Simulated gastric hydrolysis and developmental toxicity of dibutyltin bis(2‐ethylhexyl thioglycolate) in rats

Previously, dibutyltin dichloride (DBTC) was the putative toxophore for dibutyltin bis‐alkyl and bis‐thio esters. Recent chemical and toxicological data on dioctyltin bis(2‐ethylhexyl thioglycolate) suggest the thioglycolate esters of alkyltins do not generate the dichloride toxophore. Our results, using 119Sn‐nuclear magnetic resonance (NMR) spectroscopy, demonstrated that dibutyltin bis(2‐ethylhexyl thioglycolate) (DBTE) is hydrolyzed to dibutyltin chloro‐(2‐ethylhexyl thioglycolate) (DBTEC) under simulated gastric conditions. No DBTC was detected. DBTE was administered orally to presumed‐pregnant Sprague–Dawley rats in a corn oil vehicle at 2.5, 8.5, and 25.0 mg/kg/day (Gestation Day 5 [GD5] through GD19). There were no maternal deaths, no treatment‐related statistically significant reductions in feed consumption, maternal body weight or weight gain, or adverse gestational outcomes. Maternal thymus weight was significantly reduced in rats at 25 mg/kg. There were no effects on fetal growth, no dose‐dependent pattern of external, visceral, or skeletal malformations, and no increase in anatomical variations. Based on the obtained experimental data, it is concluded here that DBTE forms DBTEC, not DBTC, in the stomach, and DBTE was not teratogenic nor fetotoxic in rats, a species sensitive to DBTC. The maternal no‐observed‐adverse‐effect level (NOAEL) was 8.5 mg/kg/day, and the developmental NOAEL was 25 mg/kg/day, the high dose. The maternal LOAEL was 25 mg/kg/day based on reduced maternal thymus weight.


| INTRODUCTION
Dibutyltin substances have been studied extensively for their toxicological effects. As with several other alkyltin derivatives, the critical toxicological target of dibutyltin dichloride (DBTC), in rats, is the lymphoid cells of the thymus (Snoeij et al., 1988). This was observed with tributyltin oxide, tributyltin chloride (Bressa et al., 1991), and with dioctyltins (Miller et al., 1984;Seinen & Willems, 1976). The effect was consistently histologically characterized as lymphocyte depletion in the thymus and thymus-dependent lymphoid areas of the spleen and lymph nodes (Penninks et al., 1985). Curiously, this effect appears to be an acute effect and reversible. A single dose of DBTC in rats led to a depression in thymic weight that was maximal at 4-day postexposure and recovered fully to normal levels by Day 9 postexposure (Snoeij et al., 1989). Relative thymus weight of untreated pregnant and lactating female rats was 46% of age-matched dioctyltin dichloride (DOTC)-treated nonpregnant and nonlactating females. By postnatal Day 56, thymus weight and immune function had recovered (Menke et al., 2012). Certain butyltin substances were also shown to cause developmental effects. Developmental toxicity testing of DBTC (Ema et al., 1991) revealed an increased incidence of morphological effects; among them, a somewhat unique malformation, ankyloglossia, was produced in rat fetuses following gestational exposure to DBTC.
This work was extended to determine the most sensitive period of gestation (Ema et al., 1992;Ema & Harazano, 2000) and later included testing of dibutyltin dimaleate and dibutyltin dilaurate as well as dibutyltin oxide (Noda et al., 1993).
There were two lines of evidence that led to the hypothesis that DBTC was a putative metabolite of dibutyltin esters and the proximate toxophore for the thymic toxicity and the developmental effects.
Second, the ankyloglossia observed in fetuses exposed to DBTC (Ema et al., 1992) was also observed after exposure on Gestation Day 8 to equimolar doses of dibutyltin alkyl esters (80 μmol/kg on GD8), which supported the hypothesis that the esters were hydrolyzed to DBTC in vivo to produce the adverse fetal outcomes (Noda et al., 1993).
Existing reviews by national (Ghobrial et al., 2019;Swedish Chemicals Agency, 2018)  Water was provided ad libitum throughout the experimental period in plastic water bottles with stainless steel sipper tubes. Deepbore well water was passed through an activated charcoal filter and exposed to ultraviolet rays in the Aquaguard water filter purifier.
During mating, three rats (one male and two females) were housed in polypropylene breeding cages. The day the presence of sperm in a vaginal smear was confirmed was designated Gestation Day 0 (GD0). Mated pairs were then separated. Presumed-pregnant females were housed individually in polypropylene cages. There were 100 rats randomly assigned to each of four groups, 25 presumedpregnant females per group. Dosing solutions were prepared and administered daily by oral gavage on GD5 through GD19.
Fetuses were removed on GD20 by cesarean section. After the 4 h of contact with the acidic buffer, the hydrolysate of the starting material was extracted from the aqueous phase and characterized by 119 Sn-NMR spectroscopy. The results are summarized in Table 1 and Figure 1A-D. The 119 Sn-NMR spectrum of the recovered metabolites showed a decrease of approximately 65% of the signal at 72-74 ppm attributed to DBTE, whereas the DBTEC signal at 31-33 ppm increased by the same amount. A signal for DBTC would have been expected at 129 ppm and was not observed. The assay has a limit of detection for DBTC at less than 0.01% (Hansen, 2016). The data strongly support the conclusion that DBTC was not produced from DBTE in the acid-treatment protocol used.

