Document 70em13gZrx9eBqJjag9omOoQg

I DE b*\@.- Tdxicology and Applied Pharmacology 160, 60-75 (1999) ` Article ID taap.1999.8744. available online at http://www.idealibrary.com on I \ Dermal Absorption and Distribution of Topically Dosed Jet Fuels Jet-A, JP-8, and JP-8(100) Jim E. Riviere,*.' James D. Brooks,* Nancy A. Monteiro-Riviere,* Kamon Budsaba,? and Charles E. Smith? *Center for Cutaneous Toxicology and Residue Pharmacology, College of Veterinary Medicine, and +Department of Statistics, College of Agricultural and Life Sciences. North Carolina State University, Raleigh, North Carolina 27606 Received March 15. 1999; accepted July 2, 1999 Dermal Absorption and Distribution of Topically Dosed Jet Fuels JET-A, JP-8, and JP-8(100). Riviere, J. E., Brooks, J. D., Monteiro-Riviere, N. A., Budsaba, K., and Smith, C. E. (1999). Toxicol. Appl. Phamcol. 160,60-75. Dermal exposure to jet fuels has received increased attention with the recent release of newer fuels with novel performance additives. The purpose of these studies was to assess the percutaneousabsorptionand cutaneousdisposition of topically applied (25 pl/S an2)neat Jet-A, JP-8, and JP-8(100) jet fuels by monitoring the absorptiveflux of the marker Components "C naphthalene and 'H dodecane simultanewsly applied nonmluded to isolated perfused porcine skin Raps (IPPSF) (n = 4). Absorption of "C hexadecane was estimated from JP-8 fuel.Absorption and disposition of naphthalene and dodecane Jet fuels used in commercial and military aircraft are a complex multicomponent mixture of aromatic and aliphatic hydrocarbons. Due to their natural petroleum base, their chemical composition is not defined and the fuels are classified according to broad performance criteria such as boiling and flash points. Military fuels are designated as members of the JP Get propellant) series with JP-4 and JP-7 being the primary fuels used in earlier years. Recently, the military has fielded two new fuels [JP-8 and JP-8(100)] containing several performance additives, which result in fuels with higher flash points and improved heat sink characteristics. The primary military jet fuel is JP-8 which is essentially a kerosene-cut commercial jet fuel base, Jet-A, to which addi- were a l s ~monitored using a nonvolatile ~ p - 8fraction reflecting ex- tives are added to make up <2% of the formulation [a corro- posure to residual fuel that might occu24 h after ajet fuel spill. In allstudies,perfusate,stratum corneum,and skin concentrationswere measured over 5 h Naphthaleneabsorptionhad a clear peak absorptive flux at less than 1 h, while dodecane and hexadwane had prolonged, albeit significantly lower,absorption flux profiles. Within JP-8, the rank order of absorption for all marker components was (mean f SEM % dose) naphthalene (1.17 f 0.07) > dodecane (0.63 f 0.04) > hexadecane (0.18 k 0.08). In contrast, deposition within dosed skin showed the reverse pattern. Naphthalene a b r p - tion into perfusate was similar across all fuel types, however total sion inhibitor DC1-4A, an antistatic compound Stats 450, and an icing inhibitor diethylene glycol monomethyl ether (DI- EGME)]. In the base fuel, 13 components, present at greater than 1% vlv concentrations, comprise 29% of the formulation (see Table 1). JP-8(100) is JP-8 with an additional additive package consisting of antioxidant, chelator, detergent, and dispersant components. Because of the chemical complexity of these fuels, one cannot assess the absorption of the entire fuel, but must instead monitor absorption of individual so-called penetration into and through skin was highest with JP-8(100). Do- marker components. decane absorption and total penetration was greatest from JP-8. Although there is a relatively large reviewed database on the Absorption of both markers from aged JP-8 was lower than other fuels,yet the ratio of skin deposition to absorption was greatest for this treatment group. In most exposure scenarios, absorption into perfusate did not directly correlate to residual skin concentrations. These studiesdemonstrated different absorptionprofiles for the three marker compounds, differentialeffects of jet fuel types on naphthalene and dodecane absorption, and uncoupling of perfusate absorption from skin disposition. o 1999 ~ u ~ l e r mae s Key Words: jet fuel; naphthalene; dodecane; hexadecane; percutaneouddermal absorption; skin. toxicology and absorption after inhalational exposure to early jet fuels (Jet-A, JP-4, and JP-7) (ATSDR, 1993, and more recently to JP-8 (ATSDR, 1996;Pfaff et af.,1995; Harris et al., 1997), there is no information on the percutaneous absorption after topical exposure to any jet fuels (ATSDR, 1995, 1996). Skin sensitization assays have demonstrated some level of imtation after exposure to the older fuels that may lead to chemical bums (Clark et al., 1988; ATSDR, 1995, 1996), providing indirect evidence of dermal penetration. In reality, the potential for percutaneous absorption leading to dermal and systemic tissue exposure to components of jet fuels has not ' To whom correspondence and reprint requests should be addressed:Center for Cutaneous Toxicology and Residue Pharmacology, College of Veterinary Medicine, 4700 Hillsborough Street. North Carolina State University, Raleigh, been systematically investigated. To quantitate the dermal absorption and cutaneous distribu- tion of these fuels, we utilized a radiolabeled aromatic (I4C- NC 27606. E-mail: Jim-Riviere@NCSU.EDU. naphthalene) and an aliphatic ('H-dodecane) marker from 0041-008W99 $30.00 Copyright Q 1999 by Academic Press All rights of reproduction in any form reserved. 