Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Paraphenylenediamine (1,4-diaminobenzene) dihydrochloride (PPD), an arylamine, is one of the most common allergens among patients with allergic contact dermatitis (Schnuch et al., 1997). PPD is the most widely used primary intermediate in hair dye formulations (Corbett and Menkart, 1973). This compound is also used as photographic developing agent and as an intermediate in the manufacture of azo dyes, antioxidants, and accelerators for rubber vulcanization (Hansen et al., 1993).

The molecular mechanism behind the recognition of PPD by the immune system has not been fully elucidated. According to the classic studies by Landsteiner and Jacobs (1936), such small molecules need to form reactive metabolites, which in turn may lead to the formation of immunogenic hapten-protein conjugates and then are presented to the immune system. Therefore, an understanding of the links between bioactivation, detoxification, and covalent binding is essential to assess the potential of a substance for an immunological response. The nature and quantity of metabolites may be used as indicators for metabolic reactivity and to identify pathways involved in the detoxification of reactive metabolites. In turn, identification of the critical target proteins altered by covalent adducts may further help to unravel the various processes. In case of PPD, only one oxidation product is known so far. PPD may be oxidized to benzoquinone diimine, which, in turn may form the trinuclear dye N,N'-bis(4-aminophenyl)-2,5-diamino-1,4-quinone-diimine called Bandrowski's base (BB). Krasteva et al. (1993) reported that BB is involved in contact dermatitis to PPD.

As an arylamine, PPD may lose its reactivity through N-acetylation. Therefore, N-acetylation of PPD may be an important step for detoxification or activation of this agent. However, clear experimental evidence is lacking. Metabolic activation via acetylation has indeed recently been reported for the arylamine benzidine (Smith et al., 1992; Zenser et al., 1996). Smith et al. (1992) reported that the monoacetylated metabolite of benzidine is more toxic than benzidine itself. Arylamine N-acetylation is carried out by two isoenzymes in humans (for a review, see Meyer and Zanger, 1997). The enzymes (NATs) are classified according to their amino acid sequences and substrate specificity to N-acetyltransferase 1 (NAT1) and N-acetyltransferase 2 (NAT2) encoded by the NAT1 and NAT2 genes. Each of the two genes encodes a protein of 290 amino acids (Meyer and Zanger, 1997). Recently, an endogenous substrate for NAT1 was found. The substance, p-aminobenzoyl glutamate, was also acetylated with cytosol from human cultured keratinocytes in the presence of acetyl CoA (AcCoA; Kawakubo and Ohkido, 1998). There have been several reports on the polymorphic distribution of acetylation capacities for both enzymes NAT1 and NAT2 (Vatsis and Weber, 1993; Vatsis et al., 1995; Blömeke et al., 1997; Grant et al., 1997; Payton and Sim, 1998).

Humans are exposed to PPD by skin contact (Merk, 1988). Previously, we found acetylation capacities for 2-aminofluorene and p-aminobenzoic acid (PABA) in murine skin (Kawakubo et al., 1988, 1990). Recently, interspecies differences with regard to acetylation capacities have been observed for several compounds such as diacetylated benzidine, which is the main metabolite in murine liver, whereas monoacetylated benzidine is predominantly observed in human liver. Furthermore, even organ-specific variations in acetylation capacities have been found (Lakshmi, 1995).

In summary, this prompted us to investigate the N-acetylation capacities for PPD in the main target organ of allergic dermatitis, namely human skin and keratinocytes, to gain a better understanding of its mechanism of action in immune-mediated toxicity.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. PPD, PABA, AcCoA, dithiothreitol (DTT), sulfamethazine (SMZ), N-(4-methoxyphenyl)acetamide, and diphenylamine were purchased from Sigma Chemical Co. (St. Louis, MO). Monoacetylated paraphenylenediamine (MAPPD) was obtained as monoacetamidoaniline from Aldrich Chemical Co. (Milwaukee, WI). PABA and BB were purchased from ICN Biomedicals Inc. (Aurora, OH). N,N'-Diacetyl-PPD (DAPPD) was synthesized from acetamidoaniline using acetic anhydride and acetic acid. Monoacetamidoaniline (1.2 g) was dissolved in 12 ml of 50% acetic acid; then, acetic anhydride (2 ml) was added slowly to the mixture. After a reaction time of 24 h at room temperature, DAPPD was collected through filtration under aspiration. The product was washed twice with 10% acetic acid and distilled water. The purified product was analyzed by HPLC. All other reagents were of reagent grade.

