interaction of hydration, aging, and carbon content of soil on the evaporation and skin...

486

Journal of Toxicology and Environmental Health, Part A, 71: 486–494, 2008Copyright © Taylor & Francis Group, LLCISSN: 1528-7394 print / 1087-2620 online DOI: 10.1080/15287390801906956

UTEHInteraction of Hydration, Aging, and Carbon Content of Soil on the Evaporation and Skin Bioavailability of Munition Contaminants

Skin Absorption of Munition ContaminantsWilliam G. Reifenrath1, Harold O. Kammen1, Gunda Reddy2, Michael A. Major2, and Glenn J. Leach2

1Stratacor, Inc., Richmond, California, and 2U.S. Army Center for Health Promotion and Preventive Medicine, Aberdeen Proving Ground, Edgewood, Maryland, USA

Water plays a key role in enhancing the permeability of humanskin to many substances. To further understand its ability topotentially increase the bioavailabililty of soil contaminants, arti-ficial sweat was applied to excised pig skin prior to dosing withmunition-contaminated soils. Skin was mounted in chambers toallow simultaneous measurement of evaporation and penetrationand to control air flow, which changed the dwell time of skin sur-face water within a l-h period post application of test materials.Additional variables included type of compound, aging of spikedsoil samples, and carbon content of soil. To this end, the evaporationand 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 soilsamples aged 27 mo and 62 mo were compared to freshly spikedsoils samples. Similarly, 26DNT samples aged 35–36 mo and TNTsamples aged 18 mo were compared to freshly spiked samples.Approximately 10 mg/cm2 of radiolabeled compound was appliedin 10 mg/cm2 of soil. Radiolabel recovered from the dermis andtissue culture media (receptor fluid) was summed to determinepercent absorption from the soils. Radiolabel recovered fromvapor traps determined evaporation. Mean skin absorption of allcompounds was higher for low-carbon soil, regardless of soil ageand skin surface water as affected by air flow conditions. For26DNT, a simultaneous increase in evaporation and penetrationwith conditions that favored enhanced soil hydration of freshlyprepared samples was consistent with a mechanism that involvedwater displacement of 26DNT from its binding sites. A mean pen-etration of 17.5 ± 3.6% was observed for 26DNT in low-carbonsoil, which approached the value previously reported for acetonevehicle (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 penetra-tion, but not evaporation, consistent with a mechanism ofenhanced skin permeability from a higher hydration state of thestratum corneum. Profiles of 26DNT penetration versus air flow

conditions were exponential for freshly prepared soil samples,suggesting strong and weak binding sites; corresponding profilesof 26DNT penetration from aged samples were linear, suggestinga conversion of weak to strong binding sites. Absorption andevaporation was less than 5% for TNT and less than 1% for RDX,regardless of soil type and age. Fresh preparations of RDX inTinker soil and aged samples of TNT in Yolo soil showed a signifi-cant decrease in skin absorption with loss of surface moisture.Thepenetration rate of radiolabel into the receptor fluid was highestduring the 1–2 h interval after dosing with 26DNT or TNT. High-performance liquid chromatography (HPLC) analysis of 26DNTin receptor fluid at maximum flux indicated no metabolism orbreakdown. For TNT, however, extensive conversion to monoam-ino derivatives and other metabolites was observed. Relatively lit-tle radioactivity was found in the dermis after 26DNT and TNTapplications, and dermal extracts were therefore not analyzed byHPLC. RDX was not sufficiently absorbed from soils to allowHPLC analysis. This study has practical significance, as the use ofwater for dust control at remediation sites may have the unin-tended effect of increasing volatilization and subsequent absorp-tion of soil contaminants. Soil in contact with sweaty skin maygive the same result. Skin absorption of 26DNT from soil was over50-fold higher than the value for dryer skin and over 200-foldhigher than the value obtained when there was no sweat applica-tion. While the hydration effect was less dramatic for RDX andTNT, soil contaminants more closely matching the physical prop-erties of 26DNT may be similarly affected by hydration.

This study was conducted to determine the influence of skinsurface moisture conditions and soil aging on the in vitropercutaneous penetration of 14C-labeled hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), 2,6-dinitrotoluene (26DNT), and 2,4,6-trinitrotoluene (TNT) from soil types of varying organic con-tent. Soil samples, previously spiked with these compounds,were stored at −20°C for periods up to 5 yr. Percutaneousabsorption from these samples was determined along with thatfrom freshly spiked samples with an in vitro model using pigskin (Hawkins, 1990). Radioactivity in selected samples ofreceptor fluid from 26DNT and TNT exposures was analyzed

Received 1 October 2007; accepted 7 December 2007.Address correspondence to William G. Reifenrath, Stratacor, Inc.,

1315 So. 46th St., Bldg. 154, Richmond, CA 94804, USA. E-mail:[email protected]

SKIN ABSORPTION OF MUNITION CONTAMINANTS 487

by high-performance liquid chromatography (HPLC) to detectmetabolism or breakdown.

