Stability of the human sperm DNA methylome to folic acid fortification and short-term supplementation.

STUDY QUESTION: Do short-term and long-term exposures to low-dose folic acid supplementation alter DNA methylation in sperm? SUMMARY ANSWER: No alterations in sperm DNA methylation patterns were found following the administration of low-dose folic acid supplements of 400 µg/day for 90 days (short-term exposure) or when pre-fortification of food with folic acid and post-fortification sperm samples (long-term exposure) were compared. WHAT IS KNOWN ALREADY: Excess dietary folate may be detrimental to health and DNA methylation profiles due to folate's role in one-carbon metabolism and the formation of S-adenosyl methionine, the universal methyl donor. DNA methylation patterns are established in developing male germ cells and have been suggested to be affected by high-dose (5 mg/day) folic acid supplementation. STUDY DESIGN, SIZE, DURATION: This is a control versus treatment study where genome-wide sperm DNA methylation patterns were examined prior to fortification of food (1996-1997) in men with no history of infertility at baseline and following 90-day exposure to placebo (n = 9) or supplement containing 400 µg folic acid/day (n = 10). Additionally, pre-fortification sperm DNA methylation profiles (n = 19) were compared with those of a group of post-fortification (post-2004) men (n = 8) who had been exposed for several years to dietary folic acid fortification. PARTICIPANTS/MATERIALS, SETTING, METHODS: Blood and seminal plasma folate levels were measured in participants before and following the 90-day treatment with placebo or supplement. Sperm DNA methylation was assessed using the whole-genome and genome-wide techniques, MassArray epityper, restriction landmark genomic scanning, methyl-CpG immunoprecipitation and Illumina HumanMethylation450 Bead Array. MAIN RESULTS AND THE ROLE OF CHANCE: Following treatment, supplemented individuals had significantly higher levels of blood and seminal plasma folates compared to placebo. Initial first-generation genome-wide analyses of sperm DNA methylation showed little evidence of changes when comparing pre- and post-treatment samples. With Illumina HumanMethylation450 BeadChip arrays, no significant changes were observed in individual probes following low-level supplementation; when compared with those of the post-fortification cohort, there were also few differences in methylation despite exposure to years of fortified foods. LARGE SCALE DATA: Illumina HumanMethylation450 BeadChip data from this study have been submitted to the NCBI Gene Expression Omnibus under the accession number GSE89781. LIMITATIONS, REASONS FOR CAUTION: This study was limited to the number of participants available in each cohort, in particular those who were not exposed to early (pre-1998) fortification of food with folic acid. While genome-wide DNA methylation was assessed with several techniques that targeted genic and CpG-rich regions, intergenic regions were less well interrogated. WIDER IMPLICATIONS OF THE FINDINGS: Overall, our findings provide evidence that short-term exposure to low-dose folic acid supplements of 400 µg/day, over a period of 3 months, a duration of time that might occur during infertility treatments, has no major impact on the sperm DNA methylome. STUDY FUNDING/COMPETING INTERESTS: This work was supported by a grant to J.M.T. from the Canadian Institutes of Health Research (CIHR: MOP-89944). The authors have no conflicts of interest to declare.

Reproductive epigenetics.

Epigenetics refers to covalent modifications of DNA and core histones that regulate gene activity without altering DNA sequence. To date, the best-characterized DNA modification associated with the modulation of gene activity is methylation of cytosine residues within CpG dinucleotides. Human disorders associated with epigenetic abnormalities include rare imprinting diseases, molar pregnancies, and childhood cancers. Germ cell development and early embryo development are critical times when epigenetic patterns are initiated or maintained. This review focuses on the epigenetic modification DNA methylation and discusses recent progress that has been made in understanding when and how epigenetic patterns are differentially established in the male and female germlines, the mouse, and human disorders associated with abnormalities in epigenetic programming in germ cells and early embryos, as well as genetic and other modulators (e.g. nutrition and drugs) of reproductive epigenetic events.

