2015 Volume 3 Issue 3 Pages 92-107
Methoxyacetate (MAA), formed by the metabolisms of ethylene glycol monomethyl ether (2-methoxyethanol), di-(2-methoxyethyl) phthalate and 1,6-dimethoxyhexane, is known to be a teratogenic and testicular toxicant in experimental animals. MAA is known to inhibit histone deacetylase and is associated with lactate-carrying monocarboxylate transporter expressed in Sertoli and fetal cells. In cells of rapid division, nucleosomal histone exchanges occur through the methyl and acetyl modifications and rates of nucleic acid syntheses are elevated with consumption of cellular energy. These phenomena are considered to associate with MAA-mediated teratogenicity and phase-selective spermatocyte disorders, and also suggest a mutual adverse-outcome pathway in which MAA-mediated histone deacetylase inhibition is involved through p21 activation as the early events. In addition, a possible functional relationship of one-carbon transferring folate/S-adenosyl methionine cycle with testicular metabolisms of sarcosine and creatine is envisioned. Thus, the mechanisms underlying the MAA toxicities will be discussed in relation to the current understanding of the involvement of the epigenetic phenomena and cell-specific metabolisms.
Developmental and embryo toxicities are fields of social safety concern of chemical use. A considerable number of studies have been devoted to understand and to predict the developmental toxicity of chemicals in our environments. Possible relationship of teratogenic properties with chemical structures had been discussed with lipophilic acids1,2), sterols3) and glycol ethers4,5). The mechanistic understanding was, however, limited mostly to endocrine receptor-associated substances. With the progress of developmental biology, cellular targets are now identified with azole pesticides6,7), thalidomide8) and valproic acid (VPA)9,10).
Diverse chemicals are known to cause testicular toxicities. The mechanisms yielding the toxicities are mostly uncertain except for anti-androgenic chemicals.
Methoxyacetate (MAA) exposure has been linked to uses of industrial chemicals such as di-(2-methoxyethyl) phthalate (DMEP)11), ethylene glycol methyl ether (EGME/2-methoxyethanol(2-ME))12) and 1,6-dimethoxyhexane13,14) (Fig. 1). The use of these chemicals is now replaced to respective other derivatives less harmful on the reproductive toxicities. Although toxicities of MAA had been characterized fairly in detail in experimental animals, the mechanisms responsible for embryo- and testicular-toxicities had not yet been well clarified. Studies on a short-chain aliphatic acid, VPA, described above indicate the involvement of epigenetic events, which act combinatorially with trans-acting factors to influence gene expression, on the teratogenic action. These data also suggest the possible association of epigenetic events on the toxicological effects of MAA. Possible mechanisms of MAA toxicities are thus discussed in this context as typical epigenetic-associated events of short-chain aliphatic acid-induced reproductive toxicities to impart clues to understand the molecular mechanisms.
Metabolic pathway to methoxyacetate of 2-methoxyethanol, di-(2-methoxyethyl) phthalate and 1,6-dimethoxyhexane.
Metabolic pathways to lead to formation of methoyacetate (MAA) from 1,6-dimethoxyhexane are not yet reported in detail13). Two pathways of 1,6-methoxyhexane are expected by using a prediction tool of human CYP2E1 metabolism15). In addition, the MAA formation through two consecutive β-oxidations after the initial carboxylate production is proposed for a possible mechanism in this figure.
In a study of [14C]-2-ME administration (250 mg/kg, i.p.), 55% of the dose was recovered in the urine of Sprague-Dawley rats, in which testicular damage was identified as depletion of the spermatocyte population16). The major urinary metabolites were identified by HPLC and isotope dilution analysis, as methoxyacetate (MAA) and methoxyacetylglycine (accounting for 50 to 60% and 18 to 25%, respectively, of urinary radioactivity)17). Toxicokinetic analyses revealed a rapid conversion of 2-ME to MAA (a half-life for plasma disappearance of 2-ME = 0.6 ± 0.03 hours) and gradual clearance of radioactivity (a half-life = 19.7 ± 2.3 hours). Pretreatment of animals with an alcohol dehydrogenase inhibitor, pyrazole (400 mg/kg, i.p.), one hour prior to [14C]-2-ME treatment gave complete protection against the testicular toxicity of 2-ME. Radioactivity detected in the urine was significantly lower from the pyrazole-pretreated groups over 48 hours (18%) than from the group of 2-ME alone. Toxicokinetic analysis revealed almost complete inhibition of the conversion of 2-ME to MAA (a half-life for plasma disappearance of 2-ME = 42.6 ± 5.6 hours) and clearance of radioactivity (a half-life = 51.0 ± 7.8 hours), suggesting the role of alcohol dehydrogenase/aldehyde dehydrogenase system for the toxic metabolite formation18).
After oral administration of MAA (300 mg/kg body weight) in male Fischer 344 rats for eight days, degeneration of testicular germinal epithelia, depletion of thymic cortical lymphoid elements, and reductions in red and white blood cell counts were observed. Some of these observations were apparent to a lesser degree in rats given 100 mg/kg. These toxicological properties of MAA are remarkably similar to those of 2-ME, and thus the adverse effects of 2-ME in rats are linked as the result of biotransformation of 2-ME to MAA19).
Testicular damage was observed 24 hours after a single dose (100 mg/kg) of 2-ME. The lesion appeared localized in the primary spermatocyte. At 16 hours after a single dose of 500 mg/kg, mitochondrial damage was detected as one of the first subcellular changes. Treatment of animals with the ethoxy derivative (ethylene glycol monoethyl ether (EGEE)) resulted in a similar lesion, although higher doses and/or longer periods are necessary to yield the equivalent severity of the damage20,21).