| Maternal and developmental outcomes
The doses administered to rats in the main study were 0, 2.5, 8.5, and 25 mg/kg bw/day of DBTE. These doses were based on data from an identically designed range-finding study at doses of 0, 2.5, 8.5, 20, 35, 50, and 75 mg/kg bw/day in which a clear, statistically significant, treatment-related adverse effect on maternal weight occurred at 35 mg/kg bw/day and maternal mortalities were observed at 50 (2/10) and 75 mg/kg bw/day (6/10) (Figure 3). A probit analysis of maternal weight from the range-finding study predicted that the dose of 25 mg/kg bw/day was the maximal tolerable (defined as 80% of control weight from the range-finding study).
Maternal animals in the main study showed no treatment-related signs of toxicity. There were no mortalities nor statistically significant differences in feed consumption, maternal body weight, weight gain, or corrected weight gain across dose groups (Figures 4 and 5).
The lower thymus weight in the 25 mg/kg bw/day group was 67% of the control and was deemed a statistically significant treatment-related reduction. This reduction correlated with the reduced size of the thymus noted during the gross pathology examination and correlated with the microscopic findings of decreased cellularity in the cortex of the thymus of animals in this dose group The maternal fertility and reproductive parameters of gravid uterine weight, the number of corpora lutea, resorptions, and dead fetuses were within historical control limits for all treated groups. Calculated incidences of preimplantation and postimplantation loss were also within historical control limits for all treated groups (Table 3).
Fetal evaluations at term indicated mean litter size, sex ratio, mean fetal body weight, and mean fetal C-R length were not statistically different from controls for any DBTE group.
No increased incidence of external or visceral abnormalities was noted within any group. The observed soft tissue variations and the skeletal findings are common in rat fetuses, and there were no statistically significant differences. Skeletal observations were generally single incidences within a litter in both controls and in DBTE-treated groups, and there were no statistically significant differences.

| DISCUSSION
Read-across is often used in regulating chemical groups to fill data gaps, reduce animal use, and allow endpoint-specific toxicological data from data-rich structurally similar compounds to apply to those endpoints for a different group member for which data are lacking.
Previous hydrolysis studies of DBTE (Bautista & Herzig, 2000;Gillard-Factor & Yoder, 2000;Yoder, 2000b) used gas chromatography equipped with flame ionization detection (FID), to measure 2-ethylhexyl thioglycolate (EHTG) and 2-ethylhexanol (EH), hydrolysis breakdown products of DBTE. This analytical method did not allow for direct determination of the tin moieties. Deficiencies of this method were attributed to interference between the DBTE-derived EHTG and EH peaks, which had almost the same retention time.
The developmental toxicity studies with DBTE reported here support the conclusion that DBTE is not teratogenic nor is it a developmental toxicant in rats. These data suggest DBTE has a different reproductive outcome than either DBTC or several of the dibutyltin carboxylates, for example, the acetate and the maleate, which have been directly examined in the past (Noda et al., 1993). When compared to the World Health Organization Globally Harmonized System (WHO GHS) classification criteria, these data in conjunction with the implications of the simulated gastric hydrolysis strongly support a conclusion that DBTE should not be classified as a developmental toxicant.