Modulation of DNA binding in vivo by specific humoral immunological response: a novel host factor in environmental carcinogenesis?

To investigate the possible modulatory effect of the immune response induced by recurrent carcinogen exposure, a specific humoral immune response toward 2-acetylaminofluorene (2-AAF) was elicited in Swiss mice with repeated intraperitoneal injections of a 2-AAF-gelatin conjugate. The immunization procedure resulted in the production of specific anti-2-AAF antibodies in all treated animals. Groups of immunized and nonimmunized mice were subsequently fed 2-AAF pelleted in the diet at 50 and 150 ppm for 4 weeks. At the end of 2-AAF administration, animals were sacrificed and the content of 2-AAF-adducts in liver DNA was determined by enzyme-linked immunoadsorbent assay using a polyclonal rabbit antiserum. The comparison of the adducts levels in immunized and nonimmunized mice (receiving either the vehicle or the adjuvant alone during pretreatment) demonstrates a highly significant (p < 0.001) difference among groups, with far lower adduct levels in immunized animals. No significant difference in food consumption or liver metabolic activities was observed among experimental groups, suggesting the absence of external bias. The mechanism underlying the result observed is not yet clear; however, the experimental data strongly suggest that the specific immunological response induced by recurrent carcinogen exposure may exert a modulatory effect and act as a relevant host factor in chemical carcinogenesis.


Introduction
Manifold inherited and acquired host factors such as variance in DNA repair (1)(2)(3) and xenobiotic metabolism (4)(5)(6), age, nutrition, stress, diseases, hormonal status (7)(8), and immunological (9) and genetic factors (10,11) are known to affect the individual response to chemical carcinogens This paper was developed from a poster that was presented at the 2nd International Conference on Environmental Mutagens in Human Populations held 20-25 August 1995 (12)(13)(14). Another significant trait might be represented by the immunological response elicited by the formation of adducts to macromolecules during chronic carcinogen exposure. In this regard, previous studies demonstrated the presence of antibodies directed against benzopyrene-DNA adduct in blood sera of humans occupationally exposed to high levels of polycyclic aromatic hydrocarbons (PAHs) (15)(16) and in the urban population (17). Consequently, this trait has been proposed as a retrospective individual exposure marker in biomonitoring studies (18). On the other hand, it has not yet been elucidated whether such induced immune response may play some mechanistic role, such as modulating the effect produced by the carcinogen itself. An interesting clue to this possibility was previously provided by the observation of a protective effect of the secretory immune response to 2-acetylaminofluorene  in rabbits, possibly related to the reduction in transepithelial absorption of the carcinogen (19).
To investigate the potential significance of the humoral anticarcinogen immunity, a specific immunological response toward 2-AAF was elicited in Swiss mice. Both immunized and nonimmunized animals were subsequently challenged with a 4-week dietary exposure to the carcinogen. At the end of treatment, 2-AAF binding to liver DNA was determined in all animals and evaluated in relation to their immune status and carcinogen exposure.

Animas
Male Swiss mice (Charles River, Calco, Como, Italy), 4 weeks of age at the beginning of treatment, were maintained on a balanced standard chow (Mucedola, Milan, Italy) and tap water ad libitum. Animal care, treatments, and sacrifice were conducted in strict accordance with Directive 86/609/EEC on the protection of laboratory animals. Imnunization A 2-AAF-gelatin conjugate (20) was used as the immunogen. The immunization procedure consisted of three weekly intraperitoneal (ip) administrations of the complete immunogen (50 pl of 2-AAFgelatin conjugate 1 mg/ml together with 50 PlI of Freund's adjuvant). A further injection was delivered 14 days after the third treatment. Controls received the adjuvant alone or no treatment.

2-AAF Treatment
One week after the end of the immunization, both immunized and nonimmunized mice were fed 2-AAF pelleted in the diet at 50 and 150 ppm for 4 weeks. Parallel control groups were fed with the standard diet.
Determination of2-AAF Adducts to Liver DNA DNA Extraction. Livers frozen at -80°C were thawed and homogenized in extraction buffer (10 mM Tris-HCl, pH 8.0, 100 mM EDTA, 0.5% sodium dodecyl sulfate [SDS], 20 pg/ml RNAase) and incubated 1 hr at 370C before the addition of proteinase K (Sigma Chemical Co., St. Louis, MO) to a final concentration of 100 fig/mI. After 2 hr of incubation at 500 C, DNA was phenol extracted (21) and quantitated by 260 nm absorbance. Protein and RNA contamination was checked through the A260/A280 ratio.
Production of the Antiserum. The anti-2-AAF immune serum was produced by immunization of adult New Zeland white rabbits with six weekly subcutaneous injections of 2-AAF-gelatin (1 mg) emulsified in complete Freund's adjuvant (0.5 ml). A 2 mg booster injection was delivered 15 days after the last inoculation. One week later, the rabbits were bled from the heart.
DNA-2-AAF adducts were measured by competitive ELISA (23,24). Briefly, 96 wells were coated with 0.25 ng 2-AAFgelatin and then saturated with 200 pl of 0.5% gelatin. Standard curves were obtained by adding to each well 50 pl of rabbit antiserum diluted 1:5,000 that had been preincubated 2 hr with serial dilutions of 2-AAF-gelatin as a competitor. For 2-AAF-DNA adducts determination, 1.5 pg of DNA was used as competitor. Goat anti-rabbit IgG-horseradish peroxidase conjugate was used as the second antibody diluted 1:4,000, with o-phenylendiamine as substrate.