Experimental Samples. Human skin samples were obtained through mammoplasty reduction (Department of Plastic Surgery, University Hospital RWTH Aachen, Aachen, Germany). Normal human epidermal keratinocytes (from neonatal foreskin; lots 7245, 15321, 15352, 15513, 15604, 17106, 7F0864) from seven individual donors and keratinocyte growth medium (low-calcium, serum-free basal medium supplemented with insulin, hydrocortisone, epidermal growth factor, gentamicin, bovine pituitary extract, trypsin, and trypsin inhibitory factor) were obtained from BioWhittaker (Walkersville, MD). Keratinocytes were cultured in keratinocyte growth medium according to the manufacturer's instructions.

Preparation of Cytosolic Fractions. Subcutaneous tissues were removed from the sample. Then, the skin was cut into pieces on an ice-cooled glass plate and suspended in Tris · HCl (50 mM, pH 7.5) containing 1 mM DTT. The samples were then homogenized (Polytron tissue homogenizer; Kinematica GmbH, Lucerne, Switzerland). The lysate was centrifuged at 10,000g for 20 min. The resulting homogenate was fractionated by ultracentrifugation at 105,000g for 60 min. Keratinocytes were harvested in the second passage (60-70% confluence) and subjected to a nitrogen decompression technique as described by Kawakubo et al. (1995). In this method, the cells were suspended in Tris · HCl (50 mM, pH 7.5) containing 1 mM DTT and equilibrated with high-pressure nitrogen in a pressure-resistant stainless steel case (Parr Instrument Company, Moline, IL). The resulting homogenate was fractionated by ultracentrifugation at 105,000g for 60 min. The supernatant fractions were used as cytosols. Protein content of the samples was determined (Bio-Rad Protein Assay; Bio-Rad Laboratories GmbH, Munich, Germany).

N-Acetyltransferase Assay. The assay for N-acetyltransferase was performed as described previously (Kawakubo et al., 1995). Briefly, activity was determined in the presence of 1 mM AcCoA, except for kinetic studies, in which 500 µM AcCoA was used. Substrate concentrations were 2 mM for PPD, 800 µM for MAPPD, 300 µM PABA, and 800 µM for SMZ unless otherwise noted. The final concentration of DTT in the incubation mixture for N-acetylation of PPD and MAPPD was 1 mM. Reaction time was 30 min. The reaction was stopped by the addition of ice-cooled acetonitrile containing the internal standard for HPLC. The samples were centrifuged to precipitate remaining protein, and 50 µl of each sample was analyzed by HPLC. All assays were performed in duplicate. Resulting products were quantified with a Jasco HPLC system equipped with a guard column and a Nucleosil C18 5-µm column (4.1 × 250 mm; Alltech GmbH, Unterhaching, Germany). For separation of MAPPD and DAPPD, a mobile phase consisting of acetonitrile/25 mM ammonium acetate (pH 6.85; 8:92 v/v) was used. Detection was performed at 266 nm. Retention times were 4.12 min (PPD), 6.40 min (MAPPD), 15.3 min (DAPPD), and 41.4 min [N-(4-methoxyphenyl)acetamide, internal standard], respectively. BB was detected at 246 and 334 nm. The mobile phase consisted of acetonitrile/25 mM ammonium acetate (pH 6.85; 55:45 v/v). The retention time for BB was 7.2 min, and for diphenylamine (internal standard), it was 20.4 min. The mobile phase for N-acetylsulfamethazine consisted of MeOH/H₂O/acetic acid (25:74:1 v/v/v), and the eluent was monitored at 264 nm. The coefficient of variations for all standards were approximately 2.5%, and standard curves were linear over the range of concentrations.

Nonenzymatic Formation of BB from PPD and MAPPD. A nonenzymatic formation of BB from PPD and MAPPD (2 mM) was studied in 50 mM Tris · HCl (pH 7.5) in the absence and presence of 1 mM DTT. Samples were taken at different time points (t = 0, 30, and 60 min) and analyzed by HPLC. To avoid influences from autoxidation of PPD to BB, the experiments were immediately analyzed (delay, 2 min).

