolin is 600 μg/day/person, corresponding to an exposure of 0.01 mg/kg body weight/day (Food and Agriculture Organization/World Health Organization 1998). In France, the estimated daily intake is 3.3 μg/kg/person (Leblanc et al. 2000), which is < 1% of the ADI. The low vinclozolin dose used in the present study, 1 mg/kg body weight (V1), was higher than human food contamination levels; however, it was lower than the NOAEL combining chronic toxicity, carcinogenicity, and reproductive toxicity in rats (1.2 mg/kg body weight/day; U.S. EPA 2003).

Animals and regimens

A total of 80 specific-pathogen free (SPF) female and 32 SPF male Wistar Han rats at 8 weeks of age were purchased from Harlan France Sarl (Gannat, France) and fed with a soy-free diet (Harlan Teklad, Gannat, France). On their arrival, they were acclimatized to the animal facility conditions (22°C with constant humidity and a 12-hr light/dark period) for 4 weeks before mating. All animals were treated humanely and with regard for alleviation of suffering. They were maintained in accordance with the French Ministry of Agriculture guidelines for care and use of laboratory animals. From the acclimatization period, animals were fed a purified diet of pellets (18% casein, 46% starch, 23% sucrose, 5% corn oil, 2% cellulose, 5% mineral mixture, and 1% vitamin mixture) and supplied with water ad libitum. dissolved in corn oil (Carrefour, Dijon, France) and administered by gavage (2 mL/kg body weight). Control animals were treated with the vehicle alone. Animals were identified by a unique ear tag with a randomly assigned number.

Exposure conditions

We randomly assigned 10 females to each treatment group and mated them with male rats (five females and two males per cage) with the aim of obtaining at least 20 adult male offspring per treatment group for subsequent data comparisons and analyses. We examined females daily and caged them separately when a vaginal sperm plug was observed [defined as gestational day (GD) 1]. From GD1 until weaning [postnatal day (PND) 21], the dams were gavaged daily with the chemical preparations and then killed under anesthesia. On the day of parturition, all litter sizes were standardized to 10 offspring. From weaning (PND21) to adulthood (PND80), all animals were gavaged daily according to the various exposure protocols, and every 4 days they were weighed and inspected for anomalies.

Reproductive development end points

On PND25, we randomly selected five males from each group; the AGD was measured with a digital caliper, and the development of the penis was scored according to the procedure reported by Maeda et al. (2000). For both tests, the operator was blind to the treatment group.

F1 mating and fertility end points

On PND80, six males per group were selected for mating with new unexposed acclimatized females (Harlan France Sarl); each male was housed with one female for 4 days. As described above, females were inspected daily; when a vaginal sperm plug was found, females were housed separately until parturition. We used the following fertility end points: mating index (number of mated females ÷ number of females cohabited with males × 100), fertility index (number of pregnant females ÷ number of mated females × 100), litter size, mean weight per neonate, sex ratio at birth, and percentage of postimplantation loss (number of embryonic scars ÷ number of embryonic buttons and scars × 100).


On PND85, we anesthetized males using isoflurane (2.5%) and collected blood under heparin conditions from the dorsal aorta for subsequent hormonal tests. Animals were killed by thoracic cage opening. For each animal, both testes, epididymides and the seminal vesicles, ventral prostate, and liver were excised and trimmed of fat. We then weighed the tissues (Precisa model 125A; Precisa, Poissy, France) and calculated relative weights (weight of the organ in grams per kilogram body weight).

Sperm motility and motion characteristics

Immediately after epididymis separation, we excised approximately a 0.8-cm portion of the proximal part of the vas deferens and placed it in a Petri dish containing 4 mL of pre-warmed M199 medium (Gibco BRL, Paisley, Scotland) supplemented with 1% bovine serum albumin (Sigma-Aldrich, Saint-Quentin Fallavier, France). After incubating for 5 min at 37°C, we swirled the Petri dish to facilitate the spontaneous release of sperm from the vas deferens. We loaded 10 μL sperm suspension into an 80-μm-deep, 2X-Cel disposable sperm analysis chamber (Hamilton-Thorne Research, Beverly, MA, USA). The device was placed on the heated stage (37°C) of a computer-aided sperm analysis (CASA) integrated visual optical system (Hamilton-Thorne Research). The acquisition parameters and rates used for analysis were those of standard rat analysis setup 1 (frame rate, 60 Hz; frames acquired, 30; minimum cell size, 7 pixels; minimum contrast, 15; brightness, 2,500). Sperm were considered motile if the average path velocity (VAP) exceeded 20 μm/sec and were considered progressively motile when VAP exceeded 50 μm/sec and straightness of trajectory (STR) exceeded 80%. We analyzed a minimum of 200 motile sperm from each animal for the following variables: percentage of motile sperm, percentage of progressively motile sperm, VAP (micrometers per second), straight-line velocity (VSL; micrometers per second), curvilinear velocity (VCL; micrometers per second), lateral head displacement (ALH; micrometers), beat cross frequency (BCF; Hz), STR (%), and linearity of trajectory (LIN; %).

Epididymal sperm number

The procedure used for assessing the sperm number in the cauda epididymis was based on specific staining of sperm DNA and counting with the CASA instrument (Strader et al. 1996). We dissected out the cauda epididymis just below the point at which the vas deferens joins the epididymis at the distal corpus and at the boundary between the corpus and the proximal end of the cauda epididymis. The tissue was placed in a 50-mL plastic conical tube, chopped finely with scissors, and mixed with 25 mL physiological buffered saline plus 0.05 g/100 mL Triton X-100. The preparation was then crushed with an Ultra Turax T25 basic homogenizer (24,000 rpm 3 × 1 min; IKA, Staufen, Germany) The cauda epididymis preparation was stored frozen at −20°C. On the day of the sperm number assessment, we thawed the preparation, added 0.2-mL aliquots to vials containing the fluorescent DNA-specific dye bisbenzimide (Ident stain; Hamilton-Thorne Research, Danvers, MA, USA), and vortexed it for 30 sec. Following the manufacturer recommendations, we stained the sperm for 2 min. The preparation was vortexed again, immediately loaded into a standard-count disposable 20-μm analysis chamber (Leja, Nieuw-Vennep, the Netherlands), and placed on the CASA stage. The sperm heads were clearly illuminated under an ultra violet beam, identified (and debris ignored), and automatically counted in 10 fields using the RAT-IDENT mode of the CASA instrument. The data are expressed as the total number of spermatozoa per cauda epididymal tissue sample.


One testis and one epididymis per animal were fixed in 4% formaldehyde (Sigma-Aldrich), embedded in paraffin, cut into 4 μm-thick sections, stained with hematoxylin and eosin (Sigma-Aldrich), and examined under light microscopy.

Assays for plasma luteinizing hormone (LH), follicle-stimulating hormone (FSH), testosterone, and estradiol

One plasma sample per animal was frozen at −20°C until assayed. LH and FSH reagents (AH R002 and AH R004, respectively, from Biocode-Hycel) were supplied by OSTEOmedical GmbH (Bünde, Germany). The principle of the rat LH assay is based on competition between the LH of the rat sample and a 125I-labeled rat LH tracer for binding to a highly specific rabbit polyclonal antibody. The sensitivity of this assay is 0.14 ng/mL. The rat FSH assay is a solid-phase immunoradiometric assay offering high affinity and specificity for two different epitopes on rat FSH. The sensitivity of the assay is 0.2 ng/mL. We determined total testosterone and estradiol using the COAT-A-COUNT testosterone and estradiol radioimmunoassays (DPC France, La Garenne Colombes, France) according to the manufacturer’s protocols. The detection limits of the assays were 0.2 ng/mL and 2 pg/mL, respectively. All hormone concentrations were determined in duplicate.

RNA preparation and microarrays

We prepared RNA from individual testes obtained from six animals in each group. RNA was quantified and the quality evaluated using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, USA). The RNA integrity numbers for all RNAs were > 8. Pools of RNAs from six testes were prepared with strict equilibration of their quantity to ensure equal representation of each individual RNA. After quality control, we sent the samples to the micro array platform of NimbleGen (Reykjavik, Iceland). NimbleGen rat microarrays cover 23,456 transcripts represented by eight 60-mer oligonucleotides spotted in duplicate on the glass slide. The oligonucleotides are isothermic, which allows hybridization at a high temperature acceptable for the complete set of transcripts (70°C). The correlation between the homologous oligonucleotides in each hybridization was > 0.99. We used Cluster 3 (Eisen et al. 1998) and Treeview (Saldanha 2004) for expression clustering, and DAVID software (Dennis et al. 2003) for functional clustering.

Quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR) validation and interindividual variation

We performed qRT-PCR for 10 genes using cDNA prepared from four individual testes [four animals in each category; PCR primers are described Supplemental Material, Table 2 (doi: 10.1289/ehp.0800158.S1/)]. The reactions were carried out in a 15-μL volume using a Roche Lightcycler (Roche, Meylan, France) and an Invitrogen Sybergreen quantitative PCR kit (Invitrogen, Cergy-Pontoise, France). After 2 min at 50°C and 2 min at 94°C, each reaction was carried out for 40 cycles (94°C for 10 sec, 55°C for 20 sec, and 72°C for 20 sec when fluorescence was measured). This was followed by progressive melting by increasing the temperature from 60°C to 99°C over 10 min, with continuous fluorescence capture. We considered the resulting melting curve satisfactory if one peak only was visible on its derivative. In addition, PCR products were subjected to electrophoresis on an aga-rose gel to check for a single product of the expected size.