II. Characterization and development of the regional- and cellular-specific abnormalities in the epididymis of mice with beta-hexosaminidase A deficiency.

Beta-hexosaminidase (Hex) is a lysosomal enzyme that exists as two isoenzymes: Hex A (subunit structure alphabeta) and Hex B (betabeta). Its presence in the testis and epididymis suggests important roles for Hex and its substrates in male fertility and reproductive functions. Disruption of the Hexa gene encoding the alpha-subunit of Hex has led to the generation of a mildly affected mouse model of human Tay-Sachs disease, allowing us the opportunity to analyze the effects of isolated Hex A deficiency on epithelial cellular morphology of the male reproductive tract. At 5 weeks and at 3, 5, and 12 months, the testes, efferent ducts and epididymides of Hex A-deficient (Hexa -/-) and wild-type (Hexa +/+) mice were perfuse fixed and analyzed by routine light and electron microscopy as well as with immunocytochemistry employing antibodies to lysosomal enzymes. In the testis, the seminiferous epithelium of Hexa -/- mice appeared comparable to that of wild-type mice in appearance and topographical arrangement of its cell types at all ages examined. Also, no differences were noted for the efferent ducts. In contrast, there were striking abnormalities in the epididymides of the mutant mice; however, the abnormalities were mainly restricted to the initial segment and intermediate zone. Principal cells of these regions at 5 weeks showed a dramatic increase in the number of lysosomes as compared with those from wild-type animals, and this progressed with increasing age. Furthermore, unlike the few small lysosomes present in wild-type mice, those of Hexa -/- mice were at times enlarged and often filled the supranuclear and basal regions of these cells. In the light microscope, large, dense cellular aggregates were noted at the base of the epithelium in the proximal initial segment that corresponded in the electron microscope to two different cell types, both of which increased in size with age. One aggregate was considered to belong to narrow cells on the basis of the presence of numerous cup-shaped vesicles characteristic of these cells; they appeared to be dislocated from the upper half of the epithelium. In the distal initial segment and intermediate zone, narrow cells were readily identified, but rather than being slender as in the control animals, they were greatly enlarged and filled with pale lysosomes in mutant mice. The second type of cellular aggregate noted in the proximal initial segment corresponded to halo cells. They contained numerous small and large lysosomes and small, Golgi-related, dense, core granules characteristic of halo cells. On the basis of the large size of these cells, they appeared to be actively internalizing substances from the intercellular space. In contrast, principal and clear cells of the caput, corpus, and cauda regions did not appear to show a significant increase in number or size of lysosomes as compared with those of wild-type animals. All structures identified as lysosomes in the various cell types were immunoreactive for cathepsin D. The present data thus reveal that isolated Hex A deficiency results in region- and cell-specific abnormalities in the epididymis but in no apparent abnormalities in the testis or efferent ducts. Specific roles for Hex A that cannot be compensated for by other isozymes of Hex appear to exist within lysosomes of epithelial cells predominantly of the initial segment and intermediate zone. Taken together, the results also suggest that the inability to degrade endocytosed substrates normally acted upon by Hex A in lysosomes of principal and narrow cells leads to their accumulation, eventual fusion, and increased size.

Differential expression of ornithine decarboxylase, poly(ADP)ribose polymerase, and mitochondrial mRNAs following testosterone administration to hypophysectomized rats.