It is generally recognized that DMEP acts after in vivo hydrolysis to 2-ME as a teratogen. 2-ME is in turn metabolized to MAA, the proximate teratogen (Fig. 1). In a system of rat embryo cell culture containing DMEP, a secondary metabolite, MAA, but not of DMEP or its primary metabolites, mono-2-methoxyethyl phthalate (MMEP) and 2-ME, interfered with normal growth and development of organogenesis. These in vitro observations suggested that the reproductive toxicity of DMEP in vivo appeared after the consecutive metabolism, ester cleavage of DMEP and MMEP to 2-ME, followed by oxidation of the latter to MAA in the maternal tissue22). In studies conducted with Wistar rats on day 12 of gestation, teratogenic actions of these three agents, DMEP, 2-ME, and MAA, were equipotent on an equimolar dosage basis. There was also a striking similarity in the defects produced by these agents, mainly hydronephrosis, heart defects, and short limbs and tails. In particular all three agents produced unusual heart defects (dilated ductus arteriosus and dilated aortic arch) as well as ventral polydactyly. These results suggest the occurrence of teratogenic action by a common mechanism or component. In addition, clear protection was observed with 4-methylpyrazole, an alcohol dehydrogenase inhibitor, against 2-ME11). Similar data are also reported in mice. Embryo toxicities appeared after single oral administration of 2-ME (3.3 mmol/kg) or MAA (3.4 mmol/kg) to pregnant Crl:CD-1 ICR BR mice on gestation day 11 and induced digit malformations23).
The embryo toxic effects of 2-ME were studied in non-human primates. Macaca fascicularis females were treated daily throughout the organogenetic phase of pregnancy (days 20–45) by gavage and the fetuses collected at day 100 by Caesarean section. At the highest dose (0.47 mmol/kg), all eight pregnancies ended in death of the embryo. One of these dead embryos was abnormal, missing a digit on each forelimb. At the middle dose (0.32 mmol/kg), three of ten pregnancies ended in embryonic death24).
Therefore, MAA primarily affects tissues containing rapidly dividing cells and high rates of energy metabolism, including the testes, thymus and fetus.
Spermatogenesis is the process of germ cells, in which spermatozoa are produced from male primordial germ cells through mitosis and meiosis. The initial cells are called spermatogonia, which yield primary spermatocytes by mitosis25,26). The primary spermatocyte (diploid) divides into two secondary spermatocytes (Meiosis I) and then each secondary spermatocyte (haploid) divides into two spermatids (haploid) in the process of Meiosis II. The resultant cells develop into mature spermatozoa (haploid), also known as sperm cells. Spermatogenesis is a sequentially controlled and complex process to occur correctly, and is essential for sexual reproduction. DNA methylation and histone modification including acetylation and methylation have been implicated in the regulation of this process.
In spermatocytogenesis, a diploid spermatogonium residing in the basal compartment of the seminiferous tubules, divides mitotically and produces primary spermatocytes. Each primary spermatocyte then moves into the adluminal compartment of the seminiferous tubules.
Spermatidogenesis, which is the spermatid creation process from secondary spermatocytes, is carried out rapidly and is thus rarely seen in histological examinations. During spermiogenesis, the spermatids begin to form a tail through growing microtubules on one of the centrioles, which turns into basal body. These microtubules form an axoneme. The anterior part of the tail (called midpiece) thickens because mitochondria are arranged around the axoneme to ensure energy supply. Spermatid DNA also undergoes packaging, becoming highly condensed. The DNA is packaged first with specific nuclear basic proteins, which are subsequently replaced with protamines during spermatid elongation. The resultant tightly packed chromatin is transcriptionally inactive. The Golgi apparatus surrounds the condensed nucleus, becoming the acrosome. Maturation then takes place under the influence of testosterone, which removes the remaining unnecessary cytoplasm and organelles. Surrounding Sertoli cells in the testes phagocytose the excess cytoplasm known as residual bodies in the testes26).
2 Function of Sertoli CellsAt all stages of differentiation, the spermatogenic cells are in close contact with Sertoli cells which are thought to provide structural and metabolic support to the developing sperm cells27). The function is as follows: maintain the environment necessary for development and maturation via the blood-testis barrier, secrete substances initiating meiosis, secrete supporting testicular fluid, and secrete androgen-binding protein/sex hormone binding globulin (ABP/Shbg) which concentrates testosterone in close proximity to the developing gamete. Testosterone is needed in very high quantities for maintenance of the reproductive tract, and ABP allows a much higher level of fertility. Sertoli cells secrete hormones affecting pituitary gland control of spermatogenesis, particularly the polypeptide hormone inhibin, and also protect spermatids from the immune system of the male via the blood-testis barrier. Sertoli cells are target cells for follicle stimulating hormone (FSH).
Sertoli cells are pivotal to spermatogenesis, providing nutritional support to germ cells throughout their development28). Sertoli cells display atypical features in their cellular metabolism. The cells can metabolize various substrates, preferentially glucose. The majority of glucose is converted to lactate and not oxidized via the tricarboxylic acid cycle. Lactate is utilized as the main energy substrate by developing germ cells and has an antiapoptotic effect on these cells29). Several biochemical mechanisms contribute to the modulation of lactate secretion by Sertoli cells. These include the transport of glucose through the plasma membrane, mediated by glucose transporters, the lactate dehydrogenase-mediated interconversion between pyruvate and lactate, and the release of lactate mediated by monocarboxylate transporters (MCTs). Several factors that modulate Sertoli cell metabolism have been identified, including sex steroid hormones, which are crucial for maintenance of energy homeostasis, influencing the metabolic balance of the whole body28).
