Comparative In Vivo Toxicity of Topical JP-8 Jet Fuel and Its Individual Hydrocarbon Components: Identification of Tridecane and Tetradecane as Key Constituents Responsible for Dermal Irritation

Abstract

Despite widespread exposure to military jet fuels, there remains a knowledge gap concerning the actual toxic entities responsible for irritation observed after topical fuel exposure. The present studies with individual hydrocarbon (HC) constituents of JP-8 jet fuel shed light on this issue. To mimic occupational scenarios, JP-8, 8 aliphatic HC (nonane, decane, undecane, dodecane, tridecane, tetradecane, pentadecane, hexadecane) and 6 aromatic HC (ethyl benzene, o-xylene, trimethyl benzene, cyclohexyl benzene, naphthalene, dimethyl naphthalene) soaked cotton fabrics were topically exposed to pigs for 1 day and with repeated daily exposures for 4 days. Erythema, epidermal thickness, and epidermal cell layers were quantitated. No erythema was noted in 1-day in vivo HC exposures but significant erythema was observed in 4-day tridecane, tetradecane, pentadecane, and JP-8 exposed sites. The aromatic HCs did not produce any macroscopic lesions in 1 or 4 days of in vivo exposures. Morphological observations revealed slight intercellular and intracellular epidermal edema in 4-day exposures with the aliphatic HCs. Epidermal thickness and number of cell layers significantly increased (p < 0.05) in tridecane, tetradecane, pentadecane, and JP-8-treated sites. No significant differences were observed in the aromatic HC-exposed sites. Subcorneal microabscesses containing inflammatory cells were observed with most of the long-chain aliphatic HCs and JP-8 in 4-day exposures. Ultrastructural studies depicted that jet fuel HC-induced cleft formation within intercellular lipid lamellar bilayers of the stratum corneum. The degree of damage to the skin was proportional to the length of in vivo HC exposures. These data coupled with absorption and toxicity studies of jet fuel HC revealed that specific HCs (tridecane and tetradecane) might be the key constituents responsible for jet fuel-induced skin irritation.

Introduction

Aliphatic and aromatic hydrocarbon (HC) mixtures are the primary constituents of JP-8 jet fuel. JP-8 is the major fuel used by U.S. and NATO armed forces, and is also a multipurpose fuel used in ground vehicles, generators, heaters, and stoves (Makris, 1994). Although the component mixture can vary from batch to batch, the aliphatic HCs tend to dominate the aromatic HC in each fuel batch. On the average, JP-8 jet fuel is composed of 33–61% n-alkanes and isoalkanes, 10–45% naphtenes, 12–22% aromatics, and 0.5–5% olefins (Vere, 2003). Occupational exposures to jet fuel can occur through fuel transport, aircraft fueling and defueling, cold aircraft engine starts, aircraft maintenance, maintenance of equipment and machinery, cleaning or degreasing with fuel, and use of tent heaters (Centers for Disease Control, 1999; Subcommittee on Jet-Propulsion 8 fuel of Committee on Toxicology, 2003).

Skin can be an important route of exposure because of the potential for liquid and aerosol contact with fuel (McDougal and Rogers, 2004). The jet engine maintenance personnel wear fuel-permeable cotton coveralls to reduce the possibility of explosion due to the generation of static electricity associated with more protective clothing. Daily exposure to fuels result in saturation of the cotton cloth, resulting in an occluded environment for repeated, long-term exposure to the skin during the typical 8-hours workday (Allen et al., 2001). There are several reports that jet fuel can cause local and systemic toxic effects. Significant effects on the immune, hepatic, neurological, and respiratory systems have been observed in several animal exposure studies (Grant et al., 2000; Harris et al., 2000; Robledo et al. 2000). Neurological effects and irritant dermatitis have been reported in workers exposed to JP-8 jet fuel (Smith et al., 1997; Zeiger and Smith, 1998).

