Targeting a truncated Ho-endonuclease of yeast to novel DNA sites with foreign zinc fingers.

Ho-endonuclease of the yeast, Saccharomyces cerevisiae, initiates a mating type switch by making a site-specific double strand break in the mating type gene, MAT. Ho is a dodecamer endonuclease and shares six of the seven intein motifs with PI- Sce I endonuclease, an intein encoded by the VMAI gene. We show that a 113 residue truncated Ho-endonuclease starting at intein motif C initiates a mating type switch in yeast. Ho is the only dodecamer endonuclease with zinc fingers. To see whether they have a role in determining site specificity we exchanged them for zinc fingers of the yeast transcription factor, Swi5. A chimeric endonuclease comprising the dodecamer motifs of Ho (C-E) and the zinc finger domain of Swi5 cleaves a Swi5 substrate plasmid in vivo. A similar chimera with the zinc fingers of SpI cleaves a GC box rich substrate plasmid. These experiments delineate a catalytic fragment of Ho-endonuclease that can be fused to various DNA binding moieties in the design of chimeric endonucleases with new site specificities.

Identification of the heterothallic mutation in HO-endonuclease of S. cerevisiae using HO/ho chimeric genes.

HO-endonuclease initiates a mating-type switch in the yeast S. cerevisiae by making a double-strand cleavage in the DNA of the mating-type gene, MAT. Heterothallic strains of yeast have a stable mating type and contain a recessive ho allele. Here we report the sequence of the ho allele; ho has four point mutations all of which encode for substitute amino acids. The fourth mutation is a leucine to histidine substitution within a presumptive zinc finger. Chimeric HO/ho genes were constructed in vivo by converting different parts of the sequence of the genomic ho allele to the HO sequence by gene conversion. HO activity was assessed by three bioassays: a mating-type switch, extinction of expression of an a-specific reporter gene, and the appearance of Canr Ade- papillae resulting from excision of an engineered Ty element containing the HO-endonuclease target site and a SUP4 degrees gene. We found that the replacement of the fourth point mutation in ho to the HO sequence restored HO activity to the chimeric endonuclease.

Analysis of the DNA topoisomerase-II-mediated cleavage of the long terminal repeat of Drosophila 1731 retrotransposon.

The interaction of DNA topoisomerase II with the long terminal repeat (LTR) of the Drosophila melanogaster 1731 retrotransposon was studied. The covalent binding of topoisomerase II to the LTR was strongly stimulated by different inhibitors of the enzyme 4'-demethylepipodophyllotoxin-9-(4,6-O-2-ethylidene-beta-D-glucopy ranoside (VP-16), 4'-(9-acridinylamino)methanesulfon-m-anisidine) (m-AMSA) and an ellipticine derivative. Enzyme-mediated DNA cleavage could be observed in the absence of inhibitors and was stimulated in their presence. Cleavage occurred predominantly at sites located within or at the boundary of alternating purine/pyrimidine tracts in agreement with previous observations [Spitzner, J. R., Chung, I. K. & Muller, M. T. (1990) Eukaryotic topoisomerase II preferentially cleaves alternating purine-pyrimidine repeats, Nucleic Acids Res. 18, 1-11]. In addition, all of the cleavage sites observed in the absence of inhibitor were located in the U3 region of the LTR. The site specificity of drug-induced cleavage was studied and the conformity of the cleavage sites with previously established consensus sequences was examined. Our results suggest that DNA topoisomerase II, through its ability to alter the degree of DNA supercoiling, might be involved in the control of different functions of the LTR.

Identification of an epitope shared by the DNA-binding domain of glucocorticoid receptor and the B chain of insulin.

A monoclonal antibody (8G11-C6) generated by an auto-anti-idiotypic route and directed to a site near the ligand-binding site of the glucocorticoid receptor also binds to native insulin and the B chain of insulin but not to the A chain of insulin. The glucocorticoid receptor and the B chain of insulin, therefore, share a cross-reacting epitope. Examination of the primary sequences of the two proteins revealed a limited number of regions of identity or close homology. Several peptides representative of those regions were synthesized. A heptapeptide sequence of the B chain of insulin with homology to a sequence in the first "zinc finger" of the DNA-binding domain of the glucocorticoid receptor was identified as the cross-reactive epitope. This heptapeptide sequence is restricted to and highly conserved among insulins of various species. Homologous sequences are found in the DNA-binding domains of most steroid receptors and related DNA-binding proteins. Consistent with this is the finding that 8G11-C6 inhibits the binding of glucocorticoid receptor to DNA-cellulose.