The mRNAs of the nuclear encoded genes, ornithine decarboxylase (ODCase) and poly(ADP)ribose polymerase (PADPRP), and the mitochondrial encoded genes, cytochrome oxidase I and II (COI and COII) and ATPase 6, are differentially expressed during spermatogenesis (Alcivar et al., 1989: Biol Reprod 41:1133; 1989: Dev Biol 135:263; 1991: Biol Reprod 46:201). In this study, we use Northern blotting to examine the steady state levels of ODCase, PADPRP, COI, COII, and ATPase 6 mRNAs in testes of hypophysectomized male rats following testosterone administration. Four weeks after hypophysectomy, rats received 24 cm subcutaneous implants of testosterone-filled polydimethylsiloxane (PDS) and were killed at 3, 7, 14, 28, and 56 days thereafter. After hypophysectomy, the steady state levels for the PADPRP, COI, COII, and ATPase 6 mRNAs were not significantly different from controls, although hypophysectomy caused a 44% loss of preleptotene spermatocytes and an 88% loss of pachytene spermatocytes, the testicular cell types expressing the highest levels of these mRNAs. In contrast, the levels of the two ODCase mRNAs were greatly decreased after hypophysectomy and mirrored the number of germinal cells present in the testis. After testosterone treatment, ODCase mRNA levels remained low 3 days after treatment and gradually increased at days 14, 28, and 56. No major hybridization signal changes in PADPRP, COI, COII, and ATPase mRNA were observed after testosterone treatment. We conclude that the steady state mRNA levels for the housekeeping ODCase gene respond differently after hypophysectomy and testosterone treatment of male rats than the PADPRP and mitochondrial DNA transcripts.

Temporal gene expression is restored concomitantly with germ cells in the experimentally regressed rat testis.

The present study was designed to examine the effect of hypophysectomy and subsequent testosterone administration on germ cell numbers and germ cell- and Sertoli cell-specific mRNA levels in adult rats. Rats were hypophysectomized and 4 weeks later received 24-cm testosterone-containing polydimethylsiloxane (PDS) implants. Sham-hypophysectomized rats received an empty PDS implant. At 0 and 3 days, and at 1, 2, 4, and 8 weeks, rats were killed. One testis from each rat (n = 4/group) was used to prepare total RNA; the other testis was used to enumerate stage VII-VIII germ cells. cDNA probes for germ cell and Sertoli cell products were used to monitor germ cell- and Sertoli cell-specific mRNAs on Northern blots. Four weeks after hypophysectomy (0 days), preleptotene and pachytene spermatocytes and round and elongating spermatids were reduced in number to 54%, 12%, 1%, and 0%, respectively, of the control values. Testosterone administration caused a time-dependent increase in germ cell numbers; after 8 weeks of testosterone treatment, preleptotene and pachytene spermatocytes and round and elongating spermatids were 75%, 79%, 74%, and 22%, respectively, of control values. Lactate dehydrogenase-C, phosphoglycerate kinase-2, protamine-1, and sulfated glycoprotein-2 mRNA levels (on a per micrograms RNA basis) were 34%, 34%, less than 1%, and 580% of control values, respectively, 4 weeks after hypophysectomy and 79%, 87%, 61%, and 192% of control values, respectively, after 8 weeks of testosterone treatment. Pachytene spermatocyte and round spermatid numbers increased, while Sertoli cell sulfated glycoprotein-2 mRNA levels decreased, with respect to 4 week hypophysectomy values, as early as 3 days after implantation of testosterone capsules. In contrast, germ cell (lactate dehydrogenase-C, phosphoglycerate kinase-2, and protamine-1) mRNA levels increased to the greatest extent between 1-4 weeks after the start of testosterone treatment and, after a short lag period, reflected increases in germ cell type and number. The results indicate that cell-specific mRNAs appear concomitantly with germ cell reappearance in a time-dependent manner in the testes of testosterone-treated hypophysectomized adult rats.

DNA methyltransferase is developmentally expressed in replicating and non-replicating male germ cells.

