Stable platinum isotope measurements in presolar nanodiamonds by TEAMS.

Nanodiamonds are stardust grains commonly found in primitive meteorites. They survived the formation of the solar system and kept their own individuality. Measurements of trace-element isotopic signatures in these grains will help understanding heavy element nucleosynthesis in massive stars and dust formation from their ejecta. We have continued previous attempts to search for stable Pt isotope anomalies in nanodiamonds via trace element accelerator mass spectrometry (TEAMS). The installation of a new injector beam line at the VERA facility allowed studying low traces of stable elements in different materials. Moreover, recent experiments showed that VERA provides the required measurement precision together with a low Pt machine background. Here, we observed for the first time an indication for enhancements of 198Pt/195Pt isotope ratios in two diamond residues prepared by different chemical separation techniques from the Allende meteorite. Variations in other isotopic ratios were within analytical uncertainty, and no anomaly was identified in a third diamond fraction.

Hepatitis B surface antigen (HBsAg) derived from yeast cells (Hansenula polymorpha) used to establish an influence of antigenic subtype (adw2, adr, ayw3) in measuring the immune response after vaccination.

As a product of western world biotechnology the yeast (Saccharomyces cerevisiae) hepatitis B vaccine was introduced as antigenic subtype adw2. However, an HBsAg/adw2-vaccine may provide a good but not "optimal" immunologic response for infection with heterologous virus strains. The availability of the yeast Hansenula polymorpha HBsAg in three different antigenic forms (adw2, ayw3 and adr) enabled us to investigate the influence of variant amino acids in the binding of immune anti-HBs after vaccination. Hansenula-derived HBsAg was standardised on the basis of protein content at >95% purity. Standardisation was controlled by monoclonal anti-HBs binding in a well-conserved region. Sera were obtained after immunisation with type adw, ayw and adr vaccines. Direct binding of immune antibodies to homologous antigen (in EIA) was higher than to heterologous antigen except for the adr-related antibodies. Since the binding of the WHO reference anti-HBs was strongly reduced for the ayw and adr compared to the adw antigen, a similar binding profile for the three antigens on protein basis could result in 2-3-fold different anti-HBs level expressed in IU/l. Inhibition of Hansenula-derived HBsAg binding to solid phase monoclonal anti-HBs in enzyme immunoassays after incubation with serum anti-HBs confirmed the differential binding of serum anti-HBs with variant Hansenula-derived HBsAg. This variant (antigenic subtype) dependent reactivity of anti-HBs in immunoassays in combination with a variant specific WHO standard may limit the application of the threshold levels of 10 and 100 IU/l for seroconversion and seroprotection.

Measurement of hematological, clinical chemistry, and infection parameters from hirudinized blood collected in universal blood sampling tubes.

Hirudin, the anticoagulatory polypeptide of the leech Hirudo medicinalis, strongly inhibits thrombus formation by specifically interacting with thrombin. For diagnostic purposes, hirudin should be superior to other anticlotting compounds because it only minimally alters the mineral, protein, and cellular blood constituents. To test this hypothesis, hirudinized and routinely processed venous blood from 80 healthy volunteers and patients was subjected to a variety of automated blood tests. A strong correlation was found between the results of automated complete blood counts obtained from K(2)-ethylenediaminetetraacetic acid (EDTA) anticoagulated and hirudinized blood (1000 antithrombin units [ATU] hirudin/ml). In addition, clinical chemistry and serological infection parameters (asparlat amintransferase [ASAT], lactate dehydrogenase [LDH], sodium, and so on, and antibodies against hepatitis B and C and human immunodeficiency virus [HIV]1/2, respectively) correlated well when measured in serum as compared with hirudinized plasma. Contrary to single clotting factors, global coagulation parameters (activated partial thromboplastin time [aPTT], prothrombin time [PT]) could not be measured in hirudinized blood. Recombinant hirudin neither interfered with immunophenotyping of mononuclear cells using FACScan analysis, nor did it alter the detection of Wilms' tumor gene expression by RT-PCR technology even at high doses (5000 ATU hirudin). Thus, a hirudin-containing blood sampling tube can be designed as a universal blood sampling tube (UBT) for testing the majority of diagnostic blood parameters.

The use of hirudin as universal anticoagulant in haematology, clinical chemistry and blood grouping.