Reverse Transcription-Polymerase Chain Reaction (RT-PCR). Total RNA (High Pure RNA Isolation Kit; Roche Diagnostics, Inc.) was analyzed for the presence of mRNA for NAT1 and NAT2 in three independent human skin samples and keratinocytes. RNA concentrations were quantified based on their UV absorption at 260 nm. In total, 0.1 µg of RNA/10 µl reaction mix was reverse transcribed with an RT-PCR kit (Perkin-Elmer Cetus) according to the manufacturer's instructions. In general, the method is as follows: cDNA (1 µl) was amplified in a total volume of 50 µl consisting of Taq Gold PCR buffer, 1.5 mM MgCl₂, 0.4 µl of dNTPs (10 mM), 100 pmol of each primer, and 0.25 µl of Taq Gold (5 U/ml; Perkin-Elmer Cetus). Upstream primer 1 consisting of the 5' end (nucleotides 47-68) and either downstream primer 2 (nucleotides 908-889) or 3 (nucleotides 953-931) corresponding with the 3' end were used as described in detail by Kloth et al. (1994). With primers 1 and 2, selective for NAT1, a characteristic 861-bp DNA fragment was obtained, whereas with primers 1 and 3, selective for NAT2, a characteristic 907-bp fragment was formed. The reaction began at 95°C (9 min), followed by 35 cycles of 95°C (30 s), 51°C (30 s), and 72°C (45 s), and ended with a single final step at 72°C (3 min). The PCR products were separated by gel electrophoresis (ethidium bromide-stained 2.2% agarose). RNA samples without the addition of reverse transcriptase were used as controls for genomic DNA contamination.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

CoASAc-dependent acetylation of PPD (400 µM) was readily detected and increased linearly up to 60 min using 210 µg/ml skin cytosol (Fig. 1). The product was mainly MAPPD, but after 45 min, small amounts of DAPPD were also detected. To study whether human cytosol also showed acetylation capacities for MAPPD, kinetic analyses were performed. AcCoA-dependent formation of DAPPD was found; the amount of product increased linearly up to substrate concentrations of 400 µM (data not shown). MAPPD and DAPPD formation, as well as acetylation of PABA (800 µM), increased linearly up to 60 min with the use of 400 µM PPD and MAPPD, respectively. No such activities were measured for SMZ. For our standard experiments, we selected an incubation time of 30 min. In nine individuals, NAT activities for PPD were investigated. Analyses were performed in duplicate, and the assay was highly reproducible (variation was less than 2%). Mean values are shown in Fig. 2, because the small set excluded a meaningful statistical analysis. Activities ranged from 0.41 to 3.68 nmol/mg/min. Values for DAPPD formation ranged from 0.65 to 3.25 nmol/mg protein/min using four different cytosols (data not shown). Based on this data, PPD acetylation varied by 9-fold, and acetylation of MAPPD formation varied by 5-fold.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Representative example of kinetic analysis of PPD acetylation in human skin cytosol (). Time-dependent formation of MAPPD () and DAPPD () was measured in the presence of 400 µM PPD as substrate. DAPPD was detected after a 45-min incubation.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   PPD acetylation (2 mM) using cytosol (50 µg) from nine individuals in the absence or presence of 300 µM PABA. Data represent mean values from two independent assays. The presence of PABA significantly reduced PPD acetylation.

To characterize the NAT enzymes present in skin, the expression of NAT in skin cytosols was investigated using PABA and SMZ, substrates that are preferentially acetylated by the NAT1 and NAT2 enzymes, respectively. The presence of PABA strongly reduced the acetylation of PPD (Fig. 2). No reduction of PPD acetylation was detected when SMZ, a specific NAT2 substrate, was present. Activity toward SMZ alone was also not detected in the human skin samples and the keratinocytes (data not shown).