Spontaneous germ cell death by apoptosis occurs during normal spermatogenesis in mammals and is thought to play a role in the physiological mechanism limiting the clonal expansion of such cell population in the male gonad. In the prepubertal rat testis, the most conspicuous dying cells are pachytene spermatocytes, which are also the primary target of MAA-initiated apoptosis induced experimentally30).
3 General Profile of Hormonal Regulation of Testis FunctionsSpermatogenesis is initiated upon the pulsated surges of gonadotropin-releasing hormone (GnRH) from the hypothalamus, which leads to the secretion of FSH and luteinizing hormone (LH) produced by the anterior pituitary gland. The release of FSH into the testes will enhance spermatogenesis and lead to the development of Sertoli cells, which act as nursing cells where spermatids will go to mature after Meiosis II. LH promotes Leydig cell secretion of testosterone into the testes and blood. The secretion of FSH and LH (inducing production of testosterone) will stimulate spermatogenesis until the male dies26,31). The production of lactate is thus enhanced in the presence of FSH in Sertoli cells32).
4 Association with Stage of Spermatogenesis of MAA EffectResponses to 2-ME (EGME) and the ethoxy derivative EGEE were examined in primary mixed cultures of Sertoli and germ cells prepared from testes of immature rats, but neither 2-ME nor EGEE produced any morphological changes of toxicity at up to 50 mM for 72 hours. Their metabolites, MAA and ethoxyacetic acid (EAA), however, caused clear degeneration of the pachytene and dividing spermatocytes at 2 to 10 mM for 24 to 72 hours35).
Within the spermatocyte population, differential sensitivity was observed thus depending on the precise stage of meiotic maturation in rats treated with 500 mg/kg of 2-ME33): dividing (stage XIV) and early pachytene (stages I-II) > late pachytene (stages VIII-XIII) > mid-pachytene (stages III-VII). No obvious adverse changes were detected on preleptotene spermatocytes, spermatogonia, and Sertoli cells. Pachytene spermatocyte is at a period of maximal meiotic RNA synthesis34).
EAA was less potent than MAA, and n-propoxyacetate and n-butoxyacetate and methoxyacetylglycine, an in vivo metabolite of MAA, produced no morphological changes under these conditions. The severity of the morphological-changes in culture was paralleled with decreases in the activity of carnitine acetyltransferase and lactate dehydrogenase in the attached germ cell fraction. No evidence was provided for the conversion of 2-ME to MAA or other metabolites or for the further metabolism of MAA in culture35).
5 Androgen Receptor-associated Effects of MAAStage-specific expression of androgen receptor (AR) protein in Sertoli cells was significantly altered in MAA-treated rats as determined by AR immunohistochemistry. High AR expression was found in Sertoli cells coincident with the MAA-induced apoptosis of late-stage pachytene spermatocytes36).
MAA potentiated the AR response without significantly altering the EC50 for androgen responsiveness in the presence of AR antagonists, partially alleviating the antagonistic effect of the anti-androgens. In addition, MAA treatment markedly increased the expression of cyp17a1 (steroid 17α-hydroxylase/17,20-lyase) and ABP/Shbg in a mouse testicular Leydig cell line TM3 while suppressing Igfbp3 (insulin-like growth factor binding protein-3) expression by ~90%. Deregulation of these genes may alter the synthesis and action of androgen in a manner that contributes to MAA-induced testicular toxicity37).
In a mouse TM3 Leydig cell line stably expressing androgen receptor (TM3-AR), several Hox genes, associated with developmental processes, were altered in the presence of MAA with testosterone in a Leydig cell model38).
Exposure to 2-ME or its major metabolite, MAA, results in spermatocyte depletion and testicular atrophy in experimental animals. The site of spermatogenesis is within the seminiferous tubule. Sertoli cells support spermatogenesis by ways of synthesizing and secreting proteins, and of delivering metabolic substrates for differentiating germ cells in the seminiferous tubule lumen. One of these substrates, lactate, is preferentially consumed in spermatocytes.
Exposure to either 2-ME or MAA at 0, 3, or 10 mM for up to 12 hours had no apparent effect on viability of rat Sertoli cell. Lactate levels and rates of lactate accumulation in culture media were, however, significantly decreased in the presence of MAA, but not of 2-ME, at both 3 and 10 mM following incubation for 6, 9, and 12 hours. The results suggested that MAA-mediated inhibition of lactate production in Sertoli cells could account for the inhibitory action toward germ cells on spermatogenesis39).
Round spermatid energy metabolism is closely dependent on the presence of L-lactate in the external medium. Sertoli cells supply this L-lactate in the seminiferous tubules. L-Lactate, in conjunction with glucose, modulates intracellular Ca2+ concentration in round spermatids and pachytene spermatocytes. In the simultaneous measurements of radioactive L-lactate transport and intracellular pH (pHi) changes, Sertoli cells are shown to transport L-lactate using MCT systems40). MCT1 is associated with a trans-membrane protein, basigin, and MCT2 with another trans-membrane protein, embigin, in several cell types. Basigin, but not embigin, is co-localized and co-immunoprecipitated with both MCT1 and MCT2 in sperm. The functional investigation of MCT proteins showed the decrease in pHi with L-lactate. The pHi changes were blocked with a typical MCT-inhibitor, α-cyano-4-hydroxycinnamate, and ATP determination indicates the preference for L-lactate to D-lactate. In testis, MCT1 is located in spermatocytes, spermatids and spermatozoa, whereas MCT2 is predominantly in Sertoli cells and spermatozoa. Sertoli cells secrete lactate through MCT441). These data suggest that basigin interacts with MCT1 and MCT2 to locate them properly in the membrane of spermatogenic cells to enable sperm to utilize lactate as an energy substrate contributing to cell survival42). MCT1, MCT2 and basigin mRNAs are also expressed in human and mouse oocytes and embryos43).