Undesirable interactions between anticoagulants and diagnostic test kit procedures so far have prevented the development of a single uniform blood sampling tube. Contrary to K2-EDTA, heparin and other anticoagulants, hirudin only minimally alters blood cells and dissolved blood constituents, thus qualifying as a universal anticoagulant for diagnostic purposes. Automated complete blood counts, automated analyses of clinical chemistry analytes and immunohaematology were performed from hirudinised and routinely processed blood obtained from healthy volunteers (n=35) and hospitalised patients (n=45). Hirudin (400 ATU/ml blood) sufficiently anticoagulated blood for diagnostic purposes. The measurements of automated complete blood counts obtained from K2-EDTA-anticoagulated and hirudinised blood correlated significantly as did the measurements of 24 clinical chemistry analytes from hirudinised plasma and serum. Regression analysis revealed that the results of complete blood counts and clinical chemistry tests were predictable from the respective measurements from hirudinised blood (p=0.001). Immunohaematological tests and cross-matching from hirudinised and native blood of the same donors gave identical results. Single clotting factors, but not global coagulation analytes, could be measured from hirudinised blood. Therefore, a universal hirudin-containing blood sampling tube could be designed for automated analysis of haematological, serological and clinical chemistry analytes.

Adjuvants that enhance priming of cytotoxic T cells to a Kb-restricted epitope processed from exogenous but not endogenous hepatitis B surface antigen.

Intramuscular (i.m.) or s.c. injection of plasmid DNA encoding hepatitis B small surface antigen (HBsAg) primes potent MHC I-restricted cytotoxic T lymphocyte (CTL) responses in H-2(d) (BALB/c) and H-2(b) (C57BL/6) mice. In contrast, i.m. or s.c. injection of exogenous HBsAg particles without adjuvants primes CTL responses in 'high responder' H-2(d) but not 'low responder' H-2(b) mice. We have shown that processing of exogenous but not endogenous HBsAg generates the Kb-binding S208-215 peptide ILSPFLPL. This system allowed us to optimize conditions for stimulating murine CTL responses to exogenous antigen by identifying adjuvants that facilitate priming of Kb-restricted CTL by injecting recombinant HBsAg particles into 'low responder' H-2(b) mice. Synthetic oligodeoxynucleotides with immunostimulating sequences or the recombinant cytokine IL-12 efficiently enhanced priming of CTL to exogenous HBsAg. Hence, the adjuvanticity of DNA sequences that induce Th1 cytokines facilitate priming of MHC I-restricted T cell responses to exogenous antigen and are therefore of potential value in formulating vaccines designed to enhance CTL priming to exogenous antigen.

Methylotrophic yeast hansenula polymorpha as production organism for recombinant pharmaceuticals.

Since the onset of genetic engineering, yeasts belong to the preferred host cells for the production of heterologous proteins. They combine ease of genetic manipulation and cultivation with the ability to process and to modify the produced compounds according to a general eukaryotic scheme. Since yeasts do not contain pathogens, pyrogens or viral inclusions they constitute attractive production systems for proteins considered for therapeutic administration. At the beginning of gene technology the attention of biotechnologists focussed on the use of the best characterized species Saccharomyces cerevisiae. Insulin and hepatitis B vaccines are examples for S. cerevisiae-derived therapeutics. In recent years alternative yeast have become accessible for the techniques of modern molecular genetics and thus for potential applications in biotechnology. In this respect the methylotrophic yeast Hansenula polymorpha offers especially advantageous characteristics as host for the production of pharmaceutical proteins. As a consequence, production systems based on this yeast have been established for serum proteins, vaccines and other therapeutically important compounds. Some H. polymorpha-derived products are under preclinical or clinical trials at present and are expected to reach the market within the near future. In the following article the current status of this system is presented and discussed comparing it with other expression systems.

Processing of exogenous heat-aggregated (denatured) and particulate (native) hepatitis B surface antigen for class I-restricted epitope presentation.