Genomic methylation patterns are established during maturation of primordial germ cells and during gametogenesis. While methylation is linked to DNA replication in somatic cells, active de novo methylation and demethylation occur in post-replicative spermatocytes during meiotic prophase (1). We have examined differentiating male germ cells for alternative forms of DNA (cytosine-5)-methyltransferase (DNA MTase) and have found a 6.2 kb DNA MTase mRNA that is present in appreciable quantities only in testis; in post-replicative pachytene spermatocytes it is the predominant form of DNA MTase mRNA. The 5.2 kb DNA MTase mRNA, characteristic of all somatic cells, was detected in isolated type A and B spermatogonia and haploid round spermatids. Immunobolt analysis detected a protein in spermatogenic cells with a relative mass of 180,000-200,000, which is close to the known size of the somatic form of mammalian DNA MTase. The demonstration of the differential developmental expression of DNA MTase in male germ cells argues for a role for testicular DNA methylation events, not only during replication in premeiotic cells, but also during meiotic prophase and postmeiotic development.

Mitochondrial gene expression in male germ cells of the mouse.

Inheritance of the mitochondrial genome is known to be exclusively maternal. To determine whether the loss of paternal mitochondria could be due to a deficiency of RNA in the spermatozoal mitochondria, the expression of mitochondrial genes was studied in testicular cells at various stages of spermatogenesis and in epididymal spermatozoa. The presence of mitochondrial transcripts was examined by Northern blot analysis using probes for the following mitochondrially encoded genes: 12 S and 16 S ribosomal RNAs and a group of mRNAs including cytochrome oxidase subunits I and II (COI-COII), cytochrome b (cyt b), adenosine triphosphatase (ATPase) subunits 6 and 8, and subunit 1 of the respiratory chain NADH dehydrogenase (ND1). Comparison of total testicular RNA preparations from prepuberal (6, 8, 12, 16, 18, 20, 22, and 30 days old) and sexually mature (45 days old) mice revealed no major qualitative or quantitative differences in the levels of the mitochondrial transcripts described above. Similar results were observed from enriched preparations of type A and B spermatogonia and interstitial cells obtained from the testes of 8-day-old mice. Transcripts for COI-COII, ATPase 6, or ND1 were reduced in amount in the enriched preparations of pachytene spermatocytes, round spermatids, and residual bodies when compared to the amount in total testis or liver RNA. Transcripts of all the mitochondrial genes analyzed were present in RNA preparations isolated from sperm midpiece tails obtained after sonication of epididymal spermatozoa. These studies demonstrate that (a) during testicular development the levels of mitochondrial RNA in total testicular extracts show no major qualitative and quantitative differences; (b) the mitochondrial transcripts in enriched populations of type A and type B spermatogonia are not different from those obtained from total testes extracts; (c) mitochondrial transcript levels gradually decrease in enriched preparations of pachytene spermatocytes, round spermatids, and residual bodies; and (d) the mitochondrial rRNAs and mRNAs encoded by several mitochondrial genes can be isolated from sperm midpiece tails.

Effects of cyclophosphamide on selected cytosolic and mitochondrial enzymes in the epididymis of the rat.

The anticancer and immunosuppressive drug cyclophosphamide is extensively used in clinical practice and is known to alter fertility in man. We showed previously that treatment of male rats with low daily doses of cyclophosphamide over a 9-week period caused fetal malformations, a high rate of postimplantation loss and affected epididymal and sperm histology. In the present study, five biochemical measures of epididymal function were used to characterize further the effects of cyclophosphamide on the epididymis. For 1, 3, 6, or 9 weeks, adult Sprague-Dawley rats were gavage-fed daily with saline (control), 5.1 (low dose), or 6.8 (high dose) mg/kg of cyclophosphamide. The specific activities of the two glycolytic enzymes aldolase and lactate dehydrogenase (LDH), the mitochondrial enzyme succinate dehydrogenase, the cytosolic enzyme carnitine acetyltransferase and the lysosomal enzyme acid phosphatase were determined in cytosolic and mitochondrial subcellular fractions from four segments of the epididymis. Cyclophosphamide caused decreases in protein concentrations in all segments of the epididymis only after 6 weeks of treatment with the high dose. The specific activities of aldolase, LDH and succinate dehydrogenase did not differ from control with respect to dose or duration of treatment. In contrast, there were significant effects of cyclophosphamide on carnitine acetyltransferase and acid phosphatase specific activity. After 1 week of treatment, there was a transient dose-related decrease in the specific activity of carnitine acetyltransferase, which was most striking for the corpus epididymidis (76% of control), but which did not differ from control after 3, 6, and 9 weeks. After 6 weeks of treatment with the high dose of cyclophosphamide, carnitine acetyltransferase specific activity in the initial segment and the corpus epididymidis was elevated to 165 and 140%, respectively, as compared with the 1-week high dose values. The specific activity of acid phosphatase did not differ from control after 1 and 9 weeks of treatment. At 3 and 6 weeks, however, there was a dose-related increase in acid phosphatase specific activity for all regions of the epididymis that was most marked in the cauda after the 6-week treatment (140% of control). Therefore, low dose, daily treatment of male rats with cyclophosphamide not only alters specific enzymes in specific segments of the epididymis, but acts in a dose- and time-dependent manner. It is possible that these changes could be mediated by direct, toxic effects of the drug on the epithelium or be secondary to alterations in the spermatozoa as a result of the treatment.