60 @ DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS 61 TABLE 1 Major Components of JetA Class Alkane Alkane Alkane Alkane Alkane Alkane Alkane Alkane Aromatic Aromatic Aromatic Aromatic Aromatic Component Dodecane* Tridecane Undecane Tetradecane 2.6-Dimethylundecane Pentadecane 2-Methylundecane Heptylcyclohexane 1methylnaphthalene 2-methylnaphthalene 2.6-Dimeth ylnapthalene Naphthalene; 1,2,3,4-Tetramethylbenzene Percent of total 4.7% 4.4% 4.1% 3.0% 2.1% 1.6% 1.2% 1.O% 1.8% 1.5% 1.3% 1.1% 1.1% No. of carbons 12 13 11 14 13 15 12 13 11 II 12 10 IO No. of hydrogens 26 28 24 30 28 32 26 26 10 10 12 8 14 Mol. w t 170 184 156 198 184 212 170 182 I42 142 156 128 134 Total 28.9% t Note Jet fuel IS composed of some 228 components and these 13 only represent 28.9% of the total fuel. The marker compounds studied represent only 5.8% 4 of total. 4 * Chosen as a marker compound in this study. among the most prevalent components of JP-8 listed in Table evaporation curves was similar with no discernible "plateau" evident. As 1. To determine a constituent effect, we studied the absorption, penetration, and cutaneous distribution of these marker compounds in the in vitro isolated perfused porcine skin flap (IPPSF) when dosed from Jet-A, JP-8, and JP-8(100). This model system has been successfully utilized to model the expected, there was a first-order decay to approximately 30% of the initial mass. This remnant solution was then spiked with the radiotracer markers and used for all exposures. IPPSF studies. The IPPSF procedure has been previously documented (Riviere er al., 1986; Riviere and Monteiro-Riviere, 1991; Riviere er al., 1995b). In these studies. jet fuel mixtures were applied nonoccluded to mimic mechanism of percutaneous chemical absorption (Riviere and field exposure conditions, and experiments were conducted for a total of 5 h in Monteiro-Riviere, 1991; Riviere et al., 1995b; Qiao et al., 1996)and to extrapolatethese results to humans (Riviere et al., 1992; Wester et af., 1998; Williams et al., 1990). We also IPPSFs with 4 replicates per treatment condition. A 1 X 5 cm dosing area was drawn on the surface of the skin flap with a surgery marker. A dose, containing 25 pl of the specified jet fuel containing approximately 2 pCi of "C-naphthalene plus I O pCi of 'H-dodecane, was applied directly to the surface of the I assessed the absorption of naphthalene and dodecane from a skin flap. The specific activities of the marker compounds were sufficient that fraction of a so-called aged JP-8 "puddle" which was produced by allowing JP-8 to evaporate from an open glass petri dish left in a fume hood for 24 h. After the more volatile components evaporated,the remaining mixture should resemble residual jet fuel remaining 24 h after a jet fuel spill. Finally, to further the added radiolabeled compounds had little effect on the final concentration of naphthalene (1.21% instead of 1.1%) and dodecane (4.701% instead of 4.7%). Single label studies were initially conducted and compared to the dual-label results to test whether using this dual-label experimental design had any effect on marker absorption. No effect was detected. Perfusate samples (3 ml) were collected every 5 min for the first 40 min, probe the absorption characteristics of the primary fuel JP-8, we also studied the absorption of a longer chain aliphatic hydrocarbon, I4C-hexadecaneafter exposure in JP-8. then every 10 min until 1.5 h, and then every 15 min until termination at 5 h. At termination, several samples were taken for mass balance of the marker compounds. The surface of the dose area was swabbed twice with a 1% soap solution and gauze, and then 12 stratum corneum tape strips were collected using cellophane tape (3M Corporation, Minneapolis, MN). The entire dose MATERIALS AND METHODS area was removed. A 1 X 1 cm core of the dose area was removed and frozen for subsequent depth of penetration studies. This consisted of laying the core Chemicals. Radiolabeled "C-naphthalene (specific activity = 8.1 mCi/ sample epidermal side down in an aluminum foil boat and embedding in mmol) and 'H-dodecane (specific activity = 10,OOO mCifmmol) in methylene Tissue-Tek OCT compound (Miles, Inc., Elkhart, IN), snap freezing in liquid chloride were custom labeled by the NCI Chemical Carcinogen Reference nitrogen. followed by sectioning (40 p n J on a Reichart-Jung Model 1800 Standard Repository (Chemsyn Science Laboratories, Lenexa. KS). The "C- Cryocut (Warner Lambert, Buffalo, NY). The remaining dosed area as well as hexadecane (specific activity = 4.1 mCi/mmol) was obtained from Sigma the surrounding skin was separated from the fat and held for analysis. All Chemical (St. Louis, MO). Radiochemical purity of both compounds ranged samples (including swabs, tape strips, core sections. skin, fat, mass balance from 98 to 100%. samples, etc.) were dissolved separately in Soluene. A representative volume All jet fuels. Jet-A. JP-8, and JP-8( 100). were kindly supplied by Major T. of each sample was oxidized completely via a Packard Model 307 Tissue Miller from Wright Patterson Air Force Base. To produce an aged JP-8 Oxidizer. The 'Hand I4Csamples were counted separately on a Packard Model (puddle), JP-8 was allowed to evaporate in an open glass petri dish within a 1900TR TriCarb Scintillation Counter. fume hood for 24 h ( n = 2). The decline in mass was recorded using an &fa analysis. Data was entered into a custom IPPSF database and the analytical balance over the course of the 24-h period. The shape of the two resulting analysis reported. Since all experiments were conducted using the 62 RIVIERE ET AL. identical marker doses across all fuels. and the absolute concentrations of these marker compounds were similar. these jesults are expressed as percentage applied dose to give a representative assessment of the absorption and cutaneous penetration of a complex mixture such as jet fuel. This is appropriate since the absolute concentrations of jet fuel hydrocarbons is not fixed across all fuels due to differences that arise from the natural source of the petroleum and different refining processes. Area under the curve (AUC) in the perfusate was calculated using the trapezoidal method. Peak flux was the