MCTs are now known to belong to a solute carrier gene family, SLC16A to comprise 14 members, of which MCT1 (SLC16A1), MCT2 (SLC16A7) and MCT4 (SLC16A3) are expressed in the testis to catalyze the proton-linked transport of metabolically important monocarboxylates such as lactate, pyruvate and ketone bodies44,45).
MAA, but not sarcosine (N-methylglycine), is known as a better substrate than n-butyrate of monocarboxylate transport system in erythrocytes46), although the molecular entity is not defined. These results suggest the possible role of MCT-mediated transport on MAA delivery to spermatocytes and within embryo tissues. In turn, MAA-mediated inhibition of lactate transport to spermatocytes is likely to contribute on the exacerbation of the cellular function. Relatively long biological half-life of MAA (see section of Metabolism of 2-ME) suggests the possible involvement of slc5a8 (Na+-coupled MCT) for kidney reabsorption of MAA47).
Four SLC2A glucose-transporters, GLUT1, GLUT2, GLUT3 and GLUT8, are expressed in rat testis, but with different ontogenetic controls. The functional role of GLUT8 in the fetus is not yet established28). GLUT1 mRNA was expressed at a low level in the one-day-old testis, but increased rapidly, reaching almost adult levels five days after birth48). In contrast, expression of MCT1 and basigin mRNAs is detected in fetuses of mouse and human43,49). Low expression in fetus and newborn of GLUT transporters may explain partly the nutritional dependence of the testis to lactate than glucose, and high susceptibility toward toxicants altering lactate levels in Sertoli cells at birth50).
Transfer of one carbon group occurs in various metabolic reactions such as choline and methylfolate formation. Among them, urinary creatine and sarcosine levels have been shown to be associated with MAA toxicity.
1 Sarcosine-relatedIn rats, excretion of sarcosine was elevated one day after the treatment with 100 mg/kg 2-ME. Urinary levels of dimethylglycine and sarcosine were still higher at days 4 and 14 in rats treated with 30 and 100 mg/kg 2-ME. Primary metabolic perturbations appeared on the inhibition of choline oxidation, branched-chain amino acid catabolism and fatty acid β-oxidation pathways on the 2-ME administration51).
Both dimethylglycine dehydrogenase and sarcosine dehydrogenase (SDH) are flavoproteins, which catalyze the oxidative demethylation of dimethylglycine to sarcosine and sarcosine to glycine, respectively. During these reactions, tightly bound tetrahydrofolate pentaglutamate (THF) is converted to 5,10-methylenetetrahydrofolate pentaglutamate (5,10-CH2-THF)52) (Folate cycle in Fig. 2). The latter supports DNA synthesis through one-carbon incorporation into thymidine.
Metabolic linkage of tetrahydrofolate and S-adenosylmethionine and enzymes involved for one-carbon transfer
A possible association of histone methylation/demethylation with one-carbon transfer network is shown with a functional linkage of folate and S-adenosylmethionine (AdoMet) metabolisms (Folate cycle and AdoMet cycle). Productions of higher amounts of sarcosine and creatine are expected with increased syntheses of nucleic acid bases in this linkage system through enhanced production of 5,10-CH2-THF (Folate cycle). AdoMet inhibits MTHFR in Folate cycle, while 5-MTHF inhibits GNMT in AdoMet cycle. MAA may reduce production of 5,10-CH2-THF through SDH inhibition.
Chemicals and enzymes are shown in black and grey colors, respectively.
5,10-CH2-THF; 5,10-methylenetetrahydrofolate pentaglutamate, 5-MTHF; 5-methyltetrahydrofolate pentaglutamate, THF; tetrahydrofolate pentaglutamate, 10-formyl-THF; 10-formyl- tetrahydrofolate pentaglutamate, AdoMet; S-adenosylmethionine, AdoHomoCys; S-adenosylhomocysteine, HomoCys; homocysteine, Met; methionine, Arg; arginine, Gly; glycine, Ser; serine, Me-Histone; methylated histone, MAA; methoxyacetate,
GAMT; guanidinoacetate methyltransferase, GNMT; glycine N-methyltransferase, HDM; histone demethylase, HMT; histone methyltransferase, MTHFD; tetrahydrofolate dehydrogenase, MTHFR; methylenetetrahydrofolate reductase, SDH; sarcosine dehydrogenase, cSHMT; cytosolic serine hydroxymethyl transferase.
SDH showed a Km of 0.5 mM for sarcosine in the presence of rat liver preparation. MAA is a competitive inhibitor of SDH with a Ki of 0.26 mM53,54). Thus, accumulation of MAA may affect the level of the active folate in spermatocytes.
Simple physiological compounds such as serine, acetate, sarcosine, glycine, but not D-glucose, ameliorated 2-ME-initiated testicular toxicity. Coadministration of serine completely eliminated 2-ME-induced decreases in the daily sperm production (DSP) in rats on Day 24 following the single dose of 6.6 mmol/kg (500 mg/kg) while glucose was without effect. Acetate, sarcosine, and glycine were of similar efficacy resulting in DSP that was significantly greater than that observed in rats treated 2-ME alone55). Histopathological studies revealed that 2-ME treatment resulted in stage-specific degeneration of late stage pachytene spermatocytes 24 hours after treatment. No apparent degenerative changes occurred after concurrent treatment with serine. Similarly, serine co-treatment also prevented the decreased number of spermatids in the lumina of the seminiferous tubules on Day 24 after 2-ME exposure. All of the compounds utilized in this study are linked to oxidation pathways involving tetrahydrofolic acid as a catalyst for one-carbon moiety transfer into various biological molecules including purine and pyrimidine bases56) (Fig. 2). Serine is directly incorporated by cytosolic serine hydroxymethyl transferase (cSHMT) into 5,10-CH2-THF, which will compensate the reduced supply from SDH-mediated reaction in the presence of MAA. These results are consistent with the idea of pathogenic role of SDH in testis of MAA-treated animals.