Data analysis

Statistical tests were carried out using BMDP statistical software, version 2 for Windows (Dixon 1988). We determined statistical significance of quantitative differences between treated groups and controls and between treatment modalities using post hoc Tukey’s pairwise comparisons after an analysis of variance. For visual clarity in the tables and figures, we present only the significant differences between each compound or mixture exposure modality and controls. Only pairwise significant differences in the various treatment modalities are reported in text. Differences for qualitative variables between treatment and control groups were tested using the Pearson chi-square test. The significance level is set at 0.05. However, for some aspects of the study (as indicated in tables and figures), we used a threshold of 0.10 because of the relatively small size of the groups. For several figures in both the article and the Supplemental Material (doi: 10.1289/ehp.0800158.S1), we present box plots (rather than standard histograms showing only mean and SE values) in order to present the distributions observed for each exposure group and therefore the true level of variability.


Developmental anomalies and body and organ weights

The observed anomalies of the external genitalia are summarized in Table 1. Penile development was immature in all groups except G1, V30, and G10 + V30. Overall, we found the highest rates of genitalia anomalies for the low-dose mixture, G1 + V1, and high-dose vinclozolin, V30. The body weights and relative organ weights in control and treated animals on PND80 are presented in Table 1. The gain in body weight was unaffected by any treatment (data not shown). The relative liver weight was significantly different from control only for G10 (Table 1) and did not vary significantly among the different exposure groups studied. The relative epididymal weight was significantly lower in all treated groups compared with controls (Table 1) and significantly lower in the low-dose mixture (G1 + V1) compared with V1 (p < 0.05). Relative testis weights were significantly higher in the high-dose mixture than in the low-dose mixture (p < 0.01). The relative prostate and seminal vesicle weights were decreased with V30 compared with controls. In these cases, the addition of genistein had no influence. For these organs, we found no significant difference between all the groups exposed to the compounds.

Sperm motility and kinematic characteristics

All kinematic characteristics, except for the amplitude of ALH and LIN, were significantly altered in the treated animals. The VCL was significantly lower in all exposure groups than in controls, whereas the percentage of progressively motile sperm and the STR were significantly lower in all treated groups except G10 than in controls [Figure 1; see also Supplemental Material, Table 1 (doi: 10.1289/ehp.0800158.S1)].

Epididymal sperm number

The sperm reserve in the cauda epididymis was significantly lower than control values for G10, V30, G1 + V1, and G10 + V30, (Figure 2), and it was significantly lower for the low-dose mixture than for G1 or V1 alone (both p < 0.005). We observed the smallest mean value for the low-dose mixture (58 ± 11 vs. 104 ± 14 × 106 spermatozoa in the control; p = 0.02), which was not significantly different from that found for V30.


Treatment had no obvious effect on testis and epididymis anatomy. However, we observed small variations in tissue/organ histology [see Supplemental Material, Figure 1 (doi: 10.1289/ehp.0800158.S1)].
Hormone concentrations
On PND80, mean FSH concentrations were significantly higher for G1 + V1, G10 + V30, and V30 than for control groups [13.0 ± 5.1, 11.6 ± 2.0, and 12.3 ± 4.3 vs. 9.9 ± 2.5 ng/mL, respectively; all p < 0.05; see Supplemental Material, Figure 2 (doi: 10.1289/ehp.0800158. S1)]. Mean estradiol concentrations were significantly lower than control values only in the G1 + V1 group (3.7 ± 3.9 vs. 6.1 ± 3.6 pg/mL; p < 0.05; see Supplemental Material, Figure 2). Testosterone levels were significantly lower than controls in the G10 + V30 and G10 groups (2.0 ± 1.7 and 2.4 ± 2.6 vs. 4.0 ± 3.2 ng/mL; both p < 0.05). The LH concentration tended to be higher than control values only in the V30 group (1.34 ± 0.81 vs. 0.97 ± 0.99; p = 0.07).

Mating and fertility

Table 2 summarizes the fertility of the exposed male rats crossed with non exposed females. The mating index was markedly but not significantly lower with the low-dose mixture compared with the control and the other exposure groups. Litter sizes were significantly smaller in the G1 + V1, G10 + V30, G1, and G10 groups; the smallest litters were in the low-dose mixture group (5.3 ± 1.5 vs. 13.0 ± 1.6 in controls; p < 0.05). Litter sizes did not differ significantly among the various exposure groups. The postimplantation loss was significantly higher in the V30 and the G10 + V30 groups than in the control group.

Testis transcriptome analysis

The effects of the different treatments on gene repression/induction were strikingly different. Applying a 2-fold threshold, genistein had a generally repressive effect on gene expression in the testis, particularly at the low dose; the high vinclozolin dose (but not the low dose) had the opposite effect (Figure 3). The array results were validated by qRT-PCR with a sample of 10 genes [see Supplemental Material, Table 2 (doi: 10.1289/ehp.0800158.S1)]; consistent with other results obtained with NimbleGen arrays (Buffat et al. 2007), the correlation between the array and the qRT-PCR was > 0.8. Using the complete data set, and no thresholds, the strongest correlation was between G1 + V1 and V30 [r = 0.82; see Supplemental Table 3 (doi: 10.1289/ehp.0800158.S1); see also Figure 3]. This is in good accordance with the phenotypic observations: V30 and G1 + V1 had very similar effects on several markers. Nonsupervised hierarchical classification (Figure 4) showed perfectly correlated duplicates and illustrated the clustering of the genes modified by V30 and V1 + G1. This approach identified seven clusters of genes (Figure 4). Functional classification of the genes enabled us to identify biological functions for each sub cluster (Table 3); we functionally analyzed each subgroup of genes using DAVID, after elimination of poorly annotated factors and olfactory receptor genes that are always highly represented in rodent microarrays. Among the genes significantly clustered in KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways, some were involved in regulation of insulin pathways (cluster 1); others were involved in fructose and glucose metabolism (cluster 2). Clusters 3 and 4 comprised genes involved in inter actions between ligands and receptors (e.g., dopamine, acetylcholine, histamine, parathyroid hormone, prostanoids), with the similar composition of the two clusters indicating a quantitative rather than a qualitative effect. The same applies to clusters 5 and 6, which were highly enriched in genes encoding ribosomal proteins. These genes were strongly down-regulated by all treatments except low doses of genistein. This may signify a slowing down of protein synthesis after exposure to the compounds. Finally, cluster 7 was functionally clustered but with relatively low statistical significances, and included mainly genes of the gonadotropin-releasing hormone pathway. We observed the major effect on these genes (up-regulation) in the G1 and G10 groups.


Our study provides evidence that lifelong exposure to low-dose genistein and/or vinclozolin results in a number of anomalies of the male reproductive tract and fertility. This markedly contrasts with previous reports based on short windows of exposure to these compounds, whether gestational, neonatal, or during the puberty/adult periods, showing only mild or no reproductive alterations [see Supplemental Material, Table 4 (doi: 10.1289/ehp.0800158.S1)]. Interestingly, in the adult, despite the absence of macroscopic and microscopic modifications of the testis, we found various changes of the testis transcriptome for all exposure protocols. Furthermore, the intriguing similarities of the phenotypes/genotypes of the animals exposed to high-dose vinclozolin and those exposed to the low-dose mixture, which probably depend on complex modes of action in this chronic exposure from conception to adulthood, may suggest an antiandrogenic action of genistein at a low dose when combined with low-dose vinclozolin (with a possible “synergistic” action because each compound taken alone at a low dose had no or a weak effect). By contrast and for several phenotypes (epididymis, seminal vesicle, and ventral prostate relative weights, as well as litter size), the effect of the high-dose mixture appeared milder than the effect of high-dose vinclozolin alone. This could suggest that, in these specific cases, a “high” dose of genistein could attenuate the effects of vinclozolin. Finally, consistent with these observations, the effects observed for several end points were more pronounced with the low-dose mixture than with the high-dose mixture. Obviously our comments are purely speculative because of the descriptive design of the study, which warrants further work to approach the underlying molecular bases of these intriguing results.

Phenotype modifications with F1 vinclozolin exposures

At a vinclozolin dose of 30 mg/kg body weight, we observed a rate of developmental anomalies of the reproductive tract that are consistent with the higher rate previously reported (Gray et al. 2001); this validates our exposure protocol. At this dose, we also found some of the more detrimental effects on fertility, including a high rate of postimplantation loss (note that this end point was also high for the 1 mg/kg vinclozolin dose). High-dose vinclozolin caused a significant increase in the FSH concentration, suggesting a direct effect on the pituitary axis according to Nellemann et al. (2003). The diminished fertility, evidenced by the reduced litter size, especially with the high dose, could be due, at least in part, to decreased sperm production and motility. It could also be partly the consequence of the marked effects of vinclozolin on sperm chromatin we reported previously (Auger et al. 2004). Prenatal plus infantile exposure of male rabbits to 7.2 mg/kg/day vinclozolin increases the amount of morphologically abnormal spermatozoa, mostly due to nuclear and acrosomal defects (Veeramachaneni et al. 2006). These anomalies may result in impairment of the fertilizing ability, embryonic development, or both (Bartoov et al. 2001; Sailer et al. 1996). Whatever the period of exposure, most published studies involving low-dose vinclozolin report marginal or no effects [see Supplemental Material, Table 4 (doi: 10.1289/ehp.0800158.S1); Colbert et al. 2005; Gray et al. 1999; Hellwig et al. 2000; Shin et al. 2006], contrary to the significant adverse reproductive effects observed at high doses (Yu et al. 2004). Studies of rats exposed to vinclozolin have nevertheless revealed that dose–response relationships are not equivalent among end points (National Toxicology Program 2001). AR expression in the male reproductive system of the rat varies according to time and cell type, such that the effects of vinclozolin may vary substantially according to developmental period(s) and duration of exposure. Indeed, some of the adverse effects observed in adults may depend on changes in AR expression related to the continuous exposure. In this respect, the phenotypic changes we observed with vinclozolin exposure were not necessarily induced only in utero or during the early neonatal period. In addition, they could have been mediated by as yet undescribed actions of vinclozolin, or its metabolites, independent of antagonistic action on ARs.