The different HC of jet fuel have shown the potential for percutaneous absorption or skin retention indicating a potential source for systemic or local toxicity (Riviere et al., 1999; McDougal et al., 2000; Baynes et al., 2001). Previously, we have studied the dose-related HC absorption through skin (Muhammad et al., 2004a) suggesting that prolonged skin exposure to jet fuel may result in enhanced toxic effects. Monteiro-Riviere et al. (2001a) investigated the cutaneous toxicity of 3 jet fuels and concluded that the high-dose, fabric-soaked repeated exposure to Jet A, JP-8, and JP-8 + 100 fuels caused the greatest increase in cutaneous erythema, edema, epidermal thickness, cell layers, and rete peg depth compared with high-dose nonoccluded or low-dose exposure under occluded (HillTop chambers) and nonoccluded conditions. The ultrastructural analysis of skin exposed to 3 jet fuels revealed low-level inflammation accompanied by the formation of lipid droplets in various skin layers, mitochondrial and nucleolar changes, cleft formation in the intercellular lipid lamellar bilayers, as well as disorganization at the stratum granulosumstratum corneum interface. These changes suggest that the primary effect of jet fuel exposure is damage to the stratum corneum barrier (Monteiro-Riviere et al., 2004). Since all fuels demonstrated a similar toxicologic profile, hydrocarbon constituents—and not additives—are the primary toxic entities. Studies in our laboratory have indicated that individual and specific aliphatic HC of jet fuel are toxic to human epidermal keratinocytes (HEK) and are capable of inducing release of proinflammatory cytokines such as IL-8 (Chou et al., 2002). We have shown that IL-8 concentration increased significantly by 3- to 10-fold, with the highest increase associated with exposure to hydrocarbons in the C9–C13 chain length. In addition, the cytotoxicity of jet fuel aromatic HC depicted a dose-related response in IL-8 release at 24 hours from HEK (Chou et al., 2003).

Recently, we studied the percutaneous absorption of different jet fuel HCs through the skin previously exposed to JP-8 jet fuel (Muhammad et al., 2005). We observed 2-to 4-fold increase in absorption of short chain aliphatic (nonane, undecane, dodecane, tridecane) and aromatic (ethyl benzene, o-xylene, trimethyl benzene, cyclohexyl benzene, naphthalene, dimethyl naphthalene) HC through 1 and 4 days of JP-8 preexposed skin. Jet fuels and individual HCs may cause lipid extraction from the stratum corneum, as studied with Fourier Transform Infra Red (FTIR) spectroscopy (Muhammad et al., 2005) and transmission electron microscopy (TEM) (Monteiro-Riviere et al., 2004), which could provide a mechanism to explain the increase in HC absorption.

Since our cell culture studies suggested that the aliphatic HCs in the range of C9–C13 are more cytotoxic, we hypothesized that specific HCs may be responsible for the skin induced irritation from JP-8 exposure, and second that differences observed between jet fuels in vivo are due to additive modulation of cytotoxic hydrocarbon disposition. There are no published reports on the in vivo irritation with individual or specific HCs of jet fuels. Therefore, the objective of this study was to access the irritation caused by the individual HC (nonane, decane, undecane, dodecane, tridecane, tetradecane, pentadecane, hexadecane, ethyl benzene, o-xylene, trimethyl benzene, cyclohexyl benzene, naphthalene, and dimethyl naphthalene) at 1 and 4 days of in vivo exposures and to compare these HC toxic effects to that caused by JP-8 mixture.

Materials and Methods

Test Compounds

Eight aliphatic HCs (n-nonane (C9), n-decane (C10), n-undecane (C11), n-dodecane (C12), n-tridecane (C13), n-tetradecane (C14), n-pentadecane (C15), and n-hexadecane (C16)) all with greater than 98% purity, were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Six aromatic HC (ethyl benzene, o-xylene, 1,2,4-trimethyl benzene, cy-clohexyl benzene, 1,2-dimethyl naphthalene, all with greater than 98% purity, were purchased from Sigma Chemical Co. (St. Louis, MO, USA) except naphthalene, which was purchased from Supelco Inc. (Park Bellefonte, PA, USA). A 1.1% solution of naphthalene was prepared in ethanol. Major T. Miller from Wright Patterson Air Force Base supplied JP-8 jet fuel.