Many cell types efficiently present an epitope of the hepatitis B surface Ag (HBsAg) to murine class I-restricted CTL following an in vitro pulse with native 22-nm HBsAg particles. Processing of exogenous HBsAg particles required its cytochalasin B-insensitive uptake and acid proteolysis in an endocytic compartment, was insensitive to brefeldin A and cycloheximide, and did not involve regurgitation of antigenic peptides. In contrast, after an in vitro pulse of cells with exogenous, heat-denatured 1-micron HBsAg aggregates, only macrophages (but not other cell types tested) presented the Ld-restricted HBsAg epitope efficiently to CTL. Processing of exogenous HBsAg aggregates required its cytochalasin B-sensitive uptake, was insensitive to brefeldin A, and involved regurgitation of antigenic peptides. Processing of the two different, exogenous HBsAg preparations for class I-restricted epitope presentation thus involved alternative pathways: an "endocytic pathway" for native 22-nm particles, and a "phagocytic pathway" for denatured 1-microns aggregates. Both HBsAg preparations displayed different immunogenicity for class I-restricted CTL in vivo when delivered without adjuvants: native HBsAg particles were of high immunogenicity, and denatured HBsAg aggregates were of low immunogenicity. Class I-restricted CTL are thus primed in vivo after "endocytic processing" of native HBsAg particles as well as "phagocytic processing" of denatured HBsAg aggregates.

Exogenous hepatitis B surface antigen particles processed by dendritic cells or macrophages prime murine MHC class I-restricted cytotoxic T lymphocytes in vivo.

Injection of low doses of particulate hepatitis B surface Ag (HBsAg) into H-2d mice without adjuvants primes an Ld-restricted, S28-39-specific T cell response. This study indicates that dendritic cells (DC) and macrophages (M phi) both serve as APCs that support priming of CD8+ CTL precursors in vivo to exogenous HBsAg particles. After transfer into a syngeneic, naive host, HBsAg particle-pulsed DC, either freshly purified from skin or derived from a cloned DC line, efficiently primed class I-restricted, HBsAg-specific CTL precursors. M phi, either harvested from the peritoneal cavity or generated in macrophage-CSF-stimulated bone marrow cell cultures in vitro or derived from established, cloned M phi lines (PU5-1.8, J774A.1), pulsed with HBsAg particles in vivo or in vitro, elicited a class I-restricted, HBsAg-specific CTL response after adoptive transfer into naive hosts. The class I-restricted CTL response induced by HBsAg particle immunization was suppressed in carrageenan-treated mice, but was restored when carrageenan-treated mice were immunized with syngeneic, HBsAg-pulsed M phi. Selective elimination of M phi by liposome-incorporated dichloromethylene-diphosphonat did not suppress the induction of a CTL response of H-2d mice by HBsAg particle immunization. HBsAg-pulsed, freshly prepared DC are more potent than pulsed M phi in priming class I-restricted CTL in vivo. The relative importance of both types of APC in priming CTL remains to be resolved.

Priming of class I-restricted cytotoxic T lymphocytes by vaccination with recombinant protein antigens.

We investigated the specific priming of MHC class I-restricted cytotoxic T lymphocytes (CTL) by different protein antigen preparations in mice. The recombinant viral protein antigens tested are of potential relevance for the design of subunit vaccines. They include the hepatitis B virus (HBV) surface antigen (S-antigen), the HIV-1 gp160 envelope protein, and a chimeric HIV-1 Pr55-gag/V3-3 retrovirus-like particle. In addition, ovalbumin (OVA) was tested. The native or denatured particulate (multimeric) or monomeric form of these protein antigens was injected by various routes into mice. Class I-restricted CTL were efficiently primed by a single low-dose injection of HBV S-antigen particles or the chimeric HIV-1 Pr55-gag/V3-3 particles. After SDS-denaturation, gel-purified monomeric S-antigen and monomeric Pr55-gag/V3-3 fusion protein were still very efficient in priming CTL. CTL sensitization was not detected in a (primary or boosted) response to even high doses of native OVA or native HIV-1 gp160. Denaturation of these two antigens by detergent strikingly increased their immunogenicity for CTL. Immunization of mice with non-treated or SDS-denatured antigenic peptides representing the relevant CTL-defined epitopes of the tested protein antigens did not prime CTL. These data indicate that native, particulate and denatured, monomeric protein antigens efficiently stimulate a class I-restricted CTL response.

Hepatitis B virus small surface antigen particles are processed in a novel endosomal pathway for major histocompatibility complex class I-restricted epitope presentation.

