Thomas F. Grafton

Lack of embryotoxicity of homocysteine thiolactone in mouse embryos in vitro.

Recent work from humans and chick embryos has suggested that homocysteine may play a role in producing neural tube defects (NTDs). In an effort to determine if homocysteine is able to produce NTDs in mammalian embryos, mouse embryos were explanted on GD 8 and cultured for 44 h. When either homocysteine or homocysteine thiolactone was added to the culture medium, treated embryos developed as well as controls and had closed neural tubes. Homocysteine thiolactone was also microinjected into the amniotic sac of mouse embryos. Again, development proceeded normally with no significant increase in the number of embryos with open neural tubes at the end of the culture period. HPLC analysis of embryonic thiols 24 h after microinjection revealed a significant increase in embryonic cystathionine levels. These data suggest that homocysteine does not produce NTDs in mouse embryos cultured in vitro and that early organogenesis-stage embryos are able to metabolize homocysteine.

Alterations in maternal plasma corticosterone levels following treatment with phenytoin

The mechanisms of the embryotoxic effects of the anticonvulsant drug, phenytoin (PHT), are unknown. Glucocorticoids and PHT demonstrate similar embryopathic effects and strain sensitivity in that A/J mice are very sensitive to the embryopathic effects of synthetic glucocorticoids and PHT while C57BL/6 (B6) mice are comparatively resistant to both. It is possible that teratogenic consequences of PHT are not a result of drug interaction at the target site but are mediated indirectly by glucocorticoids. In this study, PHT was administered by intraperitoneal injection at 25 (a nonteratogenic dose) or 75 mg/kg body weight (a teratogenic dose) to pregnant A/J mice on Day 10 of gestation. Mice of the B6 strain received the drug at 75 mg/kg on Day 10. Control mice received vehicle (pH 11.0 distilled water). Dams were killed at various times after the injection; plasma samples were obtained, and corticosterone levels were determined by radioimmunoassay. In control animals, maternal plasma corticosterone levels were elevated soon after dosing but gradually declined, except for an apparent circadian rhythm effect seen in samples obtained in the afternoon. Administration of a nonteratogenic dose of PHT to A/J mice caused a temporary increase in plasma corticosterone levels which decreased to the control level between 6 and 24 hr following dosing. Treatment with a teratogenic dose in A/J mice led to plasma levels that remained elevated for the entire 48-hr period examined in this study. In B6 mice, treatment with 75 mg/kg increased plasma corticosterone levels for 24 hr, after which they declined to the control value by 30 hr. The adrenal corticosteroid response of A/J mice to PHT appeared to be much more sensitive than that of B6 mice, and there appeared to be a relationship between plasma levels of PHT and corticosterone. The lengthy increase in plasma corticosterone during organogenesis may be a factor in the increased incidence of cleft lip and palate seen after administration of PHT to A/J mice.

In vitro embryotoxicity of carbamazepine and carbamazepine-10, 11-epoxide.

Carbamazepine (Tegretol, CBZ) is an anticonvulsant drug that is very effective in the treatment of tonic-clonic seizures and is gaining acceptance as a treatment for various psychiatric disorders. The drug is embryotoxic in rodents and has been reported to produce neural tube defects in approximated 1% of prenatally exposed human offspring. It is metabolized by the cytochrome P-450 system to a stable, pharmacologically active epoxide intermediate, carbamazepine-10, 11-epoxide. It is currently unknown whether the parent compound, the epoxide intermediate or some other metabolite is the embryotoxic agent. The present study was designed to determine the embryotoxicity of CBZ and its epoxide intermediate (CBZ-E) in a rodent whole embryo culture system. Rat embryos were cultured beginning on day 9 of gestation (GD 9), and mouse embryos were cultured beginning in GD 8. All embryos were cultured for 48 hr in medium containing various concentrations of either CBZ or CBZ-E. Mice were more sensitive to the effects of CBZ than were rats. The parent compound was embryotoxic to mouse embryos at concentrations as low as 12 micrograms, but it was only embryotoxic at 60 micrograms/ml to rat embryos. CBZ-E was not embryotoxic to either species at concentrations as high as 48 micrograms/ml. These results suggest that the parent compound is the embryotoxic agent and that the epoxide intermediate plays no role in the drugs embryotoxic mechanism.