In addition to a possible role for nucleoside syntheses, association with tumor regression is suggested on sarcosine. The type I transmembrane protein with epidermal growth factor and two follistatin motifs 2 (TMEFF2) is expressed in the brain and prostate and overexpressed in prostate cancer. Several studies suggest that TMEFF2 gene is hypermethylated in cancer cells57,58). Tumor suppressor activity of TMEFF2 requires the cytoplasmic/transmembrane portion of the protein and correlates with its ability to bind to SDH and to modulate the level of sarcosine59). TMEFF2 is also expressed in mouse embryo60).
Knockdown of glycine-N-methyltransferase (GNMT), the enzyme that generates sarcosine from glycine, attenuated prostate cancer invasion. Addition of exogenous sarcosine or knockdown of SDH induced an invasive phenotype in benign prostate epithelial cells. As described above, sarcosine N-demethylation enhances the formation of 5,10-CH2-THF from THF. On the maintenance of cellular 5,10-CH2-THF levels, S-adenosylmethionine (AdoMet) is an inhibitor of methylenetetrahydrofolate reductase (MTHFR) mediating the conversion of 5,10-CH2-THF to 5-methyl-tetrahydrofolate pentaglutamate (5-MTHF)61). Local 5-MTHF levels will be increased in diminished states of AdoMet, and thus syntheses of methionine from homocysteine (HomoCys) are enhanced (Fig. 2). In addition, 5-MTHF inhibits GNMT, which is favorable to reduce the loss of AdoMet. GNMT has a major role on one-carbon transfer in folate-AdoMet linkage in various tissues61), but trace levels of GNMT are detected in fetal and tumor cells62,63). These results imply that SDH-mediated production may be a salvage pathway for 5,10-CH2-THF supplies in tissues of low GNMT activities. Growth stimulating phenomena observed in sarcosine-treated tumor cells are thus possible to link to enhanced levels of thymidine syntheses through SDH pathway. Enhanced production of sarcosine in testicular tissue may also link with the stimulation of DNA synthesis. Androgen receptor and the ERG (ets-related oncogene) fusion product coordinately regulate components of the sarcosine pathway64).
2 Creatine-relatedIn rats, a single dose of 500 mg/kg of 2-ME in the male, but not in the female, caused a rise in urinary creatine (N-(aminoiminomethyl)-N-methylglycine), together with testicular damages65). Treatment with MAA caused a significant increase in creatine excretion 0–24 hours after doses of 300, 600 and 900 mg/kg, and caused a decrease in creatinine excretion after doses of 600 and 900 mg/kg, within the first 24-hour period after dosing. The creatine:creatinine ratio for the group dosed with 900 mg/kg was still significantly elevated (0.196 ± 0.046; P < 0.05) compared to the control group (0.0414 ± 0.008), 24–48 hours after dosing66). In consistent, increase in urinary creatine excretion was also detected in rats after the administration of 1,6-dimethoxyhexane13).
In biological systems, creatine is metabolized to creatinine prior to the excretion. Creatine also exists as creatine phosphate in tissues. In general, N-phosphocreatine (PCr) was considered to represent simply an energy buffer, through its high-energy N-P bond. Reversible phosphorylation of creatine allows energy to be stored until needed, for the phosphorylation of ADP to ATP. PCr is also possible to contribute in energy metabolism as an energy shuttle. In this idea, creatine is phosphorylated in the mitochondria, and PCr then diffuses through the cytoplasm to distal sites of energy usage, where it is used locally to phosphorylate ADP. The creatine formed then returns to the mitochondria. The shuttle relies upon distinct isozymes of creatine N-phosphoryltransferase, which favor the phosphorylation and dephosphorylation of creatine, respectively67).
Guanidinoacetate methyltransferase (GAMT) catalyzes the last step of creatine biosynthesis in vertebrate animals. As a result of the GAMT-catalyzed reaction, the methyl group of AdoMet is transferred to guanidinoacetate to form creatine and S-adenosylhomocysteine (AdoHomoCys). The high levels of GAMT expression are detected in reproductive tissues68). These results, together with the presence of high levels of creatine and PCr in the male reproductive tract fluids, suggests that creatine may link to the production of 5,10-CH2-THF by way of enhanced supply of THF. Interrelationship of one-carbon supplying cofactors in mammals is shown with typical metabolic reactions (Fig. 2).
DNA and RNA syntheses occur actively during embryogenesis and spermatogenesis. One-carbon donors are required for the biosynthesis of purine and pyrimidine bases56). 10-formyl-tetrahydrofolate pentaglutamate (10-formyl-THF) serves as a donor for purine skeleton synthesis, and 5,10-CH2-THF is used for thymidine synthesis. To support the production of 10-formyl-THF and 5,10-CH2-THF, 5-MTHF taken up from plasma is first transferred to THF and then converted to 10-formyl-THF and 5,10-CH2-THF, respectively, by tetrahydrofolate dehydrogenase (MTHFD)- and serine hydroxymethyl transferase (SHMT)-mediated one-carbon addition reactions. Methionine synthase mediates the metabolism from 5-MTHF to THF, which occurs concomitantly the conversion of HomoCys to methionine (Fig. 2). AdoMet produced from methionine supports the biological methylations such as the biotransformation from glycine to sarcosine and from guanidinoacetate to creatine. Histone and protamine are rich source of arginine and glycine for substrates of methylations. Histone demethylase (HDM) and histone methyltransferase (HMT) utilize also THF and AdoMet as cofactors for their biological histone modifications, respectively.