To determine the effect of aging on skin absorption, RDXsoil samples prepared 2 and 5 yr previously were compared tofreshly spiked samples. 26DNT soil samples prepared 3 yr agoand TNT soil samples prepared 1.5 yr ago were compared totheir respective freshly prepared samples. To determine theeffect of hydration on skin absorption, artificial sweat wasapplied to the skin surface and an air flow over the skininduced the surface to dry for varying times between sweat andsoil application, thus changing the amount of moisture availablefor uptake into the soil. Alternatively for 26DNT, the timeinterval between sweat and soil application was held constant,but the time for initiating air flow was varied after the applica-tions, thus keeping the soil hydration conditions constant fromcell to cell, but changing the dwell time of moisture availablefor uptake into the stratum corneum.

Skin absorption of contaminants from soil and water hasbeen reviewed (U.S. EPA, 1992, 2001).

METHODS

Chemicals and Test Samples of Soil[14C(U)-Hexahydro-1,3-5-trinitro-1,3,5-triazine (specific

activity 2.86 mCi/mmol) was prepared by NEN Life ScienceProducts, Boston, MA. [Ring-14C(U)-2,6-dinitrotoluene (spe-cific activity 11 mCi/mmol) and [methyl-14C]-2,4,6-trinitrotolu-ene (specific activity 53 mCi/mmol) were prepared by MoravekBiochemicals, Inc., Brea, CA. Radiochemical purities deter-mined by HPLC (Supelcosil LC-18S, 25 cm × 4.6 mm ID,Supelco, Bellefonte, PA; 50% aqueous methanol [1 ml/min]Sigma-Aldrich, St. Louis, MO; ultraviolet [UV] detector at254 nm) following completion of the study were 99% for26DNT and 94% for TNT. Radiochemical purities determinedby thin-layer chromatography (TLC; (silica gel, 250 μm layer,Sigma-Aldrich; methylene chloride: acetonitrile 9:1 for RDX,hexane: ethyl acetate 1:1 for 26DNT and TNT; Instant Imager,Perkin Elmer Life Sciences, Boston) following completion of thestudy were 90% for RDX, 99% for 26DNT, and 98% for TNT.

Authentic standards for RDX, 26DNT, and TNT, werepurchased from Supelco.

Uncontaminated soils “Yolo” and “Tinker” were collected,analyzed, and provided by Carol R. Johnson, University ofCalifornia, Davis (UC Davis). The Yolo sample was collectedat the student farm on the UC Davis campus and contained34% sand, 46% silt, 20% clay, and 1.9% organic matter asdetermined by the Walkely–Black spectrophotometric method(1.1% organic carbon by calculation of 58% of organic matter).An independent analysis of Yolo soil gave 1.17% total carbonby oxidation, inorganic carbon at 0.02%, and organic carbon at1.15% by difference (Goldberg et al., 2000) The Tinker samplewas collected from Nevada County, CA (R12E, T17N, NE 1/4of the SE 1/4 of Section 22) and contained 57% sand, 38% silt,

5% clay, and 9.5% total carbon by oxidation (Carlo-ErbaCarbon-Nitrogen analyzer, Carlo Erba Strumetazione, Milan,Italy). Each soil sample reflects the major soil type in thedesignated area, as determined by soil survey maps. Soils werestored under ambient conditions in the laboratory (approxi-mately 20°C and 50% relative humidity) and ground with amortar and pestle so that the entire sample would pass throughan 80-mesh sieve. This was done to give a better representationof the entire soil sample, rather than selecting a specific sizerange from the bulk sample.

Soil samples were spiked with radiolabel so that a massof 8 mg soil would contain 8 μg compound and approximately1–2 μCi radioactivity. Soil samples (fresh preparations) werespiked in the same manner as the stored samples used for com-parison. Typically, 0.5–1.0 g soil was added to an acetone(0.35–0.5 ml) solution of the radiolabel to give a free-flowingslurry. The slurry was stirred with the closed end of a glasscapillary tube. Solvent was allowed to evaporate over 8 h orovernight. The soil sample was transferred to a fresh vial.Three aliquots (5–10 mg) were placed in liquid scintillationcounting (LSC) vials along with 1–2 ml acetone. After stand-ing 24 h for extraction, counting solution (Ultima Gold, PerkinElmer) was added to the vials for radiometric assay (model2100TR counter, Perkin Elmer). The scintillation countercontrolled interferences from static electricity and chemilumi-nescence. The values of disintegrations per minute (dpm) sodetermined were divided by the mass of the aliquot to determinea specific activity for the soil sample. If the specific activitywas not uniform, soil samples were tumbled in a tube attachedto a rotary evaporator (Buchi RE 111 Rotavapor), mixed(Scientific Products Vortex Mixer), and specific activity wasredetermined. For stored samples, the specific activity wasdetermined by this method or by sample oxidation (model 307oxidizer, Perkin Elmer) to calculate the applied radioactivedose. The specific activities of stored samples determined byacetone extraction were in good agreement with the valuesdetermined by oxidation of soil aliquots.

Fresh radiolabeled RDX preparations were used within 4 dof preparation, 9 d for 26DNT, and 5 d of TNT. Percutaneousabsorption from the 26DNT/Yolo soil preparation was redeter-mined approximately 50 d following preparation.

