Factors associated with skin cancer prevention practices in a multiethnic population.

A better understanding of factors influencing sun protection practices can improve the design and evaluation of skin cancer prevention programs. These data are from a cross-sectional survey of 756 parents with children in Grades 1 through 3, and 176 recreation program staff members in a multiethnic population in Hawaii. Questionnaires asked about skin cancer prevention practices (sunscreen use, covering up, shade seeking), knowledge, benefits and barriers, policies, and staff norms for prevention. The most important correlates of children's prevention practices were their parents' sun protection habits. Multiple regression models--which included knowledge, beliefs, program policies, and covariates related to sun protection--explained a total of between 38% and 41% of the variance in children's sun safety habits, 22% to 25% of parents' habits, and 24% of recreation staff members' sun safety habits. The models were less successful at predicting the use of hats, shirts, and shade seeking and a composite sun protection habits index. Parents and caregivers' knowledge, beliefs, and behaviors, as well as recreation program policies, are strong predictors of sun protection practices among children in Grades 1 to 3 in a multiethnic sample.

Epitope mapping studies with human anti-cytochrome P450 3A antibodies.

A subset of patients with hypersensitivity reactions to the aromatic anticonvulsants phenytoin, carbamazepine, and phenobarbital have circulating antibodies that recognize members of the rat cytochrome P450 (CYP) 3A subfamily. These antibodies do not recognize related human CYP3A proteins despite the high degree of structural similarity. To investigate the relationship between P450-mediated drug metabolism and the development of anti-P450 antibodies, we initiated epitope mapping studies by screening a library of fusion proteins constructed from rat CYP3A1 with an anti-CYP3A1-positive patient serum sample. Positive signals from colony lifts were confirmed by sodium dodecyl sulfate/polyacrylamide gel electrophoresis and immunoblotting, and a 26-amino acid sequence corresponding to amino acids 342-367 of the CYP3A1 protein (NKAPPTY-DTVMEMEYLDMVLNETLRL) was identified as containing the epitope recognized by IgG3 antibodies in this serum sample. By subjecting inserts from two clones into a second round of library construction and screening by immunoblot analysis, we further defined the epitope to EYLDMVLNETLRL. Single amino acid deletions identified DMVLNETLRL as the minimum amino acid sequence required for antibody binding. The corresponding sequence in the four human CYP3A proteins differs by only one amino acid (DMVVNETLRL) This amino acid is critical to antibody recognition as immunoreactivity of the L361V mutant is markedly reduced. Anti-CYP3A antibodies in nine of nine additional sera also recognized the 13-amino acid epitope; for five of these sera, the minimum antibody binding sequence was DMVLNETLRL. The proximity of this epitope to a region determining substrate specificity may provide the link among reactive metabolite production, hapten formation, and the production of anti-P450 antibodies in anticonvulsant-induced idiosyncratic reactions.

Human anti-endoplasmic reticulum autoantibodies produced in aromatic anticonvulsant hypersensitivity reactions recognise rodent CYP3A proteins and a similarly regulated human P450 enzyme(s)

Hypersensitivity reactions to aromatic anticonvulsants are associated with anti-liver microsomal antibodies which recognise rodent proteins. The reactivity of these antibodies, the regulation of the rodent antigens and the identity of the human autoantigen have been investigated. Dexamethasone elevated markedly the levels of an immunoreactive mouse protein(s) which exhibited a Mr (53 kDa) and inducibility consistent with the major Cyp3a product. Immunoblots conducted with hepatic microsomes from control and induced rats and purified rat P450s confirmed that these antibodies also recognised constitutive (3A2) and inducible (3A1) rat CYP3A products. Negligible reactivity was observed with microsomes from human B-lymphoblastoid cell lines expressing CYP1A1, 1A2, 2A6, 2D6, 2E1, 3A4 or epoxide hydrolase. Analysis of a phenotyped human liver bank revealed that the antibodies recognised a 52.5 kDa microsomal protein which exhibited marked heterogeneity in its expression and appeared to be regulated co-ordinately with human CYP2C8 and 3A3/4. The inter-individual variation in the expression of this protein(s) and its potential induction by anticonvulsant therapy together with an inherited deficiency in drug detoxification capacity may explain predisposition to these immunoallergic reactions.