3 Methyl Transfer-related Changes with Other Testicular ToxicantsA single dose of cadmium chloride (3.23 µmol Cd2+/kg) causing acute testicular damage in male rats also caused significant creatinuria and creatinaemia at 48 hours after dosing. Doses of cadmium, which did not cause testicular necrosis, did not cause creatinuria or creatinaemia. Surgical ligation of the pampiniform plexus also caused ischaemic necrosis of the testis and this was followed by significant creatinuria and creatinaemia. However, neither orchidectomy followed by a toxic dose of cadmium, orchidectomy alone nor sham operation caused significant creatinuria or creatinaemia. Cadmium administration induced a temporary loss of body weight which was less than that caused by food restriction. Food restriction did not cause significant creatinuria but did cause significant creatinaemia. These data suggest that the creatine is derived from the damaged testis and that measurement of urinary creatine may be a useful non-invasive means of detecting acute testicular damage caused by exposure to chemicals or mechanical impairment of blood flow69).
In Sertoli cell-enriched cultures in the presence of FSH, addition of dibutyryl cyclic AMP, mono-2-ethylhexyl phthalate or cadmium increased the secretion of creatine into the incubation medium70). Increased excretions of creatine are also reported in urine of rats treated with di-pentyl phthalate (DPP) and 1,3-dinitrobenzene66).
Co-administration of adenosyl cobalamin, one of the active vitamin B12 analogs, prevented di-(2-ethylhexyl) phthalate (DEHP)-mediated testicular specific changes including fluctuations in testicular weight71). These results may suggest the link of cellular methylation state with DEHP- and DPP-initiated testicular toxicities.
Borate-induced alterations such as epithelial vacuolization, blockage of the tubular lumen and atrophy in testis of mice. Morphometrical data showed that borate induces also enlargement of tubular diameter, epithelial height and tubular lumen72). Borate affects the DNA synthetic activity of both mitotic (spermatogonial) and meiotic (post-spermatogonial) germ cells, suggesting that the DNA synthetic activity may be affected rather than inhibited by the induction of cytotoxic DNA damage73). AdoMet is a versatile cofactor in a variety of physiologic processes. Boron binding with AdoMet is shown to be far greater than by other monoadenosine species tested by electrophoresis74).
These results on both sarcosine and creatine suggest a plausible mechanism in which altered states of cellular one-carbon transfer are involved in testicular toxicities through metabolic changes such as nucleic acid biosynthesis and also through epigenetic events such as methylation/demethylation of protein like histone and transcriptional component.
Within the eukaryotic cell nucleus, genetic information in DNA is organized in a highly conserved structural polymer, termed chromatin, which supports and controls the functions of the genome75). The fundamental repeating unit of chromatin is the nucleosome, which consists of DNA wrapped around an octamer of core histone proteins (an H3-H4 tetramer and two H2A-H2B dimers). Linker histones of the H1 class associate with DNA between single nucleosomes. Core histones are evolutionarily conserved and consist of a globular domain and a flexible charged N-terminal tail, which is covalently modified by different enzymes mainly at specific lysine and/or arginine residues. Those modifications of the histone tail include acetylation, phosphorylation, methylation and ubiquitination.
The packaging of DNA into strings of nucleosomes is one of the mechanisms allowing eukaryotic cells to tightly regulate gene expression. The ordered disassembly of nucleosomes permits RNA polymerase II to access the DNA, whereas nucleosomal reassembly impedes access, thus preventing transcription and mRNA synthesis. Disregulation of nucleosome dynamics results in aberrant transcription initiation, producing non-coding RNAs76).
Methylation/demethylation and acetylation/deacetylation of histones occur during embryo development and spermatogenesis accompanying meiosis and mitosis. Acetylation of histone is generally recognized toward transcriptionally active, while the methylation shifts mostly toward silencing.
During spermatogenesis, histones are largely replaced transiently by transition proteins and subsequently by protamines in postmeiotic cells. However, the mechanisms underlying the replacement of these histones remain unclear77). A study on sirtuins, NAD+-dependent histone deacetylases (HDACs), showed that the mutant mice had smaller testes, a delay in differentiation of pre-meiotic germ cells, decreased spermatozoa number, an increased proportion of abnormal spermatozoa and reduced fertility without apparent changes in acetylation of histone H478). The epigenetic contributions of sperm, but not spermatocyte, chromatin to embryo development are, however, considered highly limited, since nucleosomes are widely replaced by protamine in mature human sperm79).
1 MAA-mediated Modification of Histone and Other Cellular ProteinEmbryonic forelimbs on murine gestation day (GD) 12 were exposed to 3, 10, or 30 mM MAA in culture for six days to examine effects on limb morphology; limbs were cultured for one to twenty-four hours to monitor effects on the acetylation of histones (H3K9 and H4K12) and non-histone protein, p53 (p53K379), and on markers for cell cycle arrest (p21) and apoptosis (cleaved caspase-3). Exposure to MAA resulted in a significant concentration-dependent increase in limb abnormalities. MAA induced the hyperacetylation of histones H3K9Ac and H4K12Ac at all concentrations tested (3, 10, and 30 mM). Exposure to 10 or 30 mM MAA significantly increased acetylation of p53 at K379, p21 expression, and caspase-3 cleavage80).
A reproductive toxicity study of MAA suggests a target protein that is linked with mouse limb teratogenicity81). A single dose of MAA (10 mmol/kg body weight) was given by gavage on GD 11. The pregnant mice were killed at four hours after MAA treatment to examine forelimb buds of embryos. Proteins from forelimb buds GD 11 + 4-hour were analyzed by two-dimensional (2-D) sodium dodecyl sulfate-polyacrylamide gel electrophoresis technique. The 2-D gels reveal one protein with 41.6 kDa and pI 6.4. The primary sequence database search indicates the protein to 34/67-kDa laminin binding protein (LBP; P14206, SwissProt), which is encoded by p40 gene (MGI:105381). The identity was further verified by Western blotting with an antibody against the 67 kDa LBP. The results suggest that MAA treatment to pregnant mice down-regulates the LBP-p40 in the forelimb buds.