In Vivo Experimental Procedures

These experiments were conducted to mimic complete fuel HC saturation of cotton clothing in workers and to assess their effects after 1 day and with multiple exposures for 4 days with evaluation on day 5. Pigs (n = 4 for each of 1- and 4-day in vivo exposures) were sedated with an intramuscular injection of ketamine/xylazine/telazol and hairs were clipped from the back carefully 24 hours prior to topical application of the HC. Templates were created from Stomahesive (Convatec-Squibb, Princeton, NJ) within a dosing area of 1 inch² and a 0.75 cm border for attachment to the skin with Medipore tape. The 100% cotton fabric was placed directly inside the template and dosed with 300 μl, the volume required to achieve full saturation of the JP-8 and appropriate aliphatic HCs (nonane, decane, undecane, dodecane, tridecane, tetradecane, pentadecane, and hexadecane) and aromatic HC (ethyl benzene, o-xylene, trimethyl benzene, cyclohexyl benzene, naphthalene, and dimethyl naphthalene) on the cotton fabric. Nontreated fabric served as the controls while fabric soaked with ethanol served as the sham control. At 24 hours and 96 hours post dose, the templates were removed from the skin and scored for erythema as follows: 0 = no significant change; 1 = very slight erythema (barely perceptible); 2 = slight erythema, pale red in defined area; 3 = moderate to severe erythema, red in well-defined area; and 4 = severe erythema, beet red in defined area (Table 1, footnote). After scoring, two 6-mm skin biopsies were taken from each site and processed as follows: 1 was fixed in 10% neutral buffered formalin (NBF) for light microscopy (LM) and the other biopsy was embedded in OCT (Sakura Finetek, Torrance CA) and frozen in isopentane cooled by liquid nitrogen for stratum corneum lipid staining. After the biopsy sampling, all pigs were euthanized with 5 ml of Beuthanasia-D Special solution (Schering-Plough Animal Health Corp., Kenilworth, NJ). All animal procedures were approved by the NCSU IACUC.

Microscopy Techniques

For light microscopy (LM), all skin biopsies were fixed in 10% NBF, processed through graded ethanols, and embedded in paraffin blocks. Skin sections were oriented perpendicular to the intended plane of section (Monteiro-Riviere, 1991) in the block, and then transverse sections of ~6 μm thickness were cut and mounted on slides and stained with hematoxylin and eosin (H&E). Epidermal thickness excluding the stratum corneum was quantitated with an eyepiece reticle on the H&E-stained sections. Morphometric quantification was conducted using an Olympus PM-10ADS automatic photomicroscope (Olympus Optical Co., Tokyo, Japan) with a ×40 objective and calibrated ocular micrometer. The epidermal thickness and the number of nucleated epidermal cell layers (viable epidermis) were counted at 3 randomly selected areas of epidermis (devoid of microabscesses) per slide and averaged.

Transmission electron microscopy (TEM) was used to examine the stratum corneum lipids. For staining of stratum corneum lipids, frozen biopsies were sectioned (20 μm) on a cryostat, mounted on positively charged slides, air dried, and fixed in Trump’s fixative (4% formaldehyde, 1% glutaraldehyde in phosphate buffer). The sections were postfixed in 0.25% phosphate-buffered ruthenium tetroxide (Polysciences, Inc.) for 45 minutes at 4°C to preserve and stain the intercellular lipids of stratum corneum. Sections were dehydrated through graded ethanols, cleared in acetone, infiltrated and embedded in Spurr’s resin. Thin sections (800–1,000 Å) were mounted on copper grids and examined on a Philips EM208S TEM operating at an accelerating voltage of 80 kV.

Statistical Analysis

For statistical analysis, the means of erythema, epidermal thickness, and number of epidermal cell layers were calculated and the significant differences (p < 0.05) between post-treatment times, 1 day and 4 days of exposure to HC treated sites were determined using the least significant difference in the ANOVA procedure of SAS (version 8.1 for windows; SAS Institute, Cary, NC). Multiple comparisons among different treatments were conducted within each exposure length using the LSD at the p < 0.05 level of significance.

Results

Macroscopic observations (mean erythema scores) for all the tested HCs and JP-8 are summarized in Table 1. After 1-day in vivo exposure, no significant differences in erythema were observed in the HC treatments except for tridecane, tetradecane, and pentadecane compared to the other aliphatic HC, aromatic HC, ethanol, JP-8 treated, and controls. These 3 specific HCs exhibited moderate erythema. After 4 days of in vivo exposure, erythema was much greater in most treatments compared to the 1-day exposure (Table 1, Figure 6). Tridecane, tetradecane, and pentadecane erythema scores were similar to JP-8 after 4 days of exposures and significantly different from that of controls, ethanol, aromatic, and other aliphatic HCs such as decane, undecane, dodecane, and hexadecane. There were no significant differences among the tested aromatic HCs as compared to the ethanol or control (Table 1).