Aged soil preparations were contained in glass LSC vialswith polyethylene-lined caps. Caps were further sealed withparafilm and the vials were placed in a sealed polypropylenecontainer. The outer container was monitored for radioactivity.

Skin SamplesFollowing euthanasia, whole skin was harvested from the

upper back of 3 female Yorkshire pigs, weighing 25 to 55 kg.Skin samples were dermatomed to a thickness of 0.5–0.9 mmand stored for approximately 12 h at 4°C, with the visceral sidein contact with gauze soaked with RPMI tissue culture medium(Sigma, St. Louis, MO). This interval of time was not expected

488 W. G. REIFENRATH ET AL.

to alter the skin permeability properties, as the penetration of areference compound ([14C]-N,N-diethyl-m-toluamide) into thereceptor fluid was not altered (13 ± 4%, n = 9, dose = 320 μg/cm2,at t = 0 versus 15 ± 5%, n = 6, dose = 500 μg/cm2, at t = 12 h;p > .05, t-test; Reifenrath, unpublished data).

Pigs were housed in an AAALAC-accredited facility untilused acutely in an unrelated IACUC-approved protocol thatdid not involve any chemical or radioisotope exposure. Skinsamples were obtained for this study immediately after animalswere euthanized on the other project.

In Vitro Percent Absorption MeasurementsThe basic procedure consisted of obtaining freshly excised

skin, removing a portion of the dermis and subcutaneous fatwith a dermatome, and mounting the skin on a penetration cell(low-volume flow cell made from stainless steel; Reifenrathet al., 1994) so that the visceral side of the skin was in contactwith tissue culture medium perfusing the penetration cell.Using a fraction collector, medium exiting the penetration cellwas fractionated into four 1-h intervals, four 2-h intervals, andthree 4-h intervals. Samples were analyzed for radiolabeledpenetrant. An evaporation cell or donor cell was mounted onthe outer surface of the skin. Chemical or soil application wasmade to a skin area of 0.8 cm2 that was exposed by the evapo-ration cell (Hawkins, 1990). The evaporation cell maintained a600-ml/min air flow over the skin surface and containedreplaceable traps for collection of chemical, which evaporatedfrom the skin surface. Details of the procedure have beenpublished (Hawkins & Reifenrath, 1986), and used with thefollowing modifications. Pig skin was dermatomed to a thick-ness of 0.5–0.9 mm. The application area was pretreated with5 μl of artificial sweat (Fisher, 1973) prior to application of soilvehicles (since this work was completed, Stefaniak and Harvey[2007] have improved the artificial sweat formula). Thisvolume was adequate to cover the application area and wouldrepresent a mild to moderate degree of sweating. Artificialsweat was applied to the skin in all cells in a given order andthen the soil samples were applied in the same order, giving thefirst applied cells a shorter delay between sweat applicationand soil application. For aged and fresh 26DNT/Yolo soilsamples only, the experiments were repeated, keeping the timeinterval between sweat and soil application constant and at aminimum, but allowing moisture a greater residence time byvarying the start time of air flow. Twenty-four hours afterapplication, the experiment was terminated and soil particleswere removed from the skin by vacuuming the surface with asmall tube connected to a vacuum pump drawing air at 2 L/min.The skin surface was further cleaned with the application oftwo tape strips. The contents of the vacuum tube and tape stripswere placed in separate vials with LSC fluid for radiometricassay. Skin was removed from the diffusion cells and the epi-dermis was covered with a polyvinyl film (Stretch-tite plasticfood wrap for microwave ovens, Sutton, MA). A brass weight

heated to 65°C was pressed against the epidermis for 90 s. Theepidermis was then teased from the dermis with a forceps.Epidermis and dermis were placed in separate LSC vials andsolubilized with approximately l ml of tissue solubilizer (Solu-ene 350, Perkin Elmer) and heated to 50−60°C for approxi-mately 1 h. Ten milliliters of LSC counting fluid was added toeach vial for radiometric assay.

Twenty-four evaporation/penetration cells were assembledon the morning following each day that pig skin samples wereobtained. Six cells received 14C-compound/fresh Yolo soil, sixcells received 14C-compound/fresh Tinker soil, six cellsreceived 14C-compound/aged Yolo soil, and six cells received14C-compound/aged Tinker soil. Applications were made togive approximately 10 mg/cm2 soil and 10 μg/cm2 ofcompound. The soil dose was chosen to insure complete cover-age of the skin surface. For each group of six replicates, twowere obtained using skin from each of three pigs. Percutaneouspenetration values were redetermined for 26DNT/Yolosamples (aged and fresh, six replicates each) with skin from asingle pig. For runs with 26DNT and TNT, aliquots of the 1–2 hreceptor fluid (the period of maximum flux) were saved forHPLC analysis using the method described for radiochemicalpurity analysis. RDX label did not penetrate sufficiently intoreceptor fluid to permit HPLC analysis. Dermal levels of radio-label from RDX, 26DNT, and TNT were also too low forHPLC analysis.