maximum flux (% dose/min) observed at any one time point. Figure 1 depicts the experimental compartments which were analyzed in these studies where the following definitions apply. (1) Surface is the residue removed by washing the surface of the IPPSF at termination of the experiment plus the residues remaining in the dosing template. (2) Stratum corneum is the residue extracted from the outermost stratum corneum via 12 tape strips at the termination of the experiment. (3) Dosed skin is the residue that remained in the dosed skin plus the depth of penetration core taken at termination. (4) Absorption is the cumulative amount of the marker compound collected in the effluent over the course of the 5-h experiment. ( 5 ) Fat is the residue remaining in the fat when it was separated from the dermis at the end of the experiment. (6) Penetration is the summation of the label in the effluent plus skin plus fat, but not stratum corneum nor surface. (7) Evaporative loss is that label which was lost to evaporation. Our previous studies in the IPPSF indicated that the penetration estimate is the best empirical correlate to predict eventual in vivo absorption in humans (Wester et af., 1998). Statistical significance of absorption and penetration parameters were deW e d using ANOVA or by a priori-defined orthogonal contrasts where appropriate (SAS 6.12 for Windows; SAS Institute, Cary, NC) at the 0.05 level of significance. A least significance difference (LSD) procedure was used for multiple comparisons on overall tissue disposition. 3. RESULTS 1 ?here are two primary comparisons that are of interest in this FIG. 1. Schematic depicting sampling sites for which absorption and disposition data are collected. (1) Surface is the residue removed by washing study. The first is the relative absorption of the individual marker the surface of the IPPSF at termination of the experiment plus the residues compounds from jet fuel, and the second is the effect of a specific remaining in the dosing template. (2) Stratum corneum is the residue extracted jet fuel's composition on the absorption of a specific marker. from the stratum corneum via 12 tape strips at the termination of the experiment. (3) Dosed skin is the residue that remained in the dosed skin plus the Absorption of Marker Compounds depth of penetration core taken at termination. (4)Absorption is the cumulative amount of the marker compound collected in the effluent over the course of the Flgure 2 compares the absorption flux profiles for naphthalene and dodecane across all fuels, while Fig. 3 depicts all three d e n in JP-8. Naphthalene absorption peaks at less than 1 h 5-h experiment. ( 5 ) Fat is the residue remaining in the fat when it was separated from the dermis at the end of the experiment. (6) Penetration is the summation of the label in the effluent plus skin plus fat, but not stratum corneum nor surface. (7) Evaporative loss is the unaccounted label. while the aliphatic compounds are characterized by an absorption plateau at later time points. The primary effect Seen in these data is this differential absorption of the individual chemical entities, decane data. A similar pattern is seen when the disposition of an effect which overshadows any differences seen as a result of the markers in all skin compartments are examined (Fig. 5). individual fuel composition effects. Table 2 summarizes the ab- Naphthalene absorption is highest in the perfusate but para- sorption parameters of naphthalene, dodecane, and hexadecane doxically is lowest on the surface and in the stratum corneum from JP-8.It is clear from these data that the rank order of marker and dosed skin. Naphthalene total penetration, which in the absorption (or0 dose mean ? SEh4) is naphthalene (1.17 2 IPPSF should correlate to overall absorption in vivo in humans, 0.07) > dcdecane (0.63 2 0.04) > hexadecane (0.18 5 0.08) is still marginally greater than the other markers, primarily due (p < 0.05). Similarly, AUC and peak flux are both significantly to a significantly greater ( p < 0.05) partitioning into the fat different among all three markers. compartment (0.22 vs 0.08 and 0.12 for naphthalene, dodecane, However, a different perspective is achieved when the cu- and hexadecane, respectively). These data highlight the impor- taneous deposition of these markers after dosing in JP-8 is tance of the experimental perspective. If an estimate of systemic compared. There are differences in the depth of penetration exposure is desired, then naphthalene has the greatest potential of profiles evident (Fig. 4), with the rank order of compounds now the three markers studied. If local cutaneous toxicity is the desired reversed at the outermost layers of skin: hexadecane > dode- endpoint, then the reverse is true since higher amounts of the cane > naphthalene. This is especially evident with the hexa- aliphatic markers remain deposited in the skin at the termination -0.014 DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS 63 0.012 a 0.010 0.008 E f i*; ! I# 0.006 0.004 0.002 \ \ O.OO0 0 1 234 5 All Naphthalene Doses (n=16) +AU Dodecane Doses (n=l6) FIG. 2. Perfusate absorption profiles of naphthalene and dodecane (mean % dosdml 2 SEM)pooled from dosing in all jet fuel types. of an experiment. However, it should be. noted that if the compound present in the outermost stratum corneum, recovered in these studiesby 12tape snips, were totally available, then the rank order of penemtion would be hexadecane (9.84%dose) > dodecane (6.24%dose) > naphthalene (2.12%dose). ESfect of Fuel Composition The second phenomenon studied was the effect of the specific fuel formulation on absorption and deposition of the markers naphthalene and dodecane. The rank order of naph- thalene perfusate absorption based on AUC, % dose absorbed, and peak flux was JP-8(100) > Jet-A > JP-8 > JP-8 (puddle) for naphthalene and JP-8 > JP-8(100) > Jet-A > JP-8 (puddle) for dodecane (Table 