Phenotype modifications with F1 genistein exposures

Chronic exposure to dietary levels of genistein, from conception to adulthood, may have deleterious effects on male reproductive development, adult reproductive organs, and fertility. Some of our findings were in accordance with previous studies: the penis immaturity we observed on PND25 in the G10 group is similar to the delayed preputial separation after gestational and lactational exposure to 5 mg/kg genistein in the diet in Long-Evans rats (Wisniewski et al. 2003) and decreased plasma testosterone (for the “high” dose only) reported by Weber et al. (2001) for a similar dose. This result of Weber et al. (2001) reinforces the idea that genistein decreases the steroidogenic response of Leydig cells, which express both estrogen receptor (ER) α and ERβ, although the expression of these ERs was not affected in our array experiments. Svechnikov et al. (2005) reported that dietary genistein down-regulates an early step of testosterone synthesis (expression of mitochondrial P450 side-chain cleavage enzyme, which catalyses the conversion of cholesterol to pregnenolone). However, as in the study of Roberts et al. (2000), the testosterone levels we found with G1 were not significantly abnormal, suggesting that the effect of genistein on Leydig cells may be dose dependent; consistent with this notion, at the doses and with the strain we used, no difference was detectable in our microarray experiment.
Several of the adverse effects we observed are indisputably related to the estrogen agonist activity of genistein and the localization of both types of ERs that are expressed at various levels in male rat reproductive organs (Boukari et al. 2007; Mäkelä et al. 2000; Pelletier et al. 2000). The lower reproductive performance observed here probably results in part from the combination of sperm movement anomalies for the low- and high-dose exposures and the reduced epididymal reserve (only for G10). Genistein decreases sperm motion variables in vitro by inhibiting the sperm protein kinase (Bajpai and Doncel 2003), which our microarray experiments showed to be slightly decreased (reduced to 0.76 for G1). Moreover, genistein has been reported to inhibit α-glucosidase, which is known to sustain sperm motility (Lee and Lee 2001); the expression of this enzyme was also slightly reduced by G1 in our experiment (× 0.78). To our knowledge, the increased postimplantation loss we describe has not previously been reported. However, similar reproductive anomalies have recently been reported for CD-1 mice after neonatal exposure to genistein at environmentally relevant doses (Jefferson et al. 2005), and the reported genotoxicity of genistein may be involved (Stopper et al. 2005). Genistein is a topoisomerase II inhibitor (Markovits et al. 1989), but we did not find any modification of the expression of the corresponding gene in the testis. Another important consequence of the lifelong genistein exposure was the high incidence of reproductive effects, and this contrasts strongly with the absence of effects reported for shorter gestational/lactational, puberty/adult exposures [see Supplemental Material, Table 4 (doi:10.1289/ehp.0800158.S1)]. The possible potentiation of the effects of vinclozolin by exposure to low-dose genistein that we report here might be relevant for human risk assessment because of the similar exposure conditions.

Phenotype effects of the F1 exposures to the mixtures: similarities of the effects with the low-dose mixture and high-dose vinclozolin

In our experimental conditions, the type and magnitude of effects of the low-dose mixture were close to those of high-dose vinclozolin alone. Low combined doses of genistein and vinclozolin significantly increased FSH levels and decreased estradiol levels, an effect very similar to that of high-dose vinclozolin alone. This result parallels the similarities between the effects of the low-dose mixture and the high-dose vinclozolin on sperm production, motility, and kinematics and on the weights of the epididymis and seminal vesicles. These various similarities evoke a synergistic androgenic effect with the low-dose mixture. Genistein has been reported to exhibit in vitro antiandrogenic activity in addition to its well-established estrogenic activity (Rosenberg Zand et al. 2000). This type of mechanism may act in vivo and contribute to some of the anti androgenic demasculinizing effects observed here with the low-dose mixture, with a magnitude similar to that found for high-dose vinclozolin (the effects found with both exposure modalities, G1 + V1 and V30 were not significantly different for most of the end points studied). Consistent with this notion, combinations of antiandrogens at NOAEL doses have been induced significant synergistic effects in vivo (Hass et al. 2007; Kortenkamp 2008). However, more complex modes of actions cannot be ruled out because of our continuous exposure modalities and the multiple biological activities of both the compounds. The complex transcriptomic modifications caused in the adult testis illustrate this. Indeed, although the effect on gene induction of the high vinclozolin dose is generally stronger than that of the combination of genistein and vinclozolin, the overall effect is very similar. As a consequence, the phenotypic alterations may be very similar, even if the effects of the low-dose combination only seldom attain the 2× threshold.

Testis transcriptome alterations and the phenotypes observed

The gene expression profile under the influence of the EDCs genistein and vinclozolin can be summarized as follows (Figure 4). Genistein administered alone at a low dose had the smallest effects of all treatments tested. The alterations were minor, despite 1,089 genes being down-regulated (using a 2× threshold). This can be explained by the strong and exclusive deregulation of a particular cluster of genes (cluster 4), with most other clusters very similar to the controls. Cluster 4 is composed of genes involved in ligand/receptor interactions, involving numerous hormone peptides and G-protein–coupled receptors. At a higher dose of genistein, with or without high-dose vinclozolin, the alterations were greater. Presumably, the effects of genistein mask the putative effects of vinclozolin, probably indicating that high doses of genistein are the driving force in the gene expression regulation.

The effects of vinclozolin at low and high doses were similar and were strongly correlated with the effects of the low-dose mixture. This suggests that the low genistein dose is able to potentiate the effects of vinclozolin. One possible mechanism is activation by genistein of specific receptors of protein ligands induced by low-dose vinclozolin, and specific preparation of the chromatin to bind specific transcription factors. One of the candidates for such an effect is the gene Cbx3 (Chromobox, 3 homolog), whose protein has a chromodomain able to regulate chromatin dynamics toward gene induction or repression (Benetti et al. 2007).


A continuous exposure to low combined doses of genistein and vinclozolin affects male reproductive health by inducing reproductive developmental anomalies, alterations in sperm production and quality, and fertility disorders. The long-term deleterious consequences of chronic exposures to a low-dose mixture of genistein/vinclozolin suggest complex modes of action; we are currently investigating this issue further. Transcriptome analysis may uncover specific factors with potentially major effects on numerous gene targets, such as Cbx3.


Read Full Post »

Read Full Post »

Read Full Post »


Ancestral ghosts in your genome | Michael Skinner | TEDxRainier

Epigenetic Transgenerational Actions of Endocrine Disruptors by Dr. Michael K. Skinner & Matthew D. Anway

Epigenetic Transgenerational Actions of Endocrine Disruptors

Matthew D. Anway and Michael K. Skinner

Center for Reproductive Biology, School of Molecular Biosciences, Washington State University, Pullman, Washington 99164

Address all correspondence and requests for reprints to: Michael K. Skinner, Center for Reproductive Biology, School of Molecular Biosciences, Washington State University, Pullman, Washington 99164-4231. E-mail: Skinner@mail.wsu.edu.


Endocrine disruptors have recently been shown to promote an epigenetic transgenerational phenotype involving a number of disease states (e.g. male infertility). The anti-androgenic fungicide vinclozolin was found to act transiently at the time of embryonic sex determination to promote in the F1 generation a spermatogenic cell defect and subfertility in the male. When the animals were allowed to age up to 1 yr, a number of other disease states developed. This phenotype was transferred through the male germ line to all subsequent generations analyzed (F1–F4). The ability of an environmental factor (i.e. endocrine disruptor) to promote an epigenetic transgenerational phenotype impacts the potential hazards of environmental toxins, mechanisms of disease etiology, and evolutionary biology. The biological importance of the epigenetic actions of environmental agents is reviewed in the context of the primordial germ cell and development of epigenetic transgenerational phenotypes.

GENOMIC DNA CONTAINS the core of genetic information of the cell. There is a distinct pattern of gene expression throughout mammalian development that is heritable from parents to offspring. Epigenetics is defined as the molecular phenomena that regulate gene expression without alterations to the DNA sequence (1). The most studied epigenetic modification is DNA methylation of CpG nucleotides that are essential for mammalian development (2, 3, 4, 5). DNA methylation of CpG sites is used by mammals to regulate transcription of genes, alter chromosomal positioning, influence X-chromosome inactivation, control imprinted genes, and repress parasitic DNAs (1, 5, 6, 7, 8, 9). Alterations in the DNA methylation state can lead to multiple disease states including cancers (10, 11), Rett syndrome, and Prader-Willi/Angelman syndrome (11, 12, 13), male infertility (14), autism (12), and Angelman and Beckwith-Wiedemann syndromes (13). Both chemical and environmental toxins have been shown to alter DNA methylation patterns resulting in epigenetic phenotypes (14, 15).

DNA methylation patterns are established at two times during development: the lineage-specific pattern during gastrulation and the germ-line-specific pattern in the gonad after sex determination (16). The lineage-specific pattern establishes the DNA methylation for somatic cell development after fertilization. This epigenetic reprogramming is based on the genetic material transferred from the egg and sperm. Alterations in the lineage-specific epigenetic reprogramming results in developmental defects or embryonic lethality (13, 16). The germ-line DNA methylation pattern is established during gonadal development and is sex specific (16, 17, 18). Epigenetic reprogramming of the germ line is critical for imprinting (19, 20, 21, 22). Unlike the lineage-specific reprogramming, alterations in the germ-line epigenetic reprogramming can alter the heritable epigenetic information, resulting in a transgenerational phenotype (15) (Fig. 1). The embryonic period is the most sensitive for chemical and environmental effects on the epigenetics of the male germ line (15, 21, 22).