Results and Discussion
To elicit a humoral response toward 2-AAF, mice received repeated ip injections of a complete immunogen, as detailed in the previous section. Due to the low molecular weight of the hapten (insufficient to produce an effective immunogenic stimulus) a carrier molecule (gelatin) was conjugated to 2-AAF. At the end of immunization, the presence of specific IgG directed against the 2-AAF-gelatin conjugate in the sera of immunized mice was assessed by direct ELISA. Although interindividual differences in serum titers were observed, all the immunized mice examined produced humoral antibodies capable of reacting against the 2-AAF conjugate (Figure 1). Conversely, serum pools of both nonimmunized mice and mice treated with the adjuvant alone did not show any reactivity toward 2-AAF-gelatin. In all cases, no Immunized Not immunized Immunized with adjuvant only Figure 1. Reactivity of sera of immunized mice (n=37) and pooled sera of nonimmunized mice (n=20) against 2-AAF-gelatin conjugate as determined by direct ELISA. Plates were coated with 0.5 ng conjugate/well, using 50 pl of 1% gelatin aqueous solution as the blank. Sera were diluted 1:1,000. The bar represents the standard error of the mean.
reactivity was observed against the carrier protein alone (data not shown). 2AAF-DNA binding in livers of both immunized and nonimmunized mice was determined by competitive ELISA using a polyclonal rabbit antiserum. This serum was able to recognize the carcinogen adducted to gelatin and bovine, egg, or mouse serum albumin (Figure 2). Similar results were obtained with pooled sera of immunized mice (Figure 2). In both cases, no reactivity was observed with the carrier proteins alone (data not shown). As the reactivity of rabbit serum toward DNA and protein adducts by competitive ELISA was basically similar (Figure 3), the gelatin conjugate was used as quantitative standard in all determinations.
The quantitation of liver DNA adducts after 4 weeks of dietary exposure to 2-AAF demonstrated high levels of adducts in all treated animals, with a partial quantitative relationship with the administered dose (Table 1). No reactivity was observed with DNA of untreated mice (100% of reactivity in competitive ELISA), which was used as an internal experimental control. A comparison of the results obtained with immunized and nonimmunized mice  (receiving either the vehicle or the adjuvant alone) revealed a highly significant difference among groups, with far lower adduct levels in immunized mice (Table 1). No significant differences in food consumption were observed between immunized and nonimmunized mice, suggesting that all animals received comparable amounts of carcinogen at the end of treatment. Across all the experimental groups, liver weights were fairly similar; this suggested the absence of significant liver toxicity. The possible indirect effect of immunization, based on the impairment of liver enzymic activities, was investigated in satellite groups of immunized and nonimmunized mice using liver homogenates for the exogenous activation of 2-AAF to mutagen in the Salmonella reversion assay. Almost identical results were obtained with homogenates from different experimental groups (data not shown), indicating that the immunization pretreatment did not exert any significant detrimental effect on these liver activities.
The mechanism underlying the intriguing result provided by this study is not yet clear. As stated above, trivial bias due to a perturbatory effect of the immunization on liver functionality can be ruled out. Organ weight, DNA and protein content, and efficiency in the activation of 2-AAF to bacterial mutagen were fairly similar in immunized and nonimmunized mice. An unspecific perturbation related to the stimulation of the immune system is also unlikely in view of the results obtained with mice treated with the Freund's adjuvant alone; these mice experienced a significant yet unspecific immunogenic stimulation. Therefore, alternative mechanisms have to be considered. One plausible explanation lies in the high reactivity of the specific antibodies elicited by the immunization toward a variety of carcinogen conjugates. Considering that hydrophobic carcinogens are largely present in the blood stream in association with carrier proteins (25), as are their reactive metabolites (26), it is conceivable that circulating antibodies may act as scavengers that effectively lower the bioavailable fraction of the compound entering into the organism.
It can be supposed that the production of a specific humoral immunological response after chronic carcinogen exposure may be a widespread phenomenon. Production of specific antibodies has been observed in mice (27) and in humans (28,29) with prolonged exposure to DNAdamaging agents, suggesting that chronic adduct formation on serum proteins or DNA can fulfill the steric and structural conditions required for the induction of an immune humoral response. In any case, no definitive conclusion should be drawn on the biological significance of the phenomenon described herein. Further studies are required to investigate its specificity, as well as the profile of damage induced in other target tissues. Nevetheless, the data presented support the hypothesis that the immunological response induced by chronic exposures may act as a significant host factor in environmental carcinogenesis.