Human peripheral blood lymphocytes as a cell model to evaluate the genotoxic effect of coal tar treatment.

Peripheral blood lymphocytes (PBL) from psoriatic patients therapeutically exposed to polycyclic aromatic hydrocarbons (PAH) during coal tar (CT) treatment were used to evaluate the in vivo formation of benzo[a]pyrene diol epoxide(BaPDE)-DNA adducts by an ELISA technique and by the 32P-postlabeling method. Moreover, we controlled if the pretreatment with CT influences the formation of BaP-DNA adducts and the BaP metabolism in the PBL obtained from psoriatic patients, treated in vitro with BaP. Our data did not show any significant influence of the CT treatment on the levels of PAH-DNA adducts. Moreover, the use of PBL from psoriatic patients, treated in vitro with BaP, did not allow to detect significant modifications of the metabolic activation of BaP and of the ability of its metabolites to bind to DNA, before and after CT treatment. Thus, PBL do not seem to represent an useful cell model to evaluate the possible genotoxic effect of the exposure through the skin of psoriatic patients to the PAH contained in CT.


Introduction
Polycyclic aromatic hydrocarbons (PAH) are widely distributed in the environment and the exposure of humans to PAH is thought to be responsible, at least partly, for the higher lung cancer incidence in smokers than in nonsmokers, and in some groups of workers than in the general population (1,2). Epidemiological studies have also indicated an association between human occupational exposure to coal tar (CT), which is generally characterized by an high content of PAH, and skin cancer (3). Pharmaceutical grade CT solutions are also employed for therapeutical purposes, e.g., for the treatment of psoriasis, which is a common proliferative skin disease, not accompanied by malignant cell transformation, that affects 2 to 3% of the human population (4). The high exposure of psoriatic patients to genotoxic agents is documented by the exceptional levels of PAH metabolites and/or mutagenic compounds excreted in their urine, as well as by the significant levels of chromosomal damages detected in their lymphocytes (5).
PAH require metabolic activation to exert their carcinogenic activity. In particular, the metabolism of benzo[a]pyrene (BaP), which is the most widely studied carcinogenic PAH compounds, is highly stereoselective: over 30 different metabolites have been identified, among which the carcinogenic activity was related to the formation of BaP diol epoxide (BaPDE) intermediates (6).
Since the covalent binding of the reactive carcinogenic metabolites with DNA is believed to be the first step in the initiation of the carcinogenic process (7), the monitoring of PAH-DNA adducts has been included in studies of human exposures to carcinogens as a very sensitive molecular index providing the quantitative measurements of the biologically effective dose of the environmental carcinogens (8).
In experimental animals, DNA adduct levels in target tissues are related to the dose administered to the animal and the ultimate carcinogenic response (9). In human studies, target tissue samples can hardly be obtained. The relative ease with which peripheral human blood lymphocytes (PBL) can be collected prompted the use of PBL as a surrogate for the target tissues in order to detect the effects of human exposure to environmental genotoxic agents and to study the in vitro metabolism of such agents.

Materials and Methods Subjects Analyzed
Male psoriatic patients, presenting cutaneous lesions involving 20 to 100% of the body surface, excluding the genital area and hands, were analyzed. The patients were hospitalized in the Dermatology Clinic of the University of Padora (Italy). For each patient data regarding the age, smoking habits, occupational or consistent environmental exposure to PAH, medical history, domiciliary treatments for psoriasis and drugs taken during the coal tar therapy, were collected, the most relevant of which are reported in Table 1. During hospitalization, some of the patients were treated with crude coal tar (CT) either alone or in association with a CT-based paste (TP) containing 50% coal tar, sometimes in association with UV-irradiation, or with a 2% coal tar ointment (TO). Controls were male healthy subjects, 25 to 50 years old, free from chronic illness, none subjected to medical prescription or exposures to ionizing radiation or industrial chemicals at work. Nonsmokers were selected as individuals who had never smoked and smokers those consuming 15 or more cigarettes per day at least three months prior to donating the blood.