Recent investigations of the DNA methylation state of the primordial germ cells have indicated that as primordial germ cells migrate down the genital ridge, a demethylation (i.e. erasure of methylation) starts, and upon colonization in the early gonad, a complete demethylation is achieved (21, 22, 23). This has been primarily observed through the analysis of specific imprinted genes (24). During the period of sex determination in the gonad, the germ cells undergo a remethylation involving a sex-specific determination of the germ cells (Fig. 2). Although the demethylation may not require the gonad somatic cells (21), the remethylation of the germ line appears to be dependent on association with the somatic cells in the gonads (22, 23). Because of this unique property of the germ cells to undergo a demethylation and remethylation during the period of sex determination in the developing gonad, the ability of an environmental agent such as an endocrine disruptor to influence through an epigenetic process the germ line is postulated. This epigenetic effect on the germ line could reprogram the germ cell through an event such as altered DNA imprinting (25, 26). This epigenetic effect could cause a transgenerational effect on subsequent generations through the germ line. Because the remethylation of the germ line appears dependent upon the gonadal somatic cells, an alteration in somatic cell function by an agent such as an endocrine disruptor could indirectly influence the germ cell remethylation (Fig. 2). Epigenetic alterations that lead to transgenerational transmission of specific genetic traits or molecular events (e.g. imprinting) have recently been identified (6, 7, 27). These observations have led to the conclusion that a reprogramming through altered epigenetics of the male germ line is possible (15). The impact this has on human health and evolutionary biology is significant (6, 27).

Transgenerational Phenomena and Environmental Factors
Environmental effects of irradiation, chemical treatments (e.g. chemotherapy), and environmental toxins such as endocrine disruptors have been observed over the past decade. The majority of observations are simply the effects of the agent on the gestating mother (F0) and subsequent actions on the offspring associated with the F1 generation (28, 29, 30). Examples of environmental factors during embryonic development that influence the F1 generation include the effects of heavy metals causing cancer (31), abnormal nutrition that causes diabetic and uterine defects (32, 33, 34), chemical exposure (i.e. ethosuximide and benzpyrene) causing brain and endocrine defects (35, 36), and endocrine disruptors such as diethylstilbestrol (37, 38), phthalates, and dithiothreitol causing reproductive tract and endocrine defects (39, 40, 41). Environmental factors have effects on the F1 generation of a number of species including insects (42, 43, 44), fish (45, 46), birds (47), and other species (48). Therefore, exposure to a number of environmental factors in utero can cause abnormal phenotypes in the F1 generation in a number of different species. Because the F1 generation is exposed to the environmental factor, the F1 effect is not a transgenerational phenotype.

Transgenerational effects of environmental factors require effects minimally on the F3 generation (15, 49) (Fig. 1). This is because the F3 generation is the first generation not directly exposed to the environmental factor. The ability of an external agent to induce a transgenerational phenotype requires a genetic (i.e. DNA sequence) or an epigenetic (i.e. DNA methylation) phenomenon mediated through the germ line (50, 51, 52, 53). Transgenerational inheritance of an epigenetic state has been shown to occur using several mouse genetic lines and markers (6, 27, 54) and more recently with the use of monozygotic twins with epigenetic differences (55). Irradiation exposure was one of the first transgenerational phenomena observed to be transmitted through the germ line to multiple generations, often associated with mutagenesis and tumor formation (50, 51, 52, 53). The chemotherapeutic treatment of cancers has been shown to cause F1 generation effects (31, 35, 46), but the transmission to multiple generations has not been thoroughly investigated. Environmental factors do appear to promote a transgenerational susceptibility to cancer (56, 57). Gestating nutritional deficiency effects on the F1 generation have been observed (34), and recently these nutritional effects on a diabetic condition and growth defects have been shown to be transgenerational to the F2 generation (58, 59, 60). Several environmental chemical exposures have also been shown to transgenerationally affect the F2 generation including benzpyrene (36, 61), orthoaminoasotoluol (62), and dioxin (63). Environmental toxins such as endocrine disruptors have also been shown to influence the F1 generation after parental exposure (39, 46, 64, 65, 66, 67), but few have demonstrated transgenerational effects on multiple generations (15). Some evidence that diethylstilbestrol has effects in the F2 generation have been reported (68).

Endocrine Disruptors and Reproductive Toxicology

Many reports have suggested that environmental endocrine disruptors, which act to mimic estrogens or act as antiestrogens or antiandrogens, are detrimental to reproduction and may promote abnormalities such as a decrease in sperm count, an increase in testicular cancer (69, 70), and an increase in abnormalities in sex determination for many species (71). Examples of environmental endocrine disruptors that have been targeted for adverse effects on reproductive systems in humans and other animals are pesticides [e.g. dichlorodiphenyltrichloroethane (DDT) and methoxychlor] (72), fungicides (e.g. vinclozolin) (15, 73), insecticides (e.g. trichlorfon) (74), herbicides (e.g. atrazine) (75), plastics (e.g. phthalates) (76), and a range of xenoestrogens (77). Most of these chemicals are ubiquitous in the environment, resulting in daily exposure for humans and other animals. Many of these compounds and endocrine disruptors can be metabolized into both estrogenic and antiandrogenic activities (78). Recently, methoxychlor and vinclozolin have been used (66, 67) as model endocrine disruptors (72) that have estrogenic, antiestrogenic, and antiandrogenic metabolites (78).

Many environmental endocrine disruptors are weakly estrogenic and elicit their actions through the estrogen receptors. The two mammalian receptors for estrogen (ER- and ER-ß) are widely distributed throughout the reproductive tract (79, 80). ER-ß is present in higher concentrations within the fetal testis and ovary, whereas ER- is present mainly within the uterus (81, 82). During fetal testis development, ER-ß is expressed in Sertoli and myoid cells after seminiferous cord formation (83). In rats, ER-ß has also been localized to prespermatogonia, which may explain the proliferative actions of estrogen on early postnatal gonocyte cultures (84). The importance of ER- was delineated when knockout mice (85) and human males (86) lacking expression of this gene were found to be sterile. Fetal development of the testis in these experiments was not altered; however, fetal testis morphology in a double knockout remains to be examined (87). Neonatal exposure to estrogen alters the ER- and ER-ß expression during postnatal testis and hypothalamic/pituitary development (88, 89). Interestingly, neonatal exposure to the estrogenic compound diethylstilbestrol promotes abnormal testis and male reproductive tract development (90) and leads to changes in gene expression (91). Therefore, actions of estrogenic endocrine disruptors on estrogen receptors may impair normal fetal gonadal development and lead to infertility. Although the estrogen receptors are thought to have a role in testis development (92, 93, 94), the specific functions remain to be elucidated. Treatment of males with estrogens during early fetal life may alter responsiveness to androgens by changing androgen receptor (AR) expression patterns (95, 96) and/or Leydig cell function (91).

Antiandrogenic endocrine disruptors can also influence fetal gonad development. AR expression is very similar to ER-ß expression in the developing testis (82, 97). AR is detected in Sertoli, myoid, and prespermatogonial cells just after cord formation (98) and in interstitial cells late in fetal development. It is proposed that AR is present in cells that migrate from the mesonephros and enables cord formation to occur (98). Therefore, inappropriate expression or actions of AR through treatment by endocrine disruptors may affect the process of morphological sex differentiation (i.e. cord formation). Antiandrogens such as flutamide (99) or cyproterone acetate (100) administered to pregnant rats at different ages of gestation impair fertility in the male offspring. Both flutamide and cyproterone acetate block the ability of androgens and epidermal growth factor to stabilize the Wolffian duct (101). Therefore, perturbation of AR may also cause inappropriate expression and action of growth factors in the testis. A commonly used antiandrogenic endocrine disruptor is vinclozolin, which is used as a fungicide in the wine industry (102, 103). Vinclozolin has been shown to act as an environmental antiandrogen and influence gonad development and fertility (15, 67).

Epigenetic Transgenerational Actions of Endocrine Disruptors

A recent observation demonstrated that the exposure of a pregnant rat transiently to endocrine disruptors caused a spermatogenic cell defect and subfertility in the F1 generation and all subsequent generations examined (F1–F4) (15) (Fig. 1). The endocrine disruptors used were the antiandrogenic fungicide vinclozolin used in the fruit (e.g. wine) industry (73) and the pesticide methoxychlor used to replace dichlorodiphenyltrichloroethane (DDT) (78). The critical exposure period was at the time of sex determination, and the transgenerational phenotype was transmitted through the male germ line (15) (Fig. 1). The phenotype of increased spermatogenic cell apoptosis and decreased sperm numbers and sperm motility was observed in greater than 90% of all males of all the generations examined. When the animals were allowed to age up to 1 yr, additional diseases developed including cancer, prostate disease, kidney disease, and immune cell defects (Anway, M. D., and M. K. Skinner, submitted for publication). A high frequency of transmission was observed in all generations examined for all the disease states.

The frequency of the transgenerational phenotype was such that a DNA sequence mutational event could not be involved. The random nature of a DNA sequence mutation has a phenotype typically less than 1%, and this often declines in subsequent generations (50, 104). An epigenetic mechanism is involved because of the frequency of the phenotype. To support these conclusions, two genes were identified in the sperm that had altered methylation patterns associated with the transgenerational phenotype discussed (15). Therefore, the endocrine disruptors appear to induce an epigenetic transgenerational disease condition for four generations through the male germ line (15) (Fig. 1). The epigenetics appears to involve altered DNA methylation. Although most genes get reset in early embryonic development, a subset of genes called imprinted genes maintains their DNA methylation pattern that appears to be permanently programmed. In contrast to all somatic cells, the primordial germ cells undergo a demethylation during migration and early colonization of the embryonic gonad, followed by a remethylation starting at the time of sex determination in a sex-specific manner (23, 24, 105). The exposure of the pregnant mother at the time of sex determination appears to have altered the remethylation of the germ line and permanently reprogrammed the imprinted pattern of DNA methylation (15). This provides a unique epigenetic mechanism to promote a transgenerational phenotype induced by an environmental factor.