Probing the surface of Z-DNA with anti-nucleoside antibodies.

Antibodies specific for cytidine (C) and guanosine (G) were used to probe the surface of two Z-DNA conformers. When tested by ELISA, anti-G reacted with poly(dG-dC).poly(dG-dC) treated with bromine water [Br-poly(dG-dC).poly(dG-dC)] but anti-C did not. A weak reaction with anti-C was detected by dot immunobinding. In contrast, anti-C reacted strongly with poly(dG-dC).poly(dG-dC) treated with N-acetoxy-2-(acetylamino)fluorene [AAF-poly(dG-dC).poly(dG-dC)]; anti-G reacted weakly, despite the fact that most G residues had not been substituted with AAF. Neither antinucleoside bound to the B conformation of poly(dG-dC).poly(dG-dC). In competition experiments, GMP was the most efficient competitor of the reaction of anti-G with Br-poly(dG-dC).poly(dG-dC); AMP and TMP were 100-fold less efficient, and CMP did not compete to a significant extent. In contrast, the reaction of anti-Z with Br-poly(dG-dC).poly(dG-dC) was not inhibited by nucleotides. Of five possible sites recognized on guanosine by anti-G antibodies (N1, C6, O6, N7, and C8), AMP and TMP share three or their equivalent and CMP only one. The binding of anti-C to AAF-poly(dG-dC).poly(dG-dC) was inhibited best by CMP; AMP was 8 times less efficient; GMP and TMP were about 35-fold less efficient than CMP. Thus, although the amino group on the C4 position of CMP appears to be immunodominant, the capacity of GMP and TMP to inhibit the reaction indicates that other sites are also recognized in AAF-poly(dG-dC).poly(dG-dC), e.g., the exposed C5 position.(ABSTRACT TRUNCATED AT 250 WORDS)

A monoclonal antibody to the double-stranded polyribonucleotide complex poly(A) X poly(U).

A monoclonal antibody to the double-stranded polyribonucleotide complex poly(A) . poly(U) was derived from the fusion of spleen cells from immunized DBA/2 mice and the P3 X X63-Ag8 plasma cytoma. Specificity studies using radioimmunoassays showed that the anti-poly(A) . poly(U) does not cross-react with single-stranded polyribonucleotides. RNA X DNA hybrids or DNAs. In addition to RNA duplexes associating adenine and uracil, it recognizes synthetic poly(I) . poly(C) and naturally occurring reovirus RNA. It is thus directed against a conformational epitope with an absolute requirement for two polyribose phosphate chains. However, the antibody does not cross-react with poly(G) . poly(C) and is therefore able to distinguish between RNA double helices.

The mouse immune response to the double stranded polyribonucleotide complex poly(G) . poly(C).

Ten inbred strains of mice were immunized with the double stranded polyribonucleotide complex polyguanylic . polycytidylic acid [poly(G) . poly(C)]. While some immunogenic properties of this duplex were comparable to those of other nucleic acids antigens, differences were also noted. High (SJL/J, BALB/c), low (DBA/2, AKR) and intermediate responders were observed; these differences were not abolished by adsorption of the duplex to MBSA. This pattern of immune response is distinct both from that observed with two other synthetic polyribonucleotide double helices [poly(A) . poly(U) and poly(I) . poly(C)] and with single stranded DNA. The anti-poly(G) . poly(C) activity was localized in the 7S region, whether the sera came from high or low responders, from mice immunized with or without a carrier, after one or several injections. In contrast with anti-poly(A) . poly(U) sera which do not react with poly(G) . poly(C), anti-poly(G) . poly(C) exhibited poly(A) . poly(U) binding activity; no clear relationship between the two activities, however, could be demonstrated. Thus a series of immunological properties differentiates poly(G) . poly(C) not only from the natural polydeoxyribonucleotide single stranded DNA, but also, and more unexpectedly, from two other double stranded polyribonucleotide complexes. These observations suggest that the mechanism controlling the antibody response to poly(G) . poly(C) differs from that regulating poly(A) . poly(U) and/or poly(I) . poly(C), and are to be connected with the fact that the anti-poly(G) . poly(C) antibodies occurring in the sera of patients with systemic lupus erythematosus did not correlate with the antibody activities directed toward the other duplexes.