Cell and DNA Isolation
Samples of peripheral blood (20-30 ml) were collected by venapuncture in heparinized plastic syringes. The purity of the DNA, isolated from cell lysates by cycles of phenol/chloroform extraction (10,11), was checked and quantified by UV spectroscopy (the absorbance ratios of DNA at 260/230 nm and 260/280 nm were usually >2.3 and > 1.8, respectively).  Type of b  treatment  TP  TP  CT, TP  TP  TP  TP  CT, TP  CT, TP  TP  CT  TP  TP  TP  TP  CT, TP  TP  TP   CT, TP   TP  TP  TO  TO  TO  TP  TP  TP  TP  TP  TP  TP  TP  TP   TP  TP  TP  TP  TP  TP  TP  TP  TP  TP   % Body surface  with lesions  75  80  80  65  35  80  40  20  35  60  20  30  70  75  20  70  70  30  60  60  80  80  80  70  70  70  100  60  90  100  100  80   30  90  80  80  80  80  80  80  40  80   Method of detection of  BaPDE-DNA adducts  ELISA   ,, ,  DNA adducts were also detected by 32P-postlabeling as described elsewhere (11). Five pg DNA for each sample were digested for 3 hr and 30 minutes at 37°C with 250 mU micrococcal nuclease and 8m U spleen phosphodiesterase. Samples were further digested for 30 minutes with 2 pg nuclease P1. After the addition of 0.5 mM Tris-base buffer, the digested DNA was labeled with 20 pCi of 32P-ATP, 2.5 units of T4 polynucleotide kinase. After 30 minutes, the reaction was terminated by adding 40 mM potato apyrase. Purification and resolution of 32P-labeled adducts were carried out on polyethyleneimine-cellulose TLC sheets.
The isolation of the DNA adducts was also obtained by twice subsequent butanol extractions of the digested DNA samples, essentially as described by Gupta (12).
The chromatograms were visualized by autoradiography at -80°C, using intensifying screens. Adduct levels were determined by excising the marked areas of the chromatograms, the radioactivity of which was measured by Cerencov counting. Relative adduct labeling (RAL) was calculated on the basis of the amount of radioactivity on the chromatograms, the DNA amount, and the specific activity of the [32-P] ATP used for the labeling (11). In each experiment a 3H-BaPDE-DNA standard with a known modification level was included (recoveries of 80-100% were obtained).
The formation of BaPDE-DNA adducts in PBL treated in vitro for 24hr with the BaP active metabolite, 2p M (-)-BaP-7,8-dihydrodiol, was determined by quantifying by the HPLC technique the amounts of their hydrolisis products, the BaP-tetrols. To this purpose the (-)-BaP-7,8-dihydrodiol metabolites were extracted twice by adding to the culture medium an equal volume of ethyl acetate, saturated with Tris-HCl 10 mM pH 7.5 buffer, and separated by HPLC using a Resolve C18 column as previously described (13,14). IA). We found that the mean adduct level during CT treatment was 7.7 ± 4.9 adducts/ 108 nucleotides, a level comparable to those reported for subjects occupationally exposed to PAH. In nine of the 10 patients examined 2 to 5 months after the end of CT therapy, the mean adduct level was significantly lowered down to 3.3 ± 2.4 adducts/ 108 nucleotides, suggesting a relatively short persistence of the DNA damage induced by PAH in human PBL.

Results and Discussion
Detection ofPAH-DNA Adducts in PBL ofPsoriatic Patients by the 32p-Posdabeling Method The 32P-postlabeling method is perhaps at present the most widely used technique for monitoring the human exposure to PAH.
We quantified PAH-DNA adducts by the 32P-postlabeling technique using both nuclease P1 enrichment and the butanol extraction procedures. Figure 1B shows the levels of total-and anti-BaPDE-DNA adducts in the 26 PBL samples obtained from the nine psoriatic patients and in the PBL samples from the 8 healthy subjects, using the nuclease P1 enrichment procedure. No statistically significant difference was found among the mean total DNAadduct level before clinical therapy with TP (0.46 ± 0.18 adducts/108 nucleotides), after 8 days of continuos TP application (0.54 ± 0.36 adducts/108 nucleotides), and 16 days after the end of TP treatment (0.53 ± 0.26 adducts/108 nucleotides). Moreover, in healthy subjects the mean level of adducts (0.32 ± 0.24 adducts/108 nucleotides) was not statistically different from the mean levels of adducts in psoriatic patients. The autoradiograms of DNA adducts in the analyzed PBL samples revealed the presence of radioactive spots mainly in a distinct diagonal radioactive zone (DRZ) which is typical for aromatic DNA adducts (15,16). The same DRZ was found in 32P-postlabeling digests of DNA from human skin mantained in short-term organ culture after treatment with CT-ointments (17), in DNA from PBL and from lung cells of lung cancer patients (16), and was shown to consist of a multitude of DNA adducts containing aromatic hydrophobic moieties, as indicated by their chromatographic behaviour (15). It must be noticed that only in seven samples of DNA from psoriatic patients we were able to detect the specific spot corresponding to the anti-BaPDE-DNA adduct. Figure IC shows the levels of total-DNA adducts in the 18 PBL samples obtained from the nine psoriatic patients using the butanol extraction procedure. Also, in this case no statistically significant difference was found among the mean total DNA adduct level before clinical therapy with TP (0.74 ± 0.38 adducts/10 nucleotides), after 8 days of continuous TP application (0.68 ± 0.25 adducts/108 nucleotides) and 16 days after the end of TP treatment (0.79 ± 0.25 adducts/108 nucleotides).
Moreover, although the comparison between the mean total DNA-adduct levels detected by the nuclease P1 method (0.58 ± 0.31 adducts/108 nucleotides) and that detected by the butanol method (0.73 ± 0.30 adducts/108 nucleotides) did not reveal any significant difference, the autoradiograms of 18 DNA samples obtained by the butanol procedure revealed the presence of a-dditional radioactive spots not detected by the nuclease P1 method.
Comparison between the Level of Aromatic-DNA Adducts Detected by the ELISA Method and the 32P-posdabeling Tehnique in the PBL from Psoriatic Patients It must be noticed that the mean BaPDE-DNA-adduct level we detected by the ELISA method in the PBL of psoriatic patients is about 20 to 100 times higher than that we found by the 32P-postlabeling technique on the same cell system. Other authors (18,19), using ELISA methods, found in the DNA from lung tissues of lung cancer patients BaPDE-DNA levels 5 to 10 times higher than those determined by the 32P-postlabeling technique. In other studies, the postlabeling values of DNA adducts were significantly (10-100 fold) lower than those obtained by ELISA (20). The discrepancy between the results obtained by the two methods can be explained by cross reactivity of the antiserum against different DNA adducts, which enhances the responsiveness of the ELISA method as opposed to structurally related PAH adducts (19). The underestimation observed with postlabeling may also be due to some intrinsic proprieties such as poor efficiency of phosphorylation (20). Moreover, the nuclease P1 digestion of DNA, which is used in the present as well as in other studies to enhance the sensitivity of the 32P-postlabeling assay, may result in dephosphorylation of certain adducts, impeding their 32P-postlabeling, and may give an underestimation of the actual adduct level (20).