The observations that an environmental toxin (e.g. endocrine disruptor) can have an epigenetic effect on the germ line and cause a transgenerational effect on male reproduction significantly impacts our understanding of the potential hazards of these compounds to human health as well as all other mammalian species (15). These studies establish a novel mechanism of action not previously appreciated on how environmental toxins may act on a gestating mother to influence her grandchildren and subsequent generations. Elucidation of this phenomenon will allow us to better understand the true hazards of environmental toxins, identify the specific causal agents, and develop appropriate preventative and therapeutic approaches. Independent of the specific compound or agent of interest, the establishment of this potential mechanism of action is critical to our insight into the effects of environmental factors that influence embryonic development and adult reproduction.

The level of endocrine disruptors used in the recent studies (15, 66, 67, 106) (Anway, M. D., and M. K. Skinner, submitted for publication) is higher than anticipated in the environment, such that conclusions regarding the toxicology of these endocrine disruptors are not possible. However, the important factor is the identification of this novel phenomenon, that an environmental factor can promote an epigenetic transgenerational phenotype (15). Because of this observation, the potential hazards of environmental factors need to be carefully evaluated. If the exposure of your grandmother at midgestation to environmental toxins can cause a disease state in you with no exposure, and you will pass it on to your grandchildren, the potential hazards of environmental toxins need to be rigorously assessed. Transgenerational studies need to be performed in evaluating the toxicology of environmental compounds.

The epigenetic transgenerational phenotype also provides critical insights into disease etiology. Because a number of common disease states are induced (Anway, M. D., and M. K. Skinner, submitted for publication), an epigenetic component of disease now needs to be seriously considered. The fetal basis of adult-onset disease could be a result of epigenetic factors (107, 108). In the event a major epigenetic component exists, the epigenetic background of an individual may be a significant factor in susceptibility to disease development. Therefore, identification of the genes involved with altered methylation may provide essential new diagnostics to assess onset of disease. These epigenetic factors may influence the outcomes of current medical therapies such as assisted reproductive procedures (109, 110). Further analysis of the epigenetic transgenerational phenotypes and identification of specific epigenetic changes will allow new therapeutic targets and therapies to be developed to potentially prevent the onset of disease. This is a new paradigm in disease etiology that needs to be considered.

In a broader biological perspective, the ability of an environmental factor to cause a permanent genetic trait in all subsequent progeny of an effected individual can significantly impact our understanding of evolutionary biology. Currently, a DNA sequence mutation event that allows an adaptation and natural selection is considered the driving factor in evolutionary biology. However, the frequency of specific evolutionary events (110, 111) and regional influences on evolution suggest an additional epigenetic mechanism should be considered (112, 113, 114, 115). Although a DNA sequence mutational event will be important for evolutionary biology, an epigenetic component influenced by an environmental factor needs to be considered as an alternate factor that will help explain some aspects of evolutionary biology. Epigenetics is the next layer of complexity beyond the DNA sequence.


We acknowledge the assistance of Ms. Jill Griffin in the preparation of this manuscript.


This research was supported in part by grants from the U.S. Environmental Protection Agency, EPA STAR program, and from the National Institute of Environmental Health Sciences (to M.K.S.).

None of the authors have any conflicts of interest regarding the current publication.

First Published Online May 11, 2006

Abbreviation: AR, Androgen receptor.

Received August 19, 2005.

Accepted for publication October 14, 2005.

Clink on the link below to view the full article with figures.

Epigenetic Transgenerational Actions of Endocrine Disruptors by Dr. Michael K. Skinner & Matthew D. Anway

Read Full Post »

PBS Nova – Ghost in Your Genes

This is a profound & enlightening program covering the breaking field of epigenetics. It’s important to note that Dow Chemical helped fund this program but the only way you would know this is through watching it because it is mentioned at the beginning and the ending of the program. I got a DVD copy from an educator through the Howard Hughes Medical Institute. The public cannot access copies through HHMI though. This is one of the only Nova episodes that is not available to watch online. Here is why I’m guessing that Dow Chemical has not released this program for sale. The following is taken from the transcripts and has serious implications.

But in Washington state, Michael Skinner seems to have found compelling additional evidence by triggering a similar effect with commonly used pesticides. Skinner wanted to see how these chemicals would affect pregnant rats and their offspring.

MICHAEL SKINNER: And so I treated the animals, the pregnant mother, with these compounds, and then we started seeing, between six months to a year, a whole host of other diseases that we didn’t expect. And this ranged between tumors, such as breast and skin tumors, prostate disease, kidney disease and immune dysfunction.

NARRATOR: He checked that there were no genetic mutations and then proceeded to breed the rats.

MICHAEL SKINNER: The next step was for us to go to the next generation. And the same disease state occurs. So after we did several repeats, and got the third generation showing it and then a fourth generation, we sat back and realized that the phenomenon was real. We started seeing these major diseases occur in approximately 85 percent of all the animals of every single generation.

NARRATOR: His discoveries were a revelation.

MICHAEL SKINNER: We knew that if an individual was exposed to an environmental toxin that they can get a disease state, potentially. The new phenomenon is that environmental toxin no longer affects just the individual exposed but two or three generations down the line.

I knew that epigenetics existed. I knew that it was a controlling factor for DNA activity—whether genes are silenced or not—but to say that epigenetics would have a major role in disease development, so…I had no concept for that. The fact that this could have such a huge impact and could explain a whole host of things we couldn’t explain before took a while to actually sink in.

NARRATOR: Further work has revealed changed epigenetic marks in 25 segments of the affected rat’s DNA. The implications, if they apply to humans, are sobering.

The program description version of this extraordinary discovery is significantly downplayed and the name of the toxin is completely omitted from the program. It was vinclozolin (One of the most common fungicides utilized in agriculture).

Program Description

“In Washington State, a researcher finds that a toxin given to rats still affects their offspring four generations later, without producing any changes in their genes.”

Ghost in Your Genes – “Some NOVA programs are not for sale because the rights are not available.” I’m thinking that Dow Chemical owns the rights of this program. You will not find the specifics of Dr. Skinner’s research regarding pesticides and their impact on epigenetics and germ lines in the program description. (I personally don’t think that’s an accident.)

It is unbalanced in it’s emphasis on “lifestyle” choices but does cover Dr. Michael Skinner’s research on pesticide exposure and impacts on epigenetics and germ lines (transgenerational impacts). This episode also explores the research being completed on identical twins. This research is critical to understanding genetics versus environmental influence on genetic expression. This field of research will be critical to understanding the impacts from environmental toxins on human genes and the origin of disease, infertility, and birth defects. Epigenetics will show us the mechanisms through which toxins cause harm and alter us as a species.

I have discovered the transcripts!

Ghost in Your Genes – Transcripts

PBS Airdate: October 16, 2007
Go to the companion Web site

NARRATOR: Imagine sharing life with a person who seems to be you. Created from the same fertilized egg, you share exactly the same genes. So profound is their influence, everything about you appears the same: the spaces between your teeth, the way you laugh, your body language. You are, in a word, identical. Or are you?

SUSAN: As infants, they were very much alike. Their physical similarities are obvious. And all their physical milestones happened at the same time. But functioning today, for Jenna and Bridget…they’re completely different.

Jenna is enthusiastic, productive. Jenna’s going to college. She talks about it all the time now. Bridget is essentially non-verbal. She doesn’t have purposeful conversational speech. And there’s very unusual behavior. For example, she likes to spit on monitors and then rubs it in. I don’t know why, but that’s what she does.

How? How could these guys be identical and so, on such a different level, functioning-wise?

NARRATOR: So if genes don’t tell the whole story of who we are, then what does?

Scientists suspect the answer lies in a vast chemical network within our cells that controls our genes, turning them on and off.

ANDREW P. FEINBERG (Johns Hopkins University): It’s a little bit like the dark matter of the universe. I mean, we know it’s there, we know it’s terribly important, but we don’t really know all that much about how that symphony gets played out.

MARK MEHLER (Albert Einstein College of Medicine): We’re in the midst of probably the biggest revolution in biology that is going to forever transform the way we understand genetics, environment, the way the two interact, what causes disease. It’s another level of biology, which, for the first time, really, is up to the task of explaining the biological complexity of life.

NARRATOR: Ghost in Your Genes, up next on NOVA.

Major funding for NOVA is provided by the following:

For each of us, there is a moment of discovery. We understand that all of life is elemental, and as we marvel at element bonding with element, we soon realize that when you add the human element to the equation, everything changes. Suddenly all of chemistry illuminates humanity and all of humanity illuminates chemistry. The human element: nothing is more fundamental, nothing more elemental.

And by David H. Koch, and…

Discover new knowledge: HHMI.

And by the Corporation for Public Broadcasting, and by contributions to your PBS station from viewers like you. Thank you.

NARRATOR: In the early 1990s, the biggest project ever undertaken in biology captivated the world.

NEWS AUDIO: The Human Genome Project will be seen as the outstanding achievement in the history of mankind.

NARRATOR: The endeavor would reveal the chemical structure of each gene locked within our cells, the blueprint for life itself.

WOLF REIK (The Babraham Institute): The human genome is like a bible where everything was written down. The hope and the expectation was that once we had that book in front of us, and all the letters, we could just read down the pages and we would understand how the body was put together.

NARRATOR: Once the code was deciphered, scientists hoped to find the genetic cause and cure for every disease. They estimated that the human genome, the book of life, would contain around 100,000 genes.

MICHAEL SKINNER (Washington State University): And then when they started sequencing…and it popped down to 60. And then it popped down to 50. And, slowly, it went down to a much smaller number.

NARRATOR: Thirty thousand, twenty-five thousand…as the mapping drew to an end, it appeared that humans had about the same number of genes as fish and mice.

MICHAEL SKINNER: In fact, we found out that the human genome is probably not as complex and doesn’t have as many genes as plants do. So that, then, made us really question, “Well, if the genome has less genes in this species versus this species, and we’re more complex potentially, what’s going on here?”