HPLC Analysis of Radiolabel in Receptor FluidThe HPLC apparatus consisted of an automatic injector (SP

8780XR, Spectra-Physics, San Jose, CA), pump (Bio-Rad1350 series, Bio-Rad, Inc., Richmond, CA), and Spectromoni-tor III variable-wavelength UV detector set at 254 nm (LDC,Riviera Beach, FL). The system was fitted with a SupelcosilLC-18-S column (5 μm, 250 × 4.6 mm, Supelco, Inc.), guardcolumn (Supelguard LC-18-S, 2 cm), and 500-μl sample loop(Alltech, Deerfield, IL). The column outlet was connected tothe UV detector, which was connected to a radiometric flowdetector (Beta RAM model 2B, IN/US, Tampa, FL). Outputfrom the UV and flow detectors was analyzed using Win-Flowsoftware (IN/US). An isocratic solvent system (metha-nol:water, 50:50) was used at a flow rate of 1 ml per min.A standard solution of labeled 26DNT had a retention time of13.6 min and labeled TNT had a retention time of 12.0 min.Aliquots of receptor fluid (50 μl) from 14C-compound-treatedcells were directly injected into the HPLC.

RESULTS

Skin Penetration of 14C-CompoundsThe disposition of radiolabel 24 h after topical application of

14C-compounds in the two soil types is given in Table 1. Separatefresh control samples were prepared when the 27-mo and 62-moRDX samples were tested on separate days and also when the 35


and 36 mo 26DNT samples were tested. On average 89 ± 7% ofradioactivity was recovered after the soil applications.

For all soil applications, the majority of radiolabel wasrecovered as skin decontamination and much lower amounts ofradiolabel were found in the epidermis, dermis, and receptorfluid. For 26DNT and TNT applications, the majority of radio-label that penetrated through the epidermis accumulated in thereceptor fluid rather than the dermis. RDX penetration intodermis and receptor fluid hovered near zero (Table 1).

Previous studies showed that in vivo percutaneous penetra-tion can best be predicted from in vitro data by adding pene-trant residues found in the dermis to the amount of penetrantfound in receptor fluid (Hawkins & Reifenrath, 1986). Esti-mates of human in vivo percutaneous absorption were made inthis manner and these data are presented in Table 2. Asobserved in prior studies (Reifenrath et al., 2002), percutane-ous absorption was generally higher from the low-carbon Yolosoil. A higher carbon content of soil reflected a greater propor-tion of organic matter, which may favor an increased affinityfor the test compounds.

HPLC Analysis of Radiolabel in Recovered Receptor Fluid Following 26DNT and TNT Soil Exposures

First to second hour receptor fluid was analyzed for two cellstreated with 26DNT/Tinker/fresh, three cells treated with 26DNT/Tinker/aged, six cells treated with 26DNT/Yolo/fresh, and sixcells treated with 26DNT/Tinker aged soil. Only radioactivity con-sistent with starting compound could be identified. These resultsare in agreement with previous findings (Reifenrath et al., 2002).

First to second hour receptor fluid was analyzed for six cellstreated with TNT/Yolo/fresh and six treated with TNT/Yolo/aged soil (radioactivity in TNT/Tinker/aged or fresh sampleswas too low to allow HPLC analysis). For the Yolo/fresh sam-ples, 53 ± 26% of radioactivity had a retention time consistentwith TNT; 30 ± 18% had a retention time (12.90–13.45 min)consistent with monoamino metabolites (either 2-amino-4, 6-dinitrotoluene or 4-amino-2, 6-dinitrotoluene). A third peak withretention time of 3.10–3.35 accounted for 17 ± 9% of the radio-activity. This peak is not likely to be accounted for as diaminoderivatives (2,4-diamino-6-nitrotoluene and 2,6-diamino-4-nitrotoluene), since the diamino derivatives have longer retention

TABLE 1 Disposition of Radioactivity (% of Applied Dose) Following Application of C-14 Labeled Compounds (10 μg/cm2) to Excised

Pig Skin in Soils (Tinker, Yolo at Loading of 10 mg/cm2) as a Function of Preparation Age in Monthsa

Percent Applied Radioactive Dose

Cmpd/soil/date Age Evaporation Skin decon. Epidermis Dermis Receptor fluid Tot. Rec.