Early studies82) clarified the association of LBP-p40 with both the nuclear envelope and chromatin DNA in interphase nuclei, while it is bound only to chromatin DNA in mitosis. A further study83) showed the association of LBP-p40 with histones H2A, H2B, and H4, which confer tight binding of LBP-p40 to chromatin DNA in the nucleus. Chromatin DNA may become unstable by the loss of LBP-p40 as a chromatin anchoring protein. In addition, the expression of the 37/67-kDa high affinity laminin receptor was knocked out with several siRNA constructs through RNA interference in transformed liver cells (Hep3B). As the results, the message was specifically ablated and apoptosis was induced as determined by annexin V/propidium iodide staining, and by double staining with annexin V and an antibody directed against the 37/67-kDa high affinity laminin receptor. These results suggest that this protein plays a critical role in maintaining cell viability31).
Using high-density microarrays, the possible sperm-disrupting mechanisms of MAA (650 mg/kg i.p.) were assessed84). MAA treatment caused increased death of pachytene spermatocytes starting eight hours post-exposure and increasing dramatically at 12 and 24 hours post-exposure. The specific levels of tetraacetyl histone H4 (4acHIST1H4) and of diacetyl histone H3 (2acHIST1H3) in testis nuclear protein were elevated at 4, 8, and 12 hours post-exposure of MAA. Consistent with the idea of a compensatory response85), an over-presentation of a gene (histone H1 zero) was also detected in MAA-treated samples. Both 4acHIST1H4 and 2acHIST1H3 were localized primarily to elongating spermatids in testis sections from control animals. At four hours post-exposure, staining for either histone modification was dramatically increased in spermatogonia and all primary spermatocyte populations except for dividing spermatocytes. These results indicate that exposure to MAA causes increased acetylation of core histones in several testis germ cell populations, including those in prophase of meiosis, a large proportion of which die rapidly following this treatment. In consistent with MAA-related alterations of nucleosomes, clear changes of specific miRNAs were observed in testicular tissues of rats86) and monkey87) treated, respectively once with 2,000 mg/kg p.o. and 300 mg/kg p.o. for four days of EGME.
To assess the influence of MAA on cancer cells88), two immortalized human normal prostatic epithelial cell lines (RWPE-1 and pRNS-1-1) and four human prostate cancer cell lines (LNCaP, C4-2B, PC-3, and DU-145) were treated with MAA at different concentrations and for different time periods. MAA dose-dependently inhibited prostate cancer cell growth through induction of apoptosis and cell cycle arrest at G1 phase. MAA-induced apoptosis was associated with down-regulation of the anti-apoptotic gene baculoviral inhibitor of apoptosis protein repeat containing 2 (BIRC2, also named cIAP1), leading to activation of caspases 7 and 3, and turning on the downstream apoptotic events. MAA-induced cell cycle arrest (mainly G1 arrest) was due to up-regulation of p21 expression at the early time and down-regulation of cyclin-dependent kinase 4 (CDK4) and CDK2 expression at the late time. MAA up-regulated p21 expression through inhibition of HDAC activities, independently of p53/p63/p73. These findings demonstrate that MAA suppresses prostate cancer cell growth through inducing growth arrest and apoptosis. Up-regulation of p21 is also shown as a core regulated gene involved predominantly in cell cycle/apoptosis and DNA synthesis in response to HDAC inhibition in other cells89,90).
Laminin receptor 1 (37LRP/RPSA) is human equivalent of murine LBP/p-40. Down-regulation of RPSA is observed in human esophageal squamous cancer cells T.Tn and TE2 after the treatment with a HDAC inhibitor FK228 having a bicyclic depsipeptide structure91).
Critical target molecules in toxicities are likely to be distinct in embryos and spermatocytes. MAA studies described above, however, coincide with the idea that MAA-mediated HDAC inhibition results in cell cycle arrest through the enhanced expression of p21 or other unknown factors to lead to apoptosis as early events of both types of toxicities. Consistent with these observations, disruption of both HDAC1 alleles results in embryonic lethality before E10.5 due to severe proliferation defects and retardation in mouse development92). HDAC1-deficient embryonic stem cells show reduced proliferation rates, which correlate with decreased cyclin-associated kinase activities and elevated levels of the cyclin-dependent kinase inhibitors p21 and p27. Similarly, expression of p21 is up-regulated in HDAC1-null embryos. Of course, further experimental studies are necessary to envision the idea on MAA.
2 Links of Histone Deacetylation with Teratogenic and Testicular ToxicantsVPA is able to inhibit HDACs, which have fundamental impact on gene expression and therefore possible molecular targets of VPA-induced signaling cascades10). Only VPA derivatives with a teratogenic potential in mice are able to induce a hyperacetylation in core histone H4 in teratocarcinoma F9 cells93). These experiments clearly demonstrate that the functional inhibition of HDACs is related to the VPA-induced neural tube defects. A structure analogue of VPA, 2-ethylhexanoic acid that is also a metabolite of DEHP, is shown to inhibit HDACs 1, 2 and 7 in a HeLa cell system94).
Pregnant mice were treated i.p. with a teratogenic dose of boric acid (1000 mg/kg, day 8 of gestation)95). Western blot analysis and immunostaining with anti hyperacetylated histone H4 antibody were performed on preparations of embryos explanted 1, 3 or 4 hours after treatment and revealed H4 hyperacetylation at the level of somites. HDAC enzyme assay was performed on embryonic nuclear extracts. A significant HDAC inhibition activity was detected with borate.