NARRATOR: So few genes didn’t appear enough to explain human complexity. Even more startling, it turned out the same key genes that make a fruitfly, a worm or a mouse also make a human. Chimpanzees share 98.9 percent of our genome. So what accounts for the vast differences between species? Might genes not be the whole story?

Long before the genome was mapped, geneticists, like Marcus Pembrey, had caught hints of this possibility as they encountered baffling genetic conditions—Angelman syndrome, for instance.

MARCUS PEMBREY (University College London): …named after Harry Angelman, the pediatrician who first described Angelman syndrome. He referred to them as “happy puppet children,” because this described, to some extent, the features. They have a rather jerky sort of movement when they’re walking. These children have no speech; they are severely incapacitated in terms of learning but are uncharacteristically happy, and they’re smiling all the time.

NARRATION: The condition is caused by a genetic defect. A key sequence of DNA is deleted from chromosome 15.

MARCUS PEMBREY: Then we came across a paradox. At the same time, the same change, the same little deletion of chromosome 15, had been clearly associated with a quite different syndrome—much milder in terms of intellectual impairment—the Prader-Willi syndrome.

These children are characterized by being very floppy at birth, but once they started eating properly and so on, they then had an insatiable appetite and would get very, very large.

NARRATOR: Pembrey was stunned. Angelman syndrome and Prader-Willi syndrome, two completely different diseases, were caused by the same genetic abnormality.

MARCUS PEMBREY: So here we’re in a bizarre situation really. How could one propose that the same deletion could cause a different syndrome?

NARRATOR: As Pembrey looked at the inheritance pattern for the two conditions, he noticed something even stranger.

MARCUS PEMBREY: What really mattered was the origin of the chromosome 15 that had the deletion. If the deletion was on the chromosome 15 that the child had inherited from father, then you would have Prader-Willi syndrome, whereas if the deletion was inherited from the mother, you had the Angelman syndrome.

NARRATOR: It was a complete surprise that the same missing strip of DNA, depending upon its parental origin, could cause different diseases. It was as if the genes knew where they came from.

MARCUS PEMBREY: You’ve got a developing fetus manifesting this condition. How does the chromosome 15 know where it came from? There must have been a tag or an imprint placed on that chromosome, during either egg or sperm formation in the previous generation, to say, “Hi, I came from Mother.” “I came from Father, and we are functioning differently.” So that’s the key thing, that although the DNA sequence is the same, the different sets of genes were being silenced depending on whether it came from the mother or from the father.

NARRATOR: It was the first human evidence that something other than genes passed between generations. Something that could control genes directly, switch them on or off. But how exactly did these tags go about silencing a gene?

This odd strain of agouti mice provides a visual clue. Despite the difference in color and size, they’re twins, genetically identical. Both, therefore, have a particular gene, called agouti, but in the yellow mouse it’s switched on all the time.

RANDY JIRTLE (Duke University Medical Center): As a consequence, it inappropriately blocks a receptor in what’s called the satiation center of the brain, which tells mice and us when we’re full. So the yellow animals literally eat themselves into obesity, diabetes and cancer.

NARRATOR: So what switched the agouti gene off in the thin mouse? Exercise? Atkins? No, a chemical tag called a methyl molecule. Composed of carbon and hydrogen, it affixes near the agouti gene, shutting it down. Living creatures possess millions of tags like these. Some, like methyl molecules, attach to DNA directly. Other types grab the proteins called histones, around which DNA wraps, and tighten or loosen them to turn genes on or off.

JEAN-PIERRE ISSA (M.D. Anderson Cancer Center): And, in simple terms, this contact can be thought of as hugging the DNA. And if these proteins hug the DNA very tightly, then it is hidden from view for the cell. And a gene that is hidden cannot be utilized.

NARRATOR: These tags and others control gene expression through a vast network in the body called the epigenome.

RANDY JIRTLE: Epigenetics literally translates into just meaning above the genome. So if you would think, for example, of the genome as being like a computer, the hardware of a computer, the epigenome would be like the software that tells the computer when to work, how to work, and how much.

JEAN-PIERRE ISSA: Perhaps the best example of an epigenetic phenomenon…you’re actually looking at it. You see, skin and eyes and teeth and hair and organs all have exactly the same DNA. You cannot genetically tell my skin from my eyes or my teeth, yet you couldn’t really imagine that these are the same tissues.

NARRATOR: What distinguishes cells is not their genes, but how these genes are switched on or off by epigenetics.

WOLF REIK: And, as development unfolds, certain switches need to be thrown. And you can think of it as a light switch. Switch on the gene, the light is shining, the gene is active… makes the cell do a certain thing. Or the light switch is off, everything is dark. That gene is off.

And as the cells divide, the memory of whether it’s a liver cell or a brain cell, that’s brought about by these switches. And the switches are incredibly stable.

NARRATOR: But occasionally, some epigenetic switches can be flipped. To turn off the overactive agouti gene, researchers gave pregnant mothers foods rich in vitamins like B-12, or folic acid, from which they could make those methyl tags that silence genes.

The change was small, the effect huge. Fat yellow mothers gave birth to thin brown pups no longer prone to disease.

RANDY JIRTLE: This study, why it is so important is it opened the black box up and told us that this early stage of development—in the womb, basically—is linked to adult disease susceptibilities by, literally, tiny little changes in the epigenome.

NARRATOR: Agouti mice revealed the impact of an epigenetic change, one that occurred without altering a single chemical letter in the agouti gene.

It was increasingly clear that genes needed instructions for what to do, when and where. If the thousands of genes identified by the Human Genome Project symbolized the words in the book of life, it was the epigenome that determined how that book got read.

MARK MEHLER: We thought that by understanding the genetic code, we would understand life, disease, and then we’d all go home and be fine. But, in fact, the human genome project was just the beginning. What it did was it opened us up to this new world, getting us to the point where we’re understanding another level of biology which, for the first time, is up to the challenge of the biological complexity of life.

NARRATOR: If the epigenome controls the expression of our genes, could it solve the mystery of identical twins?

These rare individuals are living illustrations of the boundary point between nature and nurture. For, since their DNA is 100 percent the same, any difference should reveal the influence of the outside world.

Most identical twins appear so similar they seem the product of genes alone. Consider Javier and Carlos. Their every gesture seems the same. Or take Ana Mari and Clotilde, who show up in nearly the same red dress when, in fact, neither had a clue what the other was going to wear. They moved through life in symmetry.

CLOTILDE: When I see my sister, I see myself. If she looks good, I think, “I look pretty today.” But if she’s not wearing makeup, I say, “My god, I look horrible.”

NARRATOR: But, five years ago, symmetry appeared to break. Ana Marie was diagnosed with cancer and Clotilde was not.

CLOTILDE: I’ve been told that I am a high risk for cancer. Damocles’ sword hangs over me.

NARRATOR: In fact, it’s not unusual for one twin to get a dread disease while the other does not. But how? How can two people so alike, be so different?

Intrigued by the mystery, Spanish geneticist Manel Esteller set out, in 2005, to find the answer to that question.

MANEL ESTELLER (Spanish National Cancer Center): One of the questions of twins is, “If my twin has this disease, I will have the same disease?” And genetics tell us that there is a high risk of developing the same disease. But it’s not really sure they are going to have it, because our genes are just part of the story.

NARRATOR: Esteller suspected epigenetics was the rest. To find out, he and his team collected cells from 40 pairs of identical twins, age three to 74.

Then began the meticulous process of dissolving the cells until all that was left were the wispy strands of DNA, the master molecule that contains our genes. Next, researchers amplified fragments of the DNA, revealing both the genes and their epigenetic tags.

Those that had been turned off appear as dark pink marks on the gel. Now, notice what happens when these genes are cut out and overlapped. The epigenetic effects stand out, especially when you contrast the genes of two sets of twins who differ in age.

Here, on the left, is the overlapped DNA of six-year-old Javier and Carlos. The yellow indicates where their genes are functioning identically.

On the right, is the DNA of 66-year-old Ana Mari and Clotilde. In contrast to the younger twins’, hardly any yellow shines through. Their genome may be the same, their epigenome clearly is not.

Identical genes active in one twin maybe shut down in the other. Thus, as the years pass, epigenetic changes accumulate in twins, as in the rest of us.

MANEL ESTELLER: One of the main findings of our research is that epigenomes can change in function of what we eat, of what we smoke or what we drink. And this is one of the key differences between epigenetics and genetics.

NARRATOR: But why does the epigenome change, when the genome does not?

In Montreal, scientists Michael Meaney and Moshe Szyf believe the question contains its own answer.

MOSHE SZYF (McGill University): We have this very, very static genome, very hard to change. It could be only changed by really dramatic things like nuclear explosions or, you know, hundreds of thousands of years of evolution. On the other hand, we have the dynamic environment that changes all the time. And so what there is here is an interface between the highly dynamic world around us and the highly static genome that we have. Epigenome is an in-between creature, built in a way, to respond to changes around us.

NARRATOR: Szyf and Meaney believe that experience itself changes the epigenome. To reach this startling conclusion they studied two kinds of rats: those born to nurturing mothers who licked and groomed them intensely after birth, and those born to mothers who took a more paws-off approach.

MICHAEL MEANEY (Douglas Institute/McGill University): What we were particularly interested in is the way in which these animals might respond to stressful events. And we found the offspring of low-licking mothers, during periods of stress, show greater increases in blood pressure and greater increases in stress hormone production.

MOSHE SZYF: They will scream. They will try to bite you. Just walking into their cage, those rats will respond differently.

NARRATOR: To rule out a genetic cause, high-licking mothers were given the babies of low-licking ones and vice versa. Once again, the less-nurtured pups grew up markedly different, and not only on blood tests.

