Using Evidence Integration Techniques in the Development of Health-Based Occupational Exposure Levels.

Development of toxicology-based criteria such as occupational exposure levels (OELs) are rarely straightforward. This process requires a rigorous review of the literature, searching for patterns in toxicity, biological plausibility, coherence, and dose-response relationships. Despite the direct applicability, human data are rarely used primarily because of imprecise exposure estimates, unknown influence of assumptions, and confounding factors. As a result, high reliance is often placed on laboratory animal data. Often, data from a single study is typically used to represent an entire database to extrapolate an OEL, even for data-rich compounds. Here we present a holistic framework for evaluating epidemiological, controlled in vivo, mechanistic/in vitro, and computational evidence that can be useful in deriving OELs. It begins with describing a documented review process of the literature, followed by sorting of data into either controlled laboratory in vivo, in silico/read-across, mechanistic/in vitro, or epidemiological/field data categories. Studies are then evaluated and qualified based on rigor, risk of bias, and applicability for point of departure development. Other data (eg, in vitro, in silico estimates, read-across data and mechanistic information, and data that failed to meet the former criteria) are used alongside qualified epidemiological exposure estimates to help inform points of departure or human-equivalent concentrations that are based on toxic end points. Bayesian benchmark dose methods are used to estimate points of departure and for estimating uncertainty factors (UFs) to develop preliminary OELs. These are then compared with epidemiological data to support the OEL and the use and magnitude of UFs, when appropriate.

Subchronic Oral Toxicity of Sodium Tungstate in Sprague-Dawley Rats.

The subchronic toxicity of sodium tungstate dihydrate aqueous solution in male and female Sprague-Dawley rats was evaluated by daily oral gavage of 0, 10, 75, 125, or 200 mg/kg/d for 90 days. Measured parameters included food consumption, body weight measurements, hematology, clinical chemistry, and histopathological changes. There was a significant decrease in food consumption and body weight gain in males at 200 mg/kg/d from days 77 to 90; however, there was no effect in food consumption and body weights in females. There were no changes in the hematological and clinical parameters studied. Histopathological changes were seen in kidney of male and female and epididymis of male rats. Histopathological changes were observed in the kidneys of male and female rats dosed at 125 or 200 mg/k/d consisting of mild to severe cortical tubule basophilia in 2 high-dose groups. Histological changes in epididymides included intraluminal hypospermia with cell debris in the 200 mg/kg/d dosed male rats. Histopathological changes were observed in the glandular stomach including inflammation and metaplasia in the high-dose groups (125 or 200 mg/kg/d) of both sexes of rats. Based on histopathology effects seen in the kidneys, the lowest observable adverse effect level was 125 mg/kg/d and the no observable adverse effect level was 75 mg/kg/d in both sexes of rats for oral subchronic toxicity.

Oral toxicity of 3-nitro-1,2,4-triazol-5-one in rats.

3-Nitro-1,2,4-triazol-5-one (NTO), an insensitive explosive, was evaluated to assess potential environmental and human health effects. A 14-day oral toxicity study in Sprague-Dawley rats was conducted with NTO in polyethylene glycol -200 by gavage at doses of 0, 250, 500, 1000, 1500, or 2000 mg/kg-d. Body mass and food consumption decreased in males (2000 mg/kg-d), and testes mass was reduced at doses of 500 mg/kg-d and greater. Based on the findings in the 14-day study, a 90-day study was conducted at doses of 0, 30, 100, 315, or 1000 mg/kg-d NTO. There was no effect on food consumption, body mass, or neurobehavioral parameters. Males in the 315 and 1000 mg/kg-d groups had reduced testes mass with associated tubular degeneration and atrophy. The testicular effects were the most sensitive adverse effect and were used to derive a benchmark dose (BMD) of 70 mg/kg-d with a 10% effect level (BMDL10) of 40 mg/kg-d.

Oral toxicity evaluation of thiodiglycol in Sprague-Dawley rats.

Thiodiglycol (TDG) is the main product of sulfur mustard hydrolysis and is an environmental contaminant. Subacute and subchronic oral toxicity studies with TDG were conducted in Sprague-Dawley rats. Neat TDG was administered by gavage at doses of 157, 313, 625, 1250, 2500, 5000, and 9999 mg/kg/d, 5 days per week, for 14 days. In the 14-day study, decreased body weight and food consumption were observed at 5000 mg/kg/d. In the 90-day study, rats received neat TDG at doses of 50, 500, or 5000 mg/kg/d for 5 days per week. A fourth group served as a sham control. Individual body weight and food consumption were measured weekly. At termination of the experiment, urine, blood, and tissue samples were collected. Rats displayed significant decreased body weight with no effect on food consumption following administration of TDG at 5000 mg/kg/d. Both male and female rats showed significant increased kidney weights at 5000 mg/kg/d. The organ to body weight ratios increased significantly for liver, kidneys, testes, and brain in males and adrenals in females for 5000 mg/kg/d. At all doses of TDG, hematological and clinical parameters and tissue histopathology remained unaltered. The no observed adverse effect level (NOAEL) for oral subchronic toxicity was 500 mg/kg/d. Benchmark dose (BMD) was derived from the decreased gain in body weight that was seen in male rats. A BMD based on a 10% decrease in body weight was 1704 mg/kg/d, and the lower confidence limit on the dose BMD, the BMDL, was 372 mg/kg/d.