Table 2 represents the epidermal thickness of various studied HCs after 1 day and 4 days of in vivo exposures. The mean epidermal thickness was greater in tridecane, tetradecane, pentadecane, and JP-8 treated sites in both 1- and 4-day in vivo exposures as compared to the control. There were no significant differences in epidermal thickness for short-chain aliphatic HCs (nonane, decane) and aromatic HCs as compared to the controls. The mean number of epidermal cell layers showed an increase with longer carbon chain lengths of aliphatic HCs (Table 3). The HC with a chain length of C13–C15 (tridecane, tetradecane, pentadecane) showed a significant increase in the number of epidermal cellular layers similar to JP-8 after 4 days of in vivo exposures compared to the controls. The mean number of cell layers in the aromatic HC was similar to controls (Table 3). A comparison of the present in vivo findings with previous in vitro absorption, cytotoxicity, and IL-8 release data from our laboratory for the aliphatic HC (C11-C16) was presented in Table 4. This comparison revealed that the HC with the carbon chain length of C13 (tridecane) and C14 (tetradecane) were more toxic based on the increase in severity of the 4-day microabscesses, the increase in epidermal thickness, the increase in erythema and cytotoxicity, as well as the increase in IL-8 release in HEKs.

Microscopically, different HCs exhibited different inflammatory responses in porcine skin. The maximum inflammation was observed with tridecane and tetradecane after 1 day of in vivo exposures (Figures 1c and 1d). Intercellular and intracellular epidermal edema and focal areas of dermal inflammatory cells were present. Tridecane produced subcorneal microabscesses containing a mixed population of neutrophils and lymphocytes (Figure 2a) in all of the 1-day exposures. Tetradecane and pentadecane contained subcorneal microabscesses in 3/4 pigs in the 1-day exposure probably due to uneven contact of the saturated fabric. JP-8 did not produce subcorneal microabscesses after 1 day. Evaluation of all of the aromatic HCs were similar to controls except for slight intraepidermal edema. The most significant morphological changes occurred after 4 days. There was an increase in intracellular epidermal edema, intercellular epidermal edema, dermal papillary edema, extensive epidermal rete pegs, and dermal inflammation at the 4-day exposures (Figsure 2c–2d) as compared to 1-day HC (Figures 1b–1d and Figures 2a–2b). Microscopic observations of all the 4-day treated sites with tridecane, tetradecane, and pentadecane exhibited numerous subcorneal and intracorneal lesions containing parakeratotic cells bordered by a reforming stratum corneum and containing a mix inflammatory population of neutrophils and lymphocytes (Figures 2c–2d). JP-8-treated skin contained focal subcorneal microabscesses with neutrophils and lymphocytes. After 4 days of in vivo exposures to undecane and dodecane, focal areas of parakeratosis were exhibited. The aromatic HC-exposed sites for 4 days depicted only a slight increase in intracellular epidermal edema (Figures 3b–3d).

TEM was used to examine the ultrastructural organization of the intercellular bilipid layers between the stratum corneum layers after fixation with ruthenium tetroxide. Both aliphatic and aromatic HC caused similar damage to the SC intercellular lipids and the intensity of damage was enhanced with prolonged exposure to these HC (Figures 4b–4d and Figures 5b–5d). In control skin, the SC layers appeared normal unaltered with intact desmosomes and normal compact bilipid layers. In 1 day JP-8-treated skin (Figure 4b) no significant ultrastructural changes were observed except for small lacunae. However, in the 4-day JP-8 (Figure 5b), individual aromatic HC (e.g., o-xylene, Figure 5c) and aliphatic HCs (e.g., tetradecane, Figure 5d) exhibited large lacunae (areas with loss of electron-lucent and electron-dense lamellae).

Discussion

Repeated or prolonged exposure to jet fuels is more reflective of the occupational norm. Studies have shown that repeated dermal exposures to jet fuels result in skin irritation (Kinkead et al., 1992; Baker et al., 1999; Monteiro-Riviere et al., 2001a). However, the specific causative fuel components within JP-8 have not been identified. Previously, our laboratory has demonstrated that Jet A, JP-8, and JP-8 (100) all release pro-inflammatory cytokines such as IL-8 from normal human epidermal keratinocytes (Allen et al., 2000). However, these studies also revealed that there were no significant differences among the 3 different fuel types with respect to IL-8 release. Therefore, one may propose that the components responsible for this response must exist in all 3 fuels, eliminating performance additives as the causative mechanism (Allen et al., 2001). Additionally, differences between fuels in vivo as well as differences in absorption of components between fuels may be due to modulation of the HC dermal absorption by additives, an effect not seen in culture (Riviere et al., 1999; Baynes et al., 2000; Muhammad et al., 2004b).