We investigated the major histocompatibility complex (MHC) class I-restricted presentation of an epitope of the hepatitis B virus small surface (S) antigen particle to cloned murine cytotoxic T lymphocytes (CTL). Efficient Ld-restricted presentation of the S28-39 epitope to CTL is observed in cells of different tissue origin pulsed in vitro, either with the antigenic S28-39 12-mer S-peptide, or with particulate S-antigen. The kinetics of epitope presentation differ in S-peptide-pulsed and in S-particle-pulsed cells: while a 15-min pulse with the antigenic peptide sensitizes targets for class I-restricted CTL lysis, presentation of S-particles requires 30-60 min to sensitize cells for CTL lysis. Uptake of antigenic material and active metabolism of the presenting cell are required for processing of S-particles, but not for sensitizing targets with S-peptides. Intracellular processing and presentation of S-particles is blocked in cells treated with chloroquine, NH4Cl, primaquine, or leupeptin, but not by treatment with cycloheximide or brefeldin A. This processing pathway operates efficiently in peptide-transporter-deficient, Ld-transfected T2 cells, revealing a novel endosomal/lysosomal processing pathway for class I-restricted presentation of peptides derived from exogenous S-particles.

Selective stimulation of murine cytotoxic T cell and antibody responses by particulate or monomeric hepatitis B virus surface (S) antigen.

In the murine system, we tested in vivo the immunogenicity of different preparations of the yeast-derived surface antigen (S-antigen or S-protein) of hepatitis B virus (HBV). Native S-protein molecules self-assemble into stable 22-nm particles. BALB/c mice immunized with low doses of native S-particles without adjuvants efficiently generated an H-2 class I-restricted CD8+ cytotoxic T lymphocyte (CTL) response, and developed easily detectable serum antibody titers against conformational determinants of the native S-particle or linear epitopes of the denatured S-protein. Disruption of S-particles with sodium dodecyl sulfate and beta-2-mercaptoethanol generated p24 S-monomers. Injection of an equal dose of S-monomers into mice efficiently primed CTL, but did not stimulate an antibody response against conformational or linear epitopes of the native or denatured S-protein. In vivo priming of CTL by S-particles or S-monomers required "endogenous" processing of the antigen because the injection of an equimolar (or higher) dose of an antigenic, S-derived 12-mer peptide into mice did not prime CTL. Native (particulate) or denatured (monomeric) S-antigen injected with mineral oil (incomplete Freund's adjuvant) or aluminum hydroxide failed to stimulate a CTL response. Hence, different preparations can be produced from a small protein antigen which specifically stimulate selected compartments of the immune system.

Antibody and cytotoxic T-cell responses to soluble hepatitis B virus (HBV) S antigen in mice: implication for the pathogenesis of HBV-induced hepatitis.

Immune responses to components of hepatitis B virus (HBV) are assumed to play an essential role not only in the elimination of the virus but also in the pathogenesis of HBV-induced hepatitis. Protective humoral immunity to HBV is mediated by immune responses to HBV surface antigen (HBsAg). It is important to know which HBsAg preparations induce which type of cellular and humoral immune responses under which immunization conditions. We studied in BALB/c mice the humoral (antibody) response and the class I-restricted cytotoxic T-lymphocyte (CTL) response to different preparations of HBsAg particles: recombinant, small protein particles; plasma-derived, mixed particles formed by large, medium, and small surface proteins; and different preparations of recombinant, mixed particles formed by large and small surface proteins. Specific antibody levels appeared in the sera of immunized mice 2 to 3 weeks after immunization and were correlated with the antigen dose used for priming. HBsAg-specific antibody levels were enhanced by boost injections or by adsorbing the antigen to aluminum hydroxide. Injected in particulate form without adjuvants in the dose range of 0.1 to 10 micrograms per mouse, all HBsAg preparations tested efficiently primed specific CD8+ CTL of defined restriction and epitope specificity. Specific CTL reactivity was detectable from 5 days to more than 4 months postimmunization. In the dose range tested, it was independent of the antigen dose used for immunization and not enhanced by repeated boost injections. CTL were not elicited by HBsAg adsorbed to aluminum hydroxide. We have thus defined conditions under which HBsAg induced preferentially either a cellular immune response or a humoral immune response. These findings may be relevant for the interpretation of HBV-associated immunopathologic phenomena.

Immunization with soluble hepatitis B virus surface protein elicits murine H-2 class I-restricted CD8+ cytotoxic T lymphocyte responses in vivo.