Benzo[a]pyrene Metabolism and DNA Adduct Formation in Cultured PBL of Psoriatic Patients
The topical application of CT solution to neonatal rats was shown to induce the aryl hydrocarbons hydroxylases AHH of skin and liver (21). The AHH in the skin of psoriatic patients have a lower activity and inducibility than in control subjects (22). Therefore, we evaluated the possibility that the treatment with CT influences the BaP activation and the binding of its metabolites to DNA. PBL samples, from five healthy subjects and 10 psoriatic patients, were analyzed after treatment in vitro with 2 pM (-)-7,8-dihydrodiol for 24 hr. The metabolism of (-)-7,8-dihydrodiol to synand antistereoisomers of BaPDE was determined by monitoring the formation of their corresponding tetrols: BaP-7,10/8,9 and BaP-7/8,9,10 for (+)-anti-BaPDE, and BaP-7,9/8,10 and BaP-7,9,10/8 for (-)-syn-BaPDE. As shown in Figure 2, anti-tetrols were the predominant isomers in all the Volume 102, Supplement 9, November 1994  Figure 2. Ratio of anti/syn BaP-tetrols, formed in vitro by PBL of psoriatic patients and healthy subjects after treatment with (-)-BaP-7,8-dihydrodiol, and detected by HPLC. PBL were obtained from psoriatic patients before (1) or 4 to 10 days after the end of CT therapy (11), and from healthy subjects reanalyzed 2 to 3 times at distance of 10 days from one another (l, 11). examined PBL samples (mean ratio anti-/syn-tetrols: 23.2 ± 10.83), but significant levels of syn-tetrols, which clearly derived from the hydrolysis of (-)-syn-BaPDE, were also present. The mean ratio of anti-/syn-tetrols in healthy subjects (5.7 ± 0.35) was significantly lower than in psoriatic patients (32.1 ± 15.25) (p<0.001), whereas statistical comparison of this ratio in psoriatic patients, before and after treatment with CT, did not reveal any significant difference.
The analysis by HPLC of the DNA adducts, after incubation of lymphocytes with unlabeled 2 pM (-)-BaP-7,8-dihydrodiol for 24 hr (13), showed the formation of consistent amounts of (-)-syn-BaPDE-DNA adducts in the 15 PBL samples both from healthy subjects and psoriatic patients. The mean ratio of anti-BaPDE-DNA/syn-BaPDE-DNA adducts was 4.01 ± 3.27 pmol/'pgDNA, and no significant difference in that ratio was found either between healthy subjects and psoriatic patients, or between psoriatic patients, before and after CT treatment (13).

Conclusion
The purpose of our study was to evaluate the genotoxic risk involved in the exposure of psoriatic patients to PAH by the detec-tion of PAH-DNA adducts in their DNA isolated from PBL. Our data demonstrate no correlation between the levels of PAH-DNA adducts and the exposure to CT. With the lack of correlation between CT therapy and incidence of skin cancer in psoriatic patients (4), such data suggest that CT treatment of psoriasis should not be considered a potential genetic and carcinogenic risk for psoriatic patients. However, it must be notice that the DNAadduct levels in nontarget tissues, like PBL, could not reflect the level of adducts in skin and lung cells, both of which have been shown to be the target tissues for DNA in human PAH exposures (14,23,24). Therefore, the lack of elevated DNA adducts in PBL, although representing a relevant information related to biomonitoring, does not allow to directly predict the skin cancer risk involved in CT therapies.