Since jet fuels are complex mixtures of aliphatic and aromatic HCs, we selected a wide range of aliphatic (C9–C16) and aromatic HCs based on our previous cell culture studies (Chou et al., 2002, 2003) and observed their irritant effects in pigs after 1 and 4 days of HC exposures. Pigs were selected as our experimental model because their skin is morphologically, physiologically, and biochemically similar to human skin (Monteiro-Riviere and Riviere, 1996; Monteiro-Riviere, 2001b). Literature on the dermal toxicity of jet fuel HC is limited. Our laboratory has demonstrated the dermal absorption and distribution profiles for some of the HC (Riviere et al., 1999; Baynes et al., 2000; Muhammad et al., 2004a). In addition, we have observed the in vitro cytotoxicity and IL-8 release from HEK with different clusters of aliphatic and aromatic HCs (Allen et al., 2001; Chou et al., 2002, 2003).

The present in vivo studies suggest that topical application of JP-8 and specific aliphatic HCs on pig skin results in dermal irritation after repeated exposure for 4 days. Erythema and epidermal hyperplasia were pronounced after 4 days of repeated application of the aliphatic HC. A hydrocarbon-specific response was demonstrated by epidermal thickness and the number of epidermal cell layers, with tridecane and tetradecane having the greatest proliferative effect followed by JP-8 and pentadecane after 4 days of in vivo exposures. The short chain aliphatic HCs such as nonane, decane, and undecane produced only mild erythema after 4 days of in vivo exposures (Figure 6). These findings are in accordance with Brown and Box (1970), who studied the skin irritancy of alkanes and reported that n-decane was slightly irritant with some epidermal thickening, while n-tetradecane was more irritant with epidermal thickening. Horton et al. (1957, 1966) have demonstrated the importance of long-chain aliphatic HCs as accelerators of skin carcinogenesis. Results from Sice (1966) and Baxter and Miller (1987) demonstrated that the tumor-promoting activity of alkanes is related to their chain length, with maximum activity found in C12–C14 alkanes. The present histological findings support the in vitro studies of Allen et al. (2001), who reported that maximum IL-8 was released with C-13 from HEK and also with Chou et al. (2002) who found the IL-8 peak at mid-chain lengths of aliphatic HC.

The macroscopic and microscopic observations for JP-8 in the present study are similar to our earlier studies (Monteiro-Riviere et al., 2001a), indicating the reproducibility of in vivo effects of jet fuels. Subcorneal microabscesses were observed in all of the 4-day JP-8, and most of the tridecane-, tetradecane-, and pentadecane-treated sites. These findings are similar to Monteiro-Riviere et al. (2001a) and other kerosene studies cited in literature. Similar findings were observed in studies of Tagami and Ogino (1973) who found that after 24 hours of exposure with kerosene-soaked clothing in volunteers, intraepidermal and subcorneal vesicles containing neutrophils were present. Tridecane-, tetradecane-, and pentadecane-treated sites primarily consisted of multifocal parakeratotic lesions with an inflammatory crust consisting of a few neutrophils and lymphocytes after 4 days. Although tridecane showed this subcorneal microabscesses in all 4 pigs, tetradecane and pentadecane exhibited these microabscesses in 3/4 of the pigs after 1 day of exposure. This could be due to the fact that these lesions were focal and the biopsies did not include the specific area of damage. Serial histologic step-sectioning of the paraffin blocks did not reveal any additional subcorneal microabscesses. The typical acute subcorneal microabscesses observed after 1 day of exposure did not resemble the multifocal parakeratotic inflammatory lesions present after 4 days of exposures with the aliphatic HC. These parakeratotic lesions are different and are consistent with the healing phase of the early 1-day subcorneal microabscess. This could signify that during the 4 days of repeated exposure, the individual aliphatic hydrocarbons stimulated a hyperproliferative epidermis that contributes to the progression of the lesion to the stratum corneum, but is resistant to the development of the new subcorneal micropustules. This increase in epidermal hyperplasia seen with the 4-day exposures is typical of a chronic inflammatory process. Although JP-8 consists of a mixture of these hydrocarbons, the relative concentration of each HC is less in the JP-8 mixture, therefore explaining the presence of subcorneal microabscess taking 4 days to develop. The individual aliphatic HCs placed on the skin are more concentrated than found in the JP-8 mixture with aromatics and therefore could explain the rapid development of the hyperproliferative state and the progression of the subcorneal lesion into the stratum corneum that was observed in the histology.