MOSHE SZYF: So the conclusion from that is, it’s not the genes that the mother brings into the game. It is the behavior of the mother that has an impact on the offspring years after the mother is already gone. And the basic question was, “How does the rat remember what kind of care it received from its mother, so that it now has better or worse health conditions?”

And we reasoned that there must be some mark in genes that marks that memory.

NARRATOR: But could such a mark, capturing memory, be found? The researchers focused on a gene which lowers the levels of stress hormones in the blood. It’s active in a part of the rat’s brain called the hippocampus. By extracting and analyzing the gene, they could compare how its activity varied between low- and high-licked rats.

The difference was striking. Less nurtured rats had multiple epigenetic marks silencing the gene.

The result? With the gene less active, stress levels in neglected rats soared. In stark contrast, nurtured rats could better handle stress because they had nothing dimming the genes’ activity.

MOSHE SZYF: The maternal behavior essentially sculpted the genome of their babies. We looked at one gene; we know hundreds of genes were changed. But for me, it was a fantastic thing that just a behavior of one subject can change the gene expression in a different subject.

NARRATOR: The most surprising phase of the experiment, however, was yet to come. Szyf and Meaney injected anxious rats with a drug known to remove epigenetic marks.

MOSHE SZYF: And as we injected the drug, the gene turned on. And when it turned on, the entire behavior of the rat changed. It became less anxious. Also, it responded to stress like a normally-reared rat. And we looked at the way that gene was marked in the brain, and we saw that we actually changed the epigenetic marking of that gene.

NARRATOR: Although the work has yet to be replicated, it appears that Szyf and Meaney have linked personality traits, albeit in a rat, to the epigenome.

Could this have implications for humans? We will not know until the completion of a 10-year study, now underway, that will look at children from both nurturing and neglected backgrounds.

But even now, says Meaney, we have clues that our own upbringings produce the same effects.

MICHAEL MEANEY: If you grow up in a family that involves abuse, neglect, harsh and inconsistent discipline, then you are statistically more likely to develop depression, anxiety, drug abuse. And I don’t think that surprises anyone. But what is interesting is that you are also more likely to develop diabetes, heart disease and obesity. And the stress hormones actively promote the development of these individual diseases.

MOSHE SZYF: So, one day, we’ll be able, perhaps, to chart the pathway from child abuse to changes in the way certain genes are epigenetically marked in the brain that unfortunately affect our health years later in life.

NARRATOR: This work is controversial. Still, many scientists now believe that epigenetic changes in gene expression may underlie human diseases.

Take a disorder like M.D.S., cancer of the blood and bone marrow. It’s not a diagnosis you would ever want to hear.

SANDRA SHELBY (Medical patient): When I went in, he started patting my hand and he was going, “Your blood work does not look very good at all,” and that I had M.D.S. leukemia, and that there was not a cure for it, and, basically, I had six months to live.

NARRATOR: With no viable treatment, Sandra entered a clinical trial experimenting with epigenetic therapy. It was the result of a radical new way of thinking about the causes of diseases like cancer.

JEAN-PIERRE ISSA: If one has a genetic basis of cancer in mind, then one is simply asking, “What causes genetic damage?” Cigarette smoking, certain types of environmental exposures and radiation causes genetic damage. But, now if I come in and say, “Well, wait a minute, epigenetic damage can also cause cancer,” then you’ve got to ask “Well, why does this come about?”

NARRATOR: The trouble begins, believes Issa, when our stem cells, the master cells that create and replace our tissues, overwork.

JEAN-PIERRE ISSA: Every time a stem cell has to repair injury, it is aging a little more. And because each time a stem cell divides there is a finite chance of some sort of epigenetic damage, what we find is that in older people there’s been an accumulation of these epigenetic events that is easily measurable in DNA.

Now where does the cancer angle come from? Well, if you count age as how many times a stem cell has divided, then cancers, which copy themselves tirelessly, are awfully old tissues.

NARRATOR: As epigenetic errors pile up, the switches that turn genes on and off can go awry, creating havoc within the cell.

ANDREW FEINBERG: There are genes that help to prevent tumors that are normally active that epigenetically become silenced. Those are called tumor suppressor genes. And there are other genes, called oncogenes, that stimulate the growth of tumors. And then the tags, such as the methylation tags, come off those genes, and those genes become activated. So both ways, turning on and turning off, is a way of getting epigenetic disease.

NARRATOR: But could misplaced tags be rearranged? In 2004, Sandra and other patients began taking a drug to remove methyl tags silencing their tumor suppressor genes.

ROY CANTWELL (Medical patient): Your number one thing is, “Okay, is it going work?” And when you know that before this there was nothing, then yeah, it makes you pretty happy that there is a chance to go forward in your life.

NARRATOR: Ironically, the drug, decitabine, was tried in conventional chemotherapy in the 1970s and deemed too toxic. Today, Issa is giving his patients a dose 20 to 30 times lower.

JEAN-PIERRE ISSA: The idea of epigenetic therapy is to stay away from killing the cell. Rather, what we are trying to do is diplomacy, trying to change the instructions of the cancer cells, reminding the cell, “Hey, you’re a human cell. You shouldn’t be behaving this way.” And we try to do that by reactivating genes.

SANDRA SHELBY: The results have been incredible. And I didn’t have, really, any horrible side effects.

ROY CANTWELL: I am in remission, and going in the plus direction is a whole lot better than the minus direction.

NARRATOR: Roy has not been cured, but he has been cancer-free for two years. And he is not alone.

JEAN-PIERRE ISSA: Spectacular results—complete disappearance of the disease—can be seen in almost half of the patients that receive this drug. And 20 years ago we wouldn’t have dreamed that a drug that affects DNA methylation could have such a profound effect on patients.

NARRATOR: As epigenetic therapy takes off, so do the expectations for this new science. Many believe that a multitude of complex diseases, from Alzheimer’s to autism, may have epigenetic triggers.

Consider autism, a mysterious disorder characterized by social withdrawal. This is Bridget. She passes her day running her fingers across her computer screen. Locked in her own world, she has spent the past 13 years drifting apart from her identical twin sister, Jenna.

SUSAN : As infants, they were very much alike. Their physical similarities are obvious. And all their physical milestones happened at the same time. And then, at their first birthday party, we had a big party at the house, lots of balloons, lots of people. And I remember watching Bridget maneuver around the house as if there were nobody there. She was fixated on a balloon, which a lot of babies would be, but something struck me that she was not in tune with everybody that was there.

NARRATOR: Bridget was eventually diagnosed with severe autism. As the girls developed, so did their differences.

SUSAN : Functioning today for Jenna and Bridget…they’re completely different. Jenna is enthusiastic, productive, you know? Jenna’s going to college, talks about it all the time now. Bridget is essentially non-verbal. She doesn’t have purposeful conversational speech, so everything she does say is very memorized because she was taught over and over again.

Do you want grilled cheese?

BRIDGET: Grilled cheese?

SUSAN: Yes or no?

BRIDGET: Yes or no? No.


BRIDGET: Yes, grilled cheese. Yes.

SUSAN: You want grilled cheese, yes?


SUSAN: Good.

There’s no prescription that you get when your child’s diagnosed with autism.

And new things come that weren’t there before, new behaviors that are very problematic, that interfere with her ability to learn anything. So we don’t know, really, what the prognosis is.

NARRATOR: And, for a long time, doctors couldn’t really help. Despite millions of dollars and years of searching, no single definitive autism gene had been found.

But about a decade ago, scientists at the Kennedy Krieger Institute, in Baltimore, turned their high power imagers on the problem. They scanned the brains of both healthy and autistic children, searching for a biological cause of the disorder. One of the researchers involved was Walter Kaufmann.*

WALTER KAUFMANN (Kennedy Krieger Institute): For a long time, people questioned whether autism was a real entity, because the ways to diagnose autism had been behavior…behavioral abnormalities. And those sometimes are difficult to identify in a consistent and reliable way.

NARRATOR: But, in comparing the brain scans of identical twins discordant for autism, Kaufman finally saw the definitive data he was searching for: an area in the brain linked to learning, memory and emotions—called the hippocampus—was smaller in the twin with severe autism. But how could the same genes create different brain structures? Kaufmann asked Andy Feinberg at Johns Hopkins University.

ANDREW FEINBERG: And suddenly we were able to form an epigenetic hypothesis. And that hypothesis is that they have the same genome, but one of them maybe has an epigenetic change that’s leading to a difference in their brain that you don’t see in the other twin.

NARRATOR: Kaufmann and Feinberg are now searching for methyl marks in the DNA of identical twins discordant for autism. The work has just begun, but the hope is that by finding identical genes that differ in their expression, some causes of autism may emerge.

WALTER KAUFMANN: We know environmental stimulation, sensory stimulation, auditory, visual stimulation, have an impact on brain development and brain function. And this impact we know now is mediated, at least in part, by epigenetic mechanisms.

ANDREW FEINBERG: Epigenetic changes…generally they stand at the cornerstone between our genome—in other words, all of our genes, the development of the cells of our body—and the environment that we live in.

NARRATOR: So the environment molds our epigenomes. But might it do more?

At the far speculative edge of this new science, some are seeing evidence of an astonishing possibility, that genes may not be all that passes from generation to generation.

The evidence comes from this Swedish village huddled on the Arctic Circle. Overkalix stands out for one reason, its archives.

Olov Bygren, a Swedish public health expert, has been studying them from more than 20 years. What makes these records unique is their detail. They track births and deaths over centuries—and harvests. This is significant because, in years past, Overkalix’s location left it particularly vulnerable to crop failures and famines.

LARS OLOV BYGREN (The University of Umeå): In the 19th century this was a very isolated area. They could not have help from outside. As it was so poor, they really had a hard time when there was a famine, and they really had a good, good time when the harvests were good.

NARRATOR: Bygren was studying the connection between poor nutrition and health when he stumbled on something curious.