To compare acetylation capacities of skin cytosols and keratinocytes, NAT activity for PPD and MAPPD was studied in keratinocytes. Human keratinocytes from seven individual donors showed acetylation activities toward PPD and MAPPD. In cultured human keratinocytes, PPD acetylation ranged from 0.14 to 4.34 nmol/mg protein/min (40-fold) and MAPPD acetylation ranged from 1.31 to 6.58 nmol/mg protein/min (5-fold). Moreover, MAPPD and DAPPD formations were positively correlated in keratinocytes (Fig. 3, r = 0.93) suggesting that the two reactions were performed by the same enzyme. The observed variations between the different samples and substrates are not clear and most likely are related to the small study set. Repeated analysis excluded assay variability as a likely reason for the results. Vatsis and Weber (1993) demonstrated that interindividual variation in NAT1 activities occurs. Although preliminary, the present data suggest that such interindividual variations can also be observed in skin and keratinocytes; however, a larger study is necessary to confirm this finding.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Correlation of PPD and MAPPD acetylation in human epidermal keratinocytes. Keratinocytes from seven individual donors were used for the assay. Substrate concentrations were 2 mM (PPD-NAT) and 800 µM (MAPPD-NAT). Activity toward these substrates was significantly correlated (r = 0.93).

To investigate which genes are expressed on the mRNA level, we looked for specific mRNAs for NAT1 and NAT2 in skin samples and keratinocytes (Fig. 4). Both mRNAs were synthesized. However, no detectable activity for acetylation of SMZ, an NAT2 substrate, was observed in the cytosols from human skin and cultured primary keratinocytes.

View larger version (45K): [in this window] [in a new window]   Fig. 4.   Expression (RT-PCR) of NAT1 and NAT2 (mRNA) in human skin from three subjects. Total RNA (0.1 µg RNA/10 µl reaction mix) was reverse transcribed with an RT-PCR kit (Perkin-Elmer Cetus), and 1 µl of cDNA was amplified in a total volume of 50 µl as described in the text. A characteristic 861-bp DNA fragment was obtained for NAT1, and a 907-bp fragment was formed for NAT2. The PCR products were separated by gel electrophoresis (ethidium bromide-stained 2.2% agarose). RNA samples without the addition of reverse transcriptase were used as controls for genomic DNA contamination.

Michaelis-Menten kinetic constants for acetylation of PPD were determined using skin cytosols from two individuals (Table 1). Apparent K_m values for PPD were very similar, 1113 and 1081 µM, and apparent K_m values for MAPPD were 275 and 358 µM. In concordance with our previous experiments (see Fig. 2), a higher V_max value was observed for sample H. NAT kinetics of PPD and MAPPD were studied in the presence of PABA using cytosol H (Fig. 2). The results are shown in Fig. 5. The presence of 300 µM PABA influenced the K_m values more distinctively than V_max in both reactions, as expected for a competitive inhibitor.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Kinetic analysis of MAPPD formation in the absence () and presence () of 300 µM PABA. See Experimental Procedures for complete assay conditions. Cytosol from sample H in Fig. 2 was used. In the presence of PABA, the Km value for DAPPD formation (A) increased from 359 to 1736 µM, and that for MAPPD formation (B) increased from 1082 to 6103 µM. Vmax values changed from 4.5 to 3.8 and from 6.1 to 6.4, respectively.

We were interested in whether PPD and the acetylated metabolite (MAPPD) are stable or form BB; therefore, we studied the nonenzymatic formation of BB from PPD and MAPPD (Fig. 6). At the starting point (t = 0), there was only a small amount of BB (0.16 nmol/100 µl) observed in the system with/without DTT. After incubation at 37°C for 60 min, however, 1.9 nmol/100 µl of BB (19 µM) was formed in the absence of DTT. On the contrary, there was no statistical increase in the amount of BB in the assay system even after a 60-min incubation with the addition of 1 mM DTT (P < .05, t test). No detectable amount of BB was formed after a 60-min incubation using 2 mM MAPPD as substrate (data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Nonenzymatic formation of BB from PPD. BB formation was monitored in the absence and presence of DTT over a time period of 60 min at 37°C. The initial substrate concentration was 2 mM. Data represented are mean ± S.D. values from four separate determinations.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

PPD is an arylamine and, as such, N-acetylation may play an important role in its metabolism. PPD is a very common allergen among patients with allergic contact dermatitis. However, little is known about the metabolism of PPD in the target organ, namely in the human skin. In this study, we found that PPD was acetylated in the presence of AcCoA using skin cytosols and skin cells such as cultured primary keratinocytes. These results indicate that the biological activities of PPD may be modulated through the cutaneous N-acetylation activity.