Morphological changes in the testis and epididymis of rats treated with cyclophosphamide: a quantitative approach.

Cyclophosphamide is a widely used anticancer and immunosuppressive drug that affects fertility in men. In a previous study, we found that chronic, daily treatment of male rats with low doses of cyclophosphamide had no apparent effect on the pituitary-gonadal axis, whereas it had time- and dose-dependent effects on male reproductive organ weights, the hematologic system, and on pregnancy outcome. To determine whether cyclophosphamide induces morphological changes within the male reproductive system, a detailed qualitative and quantitative evaluation of changes in the histology of the testis and epididymis was undertaken. Adult male Sprague-Dawley rats were gavage-fed for 1, 3, 6, and 9 wk with saline (control), 5.1 (low dose) or 6.8 (high dose) mg/kg/day of cyclophosphamide; the testes and epididymides were prepared for light and electron microscopy. At the light microscopic level, the orderly process of spermatogenesis in the seminiferous tubules was not affected at any time point with either dose of the drug. A number of time-dependent drug-induced changes in the histology of the epididymis, however, were apparent: 1) an increase in the relative number and a change in the distribution of halo cells in the caput epididymidis, 2) an increase in the number and size of clear cells in the caput and/or cauda epididymidis, and 3) an increase in the size of clear cells in both the caput and cauda epididymides; these changes were time dependent. At the electron microscopic level, there was a dose-dependent, two- to threefold increase in the number of spermatozoa with abnormal flagellar midpieces in the lumen of both the caput and cauda epididymides. Although the 9 plus 2 axonemal complex and the 9 outer dense fibers were present and appeared normal, the close approximation of these two structures was lost in these abnormal spermatozoa. Such abnormal flagellar midpieces were also found in the testes of control and treated rats. Electron microscopic examination of the testis revealed that both Sertoli and Leydig cells were normal in appearance. The type and timing of the effects of cyclophosphamide on the histology of the testis and epididymis suggest that the drug could be affecting germ cells by 1) inducing changes in the developing spermatozoa in the testis, some of which are seen microscopically in the epididymal lumen, and/or 2) affecting epididymal morphology and function.

A time-course study of chronic paternal cyclophosphamide treatment in rats: effects on pregnancy outcome and the male reproductive and hematologic systems.