Aromatic HC such as ethyl benzene, o-xylene, trimethyl benzene, cyclohexyl benzene, naphthalene, and dimethyl naphthalene did not produce any macroscopic or significant microscopic changes in epidermal thickness or lesions after 1 or 4 days of in vivo exposures. In a 13-week inhalation study, rats and beagle dogs were exposed to mixed xylenes, with no changes in body weight gain observed relative to controls, and no xylene-related alterations in histopathology (Carpenter et al., 1975). Jenkins et al. (1970) conducted inhalation studies on rats, guinea pigs, dogs, and monkeys to the vapors of o-xylene for 8 hours per day, 5 days per week, for 6 weeks, or for 90 or 127 days continuously, reporting no effects seen on body weight gain, hematological parameters, blood chemistry, or tissue histology.

Biophysical, morphological and biochemical data indicate that the stratum corneum forms a continuous sheath of protein-enriched corneocytes embedded in an intercellular matrix enriched in nonpolar lipids and organized as lamellar lipid layers (Elias, 1983; Landmann, 1986; Elias and Menon, 1991). Lipid layers in the stratum corneum interstices were first visualized by freeze-fracture technology (Elias et al., 1977; Landmann, 1986). However, this method bypasses the more polar domains because of the preferential deviation of the fracture plane to the most hydrophobic surfaces (Hou et al., 1991). The use of ruthenium tetroxide fixation allows for detailed studies of the spatial organization of the intercellular lipids of the stratum corneum (Madison et al., 1987; Fartasch et al., 1993; Fartasch and Ponec, 1994). In the present study, we observed intercellular disruption, loss of cohesion between the layers, and extraction of lipids from some layers. Specific individual HCs caused the same magnitude of damage to stratum corneum, as did the JP-8. Similar disorganization of bilayers have been reported in jet fuels-treated porcine stratum corneum (Monteiro-Riviere et al., 2004), in humans after application of sodium dodecyl sulfate or acetone irritants (Fartasch, 1997), in humans after kerosene exposure that induced large lacunae in the horny layers (Lupulescu et al., 1973), in murine stratum corneum after topical application of silicone (Menon and Ghadially, 1997), in porcine selective lipid extraction studies in different body regions (Monteiro-Riviere et al., 2001c), and in petrolatum-treated murine skin (Ghadially et al., 1992). It is well known that lipid extraction severely affects the biophysical properties of the horny layer (Wolfram et al., 1972). The present findings clearly indicate that JP-8 and specific individual HC can extract the lipids from the stratum corneum even after 24-hour exposure. These findings support the proposed mechanism of lipid extraction for the increase in HC absorption through the JP-8 preexposed skin (Muhammad et al., 2005). Although aromatic HCs such as o-xylene did not produce macroscopic or microscopic skin lesions, they had the potential to extract lipids out of stratum corneum of skin (Figures 4c and 5c). Thus, aromatic HCs can lead to an increase in percutaneous absorption on subsequent dermal exposures to themselves or to other chemicals (Muhammad et al., 2005). The ability of aromatic HCs to extract lipids out of stratum corneum of skin as investigated with TEM in the present study confirms our previous finding of stratum corneum delipidization with these aromatic HCs investigated with FTIR (Muhammad et al., 2005).

Macroscopic and microscopic findings in the present study indicate that individual aliphatic HCs (tridecane, tetradecane, pentadecane) are the principal source of jet fuel-induced irritation. A comparison of absorption and toxicity data for aliphatic HCs is shown in Table 4. By comparing all these data, it is evident that tridecane is the last HC that is absorbed through the skin, and tetradecane is the first HC retained in the skin. It appears that this transition between absorption and retention of aliphatic HC coincides with a HC’s propensity to induce skin irritation. Furthermore, the erythema, increase in epidermal thickness, and the microabscesses produced by tridecane and tetradecane alone (not as a mixture as in JP-8) after 1 and 4 days of in vivo exposures duplicate the same type of irritation produced with the JP-8 mixture. Based on these studies it is postulated that tridecane and tetradecane may be the 2 most important hydrocarbons responsible for jet fuel-induced skin irritation.