It appeared that a famine might affect people almost a century later, even if they had never experienced a famine themselves. If so, past and future generations might be linked in ways no one had imagined.

Wondering if epigenetics might explain the phenomenon, Bygren sent his research to geneticist Marcus Pembrey.

MARCUS PEMBREY: I was terribly excited to get this, completely out of the blue. And for the first time it seemed that there was some data that we could then start to explore, so that was the beginning of our collaboration.

NARRATOR: Overkalix offered Pembrey a unique opportunity to see if the events that happened in one generation could affect another decades later.

MARCUS PEMBREY: Olly first reported that the food supply of the ancestors was affecting the longevity or mortality rate of the grandchildren, so I was very excited. I responded immediately.

NARRATOR: Pembrey suspected the incidence of one disease, diabetes, might show that the environment and epigenetics were involved. So Olov trawled the records for any deaths due to diabetes and then looked back to see if there was anything unusual about the diet of their grandparents.

MARCUS PEMBREY: A few months later he emailed me to say that indeed they had shown a strong association between the food supply of the father’s father and the chance of diabetes being mentioned on the death certificate of the grandchild.

NARRATOR: In fact, a grandson was four times more likely to die from an illness related to diabetes if his grandfather had plenty of food to eat in late childhood.

This was one of the first indications that an environmental exposure in a man, one that did not cause a genetic mutation, could directly affect his male offspring.

MARCUS PEMBREY: It really did look as if there was some new mechanism transmitting environmental exposure information from one generation to the next.

NARRATOR: Because these ideas were so heretical, Pembrey knew the results could be dismissed as nothing more than a curiosity. To bolster the research, he needed to find out how a trans-generational effect impacted each sex and if it was linked to a specific period of development.

MARCUS PEMBREY: We wanted to tease out when you could trigger, in the ancestor, a trans-generational response.

NARRATOR: So he and Bygren went back to the data. The more they looked, the more patterns started to appear.

MARCUS PEMBREY: We were able to look at the food supply every year, in the grandfather and the grandmother, from the moment they were conceived right through to the age of 20. We found that there are only certain periods in the ancestors’ development when they can trigger this trans-generational response. They’re what one might call sensitive periods of development.

NARRATOR: They discovered that when a famine was able to trigger an effect was different for the grandmother than the grandfather. The grandmother appeared susceptible while she, herself, was still in the womb, while the grandfather was affected in late childhood.

MARCUS PEMBREY: And the timing of these sensitive periods was telling us that it was tied in with the formation of the eggs and the sperm.

NARRATOR: This suggested what might be happening. Perhaps environmental information was being imprinted on the egg and sperm at the time of their formation.

At last a sharper picture was beginning to emerge. The next step was to compile their findings. Bygren drew up a rough diagram and sent it to Pembrey.

MARCUS PEMBREY: Hand-drawn…this is what Olly sent me, you know he was too excited to wait for the thing to be drawn out properly. You know, he sent me the data, and, in fact, I was recovering from having something done on my heart, so he sent it saying, you know, “I hope this helps you get better quickly,” you know? Because it was so exciting.

NARRATOR: When Pembrey looked at the diagram, he was immediately struck by seemingly bizarre connections between gender, diet and health, connections that were most pronounced two generations later. Men, for example, who experienced famine at around age 10, had paternal grandsons who lived much longer than those whose grandfathers experienced plenty. Yet, women who experienced famine while in the womb, had paternal granddaughters who died on average far earlier.

MARCUS PEMBREY: Once I had plotted out the full extent of those results, it was so beautiful and such a clear pattern, I knew then, quite definitely, that we were dealing with a trans-generational response. It was so coherent—and that’s important in science, that the effect was coherent in some way—was tying in when eggs and sperm were being formed.

NARRATOR: The diagram showed a significant link between generations, between the diet in one and the life expectancy of another.

OLOV BYGREN: When you think that you have found something important for the understanding of the seasons itself, you can imagine that this is something really special.

MARCUS PEMBREY: This is going to become a famous diagram, I’m convinced about that. I get so excited every time I see it. It’s just amazing. Every time I look at it, I find it really exciting. It’s fantastic.

NARRATOR: Much about these findings puzzles researchers. Why, for instance, does this effect only appear in the paternal line of inheritance? And why should famine be both harmful and beneficial, depending on the sex and age of the grandparent who experiences it?

Nonetheless, it raises a tantalizing prospect: that the impact of famine can be captured by the genes, in the egg and sperm, and that the memory of this event could be carried forward to affect grandchildren two generations later.

MARCUS PEMBREY: We are changing the view of what inheritance is. You can’t, in life, in ordinary development and living, separate out the gene from the environmental effect. They’re so intertwined.

NARRATOR: Pembrey and Bygren’s work suggests that our grandparents’ experiences effect our health. But is the effect epigenetic? With no DNA yet analyzed, Pembrey can only speculate. But in Washington state, Michael Skinner seems to have found compelling additional evidence by triggering a similar effect with commonly used pesticides. Skinner wanted to see how these chemicals would affect pregnant rats and their offspring.

MICHAEL SKINNER: And so I treated the animals, the pregnant mother, with these compounds, and then we started seeing, between six months to a year, a whole host of other diseases that we didn’t expect. And this ranged between tumors, such as breast and skin tumors, prostate disease, kidney disease and immune dysfunction.

NARRATOR: He checked that there were no genetic mutations and then proceeded to breed the rats.

MICHAEL SKINNER: The next step was for us to go to the next generation. And the same disease state occurs. So after we did several repeats, and got the third generation showing it and then a fourth generation, we sat back and realized that the phenomenon was real. We started seeing these major diseases occur in approximately 85 percent of all the animals of every single generation.

NARRATOR: His discoveries were a revelation.

MICHAEL SKINNER: We knew that if an individual was exposed to an environmental toxin that they can get a disease state, potentially. The new phenomenon is that environmental toxin no longer affects just the individual exposed but two or three generations down the line.

I knew that epigenetics existed. I knew that it was a controlling factor for DNA activity—whether genes are silenced or not—but to say that epigenetics would have a major role in disease development, so…I had no concept for that. The fact that this could have such a huge impact and could explain a whole host of things we couldn’t explain before took a while to actually sink in.

NARRATOR: Further work has revealed changed epigenetic marks in 25 segments of the affected rat’s DNA. The implications, if they apply to humans, are sobering.

MICHAEL SKINNER: What this means, then, is what your grandmother was exposed to when she was pregnant could cause a disease in you—even though you’ve had no exposure—and you’re going to pass it on to your great-grandchildren.

NARRATOR: And if a pesticide can generate such effects, what about stress, smoking, drinking? To some, the epigenetic evidence is compelling enough already to warrant a public note of caution.

RANDY JIRTLE: We’ve got to get people thinking more about what they do. They have a responsibility for their epigenome. Their genome they inherit. But their epigenome, they potentially can alter, and particularly that of their children. And that brings in responsibility, but it also brings in hope. You’re not necessarily stuck with this. You can alter this.

NARRATOR: Might our lifestyle choices resonate down the ages, effecting people yet unborn? Such ideas remain, to say the least, controversial. But one thing many in the field can agree upon is the need to take a cue from the Human Genome Project and launch a similar effort, this time to decipher the epigenome.

ANDREW FEINBERG: Mapping the human epigenome is the most important thing that we could do right now, as a big project in science, because it will tell us some very important things about why organisms function the way they do, why cells have the behavior they do.

JEAN-PIERRE ISSA: We now know how many genes we have, but we really don’t know how they are regulated inside the cells. And mapping the epigenome will give a much better understanding of this particular process.

NARRATOR: The hurdle is that, unlike the genome, which is the same in every cell, the epigenome varies from tissue to tissue, between individuals and over time.

MARK MEHLER: If you thought sequencing the human genome took years and was difficult, you’re talking about levels of complexity that will dwarf anything we knew about the human genome. But it’s crucial, it’s essential. That’s the way that the future is going to unfold. So in a sense, the Human Genome Project was just the beginning.

NARRATOR: The end may be the realization that the code of life is more complex and interactive than we ever imagined.

MARCUS PEMBREY: I’ve thought of nothing else really for the last five years.

It is said, the first time, you know, one had a photograph of the Earth, you know this sort of delicate thing sailing through the universe, you know, it had a huge effect on the sort of “save-the-planet” type of feeling, you know? I’m sure that’s part of why the future generation think in a planetary way, because they’ve actually seen that picture, you know? And this might be the same. It may get to a point where they realize that you live your life as a sort of guardian of your genome. It seems to me you’ve got to be careful of it because it’s not just you. You can’t be selfish because you can’t say, “Well I’ll smoke,” or “I’ll do whatever it is because I’m prepared to die early.” You’re also looking after it for your children and grandchildren. It is changing the way we think about inheritance forever.

*This study was led by the Kennedy Krieger Institute’s Dr. Wendy Kates, who is now (late 2007) at the State University of New York at Upstate Medical University. NOVA would like to thank Dr. Kates for providing access to the autistic twins featured in the program.

On the Ghost in Your Genes Web site, an expert answers viewer questions about the growing field of epigenetics and its potential. Find it on PBS.org.

Educators and other educational institutions can order this or other NOVA programs for $19.95 plus shipping and handling. Call WGBH Boston Video at 1-800-255-9424.

NOVA is a production of WGBH Boston.

Major funding for NOVA is provided by the following:

For each of us, there is a moment of discovery. We understand that all of life is elemental, and as we marvel at element bonding with element, we soon realize that when you add the human element to the equation, everything changes. Suddenly all of chemistry illuminates humanity, and all of humanity illuminates chemistry. The human element: nothing is more fundamental, nothing more elemental.

And by David H. Koch, and…

Discover new knowledge: HHMI.

And by the Corporation for Public Broadcasting, and by contributions to your PBS station from viewers like you. Thank you.

Read Full Post »