Chemical synthesis and biological activity of rat INSL3.

The recently identified protein, insulin 3 (INSL3), has structural features that make it a bona fide member of the insulin superfamily. Its predicted amino acid sequence contains the classic two-peptide chain (A- and B-) structure with conserved cysteine residues that results in a disulphide bond disposition identical to that of insulin. Recently, the generation of insl3 knockout mice has demonstrated that testicular descent is blocked due to the failure of a specific ligament, the gubernaculum, to develop. The mechanism by which INSL3 exerts its action on the gubernaculum is currently unknown. The purpose of this study was to, for the first time, synthesize rat INSL3 and test its action on organ cultures of foetal rat gubernaculum. INSL3 also contains a cassette of residues Arg-X-X-X-Arg within the B-chain, a motif that is essential for characteristic activity of another related member of the superfamily, relaxin. Hence, the relaxin activity of rat INSL3 was also tested in two different relaxin bioassays. The primary structure of rat INSL3 was determined by deduction from its cDNA sequence and successfully prepared by solid phase peptide synthesis of the two constituent chains followed by their combination in solution. Following confirmation of its chemical integrity by a variety of analytical techniques, circular dichroism spectroscopy confirmed the presence of high beta-turn and alpha-helical content, with a remarkable spectral similarity to the synthetic ovine INSL3 peptide and to synthetic rat relaxin. The synthetic rat INSL3 bound with very low affinity to rat relaxin receptors and had no activity in a relaxin bioassay. Furthermore, it did not augment or antagonize relaxin activity. The rat INSL3 did however induce growth of foetal rat gubernaculum in whole organ cultures demonstrating that INSL3 has a direct action on this structure.

Recognition of the minimal epitope of monoclonal antibody Tau-1 depends upon the presence of a phosphate group but not its location.

The major constituent of the paired helical filaments (PHFs) of Alzheimer's disease is the abnormally phosphorylated form of the microtubule-associated protein, tau. Monoclonal antibody (mAb) Tau-1 is used extensively to stain normal human tau, and tau isolated from the brains of Alzheimer's disease patients after dephosphorylation. We used a panel of 6 synthetic peptides to localize the minimal epitope of Tau-1 between amino acids 192-204. All 4 serine residues within this fragment were later phosphorylated individually by chemical methods, and it was shown that none of the peptides carrying a single phosphate group were recognized by the antibody. The serines included those that are probably not transformed in AD and consequently, conclusions drawn about malfunctioning kinase activity, based on Tau-1 immunoreactivity, can be extremely misleading. The recognition was restored at a decreased level when one of the serines was replaced by an alanine residue. mAb AT8 was made by immunizing with the PHFs and was reported to recognize the same region of the protein in a phosphorylated form. AT8 did not, however, cross-react with any of the singly phosphorylated peptides, indicating that the recognition site of the two antibodies are not entirely complementary or the binding to AT8 needs multiple phosphorylation of the antigen. The abolished recognition of the phosphorylated peptides cannot be attributed to a conformational change due to phosphorylation, since all peptides exhibited reverse-turn secondary structures, as indicated by circular dichroism (CD) spectroscopy. Anti-tau mAbs may distinguish between phosphorylated and non-phosphorylated forms of epitopes regardless of the location of the phosphate group.

Synthesis and antitumor and antiviral properties of 5-halo- and 5-(trifluoromethyl)-2'-deoxyuridine 3',5'-cyclic monophosphates and neutral triesters.

The title diesters (11-15; halo substituents F, Cl, Br, I) were prepared by DCC-induced cyclization of the precursor 5'-monophosphate or direct halogenation of the 2'-deoxyuridine 3',5'-cyclic monophosphate. Antitumor activities of 11-15 in cell systems (L1210 and Raji/0) were compared to those of the corresponding nucleosides and 5'-monophosphates. Thus, the 5-F- and 5-CF3-2'-deoxyuridines proved to be highly active derivatives [ID50 values (microgram/mL) for L1210, 0.002 and 0.06, respectively], with the 5'-monophosphates showing comparable potencies. The corresponding 3',5'-cyclic monophosphate diesters were 20-30 times less potent but nonetheless highly cytostatic. All derivatives including 11-15 had greatly increased ID50 values for the thymidine kinase deficient (TK-) L1210 and Raji cells. The 3',5'-cyclic diesters (11-15) evidently are not efficient prodrug sources of the nucleoside 5'-monophosphates in TK- cells. They also proved to be 100- to 2000-fold less efficient inhibitors of L1210 thymidylate synthetase than were the 5'-monophosphates. The 5-substituted 2'-deoxyuridines and their 5'-monophosphates were potent inhibitors of herpes simplex virus (MIC50 mostly 0.07-10 micrograms/mL) and vaccinia virus (MIC50 0.07-0.2 microgram/mL), with antiviral activity decreasing in the order 5-I, 5-Br greater than 5-CF3 greater than 5-Cl greater than 5-F. The 3',5'-cyclic monophosphates (11-15) were for the most part 10- to 40-fold less active than the 5'-monophosphates in the virus assay systems (e.g., MIC50 for the 5-Br and 5-I derivatives ranged 1-20 micrograms/mL). By contrast 11-15 were considerably more potent inhibitors of vaccinia virus growth (MIC50 0.4-2 micrograms/mL). As the neutral 3',5'-cyclic methyl phosphate triesters (16-18), the 5-I and 5-Br compounds were less potent in antiviral and cytostatic agents than the 3',5'-cyclic diesters, while the 5-iodo benzyl triester was in several cases as active as the 3',5'-cyclic diester. The title compounds (11-15) appear to require extracellular hydrolysis to the nucleoside before functioning as antitumor or antiviral agents.