3). Figure 6 depicts the comparative perfusate absorption profiles, and Figure 7 illustrates overall tissue deposition. A dichotomy similar to that discussed above is seen when absorption parameters are compared to skin deposition. Naphthalene and dodecane disposition in skin com- 0.016 RIVIERE ET AL. 0.014 0.012 0.010 7 14 0.008 c I & 0.006 0.004 T i. 0.002 0.000 0123 4 5 -Naphthalene from JP-8 ( n 4 ) +Hexadme from JP-8 (n4) -+Dodecane from JP-8 (n4) FIG. 3. Comparison of naphthalene, dodecane, and hexadecane absorption into perfusate (mean % dose/ml -t SEM)after dosing in JP-8 jet fuel. partments did not exactly parallel that seen in perfusate. How- ever, despite these differences in local skin disposition, overall penetration still tracks the pattern in the perfusate with naph- thalene from JP-8(100) statistically greater ( p < 0.05) than JP-8 or JP-8 (puddle). For dodecane, absorption from JP-8 is greater ( p < 0.05) than the other fuels. This apparent dichotomy between skin deposition and absorption is further explored in Fig. 8 which presents histograms of the dosed skidabsorption and absorptioddosed skin ratios for naphthalene and dodecane in Jet-A, JP-8, JP-8 (puddle), and JP-8 (100). The major effect seen with this analysis is the significantly different ratios for dodecane disposition from JP-8 (puddle), and again the differences between the naphthalene and dodecane markers. Statistical Comparisons Table 4 provides an alternative perspective and statistical summary of the role of naphthalene and of dodecane seen in t i ~ DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS 65 TABLE 2 Comparison of Three Marker Compounds from JP-8 Mean (SEM) of Each Parameter ~ (n = 4) AUC (%Dh/ml) Absorbed (%D) Peak flux (%D/min) 0.0199 (0.0020)* 1.17 (0.07)* 0.015 (0.003)* 'H-Dodecane (n = 4) AUC (%D-Wml) Absorbed (%D) Peak flux (%D/min) 0.0107 (O.OOO9)' 0.63 (0.04)' 0.0036(0.0004)n ''C-Hexadecane (n = 4) more significant after a log transformation, but again this is not due to the two contrasts of interest. The last three columns in Table 4 examine the tissue disposition of dodecane in a similar manner. For contrast 1, all skin layers, AUC, and peak flux are significant with the exception of fat. For contrast 2, stratum corneum, absorption, penetration, AUC, and peak flux are significant. The overall p values are significant at all the places that contrast 2 is, but also indicate mean differences in dosed skin and evaporative loss. Similar results are found by the log responses. Note that the LSD approach in Fig. 7 and contrast 2 in Table 4 agree in indicating significant differences between JP-8 and JP8(100). However the LSD results from Fig. 7 are pairwise and cannot easily give the comparable results between Jet-A and the JP-8 fuels (contrast 1). AUC (%D-h/ml) Absorbed (%D) 0.0017 (0.0003)c 0.18 (0.08)' DISCUSSION Peak flux (%D/rnin) 0.0011 (O.OoO2)' f These studies highlight the complexitiesinherent in studying I Note. Means with the same letter are not significantly different. the percutaneous absorption of complex mixtures such as jet 1 fuel. This complexity arises both from the heterogeneity of the chemicals comprising fuel as well as from the different inter- Fig. 7, A and B, respectively, and in Table 3. Two orthogonal pretations obtained if systemic absorption vs local cutaneous contrasts on the four fuel treatments [Jet-A, JP-8, JP-8 (pud- disposition are the endpoints of concern. dle), JP-8(100)] are used to examine two questions. (1) Is the Based on analyses of the fuel stock used in these studies, mean response to Jet-A different from the average of the three dodecane is a primary component present at 4.7% (v/v). Naph- JP-8 treatments, namely contrast 1 [3, - 1, - 1, - l]? (2)Is the thalene is a primary aromatic constituent of jet fuel at 1.1%. In mean response of JP-8 different from the mean of JP-8(100), contrast, hexadecane is a minor component being present at namely contrast 2 [0, 1, 0,- l]? For answering these specific less than 1%. There are a total of 228 characterized major a priori questions based on the structure of the fuel treatments, nonadditive hydrocarbon constituents of these fuels. It is im- orthogonal contrasts are statistically preferable to multiple possible to characterize the absorption of even a fraction of comparison techniques such as LSD (Swallow, 1984; Hsu, these constituents at the level of detail presented in this inves- 4 1996). While the LSD procedure allowed pairwise compari- tigation. Being relatively nonvolatile, the long-chain hydrocar- sons among the fuels, for question 1, Jet-A (a commercial fuel) bons typified by dodecane and hexadecane make up a large is compared to the average of the three military fuels. For percentage of this mixture. As seen in these studies, they have question 2, a specific pairwise comparison examines the role of the greatest surface concentrationsat the end of an experiment additives in the JP-8 fuel series. The overall p value from the (Fig. 5) and tend to have a larger fraction of dose deposited in F statistic in ANOVA for the four jet fuels is also presented the skin. However, they have a decreased systemic absorption and denoted as ALL. Note that the overall F may fail to detect as characterized by perfusate fluxes and total penetration. Be- some differences that do exist, and that are indicated by the cause the fractional absorption is almost an order of magnitude orthogonal contrasts. greater for naphthalene than for dodecane, absolute naphtha- In Table 4, the tissue disposition of naphthalene is analyzed lene mass absorbed (% dose X concentration in fuel) is still using the two contrasts in columns 2 and 3. The only signifi- two-fold greater than dodecane even if the latter compound is cant differences seen are for contrast 2 in fat and penetration. present at close to 5 times the concentration.A similar ordering Surface indicates some difference ( p > 0.10). For