Human anti-cytochrome P450 antibodies in aromatic anticonvulsant-induced hypersensitivity reactions.

Aromatic anticonvulsants such as phenytoin, phenobarbital and carbamazepine are associated with a hypersensitivity syndrome (fever, rash lymphadenopathy, hepatitis) suggestive of an immune component. We have identified immunoglobulin G antibodies in the sera of nine affected patients which recognize a 53-kD protein which is constitutively expressed and PB inducible in rat liver microsomes. No such reactivity was observed in sera from healthy controls, patients on chronic phenytoin therapy without toxicity or patients with hepatic failure not receiving anticonvulsants. Using highly purified rat hepatic cytochrome P450, P450 3A1 was identified as the major antigenic species, whereas less intense reactivity was noted with P450 2C11. P450 2C6 and 3A2 were minor antigens in some patients. In all patients, the apparent constitutive and phenobarbital-inducible expression of the antigen was a composite effect of antibodies reacting with at least two isozymes, one of which was constitutively expressed and the other PB inducible. In human liver, a 53-kD antigen was expressed to a greater extent in microsomes from a patient with a fatal hepatotoxic reaction to phenytoin compared to microsomes from normal liver or from a sulfonamide hepatitis patient. Western blotting with microsomes prepared from lymphoblastoid cell lines transfected with different human hepatic cytochromes P450 failed to identify P450s 1A1, 1A2, 2A3, 2B6, 2C9, 2D6, 2E1, 3A4 or epoxide hydrolase as the target antigen. Identification of the antigen will be important in understanding the relationship between drug metabolism and the subsequent immune response in the pathogenesis of these rare but severe forms of drug toxicity.

Growth of human lymphoma cells in SCID mice.

Two EBV-negative human lymphoma cell lines raised in this laboratory and peripheral blood cells of a patient with large cell lymphoma in leukemic phase were injected intravenously or intraperitoneally into C.B.17 SCID mice. One line (OCI-LY18) was derived from the pleural fluid of a patient with a large cell, immunoblastic malignant lymphoma. Cells of this line are of B cell origin and characterized by multiple rearrangements of the JH locus. The second line (OCI-LY17) was grown from the peripheral blood of a patient with a large cell lymphoma of T-cell phenotype which has characteristic rearrangement of the T-cell receptor beta chain. The cells directly obtained from a patient with large cell non-cleaved malignant lymphoma were of B-cell origin. Animals carrying OCI-LY18 developed large tumor masses within 6-8 weeks of inoculation. The tumors were detected in the intestine, mesentery, retroperitoneum, lymph nodes, spleen, lung and kidney. The masses resembled the primary tumor with respect to histological appearance and immunological phenotype. It was possible to generate secondary cell lines from the tumors found in the inoculated SCID mice. Injection of one of these secondary cell lines into SCID mice resulted in the rapid development of lymphoma and as few as 10 cells were sufficient to establish the disease in the inoculated animals. In contrast cells of OCI-LY17 produced small tumor aggregates that did not appear to progress over time and did not cause death of the animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Exogenous but not endogenous EBV induces lymphomas in beige/nude/xid mice carrying human lymphoid xenografts.

Epstein--Barr virus (EBV) is a latent human herpes virus that growth-transforms EBV receptor/CD21+ B cells and is associated with several high-frequency malignancies. Reactivation of latent EBV occurs in approximately 1/3 of organ graft recipients and a majority of AIDS patients; EBV-positive B lymphoproliferative lesions represent often fatal complications in organ transplantation and late-stage AIDS. Although such lymphomas can arise from endogenous virus, the high tumor risk in EBV-seronegative transplant recipients implies de novo infection, in particular virus transmission with intra-graft B lymphocytes. Since SCID mice engrafted with human lymphocytes (SCIDhum) typically develop endogenous EBV+ (human) tumors in their graft it is difficult to study exogenous virus transmission in this model. We here demonstrate that beige/nude/xid mice engrafted with human lymphoid cells (BNXhum) selectively accept human B but not T cell grafts. Unexpectedly these mice fail to develop endogenous lymphomas observed in SCIDhum mice engrafted in parallel. However, injection of as few as less than 500 EBV particles produces rapidly fatal, polyclonal lymphomas in BNXhum animals. This virus sensitivity of BNXhum approaches conditions for EBV transmission with organ grafts.