Toxicity assessment of 4-amino-2-nitrotoluene.

4-Amino-2-nitrotoluene (4A2NT; CAS 119-32-4) is a degradation product of 2,4-dinitrotoluene. The toxicity data on 4A2NT are limited. Therefore, we collected toxicity data from rats to assess environmental and human health effects from exposures. The approximate lethal dose for both sexes was 5000 mg/kg. A 14-day toxicity study in rats was conducted with 4A2NT in the feed at concentrations of 0, 125, 250, 500, 1000, and 2000 ppm. Based on a 14-day oral dose range toxicity study with 4A2NT in the feed, 2000 ppm was selected as highest concentration for a subsequent 90-day study. An oral 90-day subchronic toxicity study in rats was conducted with concentrations of 0, 500, 1000, or 2000 ppm of 4A2NT in the feed. The calculated consumed doses of 4A2NT in the feed were 0, 27, 52, or 115 mg/kg/d for males and 0, 32, 65, or 138 mg/kg/d for females. A no-observed adverse effect level could not be determined. The lowest observed adverse effect level was 27 mg/kg/d for males and 32 mg/kg/d for female rats based upon decreased body weight gain. The decreased body weight gain in male rats was the most sensitive adverse event observed in this study and was used to derive a benchmark dose (BMD). A BMD of 23.1 mg/kg/d and BMD with 10% effect level of 15.5 mg/kg/d were calculated for male rats, which were used to derive an oral reference dose (RfD). The human RfD of 1.26 µg/kg/d was derived using current United States Environmental Protection Agency guidelines.

Oral bioavailability of cyclotrimethylenetrinitramine (RDX) from contaminated site soils in rats.

Cyclotrimethylenetrinitramine (RDX), a commonly used military explosive, was detected as a contaminant of soil and water at Army facilities and ranges. This study was conducted to determine the relative oral bioavailability of RDX in contaminated soil and to develop a method to derive bioavailability adjustments for risk assessments using rodents. Adult male Sprague-Dawley rats preimplanted with femoral artery catheters were dosed orally with gelatin capsules containing either pure RDX or an equivalent amount of RDX in contaminated soils from Louisiana Army Ammunition Plant (LAAP) (2300 microg/g of soil) or Fort Meade (FM) (670 microg/g of soil). After dosing rats, blood samples were collected from catheters at 2-h intervals (2, 4, 6, 8, 10, and 12) and at 24 and 48 h. RDX levels in the blood were determined by gas chromatography. The results show that the peak absorption of RDX in blood was 6 h for neat RDX (1.24 mg/kg) and for RDX from contaminated soil (1.24 mg/kg) of LAAP. Rats dosed with RDX-contaminated FM soil (0.2 mg/kg) showed peak levels of RDX in blood at 6 h, whereas their counterparts that received an identical dose (0.2 mg/kg) of neat RDX showed peak absorption at 4 h. The blood levels of absorbed RDX from LAAP soil were about 25% less than for neat RDX, whereas the bioavailability of RDX from FM soils was about 15% less than that seen in rats treated with neat RDX (0.2 mg/kg). The oral bioavailability in rats fed RDX in LAAP soil and the FM soil was reduced with the neat compound but decrease in bioavailability varied with the soil type.

Interaction of hydration, aging, and carbon content of soil on the evaporation and skin bioavailability of munition contaminants.