Immunization with soluble proteins only rarely induces a specific response of CD8+ CTL. We describe experiments that demonstrate the efficient and specific in vivo priming of CTL in BALB/c mice immunized with soluble hepatitis B virus (HBV)-derived surface (S) protein. A single (s.c., i.p. or i.v.) injection of a low dose (30 ng to 3 micrograms per mouse) of recombinant S protein particles without adjuvants induced a CTL response. This specific cytotoxic response was read out against a panel of different S protein-expressing transfected mouse cell lines. Effector cells of this response were Ld-restricted, CD3+ CD4- CD8+ CTL. H-2d/Ld+ (BALB/c, C.B-17) mice were responders; H-2d/Ld- (dm2) mutant mice and H-2b (C57BL/6) mice were nonresponders. Injections of various dosages of a S protein-derived, immunogenic, synthetic peptide into BALB/c mice by various routes did not prime CTL. After incorporation of S protein particles into IFA or aluminum hydroxide, these protein Ag lost their ability to specifically stimulate CTL in vivo. After priming of mice with S protein emulsified in IFA or adsorbed to aluminum hydroxide boost injections with native S protein particles were inefficient in stimulating a specific CTL response. These findings are of relevance for the design of synthetic subunit vaccines for which specific stimulation of CD8+ T effector functions is desired.

Heterologous protein production in yeast.

The exploitation of recombinant DNA technology to engineer expression systems for heterologous proteins represented a major task within the field of biotechnology during the last decade. Yeasts attracted the attention of molecular biologists because of properties most favourable for their use as hosts in heterologous protein production. Yeasts follow the general eukaryotic posttranslational modification pattern of expressed polypeptides, exhibit the ability to secrete heterologous proteins and benefit from an established fermentation technology. Aside from the baker's yeast Saccharomyces cerevisiae, an increasing number of alternative non-Saccharomyces yeast species are used as expression systems in basic research and for an industrial application. In the following review a selection from the different yeast systems is described and compared.

High-level expression of foreign genes in Hansenula polymorpha.

The methylotrophic yeast Hansenula polymorpha belongs to a limited number of non-Saccharomyces yeast species used as hosts for heterologous gene expression. It has successfully been applied for the production of hormones, antigens and enzymes. The system excells by mitotically stable recombinant strains, high productivity and faithful processing of the produced polypeptides. The favourable characteristics of this microorganism for protein production at an industrial scale are described in the following article focusing on some recent representative examples.

Simultaneous expression of the S and L surface antigens of hepatitis B, and formation of mixed particles in the methylotrophic yeast, Hansenula polymorpha.

An expression system has been developed for the methylotrophic yeast Hansenula polymorpha and used to co-express both the L (preS1-S2-S) and S hepatitis B surface antigens (HBsAg) under the control of strong methanol-inducible promoters derived from the methanol oxidase and from the formate dehydrogenase genes. A unique feature of this H. polymorpha expression system is the possibility of integrating up to 100 copies of an expression cassette via a multimeric integration mechanism. Several multimeric integrants containing various numbers of L and S expression cassettes were constructed to give a spectrum of strains characterized by different L to S ratios. The expression level of S antigen was 5-8% of the total soluble cell protein. Analysis by sucrose and CsCl density gradient centrifugation and by particle-specific immunoassays demonstrated that the synthesized HBsAg spontaneously assembled into composite subviral particles containing both S and L proteins. Only a minor portion of the L protein was found to be glycosylated. These H polymorpha-derived composite particles can be used for the production of a hepatitis B virus vaccine with the potential for improved immunogenicity due to the presence of a wider spectrum of epitopes and negligible glycosylation.

Constitutive expression of multiple growth factor genes by melanoma cells but not normal melanocytes.

In a panel of metastatic melanoma cell lines we found steady-state mRNA transcripts for multiple growth factors including basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF)-A, PDGF-B, transforming growth factor (TGF)- beta 1, TGF- alpha, melanoma growth-stimulating activity (MGSA), interleukin (IL)-1 alpha, and IL-1 beta but not insulin-like growth factor (IGF)-1 or IGF-2. Expression of growth factor genes was constitutive because prior to RNA extraction melanoma cells were maintained in a chemically defined culture medium free of exogenous growth factors. Each of four cell lines had an individual pattern of expression of either two, four, five, or seven growth factors; however, all cell lines shared expression of the bFGF gene. Two strains of normal melanocytes expressed TGF- beta 1 but not bFGF, PDGF, TGF- alpha , or MGSA mRNA at detectable levels. We tested growth-modulatory effects of the growth factors most frequently expressed by melanoma cells (bFGF, TGF- alpha, TGF- beta, PDGF). None of these stimulated melanoma cell growth consistently, whereas exogenous, acid-activated TGF- beta inhibited melanoma growth at concentrations greater than 10 ng/ml, suggesting that bioactive TGF- beta may represent a physiologic growth inhibitor. Neither neutralizing antisera to PDGF or TGF- alpha nor a monoclonal antibody to the epidermal growth factor (EGF)-receptor inhibited melanoma cell growth. Our results indicate that multiple growth factors are expressed simultaneously and constitutively by melanoma cells but not normal melanocytes in culture. Expression of bFGF is a common feature underscoring the significance of bFGF as an autocrine factor for melanoma cells as described earlier. Secreted PDGF and TGF- alpha are apparently not involved in or not essential for autocrine growth stimulation of melanoma cells.