Some HDAC inhibitors have been related to teratogenic effects in rodents. Three HDAC inhibitors (apicidin, MS-275 and sodium butyrate) were tested on mouse development and their activities on embryonic histonic and nonhistonic proteins. Pregnant mice were treated i.p. with 10 mg/kg body weight apicidin, 25 mg/kg MS-275, 2000 mg/kg butyrate or with the vehicle alone on day 8 post coitum. Embryos were extracted 1, 2, or 3 hours after treatment and Western blotting (using antibodies against hyperacetylated histone H4, acetylated lysine, or tubulin) and immunohistochemistry on hyperacetylated histone H4 were performed. Fetuses, explanted at term of gestation, were double stained for bone and cartilage to detect skeletal abnormalities. The HDAC inhibitors studied were teratogenic. The specific axial skeletal malformations were fusions or homeotic respecifications. These molecules induced hyperacetylation restricted to somitic histones96).
Rodent exposure to butyl paraben and propyl paraben adversely affected testosterone synthesis and male reproductive function, while the methyl paraben and ethyl paraben had the diminished effects97,98). Alkyl parabens have weak estrogenic activities and their relative potencies on reproductive toxicities are consistent with their uteotropic effects in vitro99). However, controversy exists over the estrogenic potential of parabens and the possible connection with mitochondrial events is proposed100). Butyrate, formed from butyl paraben after the hydrolysis and oxidation, is known as an inhibitor of histone deacetylase. Other mechanisms including mitochondrial energy crisis and histone disorder are thus possible to contribute on the reproductive toxicity.
Doxorubicin is known to have testicular toxicity. Using an isogenic pair of lung adenocarcinoma cell lines; A549 (wild-type) and A549DOX11 (doxorubicin resistant), the roles of epigenetics and miRNA were studied in resistance/response of non-small cell lung cancer cells to doxorubicin. Clear differential expression of epigenetic markers whereby the level of HDACs 1, 2, 3 and 4, DNA methyltransferase, acetylated H2B and acetylated H3 were observed and lower in A549DOX11 compared to A549 cells. Specific miRNAs were also dysregulated in A549DOX11 cells compared to A549 cells101).
Experiments of knockout mice lacking HDAC genes indicate highly specific functions for the individual isoforms during developments. Class I and IIa HDACs are shown to be associated with heart development, DNA repair, skeletogenesis, skeletal muscle and endothelial function102). Sirtuins (SIRTs) are class-III NAD-dependent HDACs involving in various physiological processes including energy metabolism and life span103). Inactivation of SIRT1 in the mouse leads to male sterility. Fetal testis development appears normal in SIRT1 null mice. In contrast, the first round of spermatogenesis discontinues before the completion of meiosis with abundant apoptosis of pachytene spermatocytes abnormal Leydig and Sertoli cell maturation, and strongly reduced intratesticular testosterone levels. This phenotype is shown to be the consequence of diminished expression of hypothalamic GnRH and of reduced LH levels, rather than having an intrinsic effect on male germ cells per se, SIRT1 thus regulates spermatogenesis at postnatal stages through controlling hypothalamus-pituitary gonadotropin signaling104). These data indicate the diverse epigenetic mechanisms for the embryo development and spermatogenesis.
In past, chemical-initiated developmental and reproductive toxicities had been described mostly in connection with direct hormonal modulations. The data shown above, however, indicate the possible roles of non-classical events such as DNA methylation/demethylation, histone modifications, miRNA, other non-coding RNAs, nucleic acid syntheses and mitochondrial energy modification on chemical-mediated toxicities.
In addition to the biological interaction described above, a pharmacokinetic interaction may possibly occur on MAA-initiated testicular toxicity. If MAA is a good substrate of testicular MCT system in similar to the case with erythrocytes46), MAA may be efficiently delivered, like lactate, from Sertoli cells to germ cells (Fig. 3). This would enhance the chance for MAA to interact with SDH and nucleosomal components and to cause the disorders. Defective supply of 5,10-CH2-THF in target tissue is also depicted as a possible cause of the disorders. These events deteriorating nucleic acid synthesis coincide with developmental-phase selective appearances of MAA toxicity at pachytene spermatocytes and embryo cells. Consistent with this idea, in vitro folate deficiency is shown to induce imbalance of deoxynucleotide pool and apoptosis in Chinese hamster ovary cells106). Increased stimuli for cell division like p21 activation and defects of DNA synthesis occur simultaneously. These discordant events are possible to direct cells of rapid division to evoke an apoptotic signal. Of course, the roles of these events are necessary to be established in cellular and intact organisms.
A possible scheme of cellular metabolic changes to lead to MAA-mediated testicular toxicity
MAA shows slow plasma clearance possibly due to the kidney reabsorption, although MAA is excreted in urine as MAA and the glycine conjugate. Only trace amounts of MAA is considered to enter, if any, in β-oxidation pathway from results of the low extent of the metabolism to carbon dioxide105) and metabolite characterization17).
HDAC; histone deacetylase, MAA; methoxyacetate, MCT; monocarboxylate transporter, THF; tetrahydrofolate pentaglutamate, 5,10-CH2-THF; 5,10-methylenetetrahydrofolate pentaglutamate.
This review focuses on the mechanism of MAA-mediated embryo- and testicular-toxicities, but the view discussed in this context may also be applicable for other toxicants including phthalates to understand the underlying mechanisms of developmental and reproductive toxicities occurring at organs and tissues with rapid cell division. Rapid and extensive progress on the studies of epigenetic-associated cell replication and silencing would help us to understand the molecular mechanisms of chemical-mediated developmental toxicities for the safety evaluations.