Water plays a key role in enhancing the permeability of human skin to many substances. To further understand its ability to potentially increase the bioavailability of soil contaminants, artificial sweat was applied to excised pig skin prior to dosing with munition-contaminated soils. Skin was mounted in chambers to allow simultaneous measurement of evaporation and penetration and to control air flow, which changed the dwell time of skin surface water within a l-h period post application of test materials. Additional variables included type of compound, aging of spiked soil samples, and carbon content of soil. To this end, the evaporation and skin penetration of C-14 labeled hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), 2,6-dinitrotoluene (26DNT), and 2,4,6-trinitrotoluene (TNT) were determined from two soil types, Yolo, having 1.2% carbon, and Tinker, having 9.5% carbon. RDX soil samples aged 27 mo and 62 mo were compared to freshly spiked soils samples. Similarly, 26DNT samples aged 35-36 mo and TNT samples aged 18 mo were compared to freshly spiked samples. Approximately 10 microg/cm(2) of radiolabeled compound was applied in 10 mg/cm(2) of soil. Radiolabel recovered from the dermis and tissue culture media (receptor fluid) was summed to determine percent absorption from the soils. Radiolabel recovered from vapor traps determined evaporation. Mean skin absorption of all compounds was higher for low-carbon soil, regardless of soil age and skin surface water as affected by air flow conditions. For 26DNT, a simultaneous increase in evaporation and penetration with conditions that favored enhanced soil hydration of freshly prepared samples was consistent with a mechanism that involved water displacement of 26DNT from its binding sites. A mean penetration of 17.5 +/- 3.6% was observed for 26DNT in low-carbon soil, which approached the value previously reported for acetone vehicle (24 +/- 6%). 26DNT penetration was reduced to 0.35% under dryer conditions and to 0.08% when no sweat was applied. When soil hydration conditions were not varied from cell to cell, air flow that favored a longer water dwell time increased penetration, but not evaporation, consistent with a mechanism of enhanced skin permeability from a higher hydration state of the stratum corneum. Profiles of 26DNT penetration versus air flow conditions were exponential for freshly prepared soil samples, suggesting strong and weak binding sites; corresponding profiles of 26DNT penetration from aged samples were linear, suggesting a conversion of weak to strong binding sites. Absorption and evaporation was less than 5% for TNT and less than 1% for RDX, regardless of soil type and age. Fresh preparations of RDX in Tinker soil and aged samples of TNT in Yolo soil showed a significant decrease in skin absorption with loss of surface moisture. The penetration rate of radiolabel into the receptor fluid was highest during the 1-2 h interval after dosing with 26DNT or TNT. High-performance liquid chromatography (HPLC) analysis of 26DNT in receptor fluid at maximum flux indicated no metabolism or breakdown. For TNT, however, extensive conversion to monoamino derivatives and other metabolites was observed. Relatively little radioactivity was found in the dermis after 26DNT and TNT applications, and dermal extracts were therefore not analyzed by HPLC. RDX was not sufficiently absorbed from soils to allow HPLC analysis. This study has practical significance, as the use of water for dust control at remediation sites may have the unintended effect of increasing volatilization and subsequent absorption of soil contaminants. Soil in contact with sweaty skin may give the same result. Skin absorption of 26DNT from soil was over 50-fold higher than the value for dryer skin and over 200-fold higher than the value obtained when there was no sweat application. While the hydration effect was less dramatic for RDX and TNT, soil contaminants more closely matching the physical properties of 26DNT may be similarly affected by hydration.

Developmental toxicity of thiodiglycol in Sprague-Dawley rats.

Thiodiglycol (TG), a hydrolysis product of sulfur mustard (HD), is a potential contaminant of soil and water at certain military sites. To establish developmental toxicity criteria for TG, an oral developmental toxicity study was conducted in Sprague-Dawley rats. Neat thiodiglycol (99.9 %) was administered orally to mated female rats from gestation days (GDs) 5 through 19. The day of positive mating was considered day 0. A pilot study was conducted with TG at dose levels 250, 500, 1,000, 2,000, or 5,000 mg/kg to select suitable doses for the main study. In the main study, three groups of rats (25/group) received TG by gavage at dose levels of 430, 1,290, or 3,870 mg/kg/day. A fourth group served as a sham control. On day 20 of gestation, all females were euthanized and a cesarean section performed. Litters were examined for soft tissue and skeletal alterations. Maternal toxicity was limited to dams receiving TG at 3,870 mg/kg/day. At this dose, body weights and food consumption were reduced during certain periods of gestation. Fetuses derived from those dams exhibited a nonstatistically significant increased incidence of variations when compared to controls. Fetal body weights in the 3,870 mg/kg/day group were significantly lower than controls. There was no increased incidence of anomalies when thiodiglycol-treated fetuses were compared to controls. It was concluded that TG did not produce terata. Developmental toxicity (decreased fetal weights and associated delays in development) occurred only at the maternally toxic dose of 3,870 mg/kg. It appears that 1,290 mg/kg/day could be considered no observed adverse effect level (NOAEL) for oral developmental toxicity. The lowest observed adverse effect level (LOAEL) was 3,870 mg/kg for maternal toxicity.

Genotoxicity assessment of hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX).