RDX/Tinker/96 62 m 0.04 [0.02] 97 [16] 0.32 [0.16] 0.03 [0.02] 0.08 [0.06] 99 [16]RDX/Yolo/96 62 m 0.08 [0.03] 71 [25] 0.42 [0.26] 0.02 [0.02] 0.16 [0.12] 73 [24]RDX/Tinker/01 fresh 0.16 [0.04] 92 [11] 0.60 [0.23] 0.06 [0.05] 0.10 [0.09] 95 [11]RDX/Yolo/01 fresh 0.21 [0.07] 87 [28] 0.78 [0.36] 0.06 [0.03] 0.42 [0.45] 90 [28]RDX/Tinker/99 27 m 0.06 [0.01] 93 [3] 0.45 [0.18] 0.06 [0.04] 0.14 [0.16] 96 [4]RDX/Yolo/99 27 m 0.08 [0.02] 82 [5] 0.60 [0.19] 0.06 [0.04] 0.23 [0.29] 85 [4]RDX/Tinker/01b fresh 0.07 [0.01] 82 [15] 0.20 [0.12] 0.02 [0.01] 0.02 [0.02] 84 [15]RDX/Yolo/01b fresh 0.12 [0.03] 81 [13] 0.53 [0.13] 0.02 [0.02] 0.09 [0.12] 83 [13]DNT/Tinker/99 35 m 4.21 [0.85] 80 [2] 0.38 [0.14] 0.16 [0.07] 1.34 [0.90] 88 [3]DNT/Yolo/99 35 m 6.08 [2.24] 66 [7] 1.52 [0.47] 0.67 [0.27] 9.23 [3.79] 86 [2]DNT/Tinker/01 fresh 4.31 [1.98] 82 [10] 0.27 [0.17] 0.11 [0.11] 0.86 [1.11] 90 [14]DNT/Yolo/01 fresh 2.82 [2.10] 76 [19] 0.92 [0.99] 0.58 [0.79] 5.97 [9.18] 88 [6]TNT/Tinker/00 18 m 0.27 [0.08] 95 [2] 0.33 [0.13] 0.08 [0.07] 0.47 [0.24] 99 [3]TNT/Yolo/00 18 m 0.34 [0.07] 84 [4] 1.52 [0.27] 0.12 [0.05] 2.81 [1.02] 93 [3]TNT/Tinker/02 fresh 0.74 [0.27] 89 [15] 0.33 [0.10] 0.07 [0.05] 0.47 [0.16] 92 [16]TNT/Yolo/02 fresh 0.27 [0.09] 74 [5] 1.16 [0.33] 0.16 [0.13] 3.93 [2.35] 83 [2]DNT/Yolo/99 35 m 8.84 [1.15] 55 [5] 2.16 [0.17] 0.65 [0.16] 16.9 [3.5] 88 [3]DNT/Yolo/01b fresh 8.21 [2.04] 54 [11] 1.65 [0.61] 0.52 [0.22] 15.5 [10.8] 83 [1]DNT/Yolo/01/NS fresh 0.43 [0.26] 90 [1] 0.04 [0.02] 0.06 [0.01] 0.01 [0.01] 91 [1]

Note. aValues represent the mean and standard deviation [in brackets] of 6 replicates, two from each of three pigs, except for RDX/Tinker/96,RDX/Tinker/01, RDX/Yolo/96, DNT/Yolo/99, DNT/Yolo/01, and TNT/Yolo/00, where N = 5; and DNT//Yolo/01/NS (no sweat), where N =2. DNT = 2,6-dinitrotoluene. Skin was decontaminated (Skin Decon.) by vacuuming soil particles from the skin surface. Total Recovery (Tot.Rec.) also includes small amounts of radioactivity recovered from tape strips of the skin surface, from rinses of the soil applicator, and fromrinses of equipment.

bSeparate fresh control preparations and results.


times under these conditions (4.83–6.33 min; 4.47–5.77 min). Inaddition, previous studies have shown that diamino metabolitesare not produced when the monoamino derivatives are applied tothe skin. For the Yolo/aged samples, 61 ± 20% of radioactivityhad a retention time consistent with TNT; 17 ± 16% had a reten-tion time (12.90–13.45 min) consistent with monoamino metab-olites; and the third peak with retention time of 3.20–3.35accounted for 16 ± 5% of the radioactivity. In addition to skinmetabolism of TNT, it is possible that soil microbial metabolismmay be operative; however, the similar pattern of metabolitesbetween stored and fresh soil samples (where prolonged acetoneexposure in the slurry method of preparation may inhibit soilmicrobes) and between soil and acetone applications of TNT(Reifenrath et al., 2002) would agrue against this.

As previously observed (Reifenrath et al., 2002), TNTradiolabel in receptor fluid was extensively metabolized ordegraded during the absorption process, with reduction tomonoamino derivatives being the likely reaction. An additionalunknown product(s) was also observed with a retention time sig-nificantly different from diamino derivatives. As before, degrada-tion products of 26DNT were not found in the receptor fluid

(Reifenrath et al., 2002). RDX did not penetrate sufficiently fromsoils, and radioactivity levels were too low for HPLC analysis.