an overall of naphthalene and dodecane was reported in preliminary in effect (column 4). penetration and surface also have a p < v i m rat absorption studies of JP-8 components (McDougal and 0.10. The use of log responses give a similar result. Note that Miller, 1998). The magnitude of the differences in total ab- the p value for surface and contrast 1 now becomes more sorption between naphthalene and dodecane overshadow any significant at p = 0.036 as the variance is stabilized by the log effect that fuel composition has on marker absorption as can be transformation. In a similar manner, the last two rows (AUC appreciated by examining the composite absorption profiles in and peak flux) are compared with Table 3. While the overall p Fig. 2. value indicates a difference in AUC means, it is not primarily A limitation of any short duration study of this design is that due to the two contrasts examined. For peak flux there is a steady-state conditions are impossible to achieve due to bio- bigger spread in the SEM and the overall p value becomes logical limitations of the model coupled with the volatility of RIVIERE ET AL. t0.015 T \ 0.010 0.005 I t' O.OO0 0 320 640 960 1280 1600 1920 2240 2560 2880 3200 3520 3840 4160 4480 4800 Depth (urn) - Naphthalenefrom JP-8 ( n 4 ) -Dodecane from JP-8 (n-1) *Hatadccane horn JP-8 (n=4) FIG.4. Depth of penetration of naphthalene, dodecane, and hexadecane into skin (mean % dose 2 SEM)after dosing in JP-8. the hydrocarbons being investigated. This is especially evident when interpreting tissue concentrations at a single 5 h time point at the termination of a study. Previous IPPSF studies have taught that an accurate estimate of potential in vivo absorption is obtained by combining skin and fat residues with absorbed perfusate flux (Wester et ai., 1998; Williams et ai., 1990). This is the penetration parameter reported in this work. Furthermore, skin concentrations at any single time point must be interpreted in the context of a continuum (Williams and Riviere, 1995; Manitz et ai., 1998). Finally, it is expected that some fraction of dose in skin will remain nonabsorbed since a concentration gradient may no longer exist to deliver the compound if the compound preferentially partitions in the skin and forms a depot. These residual skin concentrations could have toxicological significance for a bioactive molecule. The full prediction and description of skin concentrations is obviously more complex. Various pharmacokinetic models are being developed to assess such phenomenon (Manitz et al., 1998; , a a 4.64 a 100 95.89 b 80 -8 6 0 140 e, 20 0 Surface a 8.63 I I I Evaporative Loss 8 H6 &it 4 2 0 Stratum Corneum 1.o a 0.91 a 0.8 0.22 - B-0.6 o.2 1k 0.4 k 0.1 0.2 Dosed Skin 0.0 Fat 1.4 1.2 1.0 0.8 '10.6 0.4 0.2 0.0 a 1.17 Absorption NNaphthalene from P - 8 (n=4) 1.5 B" I 2 0.5 0.0 0Dodecane from JP-8 (n=4) a 1.47 Penetration H Hexadecane from JP-8 ( n 4 ) I FIG. 5. Pattern of tissue deposition and absorption of naphthalene. dodecane. and hexadecane dosed in JP-8 fuel. All values are expressed as mean 'Ti dose 2 SEM.Values with different letters are significantly different (p < 0.05). 67 68 RIVIERE ET AL. TABLE 3 Jet Fuel Comparison Mean (SEM)of Each Parameter Jet-A ( n = 4) JP-8 ( n = 4 ) JP-R(Puddle) ( n = 4) JP-8( 100) (n = 4) All 16 Doses ( n = 16) I4C-Naphthalene AUC (%D-hlml) Absorbed (%D) Peak flux (%D/min) 'H-Dodecane AUC (%D-h/ml) Absorbed (%D) Peak flux (%D/min) 0.025 (0.002)' 1.49 (0.18)" 0.015 ( 0 . 0 0 I ) a b 0.0048 (0.0005)" 0.29 (0.04)b 0.0017 (0.0002)k 0.020 (0.002)" I . 17 (0.07)a 0.015 (0.003)'h 0.0107 (O.ooo9)' 0.63 (0.04)' 0.0036 (O.OoO4)' 0.015 (0.003)h 1.11 (0.16)d 0.008 (O.OOl)b 0.0039 (O.ooo4)' 0.27 ( 0 . 0 7 ~ ~ 0.0014 (0.0002)' 0.026 (0.003)a 1.63 (0.29)a 0.016 (0.003)a 0.0061 (0.0007)b 0.35 (0.04)b 0.0024 (O.OOO1)b 0.021 co.oo2)" 1.35 (0.10)" 0.013 (0.001)" 0.0064 (O.oo08)B 0.38 (0.04)' 0.0023 (0.0002)' Nore. Means with the same letter are not significantly different. Lowercase letters represent comparisons within parameters across fuel types. Uppercase letters represent comparisons only within the last column across markers. McCarley and Bunge, 1998; Riviere et al., 1995a; Williams perfusate absorption. Evaluation of perfusate fluxes alone, a and Riviere, 1995), however a discussion of such kinetic in- procedure often done with in virro diffusion cell studies, would teractions is beyond the scope of the present work. The impor- mask the significant absorption which ultimately could occur tant point to be taken from these studies is that compound with dodecane and especially hexadecane when skin depots are absorption and penetration must be assessed in all skin com- factored in. partments since only looking at the absorbed fraction (e.g., These experimental data suggest that both aliphatic com- perfusate) may produce very misleading conclusions. pounds have a propensity for partitioning into skin and forming One can appreciate the difference seen for a rapidly pene- relatively slowly equilibrating pools that ultimately could be trating compound like naphthalene vs slower penetrating sub- absorbed. Previously, hexadecane has been reported to bind to stances such as dodecane and hexadecane. It is not surprising the stratum corneum (Singh, 1998), a phenomenon which I that a low molecular weight aromatic such as naphthalene would explain the elevated hexadecane levels seen in stratum \ would have an enhanced absorption compared to less volatile corneum (Fig. 5). In contrast, naphthalene is rapidly absorbed long-chain aliphatic hydrocarbons such as dodecane and hexa- and has minimal propensity to stay in skin, although it tends to decane. Based on data for these three compounds, the percu- partition into subcutaneous fat after absorption. The volatility taneous absorption is relatively low with the greatest estimated of these compounds could also be an important factor in penetration being only 