Production and characterization of recombinant human anti-HBs Fab antibodies.

Recombinant human Fab antibodies were generated with different reactivities against the hepatitis B virus surface (HBs) antigen. To isolate the antibodies, a method was used that combined transformation of human B cells by Epstein-Barr virus (EBV) infection with a primer-vector system developed for isolating DNA fragments of human Ig Fab portions. With this method, monoclonal and oligoclonal cell lines producing anti-HBs antibodies were established and three anti-HBs Fab antibodies were isolated from two of these cell lines. From analysis of affinity characteristics, immunohistochemical activity, and cytolysis activity, these three Fab antibodies were classified into three different groups. The first group had high affinity for HBs, the second had the ability to kill HBV-infected cells, and the third was applicable to immunohistochemical staining with HBV-infected cells. The combined effect of these antibodies was also investigated by complement-dependent cytotoxicity assay.

Haemopoietic biglycan produced by brain cells stimulates growth of microglial cells.

We have recently found that soluble biglycan purified from rat thymic myoid cells had haemopoietic activity capable of inducing preferential growth and differentiation of monocytic lineage cells from various haemopoietic sources, including brain microglial cells. In the present study, to understand developmental mechanisms of microglial/monocytic cells in the brain, we have attempted to identify haemopoietic activity of the brain biglycan. The mRNA and the immunological epitope of biglycan were detected in the rat brain homogenates and several rat glial cell lines. Immunohistochemical study showed that several different types of brain cells produced biglycan. During development biglycan synthesis in the brain appeared to be increased. The brain haemopoietic biglycan was easily separated by DEAE-Sepharose chromatography from the macrophage colony stimulating factor (M-CSF) which was concomitantly produced from the brain cells. The brain haemopoietic biglycan, purified through immunoaffinity column, indeed stimulated growth of primarily cultured microglial cells. Taken together, these results suggest that the haemopoietic biglycan plays an important role in generating brain-specific circumstances for development of microglial/monocytic cells.

cDNA cloning and deduced amino acid sequence of prothrombin activator (ecarin) from Kenyan Echis carinatus venom.

The complete amino acid sequence of ecarin is deduced from the nucleotide sequence of a cDNA clone isolated by screening a venomous gland cDNA library of Kenyan Echis carinatus. The cDNA sequence with 2379 base pairs encodes an open reading frame of 616 amino acids with a remarkable sequence homology to the putative precursor protein of trigramin from Trimeresurus gramineus venom (61% identity) and a large hemorrhagin, jararhagin, from the pit viper Bothrops jararaca venom (62% identity). Thus, ecarin, as well as jararhagin and trigramin, is translated as a precursor protein, which may be processed posttranslationally. The ecarin proprotein has a "cysteine switch" motif (-Pro-Lys-Met-Cys-Gly-Val-) similar to that involved in the activation of matrix metalloproteinase zymogens. The processed mature protein consists of 426 amino acid residues (residues 191-616), showing the strongest sequence similarity with that of Russell's viper venom factor X activator (RVV-X) heavy chain (64% identity). Like RVV-X heavy chain, ecarin contains metalloproteinase, disintegrin, and cysteine-rich domains. The metalloproteinase domain has a typical zinc-chelating sequence (-His-Glu-Xaa-Xaa-His-Xaa-Xaa-Gly-Xaa-Xaa-His-), as found in crayfish astacin. In the disintegrin domain of ecarin, the Arg-Gly-Asp sequence is replaced by Arg-Asp-Asp, as found in the disintegrin domains of RVV-X heavy chain (Arg-Asp-Glu) and a guinea pig sperm fusion protein, PH-30 beta (Thr-Asp-Glu). These findings show that while there are structural and evolutionary relationships among these proteins, each has a unique functional activity.

Heterogeneous expression of human antibody lambda chains by concanavalin A-resistant hybridomas lead to changed antigen binding.

Human HB4C5 hybridoma cells produce a lung cancer-specific human monoclonal antibody that possesses a lambda light chain on which a N-linked carbohydrate chain is attached at the first complementary determining region. Up to six light chains of various sizes are secreted by the original and concanavalin A (ConA)-resistant HB4C5 clones. Two of these six light chains are derived directly from the HB4C5 original and are 30 and 32 kDa in size. The structural difference between these two light chains has been shown to be a N-linked carbohydrate located at a single site. The other four variants, which are derived from ConA-resistant variants, have light chain sizes ranging from about 26 to 29 kDa. No changes in the secretory forms were observed when glycosylation was inhibited by the addition of tunicamycin, indicating that the heterogeneous light chain forms produced by these ConA-resistant variants resulted from the size differences of core polypeptides and not from N-glycosylation. Sequence analysis from one of the variant light chains revealed that its V lambda gene segment differs markedly from identified human V lambda gene segments (at most, 60% homology with known human V lambda subgroups) and, furthermore, the variant light chain shares 80% homology with the rabbit V lambda gene. In addition, the antigen binding ability of these variant antibodies changed significantly in both specificity and affinity. These results suggest that heterogeneous light chains expression by ConA-resistant variants may provide a new way for developing an antigen-specific human monoclonal antibody repertoire.