The variability of skin absorption values obtained withfreshly spiked and aged soil samples (Table 2) did not permit astatistically significant discrimination of mean values. Themean percutaneous absorption value for freshly prepared26DNT/Yolo soil (7 ± 10%, Table 2) was lower than previ-ously determined (16 ± 5%, Reifenrath et al., 2002) and morevariable. An examination of the data from individual cellsrevealed that receptor fluid values decreased with increasingtime between sweat application and soil application (artificialsweat was applied to all cells in a given order and then the soilsamples were applied in the same order, giving the first appliedcells a shorter delay between sweat application and soil appli-cation). Moisture was increasingly lost from the skin surface asthe time interval rose. When plots of this time interval versusskin absorption and evaporation of 26DNT were made (initialexperiment, Figures 1 and 2), significant loss of moisture wasassociated with lower skin absorption, regardless of soil type or

TABLE 2 Percutaneous Absorption of 14C-Radiolabeled RDX, 26DNT,

and TNT from Freshly Spiked and Aged Soils

Sample/soil/year of preparation

Elapsed time (mo)

Percutaneous absorptiona

(% applied dose) n

RDX/Tinker/96 62 0.11 ± 0.08 5RDX/Tinker/01 Fresh 0.15 ± 1.14 5RDX/Yolo/96 62 0.19 ± 0.14 5RDX/Yolo/01 Fresh 0.49 ± 0.48 6RDX/Tinker/99 27 0.20 ± 0.19 6RDX/Tinker/01 Fresh 0.04 ± 0.02 6RDX/Yolo/99 27 0.29 ± 0.33 6RDX/Yolo/01 Fresh 0.11 ± 0.14 626DNT/Tinker/99 35 1.5 ± 0.9 626DNT/Tinker/01 Fresh 1.0 ± 1.2 626DNT/Yolo/99 35 9.9 ± 4.0c 526DNT/Yolo/01 Fresh 6.6 ± 10.0c 526DNT/Yolo/99 36 17.5 ± 3.6c 626DNT/Yolo/01 Fresh 16.0 ± 11.0c 626DNT/Yolo/01 (no sweat)

Fresh 0.08 ± 0.02 2

TNT/Tinker/00 18 0.54 ± 0.28 6TNT/Tinker/02 Fresh 0.54 ± 0.18 6TNT/Yolo/00 18 2.9 ± 1.0 6TNT/Yolo/02 Fresh 4.1 ± 2.5 6

Note. Values are mean and standard deviation of n replicates ofapplications of 10 μg/cm2 compound and 10 mg/cm2 soil.

aDermis plus receptor fluid.cInitial exposure is 35 mo aging; second exposure is 36 mo aging.

FIG. 1. (A) Effect of Yolo soil aging (35 mo) and hydration on absorptionof 2,6-dinitrotoluene (26DNT) with excised pig skin (initial experiment). (B)Effect of Yolo soil aging (35 mo) and hydration on evaporation of 2,6-dinitrotoluene (26DNT) with excised pig skin (initial experiment).

R2 = 0.8017

R2 = 0.9148

0

5

10

15

20

25

A

B

0 5 10 15 20 25 30 35 40Time (min) between sweat and soil applications

(more water evaporated with increased time)

Ski

n A

bso

rpti

on

(%

ap

plie

d d

ose

)

26DNT/Yolo/fresh prep.(dashed line)

26DNT/Yolo/aged (solid line)

2

0

4

6

8

10

12

Eva

po

rati

on

(%

ap

plie

d d

ose

)

0 5 10 15 20 25 30 35 40

Time (min) between sweat and soil applications(more water evaporated with increased time)

26DNT/Yolo/fresh prep.(dashed line)26DNT/Yolo/aged (solid line)

R2 = 0.5152

R2 = 1


age, and lower evaporation for fresh soil samples. Fresh sam-ples were best fitted with nonlinear curves, while the aged sam-ples changed in a linear fashion. Freshly prepared samples mayhave loosely bound and tightly bound reservoirs of 26DNT, andaging may convert the former to the latter. The simultaneousincrease in both evaporation and penetration with higher mois-ture conditions was consistent with the predominant effect ofmoisture on the release of 26DNT from freshly prepared soils.Moisture was shown to enhance the volatilization of pesticidesfrom soils (Igue et al., 1972). If the effect of moisture had beenonly to increase skin permeability, evaporation would not haveelevated or perhaps decreased as penetration rose. When thetime interval between sweat and soil application was kept con-stant and at a minimum, but allowing moisture a greater resi-dence time by varying the start time of air flow, skin absorptionof 26DNT from Yolo soil significantly increased with the timelength of no air flow and evaporation significantly decreased forthe fresh preparation, with no significant effect on the aged

preparation (second experiment, Figure 3). For the freshpreparation in this experiment, the predominant effect of mois-ture was consistent with increased skin permeability.

For the freshly prepared RDX/Tinker samples, the loss ofmoisture significantly correlated with a decrease in skin absorp-tion (Figures 4 and 5), with no significant effect on the 27- and62-mo aged samples. Comparing the fresh samples (Figures 4and 5), the exponential portion of the fresh sample in Figure 5may have been missed because there were no data points earlierthan 22 min. The experiments with RDX/Yolo samples wereinconsistent, with a significant decline in skin absorption for thefresh sample and not the 62-mo sample (Figure 6), but no signif-icant effect for the 27-mo and fresh samples (Figure 7). Compar-ing the fresh samples (Figures 6 and 7), the exponential portionof the fresh samples in Figure 7 may again have been missedbecause there were no data points earlier than 20 min. Evapora-tion of RDX was always less than 0.5% from the skin surface,and soil sample type, age, and moisture did not show consistenteffects on this mode of loss (plots not shown).