Lack of attenuation of valproic acid-induced embryotoxicity by compounds involved in one-carbon transfer reactions

Previous reports have indicated that valproic acid (VPA) is a developmental toxicant in vitro as well as in vivo, producing primarily neural tube defects. The mechanism for the drugs embryotoxic effects is unknown; however, work from our laboratory over the last few years has indicated that addition of various folate derivatives did not significantly decrease the incidence of VPA-induced neural tube defects. It is possible that some compound involved in one-carbon transfer reactions other than folate could protect against VPA embryotoxicity. We have examined the ability of l- or d-serine, sodium formate and l-methionine to decrease VPA-induced embryotoxicity. CD strain rat embryos were cultured for 48 hr beginning on day 9 of gestation. Each compound was examined for embryotoxicity, and a non-embryotoxic concentration was added to the culture medium 3 hr before the addition of 150 mug VPA/ml. Neither l-serine (300 mug/ml), d-serine (300 mug/ml), sodium formate (600 mug/ml) or l-methionine (1.12 mg/ml) were able to decrease VPA-induced developmental toxicity in vitro. The morphological scores of the embryos were not increased, nor was the frequency of embryos with open neural tubes decreased by treatment with any of these compounds. These data suggest that the mechanism for VPA-induced embryotoxicity does not involve compounds involved in one-carbon transfer reactions.

Antisense modulation of 5,10-methylenetetrahydrofolate reductase expression produces neural tube defects in mouse embryos

The role of folate metabolism in producing neural tube defects (NTDs) in humans is unknown. In the current study, antisense oligodeoxyribonucleotide technology was utilized to disrupt normal expression of the gene for 5,10-methylenetetrahydrofolate reductase (MTHFR) in organogenesis-stage mouse embryos. Two different antisense probes were microinjected into the amniotic sac of gestation day (GD) 8 mouse embryos with PBS or scrambled sense oligodeoxyribonucleotides injected into control embryos. Concentration-dependent increases in the frequencies of embryos with NTDs were observed for both antisense sequences. The level of mRNA for MTHFR was decreased in embryos treated with the higher concentration of one antisense sequence, indicating that the sequence is able to decrease gene expression. 5-methyltetrahydrofolate, the product of the MTHFR reaction, was able to decrease the incidence of antisense-induced NTDs, but co-injection with L-methionine did not. These results suggest that reduced expression of MTHFR may play a role in producing NTDs.

Partial attenuation of hydroxyurea-induced embryotoxicity by deoxyribonucleotides in mouse and rat embryos treated in vitro

Hydroxyurea (HU) is a well known developmental toxicant in all animal species tested. It inhibits DNA synthesis, and addition of deoxycytidine monophosphate (dCMP) has been shown previously to attenuate the developmental toxicant effects of HU in vivo. The purpose of the present investigation was to determine whether addition of deoxyadenosine monophosphate (dAMP) or dCMP would attenuate HU-induced embryotoxicity using a rodent whole embryo culture system. Rat embryos were removed on the morning of day 10 of gestation, and mouse embryos were removed on day 8 of gestation (day 0 = the day a vaginal plug was found). Embryos were treated with various concentrations of HU (up to 500 mug/ml) for 1 hr at 37 degrees C before being washed and cultured for 43 hr in rat serum containing dAMP or dCMP. At the end of the culture period, six endpoints were evaluated for each viable embryo: morphological score; number of somite pairs; crown-rump and head lengths, DNA and protein contents. HU (300 mug/ml) significantly decreased values for all endpoints in embryos from both mice and rats; however, mouse embryos appeared to be more sensitive to the effects of the drug. dAMP and dCMP alone produced some embryotoxicity at high concentrations. The combination of HU + dAMP had no consistent beneficial effect in rat embryos and no effect on mouse embryos. The combination of HU + dCMP improved growth and development slightly. As determined by HPLC analysis, HU treatment (300 mug/ml for 1 hr) decreased all nucleotide pools. Subsequent treatment with dAMP increased all pool levels, although these levels remained below those of control embryos. These results suggest that the developmental toxicant effects of HU are not due solely to alterations in deoxyribonucleotide pool levels.