Hiroko Tada

AN IMPROVED COLORIMETRIC ASSAY FOR INTERLEUKIN 2

Mosmanns method for measuring the number of viable cells with a tetrazolium salt, 3-(4,5-dimethylthiazol-2-yl)-2,5-di-phenyltetrazolium bromide (MTT), was modified to make it possible to measure a large number of interleukin 2 (IL-2) samples at one time with less labor and more accuracy. Each step of the method was examined in detail and modified (the modified MTT method). An IL-2-dependent mouse natural killer cell line, NKC3, was used as an indicator cell line. The incubation period before adding MTT was reduced to 24 h, A solution of 10% sodium dodecyl sulfate-0.01 N HCl was used to dissolve the MTT formazan produced. We have compared the values obtained by the modified MTT method and the conventional [3H]thymidine method (3H-TdR method), and confirmed that the estimates of IL-2 content were almost equal. The variation of IL-2 content measured by both methods was within 5% in terms of the standard error.

Comparison of the biological properties of purified natural and recombinant human interleukin-2

We compared the biological properties of the purified recombinant human IL-2 derived from E. coli with those of purified natural IL-2. Both had almost the same specific in vitro activities on a weight basis to support long-term proliferation of IL-2 dependent human peripheral blood lymphocytes, a mouse killer T cell line, and a mouse natural killer cell line; induce killer cells in normal mouse spleen cells; and induce antibody forming cells in nude mouse spleen cells. No differences in these biological activities were found between two forms of natural IL-2 that were separable by reverse phase high performance liquid chromatography.

Efficient production of a functional mouse/human chimeric Fab′ against human urokinase-type plasminogen activator by Bacillus brevis

Abstract Expression/secretion vectors for the production of Fab′ and single-chain (sc) Fab′ by Bacillus brevis have been constructed. For the production of Fab′, the cDNAs encoding the L chain and Fd′ fragment (Fd with the hinge region) of a mouse-human chimeric Fab′ against human urokinase-type plasminogen activator were fused directly with the translation-start and signal-peptide-encoding regions of the mwp gene, the gene for one of the major cell-wall proteins of Bacillus brevis. The two fused genes were placed tandemly downstream from the promoter of the cell-wall protein gene operon (cwp) of B. brevis. For the production of scFab′, the two cDNAs were linked with a synthetic oligonucleotide encoding a flexible peptide linker of 17 or 24 amino acids, and fused with the translation start and signal-peptide-encoding regions of the mwp gene. Fab′ was efficiently produced by B. brevis, being accumulated at a level of 100u2009mg/l in the culture medium in a simple shake-flask culture, which is the highest level obtained so far for a gram-positive bacterium. On the other hand, the scFab′ remained at a level of a few milligrams per liter in the culture medium. The Fab′ produced by B. brevis showed comparable antigen-binding activity to that of the parental antibody. The L chain and Fd′ fragment, constituting the Fab′, had the correct N-terminal amino acid sequences. These results indicate that B. brevis is a very promising host for the production of native Ig fragments.

Expression and characterization of a chimeric bispecific antibody against fibrin and against urokinase-type plasminogen activator

We have produced a chimeric bispecific antibody that has dual specificity of human fibrin and urokinase-type plasminogen activator (u-PA). Complementary DNAs for variable regions of both anti-fibrin and anti-u-PA antibodies were cloned from two murine hybridomas secreting respective antibodies using polymerase chain reaction (PCR) techniques, and joined to cDNAs for human constant regions to form chimeric antibody genes. Both of two expression vectors for chimeric anti-fibrin and chimeric anti-u-PA antibodies were sequentially introduced into Chinese hamster ovary cells, and stable transfectants secreting the chimeric bispecific antibody were obtained. The highest producer transfectant (SULF/C2-30) secreted high level (about 40 micrograms ml-1) of total chimeric IgG and about 2% of the IgG had the bispecific activity of binding with both antigens. The chimeric bispecific antibody was purified by a combination of affinity chromatographies employing antigen-coupled columns and hydroxyapatite high-performance liquid chromatography. The purified chimeric bispecific antibody significantly enhanced the thrombolytic potency of single chain u-PA in an in vitro clot lysis assay as well as the original murine bispecific antibody.

Purification and partial sequence analysis of human interleukin-2 derived from peripheral blood leukocytes

Human peripheral blood leukocyte-derived interleukin-2 (IL-2) was resolved by DEAE-cellulose column chromatography into three peaks of activity, IL-2A, B, and C, with isoelectric points of 7.2, 6.6, and 7.9, respectively. IL-2 A, B, and C were further purified by reverse phase high performance liquid chromatography and resolved into two apparently homogeneous peaks each with identical molecular weight: A-1 and A-2 (Mr17000); B-1 and B-2 (Mr17500); and C-1 and C-2 (Mr14400). The amino acid compositions and partial NH2-terminal amino acid sequences of these molecular species were consistent with those predicted from IL-2 cDNA sequences derived from Jurkat and peripheral blood leukocytes.

An enzyme immunoassay for human lymphotoxin.

A highly sensitive and specific enzyme immunoassay (EIA) for human lymphotoxin (hLT) has been developed. The assay is based upon a sandwich system employing two kinds of anti-hLT antibodies with neutralizing activity. One of them was mouse monoclonal antibody raised against Escherichia coli-derived recombinant hLT with a deletion of 20 amino-terminal amino acids and used as labelled antibody. The other was rabbit antibody raised against the carboxyl-terminal portion of hLT and used as solid-phase antibody. The EIA employing such a combination was able to detect less than 50 pg/ml of hLT, showing that this method was approximately 5-10 times higher sensitivity than the conventional bioassay employing L929 cell-lysis. The mean recovery of hLT added to serum specimens was 101% and the coefficients of variation were 3.3-7.8% (intra-assay) and 2.9-17.2% (interassay). There was a good correlation between the present EIA and the bioassay (r = 0.93).