1.5% of the applied dose. However, determining skin concentrations. Both dodecane and hexade- estimating total in vivo absorption is not only a function of the cane have significantly higher residual surface concentrations applied dose, but also of the surface area of skin to which the which result in longer surface residence times on skin, and thus jet fuel is exposed. Thus exposure to large surface areas for in effect longer dosing times. These different mechanisms of prolonged contact times could potentially result in significant absorption and deposition impact on interpreting what absorp- systemic absorption. tion and penetration mean. Compounds with enhanced skin There is limited data in the literature on the percutaneous residence, if inherently bioactive, might be expected to have a absorption of these three hydrocarbons in other experimental greater propensity for inducing skin initation than those com- models or vehicles. Percutaneous absorption studies of neat pounds which do not form skin depots. A11 of these hydrocar- naphthalene in humans or laboratory animals had not been bons, and especially the composite fuels, have the potential to reported in a recent toxicological profile (ATSDR, 1993). extract epidermal lipids thereby altering the barrier properties Absorption was inferred from the occurrence of toxicity after of skin. Such an effect was reported in a preliminary study of dermal application. A study of neat naphthalene absorption in JP-8 toxicity (Singh, 1998) and is consistent with toxicological rats suggested that 50% of the applied dose was excreted in findings in our laboratory that transepidermal water loss urine by 12 h, with the predominant urinary metabolites being (TEWL) increased with chronic jet fuel exposure (unpublished 2,7- and 1,2-dihydroxynaphthaIene (Turkall et al.. 1994). data). Naphthalene was also included in an in virro monkey skin These findings suggest that if toxicology studies of individ- study comparing the percutaneous absorption in acetone using ual components are conducted, the disposition of an aromatic a series of aliphatic compounds (Sartorelli et al., 1998). Log such as naphthalene cannot be used to extrapolate absorption octanovwater partition coefficients were correlated to perme- estimates to a long-chain aliphatic such as dodecane. The ability constant, with vapor pressure and molecular weight not optimal approach would be to select marker compounds that adding any significant increase in correlation. In the context of are characteristic of groups of fuel components based on struc- the present studies, these correlations would be to the observed ture-activity relationships, and then assess absorption, cutane- 0.020 DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS 69 0.015 z Iz 0.010 5 2 g 0.005 0.000 0 -Jet-A (n4) 0.004 123 4 5 Hours -P-8 (n=4) 4-JP-s(Puddle) ( ~ 4 ) *JP-8(100) (n=4) 0.003 2 I 5 0.002 5 z0L 0.001 0.000 0 -Jet-A ( n 4 ) 1 +P-8 (n=4) 23 Hours +P-l(PuddIe) (n=4) 45 *P-8(100) (n=4) FIG. 6. Naphthalene (a) and dodecane (b) perfusate absorption (mean % doselml 2 SEM) after dosing in Jet-A, JP-8, JP-8(puddle), and JP-8(100) jet fuels. ous disposition, as well as toxicological parameters (Basak and Grunwald, 1998). A surprising finding was the difference in naphthalene and dodecane absorption and skin deposition between the three fuels studied, since these fuels are predominantly composed of similar hydrocarbon components with different additive combinations. Absorption of dodecane was greater from JP-8 than from Jet-A or JP-8(IOO), while overall penetration of naphthalene was greatest from JP-8(100). Aging of the JP-8 by 24 h evaporation decreased the absorption for both of these markers compared to the other fuel types. These findings suggest that one or more of the additives combined with Jet-A to make JP-8 modulated dodecane absorption but had minimal effect on naphthalene absorption. The total additive package [corrosion inhibitor DC 1-4A, antistatic compound Stats 450. the deicing additive diethylene glycol monomethyl ether (DIEGME)] composes only a small percentage of the final fuel. A common attribute of these three Surface 0.7 T 0.6 aa 0.65 0.64 0.5 d- 0.4 0.3 k 0.2 r 0.1 0.0 Stratum Corneum 0.16 T a 0.151 I loo, a 95.13 a 95.89 a 95.24 a 94.30 I 40 e 20 Evaporative Loss (a) Naphthalene a 0.40 Dosed Skin Fat I 2.0 1.5 IPI 0.5 a Ta 2'5 1.63 a 2.18 0.0 I I Jct-A (n=4) Absorption E3JP-8 (n-4) JF'-8(Puddle) ( n 4 ) Penetration HJP-8(100)(n-4) FIG. 7. Pattern of tissue deposition and absorption of naphthalene (a) and dodecane (b) dosed in Jet-A. JP-8, JP-8(puddle),and JP-8(100)jet fuels. .411values are expressed as mean ?& dose t SEM. Values with different letters are significantly different (p < 0.05). 70 B `T DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS a 5.07 . ab 71 I 5 !` I 'k 2 1 0 Surface aa 5.13 5.16 Stratum Corneum - U 0.40 0.40 Evaporative Loss (b) Dodecane I 0.7 T I Jet-A (n=4) Dosed Skin a 0.63 Absorption C4 JP-8 (n4) I 1 1.2 1.0 H o.8 I 0.6 E & 0.4 0.2 0.0 JPP-8(Puddle()n=4) FIG. l d o n t i n u e d Fat a 1.11 Penetration 6P-8(100) (n-4) 72 RIVIERE ET AL. h 0 It 5 rm I II 0N 2 0 m 0 d i 0 rs '03 2 s7 E n 'c) - z s0 0 9 00 7 5 0 m I av -.0- 2 .5P s1 na DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS 73 TABLE 4 Statistical Significance (p Values) of Napthalene and of Dodecane Absorption and Skin Deposition from Four Jet Fuels Napthalene 12 Contrasts All 1 Dodecane 2 All Surface Stratum corneum Dosed skin Fat Absorption Penetration Evaporative loss AUC Peak flux Surface Stratum corneum Dosed skin Fat Absorption Penetration Evaporative loss AUC Peak flux .06 .18 .32 .72 .43 .61 .98 .11 .49 .OX .24 .44 .96 .34 .48 .99 .11 .27 Original Scale .06 .08 .66 .53 . I 1 .30 .045 .I7 .12 .22 .026 .07 .10 .38 .I2 -027 .64 . I O Ln scale .I3 .w .38 .52 .11 .35 .07 .28 .I7 .24 . O S .M .09 .38 .20 .027 .71 .043 .021 .OM .005 .24 .048 .001 .003 .013 .017 .006 .026 -001 .24 814 .ooo1 .004 .OX .015 .6 1 .024 .7--7 .15 .002 .OOO5 .I3 .0002 .m .59 .012 .25 .I6 .015 .002 .I5 .001 .029 .07 .017 .022 .21 .OOO8 .0002 .003 .OOO1 .0003 .033 .012 .006 .21 .005 .0003 .w .OOO1 .0002 Note. Contrasts: (Jet-A, JP-8, JP-I(hddle), JP-8(100)); I: Jet-A vs Others (3, -1. - 1, -1); 2: JP-8 vs JP-8(100) (0,1, 0, -1): All: Overall p value from ANOVA for four jet fuels. Bold p values are 50.05. compounds is their relative hydrophilicity compared to the hydrophobic environment of the hydrocarbon jet fuel. DC1-4A is a linoleic acid derivative. Linoleic acid has previously been identified as a skin penetration enhancer for topically applied drugs (Aungst et al., 1986; Mahjour et al., 1989; Schneider et al., 1996). The potential for DIEGME to function as a pharmaceutic enhancer has not been investigated, although in the preliminary study quoted above (McDougal and Miller, 1998), DIEGME had an order of magnitude greater transdermal flux of any hydrocarbon component of JP-8. Enhancers generally have their greatest effect on more nonpolar compounds for which the stratum corneum is a true barrier to absorption. These studies would seem to suggest that dodecane would be more susceptible to enhancement than naphthalene. Individual component absorption studies would have to be assessed to further probe this effect. Of equal significance is the finding that some of these effects (e.g., dodecane) were apparently reversed in JP-8(loo), suggesting further interactions with this more complex additive package. For naphthalene, the additive package which transforms JP-8 into JP-8(100) seemed to have increased its penetration. These findings are not totally unexpected, since we have reported on similar complex interactions when absorption of various organics (parathion, carbaryl. benzidine) was studied from defined chemical mixtures (Baynes et al., 1996. 1997; Qiao er al., 1996; Williams er al., 1996). Possible interactions include chemical binding, chemical effects from surfactants, stratum corneum lipid interactions from enhancers, solventmediated lipid extraction, low level irritation, and vasomodulation. Dankovic (1989) had reported on prolonged dermal residence times for benzo[a]pyrene when applied in a mixture of 11 polycyclic aromatic hydrocarbons compared to application in a volatile solvent. All of this previous work suggests that such subtle changes in absorption from mixtures is not unexpected. Finally, the absorption of these additives was not investigated in this study, and their potential biological effects on skin could not be assessed. Since these additives are more polar than the hydrocarbon fuel constituents, they would be expected to be more sensitive to modulation by fuel composition. Further speculation on the mechanism of these jet fuel interactions, and on identifying responsible components, is not warranted at this time but requires further probing. It must be stressed that these studies were designed only to probe whether absorption of two marker hydrocarbon components of jet fuel were affected by addition of performance additives. It is intriguing that there were some significant differences in absorption of the two markers studied. The absorption of both markers was reduced from the aged JP8(puddle).This is not totally unexpected since the composition of this fuel would be expected to be the most different from the 74 RIVIERE ET AL. other fuels since a large fraction of its most volatile components are lost. This could alter partitioning of the markers from the fuel into the stratum corneum lipids. Second, one must also stress that selection of naphthalene and dodecane was not based on any toxicological property, but only by their virtue of representing a fraction of fuel. It is possible that if any cutaneous toxicity were produced from jet fuel exposure that the additive components themselves could be responsible. If the markers studied, naphthalene and dodecane, were affected by fuel composition, it is entirely feasible that the disposition of the more polar additives could be modulated. This modulation Clark, C. R., Walter, M. K., Ferguson, P. W., and Katchen. W. (1988). Comparative dermal carcinogenesis of shale and petroleum-derived distillates. Toxicol. Ind. Health 4, 11-22. Dankovic, D. A., Wright. C. W., Zangar. R. C., and Springer, D. L. (1989). Complex mixture effects on the dermal absorption of benzo[a]pyrene and other polycyclic aromatic hydrocarbons from mouse skin. J. Appl. Toxicol. 9,239-244. Harris, D. T., Sakiestewa. D., Robledo, R. F., and Witten, M. (1997). Shortterm exposure to JP-8 jet fuel results in long term immunotoxicity. Toxicol. Ind. Health. 13, 559-570. Hsu, J. C. (1996). Multiple Comparisons: Theory and Merhods. Chapman and Hall, London, UK. could alter the propensity to induce cutaneous toxicity. This Mahjour, M., Mauser, B. E., and Fawzi, M. B. (1989). Skin permeation hypothesis has not been evaluated in this work. In conclusion, assessment of dermal absorption of jet fuels is extremely complex. We have characterized the absorption of three different hydrocarbon marker compounds and demonstrated significant differences in individual component absorption and skin deposition. It must be stressed that we are only looking at 3 of the 228 hydrocarbon components that comprise enhancement effects of linoleic acid and z o n e on narcotic analgesics. Inr. J. Pharm. 56, 1-11. Manitz, R., Lucht, W., Strehmel, K., Weiner, R., and Neubert, R. (1998). On mathematical modeling of dermal and transdermal drug delivery. 