We have found previously that daily treatment of male rats for 11 wk with low doses of the anticancer drug cyclophosphamide had no apparent effect on male reproductive organ weights, epididymal sperm counts, or serum hormones at the end of the treatment period; yet, upon breeding to untreated females, these males produced a high rate of post-implantation loss and fetal anomalies. The present study was designed to investigate the time course and dose response of the effects of chronic cyclophosphamide treatment on the male reproductive and hematologic systems. Male Sprague-Dawley rats were gavage-fed for 1, 3, 6 and 9 wk with saline (control), or 5.1 (low dose) or 6.8 (high dose) mg/kg/day of cyclophosphamide. After each of the treatment periods, males were mated to determine the effect on pregnancy outcome, then killed, and the effects on the male reproductive and hematologic systems were assessed. After 6 wk of treatment, a sharp increase in mortality was found between the 5.1 and 6.8 mg/kg/day doses of cyclophosphamide. The high dose of cyclophosphamide induced higher levels of pre- and post-implantation loss but fewer fetal anomalies than did the low dose. The low dose of cyclophosphamide did not affect reproductive organ weights; in contrast, the high dose caused decreases in epididymal, ventral prostate, and seminal vesicle weights after 3, 6, and 9 wk. Testicular and epididymal sperm counts were decreased in a dose-dependent manner after 3 wk; in addition, the high dose led to a decrease in epididymal sperm counts after 6 wk of treatment. Another rapidly proliferative tissue, the bone marrow, was dramatically affected by both doses of cyclophosphamide at all time points, with leukocyte counts decreasing to 40% of control by 1 wk. After 9 wk of treatment, effects on the male reproductive system were less marked, compared to earlier time points, whereas those on the hematologic system and pregnancy outcome persisted. Thus chronic low-dose treatment of male rats with cyclophosphamide not only had early and striking effects on the bone marrow and the pregnancy outcome but also affected the male reproductive system in a clear time- and dose-dependent manner.

Chronic low dose cyclophosphamide treatment of adult male rats: effect on fertility, pregnancy outcome and progeny.

Cyclophosphamide is an anticancer and immunosuppressive agent commonly used in men of reproductive age. The relationship between the effects of paternal cyclophosphamide treatment on the male reproductive system and the pregnancy outcome is unknown. To study this relationship, adult male Sprague-Dawley rats were administered saline or cyclophosphamide (1.4, 3.4, and 5.1 mg/kg) daily for 11 wk by gavage. Each male was mated weekly with two females in proestrous; 20 days later, the females were caesarean-sectioned and the number of corpora lutea, resorptions, and normal and abnormal fetuses were noted. After 11 wk of treatment, none of the drug-treated males showed any significant difference compared to controls with respect to male reproductive organ weights, serum testosterone, luteinizing hormone or follicle-stimulating hormone, epididymal sperm counts or fertility. Despite the apparent minimal effects of the treatment regimen on the male reproductive system, there were a number of effects on pregnancy outcome. There was a dose-dependent increase in preimplantation loss at 5-6 wk that was not evident at other times, a progressive dose-dependent increase in postimplantation loss starting at 2 wk, and an increase in malformed and growth-retarded fetuses at 3-4 and 7-9 wk. These results indicate that low dose chronic cyclophosphamide treatment of the male rat can affect the outcome of his progeny; such effects are seen in the absence of any apparent alteration of a number of measures of male reproductive function.

Paternal cyclophosphamide treatment of rats causes fetal loss and malformations without affecting male fertility.

The use of cytotoxic, mutagenic and carcinogenic agents as treatment for various types of cancer may be particularly hazardous in men of reproductive age as there exists the possibility that this may lead to congenital malformations in the progeny. Such agents can affect fertility and other aspects of male reproductive function, for example, treatment with anti-cancer drugs such as cyclophosphamide has been associated with oligozoospermia, azoospermia and increased levels of serum follicle-stimulating hormone (FSH). Depending on the cumulative dose and the duration of treatment, spermatogenesis often returns but this may take years. The relevance of the effects of such chemicals on the male reproductive system to the offspring is poorly understood. We have set out to determine whether present tests of male reproductive function (that is, endocrine status, numbers of spermatozoa, fertility) can predict deleterious effects of a paternally administered agent on the offspring. Here, we report that chronic administration in rats of low doses of the widely used drug cyclophosphamide had minimal effects on the male reproductive system and fertility, but resulted in malformations and retardation of growth in the surviving fetuses and a high frequency of fetal death. Thus, adverse effects on the fetus cannot be predicted from the effects of a drug on the male reproductive system.