For TNT in Yolo soil, loss of moisture correlated with lowerskin absorption from the aged sample, but the effect was not

FIG. 2. (A) Effect of Tinker soil aging (35 mo) and hydration on absorptionof 2,6-dinitrotoluene (26DNT) with excised pig skin. (B) Effect of Tinker soilaging (35 mo) and hydration on evaporation of 2,6-dinitrotoluene (26DNT)with excised pig skin.

B

R2 = 0.8309

R2 = 0.7867

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0 5 10 15 20 25 30 35 40 45 50Time (min) between sweat and soil applications

(more water evaporated with increased time)

Ski

n A

bso

rpti

on

(%

ap

plie

d d

ose

)

26DNT/Tinker/fresh prep.(dashed line)

26DNT/Tinker/aged (solid line)

26DNT/Tinker/fresh prep.(dashed line)26DNT/Tinker/aged (solid line)

A

Eva

po

rati

on

(%

ap

plie

d d

ose

)

4

3

2

1

5

6

7

8

9

00 10 20 30 40 50

Time (min) between sweat and soil applications(more water evaporated with increased time)

R2 = 0.6332

R2 = 0.1593

FIG. 3. (A) Effect of Yolo soil aging (36 mo) and air flow (hydration) onabsorption of 2,6-dinitrotoluene (26DNT) with excised pig skin (secondexperiment). (B) Effect of Yolo soil aging (36 mo) and air flow (hydration) onevaporation of 2,6-dinitrotoluene (26DNT) with excised pig skin (secondexperiment).

R2 = 0.1504

R2 = 0.8822

0

2

4

6

8

10

12

14

0 10 20 30 40 50Time (min) of no air flow (increased time of no

air flow increased hydration)

Eva

po

rati

on

(%

ap

pl.

do

se)

26DNT/Yolo/fresh prep. (dashed line)


B

R2= 0.4311

R2 = 0.8548

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35 40 45 50Time (min) of “no air flow” (increased time of no

air flow increased hydration)

Ski

n A

bso

rpti

on

(%

ap

pl.

do

se)

26DNT/Yolo/fresh prep.(dashed line)


A


FIG. 4. Effect of Tinker soil aging (62 mo) and hydration on absorption of RDX with excised pig skin.

R2 = 0.8619

R2 = 0.11

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

0 5 10 15 20 25 30 35 40 45Time (min) between sweat and soil applications (more water

evaporated with increased time)

Ski

n A

bso

rpti

on

(%

ap

pl.

do

se) RDX/Tinker, fresh prep. (dashed line)

RDX/Tinker, aged (solid line)

FIG. 5. Effect of Tinker soil aging (27 mo) and hydration on the absorption of RDX with excised pig skin.

0

0.1

0.2

0.3

0.4

0.5

0.6



Ski

n A

bso

rpti

on

(%

ap

pl.

do

se)

R2 = 0.443

R2 = 0.678

RDX/Tinker, aged (solid line)

RDX/Tinker, fresh prep. (dashed line)

FIG. 6. Effect of Yolo soil aging (62 mo) and hydration on absorption of RDX with excised pig skin.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Ski

n A

bso

rpti

on

(%

ap

pl.

do

se)

RDX/Yolo, fresh prep. (dashedline)

RDX/Yolo, aged (solid line)

R2 = 0.8156

R2 = 0.2783




significant for the fresh preparation (Figure 8). Evaporation ofradiolabeled TNT was less than 0.5% for all preparations, withno significant correlations with loss of moisture (plots notshown). Skin absorption and evaporaton of TNT from Tinkersoil were generally less than 1%, and moisture and soil age didnot show consistent effects (plots not shown).

DISCUSSIONFor a series of 10 munition contaminants applied to skin in

acetone vehicle (Reifenrath et al., 2002), there existed a posi-tive correlation between percutaneous absorption versus watersolubility over the range of 0 to 500 mg/L (ppm). When thesecompounds were applied in soils to skin pretreated with artifi-cial sweat, a positive correlation was also found betweenpercutaneous absorption versus both water solubility (S) andvapor pressure (Vp). The water solubility of 26DNT (160 mg/L)

was comparable to that of TNT (129 mg/L), and roughlythreefold higher than that for RDX (51 mg/L). However, thevapor pressure for 26DNT (567) was roughly 100-fold higherthan the value for TNT (5.5 × 10−6 torr at 25°C) and RDX wasessentially nonvolatile. One would expect the soil contaminant24DNT (Vp = 217 × 10−6 torr at 25°C, S = 157 mg/L) tobehave as 26DNT. Further research is needed to determinewhether compounds of other structural types behave in soils asthe munition contaminants, especially those with water solubil-ity and vapor pressure equal to or somewhat greater than thoseof 26DNT. Dimethyl phthalate (Vp = 7000 × 10−6 torr at 25°C[Stull, 1947], S = 4300 mg/L [Merck Index, 1996]) and diethylphthalate (Vp = 500 × 10−6 torr at 25°C [Stull, 1947],S = 7000 mg/L [Lyman, 1990]) would be commercial com-pounds of interest. VX (O-ethyl S-[2-(diisopropylamino)ethyl]methylphosphonothioate, Vp = 700 × 10−6 torr at 25°C,S = 31000 mg/L [Daugherty et al., 1992]) would be a

FIG. 7. Effect of Yolo soil aging (27 mo) and hydration on the absorption of RDX with excised pig skin.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Ski

n A

bso

rpti

on

(%

ap

pl.

do

se)

R2 = 0.0012

R2 = 0.0508

0 5 10 15 20 25 30 35 40 45 50Time (min) between sweat and soil applications (more water


RDX/Yolo, aged (solid line)RDX/Yolo, fresh prep. (dashed line)

FIG. 8. Effect of Yolo soil aging (18 mo) and hydration on absorption of TNT with excised pig skin.