SV40-transfected human melanocyte sensitivity to growth inhibition by the phorbol ester 12-O-tetradecanoylphorbol-13-acetate.

Normal human melanocytes, which require the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) for growth in culture, were transfected with a SV40 T-antigen-containing plasmid by using the technique of electroporation, and were scored for colonies of morphologically altered cells. The frequency of transformed colonies was higher when selection was done in the absence rather than the presence of TPA in the medium. Three cell lines derived from transformed colonies were characterized. All show an enhanced growth rate compared to parental cells, anchorage independence, loss of dependence on medium supplements for growth, chromosomal abnormalities, an extended life span, and growth inhibition by TPA. They express nerve growth factor receptor and Mr 97,000 protein, melanotransferrin, two antigens usually associated with melanocytic cells, but the transformed cells are not pigmented. The three cell lines underwent crisis at about passage 10 posttransfection; one cell line recovered and appears to have unlimited growth potential. None of the cell lines is tumorigenic. They should be interesting models for studying multistage carcinogenesis in human cells and transcriptional activation by TPA.

Tumor promoters induce a transient expression of tumor-associated genes in both basal and differentiated cells of the mouse epidermis.

The effect of tumor promoters on the in vivo expression of tumor-associated, overexpressed genes was studied. Two of the tumor-associated genes, mal 1 and mal 2 were overexpressed already in the benign papilloma stage of mouse skin carcinogenesis. Overexpression of the other two genes, mal 4 and transin, was specific for the malignant state. Treatment of the normal adult epidermis with the complete tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA) and the incomplete, second-stage promoter 12-O-retinylphorbol-13-acetate (RPA) enhanced transiently the expression of the mal sequences and transin. Fractionation of the adult epidermis on Percoll gradients into basal cells and differentiated, suprabasal cells showed that expression of the mal sequences was enhanced by TPA in both basal and differentiated cells. In contrast, transin expression, which was undetectable in cells of the normal epidermis, was enhanced in only the basal cells of the TPA-treated epidermis. The non-tumor-promoting hyperplastic agent, ethylphenyl propiolate (EPP), applied to the skin at a hyperplastic dose level did not enhance the expression of the mal 4 or transin sequences in the epidermis and had only a slight enhancing effect on the levels of mal 1 and mal 2 transcripts in the epidermis. Our results suggest that the observed stimulated expression of mal 1 and mal 2 is related to proliferative processes, whereas stimulated expression of mal 4 and transin reflects tumor-promoter-specific responses.

Molecular cloning of sequences activated during multi-stage carcinogenesis in mouse skin.

Recombinant DNA techniques were employed to isolate sequences that were activated during multi-stage carcinogenesis in the skin of NMRI mice. Differential screening of 5000 cDNA clones made from poly(A)+ RNA of squamous cell carcinomas induced by 7,12-dimethylbenz[a]anthracene (DMBA) and the tumor promoter 12-O-tetradecanoyl-phorbol-13-acetate (TPA) resulted in the identification of 35 cDNA clones displaying a different hybridization signal. Eight cDNA clones, three of which proved to be identical, were used as probes in transfer hybridization experiments. These cDNA clones, designated pmal-1 to pmal-6, showed a strong signal with carcinoma RNA but either a weak or no signal with RNA from normal epidermis. Clone pmal-2 contained repetitive sequences as shown by Southern analysis resulting in a smeared RNA hybridization signal in RNA blots whereas the other five clones corresponded to discrete size classes of mRNA. Studies on the expression pattern of mal-1,-2 and -3 related sequences revealed transcriptional activation already in the benign papilloma stage of multi-step carcinogenesis.