1.Pharm. Sei. 87, 873-879. McCarley, K. D., and Bunge. A. L. (1998). Physiologically relevant onecompartment pharmacokinetic models for skin. I. Development of models. J. Pharm. Sei. 87,470-481. these fuels and there is no reason to believe that these specific McDougal, J. N., and Miller, T. E. (1998). Dermal Absorption 0fJP-8 and its components are toxicologically significant. However, before a complete risk assessment can be made, some estimate of der- Componenrs. Proceedings of AFOSR JP-8 Jet Fuel Toxicology Workshop, Tucson, AZ. mal absorption and deposition is required. Second, it is critical Haff, J., Parton, K., Lantz, R., Chen, H., Hays, A., and Witten, M. (1995). that when making these assessments, both absorption into perfusate as well as penetration into skin be assessed to obtain Inhalation exposure to JP-8 jet fuels alters pulmonary function and substance P levels in Fischer 344 rats. J. Appl. Toxicol. 15, 249-256. 'f a complete picture of disposition. Finally, the effects of fuel Qiao, G.L., Brooks, J. D., Baynes, R. L., Monteiro-Riviere, N. A,, Williams, P. L., and Riviere, J. E. (1996). The use of mechanistically defined chemical '.. additives on the disposition of naphthalene and dodecane sug- mixtures (MDCM) to assess component effects on the percutaneous absorp- gests that all fuels cannot be treated as one and furthermore, the tion and cutaneous disposition of topicallyexposed chemicals. I. Studies mechanisms of these specific interactions deserve clarification. with parathion mixtures in isolated perfused porcine skin. Toxicol. Appl. Pharmacol. 141,473-486. ACKNOWLEDGMENT This work was supported by the U.S. Air Force Office of Scientific Research, Grants F49620 and FQ867 i. REFERENCES ATSDR (1993). Draft Toxicological Rofile for Naphthalene - Update. Agency for Toxic Substances and Disease Registry, Atlanta, GA. ATSDR (1995). Toxicological Profile for Jet Fuels (JP4 and JW). Agency for Toxic Substances and Disease Registry, Atlanta, GA. ATSDR (1996). Draft Toxicological Profile for Jet Fuels (JP-5 and JP-8). Agency for Toxic Substances and Disease Registry, '4tlanta. GA. Riviere, J. E., Bowman, K. F., Monteiro-Riviere, N. A,, Dix, L. P.,and Carver, M.P. (1986). The isolated perfused porcine skin flap (IPPSF). I. A novel in vitro model for percutaneous absorption and cutaneous toxicology studies. Fundam. Appl. Toxicol. 7, 444-453. Riviere, J. E., Brooks, J. D.. Williams, P. L., and Monteiro-Riviere, N. A. (1995a). Toxicokinetics of topical sulfur mustard penetration, disposition, and vascular toxicity in isolated perfused porcine skin. Toxicol. Appl. P h a ~ ~ ~ ~13c5o, 2l 5. -34. Riviere, J. E., Monteiro-Riviere, N. A.. and Williams, P. L.(1995b). Isolated perfused porcine skin flap as an in vitro model for predicting transdermal pharmacokinetics. Eur. J. Pharm. Biopham. 41, 152-162. Riviere, J. E., and Monteiro-Riviere, N. A. (1991). The isolated perfused porcine slun flap as an in vitro model for percutaneous absorption and cutaneous toxicology. Crirical Reviews in Toxicol. 21, 329-344. Aungst, B. J., Rogers. N. J., and Shefter, E. (1986). Enhancement of naloxone penetration through human skin in vitro using fatty acids, fatty alcohols. surfactants, sulfoxides and amines. In?. J. Pharm. 33, 225-234. Basak, S. C., and Grunwald, G.D. (1998). Clustering ofJP-8 Constituents into Strucrurally Dissimilar Groups: A Novel Computational Strategy for Predictive Toxicology. Proceedings of AFOSR JP-8 Jet Fuel Toxicology Work- shop, Tucson, AZ. Baynes, R. E., Brownie, C., Freeman, H., and Riviere, J. E. (1996). In vitro percutaneous absorption of benzidine in complex mechanistically-defined chemical mixtures. Toxicol. Appl. Pharmacol. 141, 497-506. Baynes, R. E., and Riviere. J. E. (1997). Inert ingredients in pesticide formulations influence the dermal absorption of carbaryl. .4m. J. Vet. Res. 59, 159-169. Riviere, J. E..Williams, P. L., Hillman, R., and Mishky, L. (1992). Quantita- tive prediction of transdermal iontophoretic delivery of arbutamine in humans using the in vitro isolated perfused porcine skin flap (IPPSF). J. Pharm. Sei. 81, 504-507. Sartorelli, P., Aprea. C., Cenni, A,, Novelli, M. T., Orsi, D., Palmi, S., and Matteucci, G. (1998). Prediction of percutaneous absorption from physiochemical data: A model based on data of in vitro experiments. Ann. Occup. Hyg. 42, 267-276. Schneider, I. M.. Dobner. B., Neubert, R., and Wohlrab, W. (1996). Evaluation of drug penetration into human skin ex vivo using branched fatty acids and propylene glycol. Inr. J. Pharm. 145, 187-196. Singh. J. (1998). Skin Permeability, Biophysics, and Irritation from JP-8. Proceedings of AFOSR JP-8 Jet Fuel Toxicology Workshop, Tucson, AZ. l 1 i I i ! DERMAL ABSORPTION AND DISTRIBUTION OF JET FUELS 75 Swallow, W. H.(1984). Those overworked and oft-misused mean separation procedures-Duncan's, LSD, etc. Plunr Dis.68,919-921. relevant pharmacokinetic model of xenobiotic percutaneous absorption utilizing the isolated perfused porcine skin flap (IPPSF). J. Phann. Sci. 79, Turkall, R. M., Skowronski, G. A., Kadry, A. M.,and Abdel-Rahman, M. S. 305-311. (1994). A comparative study of the kinetics and bioavailability of pure and Williams, P. L., and Riviere, J. E. (1995). A biophysically-based dermato- soil-adsorbednaphthalene in dermally exposed rats. Arch. Environ. Contam. Toxicol. 26, 504-509. pharmacokinetic compartment model for quantifying percutaneous penetration and absorption of topically applied agents. J. Phann.Sci. 84,599-608. Wester, R. C.. Melendres, J., Sedik, L., Maibach, H. I.. and Riviere, J. E. Williams, P. L.,Thompson, D., Qtao, G. L., Monteiro-Riviere,N. A., Baynes, (1998). Percutaneous absorption of salicylic acid, theophylline, 2.4-dimeth- R. L., and Riviere, J. E. (1996). The use of mechanistically defined chemical ylamine, diethyl hexylphthalic acid and paminobenzoic acid in the isolated mixtures (MDCM) to assess component effects on the percutaneous absorp , perfused porcine skin Rap compared to man. Toxicol. Appl. Phamcol. 151, tion and cutaneous disposition of topicallyexposed chemicals. I. Develop- 159-165. ment of a general dermatopharmacokinetic model for use in risk assessment Williams, P. L., Carver, M. P., and Riviere, J. E. (1990). A physiologically Toxicol. Appl. P h a m c o l . 141,487-496.