Two isoenzymes, NAT1 and NAT2, are known to acetylate arylamines. They show different, although overlapping, substrate specificity. Previously, we found that human and murine skin possesses NAT activities comparable to those of liver when PABA, an NAT1 specific substrate, was selected as the acceptor (Kawakubo et al., 1988, 1990). To find out which enzyme was involved in the metabolism of PPD, we performed competition experiments. The parallel presence of PABA but not SMZ, an NAT2-specific substrate, reduced acetylation capacities of PPD and MAPPD. Furthermore, no detectable activity for N-acetylation of SMZ alone was observed in cytosol from keratinocytes and human skin. In contrast to that, we found that mRNAs of both genes were synthesized in the samples. These data suggest that PPD and MAPPD are acetylated by the same enzyme, namely NAT1. The observed correlation between the formation of MAPPD and DAPPD further supports this hypothesis. The observed discrepancy between the presence of mRNA for NAT2 and no detectable catalytic activity was also found for human bladder and mammary gland by other investigators (Kloth et al., 1994; Sadrieh et al., 1996; Stanley et al., 1996). Altogether, the catalytic results fit with the known organ distribution of NATs in which NAT 1 activity is ubiquitously distributed and NAT2 activity is exclusively found in liver, intestine, and lung (Coroneos and Sim, 1993).

Various biological effects of PPD and its autoxidation product, BB, have been reported. PPD itself showed little mutagenic effects, whereas its reactivity increased after the treatment with hydrogen peroxide (Bracher et al., 1990) or microsomes (Ames et al., 1975; Rojanapo et al., 1986). These effects have been explained by the development of BB via quinonediimine formation of PPD (Munday, 1992). Similar to these mutagenic effects, Krasteva et al. (1993) found in reference to immunogenic effects an essential role of BB for the allergic reaction of PPD. Our own data revealed that the formation of BB from PPD increased rapidly under aerobic conditions. This suggests that possibly a large amount of PPD may be oxidized to BB at the surface of the skin. Only a limited amount of PPD, less than 1% (Ames et al., 1975), may penetrate the horny layer and reach the epidermis. A series of enzymes could then participate in oxidative metabolism of PPD and its derivatives; possible examples are ceruloplasmin (Frieden and Hsieh, 1976), myeloperoxidase (Pember et al., 1983), prostaglandin synthase (van der Ouderaa et al., 1977), cytochromes P-450 (Hrycay and O'Brien, 1971), and the cytochrome c/cytochrome oxidase system (Munday, 1992). However, anaerobic processes such as glycolysis and azoreduction may predominate in the skin (Frienkel, 1960, Collier et al., 1993). In addition, we demonstrated that skin has very high acetylation capacities (Kawakubo et al., 1988 and 1990). According to this, oxidative processes may not easily occur in the epidermis except for cells with presumably high oxidation capacities such as Langerhans cells. Therefore, PPD may be acetylated first and the resulting products are then possibly the main substrates for an oxidative metabolism in the skin.

As mentioned, Krasteva et al. (1993) reported that BB is responsible for the common allergic reactions of PPD; therefore, we were interested in studying whether acetylation of PPD may be a detoxification reaction for allergic contact dermatitis. Our in vitro results did not indicate that acetylated-PPD can be transformed to BB. Assuming that MAPPD is also a poor substrate for enzymatic BB formation, we now consider acetylation of PPD to be a detoxification reaction. Furthermore, acetylation capacities of PPD showed variation (9-fold). This may be based on the recently described genetic polymorphisms in the NAT1 gene (Vatsis and Weber, 1993; Grant et al., 1997); therefore, acetylation status may influence individual susceptibility to the effect of PPD. A large molecular epidemiological study to support these findings is under way.

In summary, we demonstrated that PPD can be acetylated in human skin cytosols and keratinocytes. The responsible enzyme is presumably NAT1. Catalytic activities toward an NAT2 specific substrate were not detected, although mRNA for NAT2 was synthesized.