0

1

2

3

4

5

6

7

8

Ski

n A

bso

rpti

on

(%

ap

pl.

do

se)

R2 = 0.6789

R2 = 0.094

0 5 10 15 20 25 30 35Time (min) between sweat and soil applications (more water


TNT/Yolo, fresh prep. (dashed line)

TNT/Yolo, aged (solid line)


compound of military interest. There are obviously many morecompounds of higher S and Vp, but if compounds become toopolar, their ability to penetrate the lipid barrier of the skindecreases (e.g., thiourea, < 5%; Franz, 1978). Likewise, ifvapor pressure becomes too high, the residence time on theskin decreases, thereby limiting skin penetration.

In summary, sample age did not significantly alter meanpercutaneous absorption values for 26DNT, TNT, and RDX orthe pattern of metabolites seen after 26DNT and TNT soilapplications; however, the variation of skin absorption andevaporation from soils was found to be a function of the com-pound, skin surface moisture levels, preparation age, and soilcarbon content.

REFERENCESDaugherty, M. L., Watson, A. P., and Vo-Dinh, T. 1992. Currently available

permeability and breakthrough data characterizing chemical warfare agentsand their simulants in civilian protective clothing materials. J. Hazard.Mater. 30:243–267.

Fisher, A. A. 1973. Contact dermatitis, 2nd Ed., p. 27. Philadelphia: Lea andFibiger.

Franz, T. J. 1978. The finite dose technique as a valid in vitro model for the studyof percutaneous absorption in man. Curr. Problem. Dermatol. 7:58–68.

Goldberg, S., Lesch, S. M., and Suarez, D. L. 2000. Predicting boron adsorp-tion by soils using soil chemical parameters in the constant capacitancemodel. Soil Sci. Soc. Am. J. 64:1356–1363.

Hawkins, G. S. 1990. Methodology for the execution of in vitro skin penetra-tion determinations. In Methods for skin absorption, eds. B. W. Kemppainenand W. G. Reifenrath, pp. 67–80.Boca Raton, FL: CRC Press.

Hawkins, G. S., and Reifenrath, W. G. 1986. The influence of skin source, pen-etration cell fluid and partition coefficient on in vitro skin penetration.J. Pharm. Sci. 75:378–381.

Igue, K., Farmer, W. J., Spencer, W. F., and Martin, J. P. 1972. Volatility oforganochlorine insecticides from soil: II. Effect of relative humidity and soilwater content on dieldrin volatility. Soil Sci. Soc. Am. Proc. 36:447–450.

Lyman, W. J. 1990. Solubility in water. In Handbook of chemicalproperty esti-mation methods, eds. W. J. Lyman, W. F. Reehl, and D. H. Rosenblatt,pp. 2–29. Washington, DC: American Chemical Society.

Merck Index Monographs. 1996. 12th ed., eds. S. Budavari, M. J. O’Neil,A. Smith, P. E. Heckelman, and J. F. Kinneary, p. 550. WhitehouseStation, NJ: Merck.

Reifenrath, W. G., Lee, B., Wilson, D. R., and Spencer, T. S. 1994. A compar-ison of in vitro skin-penetration cells. J. Pharm. Sci. 83:1229–1233.

Reifenrath, W. G., Kammen, H. O., Palmer, W. G., Major, M. M., and Leach,G. J. 2002. Percutaneous absorption of explosives and related compounds.An empirical model of bioavailability of organic nitro compounds fromsoil. Toxicol. Appl. Pharmacol. 182:160–168.

Stefaniak, A. B., and Harvey, C. J. 2007. Dissolution of materials in artificialskin surface film liquids. Toxicol. In Vitro 20:1265–1283.

Stull, D. R. 1947. Vapor pressures of pure substances. Organic and inorganiccompounds. Ind. Eng. Chem. 39:517–549.

U.S. Environmental Protection Agency. 1992. Dermal exposure assessment:Principles and applications. Washington, DC: Office of Health and Envi-ronmental Assessment. EPA/600/8–91/011B.

U.S. Environmental Protection Agency. 2001. Risk assessment guidance forSuperfund, Vol. 1, Human health evaluation manual (Part E, SupplementalGuidance f or Dermal Risk Assessment). EPA/540/R/99/005. Washington, DC.

interaction of hydration, aging, and carbon content of soil on the evaporation and skin...

Documents