Hexahydro-1,3,5-trinitro-1,3,5-triazine, a polynitramine compound, commonly known as RDX, has been used as an explosive in military munitions formulations since World War II. There is considerable data available regarding the toxicity and carcinogenicity of RDX. It has been classified as a possible carcinogen (U.S. Environmental Protection Agency, Integrated Risk Information System, 2005, www.epa.gov/IRIS/subst/0313.htm). In order to better understand its gentoxic potential, the authors conducted the in vitro mouse lymphoma forward mutation and the in vivo mouse bone marrow micronucleus assays. Pure RDX (99.99%) at concentrations ranging from 3.93 to 500 microg/ml showed no cytotoxicity and no mutagenicity in forward mutations at the thymidine kinase (TK) locus in L5178Y mouse lymphoma cells, with and without metabolic activation. This finding was also confirmed by repeat assays under identical conditions. In addition, RDX did not induce micronuclei in mouse bone marrow cells when tested to the maximum tolerated dose of 250 mg/kg in male mice. These results show that RDX was not mutagenic in these in vitro and in vivo mammalian systems.

Toxicity assessment of thiodiglycol.

Sulfur mustard (HD) undergoes hydrolysis to form various products such as thiodiglycol (TG) in biological and environmental systems. TG is a precursor in the production of HD and it is also considered as a "Schedule 2" compound (dual-use chemicals with low to moderate commercial use and high-risk precursors). Several toxicological studies on TG were conducted to assess environmental and health effects. The oral LD(50) values were >5000 mg/kg in rats. It was a mild skin and moderate ocular irritant and was not a skin sensitizer in animals. It was not mutagenic in Ames Salmonella, Escherichia coli, mouse lymphoma, and in vivo mouse micronucleus assays, but it induced chromosomal aberrations in Chinese hamster ovarian (CHO) cells. A 90-day oral subchronic toxicity study with neat TG at doses of 0, 50, 500, and 5000 mg/kg/day (5 days/week) in Sprague-Dawley rats results show that there are no treatment-related changes in food consumption, hematology, and clinical chemistry in rats of either sex. The body weights of both sexes were significantly lower than controls at 5000 mg/kg/day. Significant changes were also noted in both sexes in absolute weights of kidneys, kidney to body weight ratios, and kidney to brain weight ratios, in the high-dose group. The no-observed-adverse-effect level (NOAEL) for oral toxicity was 500 mg/kg/day. The developmental toxicity conducted at 0, 430, 1290, and 3870 mg/kg by oral gavage showed maternal toxicity in dams receiving 3870 mg/kg. TG was not a developmental toxicant. The NOAEL for the developmental toxicity in rats was 1290 mg/kg. The provisional oral reference dose (RfD) of 0.4 mg/kg/day was calculated for health risk assessments. The fate of TG in the environment and soil showed biological formation of thiodiglycalic acid with formation of an intermediate ((2-hydroxyethyl)thio)acetic acid. It was slowly biodegraded under anaerobic conditions. It was not toxic to bluegill sunfish at 1000 mg/L and its metabolism and environmental and biochemical effects are summarized.

Percutaneous absorption of explosives and related compounds: an empirical model of bioavailability of organic nitro compounds from soil.

The percutaneous absorption potentials of (14)C-labeled 2,4,6-trinitrotoluene (TNT), trinitrobenzene, 2,4-dinitrotoluene (2,4-DNT), 2,6-dinitrotoluene (2,6-DNT), 2-amino-4,6-dinitrotoluene, 4-amino-2,6-dinitrotoluene, 2,4-diamino-6-nitrotoluene, 2,6-diamino-4-nitrotoluene, N-methyl-N,2,4,6-tetranitrobenzamine, hexahydro-1,3,5-trinitro-1,3,5-triazine, octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine, and 2,2-thiobis(ethanol) were determined from two soil types, Yolo having 1.9% carbon and Tinker having 9.5% carbon. TNT skin absorption from another low-carbon soil (Umatilla) was also determined. Approximately 10 microg/cm(2) of radiolabeled compound was applied in 5 microl of acetone or 10 mg/cm(2) of soil to excised pigskin mounted in skin penetration-evaporation chambers. Absorption from acetone served as a control. Radiolabel recovered from the dermis and tissue culture media (receptor fluid) was summed to determine the percentage of absorption from the soils. For each compound, percentage absorptions of radiolabel were highest from acetone solution and lowest from Tinker soil, with the results from Yolo soil being intermediate. Skin absorptions of TNT from Yolo and Umatilla soils were similar. For TNT in all vehicles, the penetration rate of radiolabel into the receptor fluid was highest during the 1- to 2-h interval after dosing. HPLC analysis of TNT radiolabel in receptor fluid at maximum flux indicated extensive conversion to monoamino derivatives and other metabolites. For 2,4-DNT and 2,6-DNT applications in Yolo soil, percentage absorptions approached those obtained from acetone applications. After 2,4-DNT and 2,6-DNT applications (acetone and soils), HPLC analysis of radiolabel in receptor fluid during the period of maximum flux revealed no significant metabolites. Skin absorption of the nitro compounds from soils was found to correlate with the compound's water solubility and vapor pressure. These findings formed the basis of an empirical model to predict skin bioavailability.