ORIGINAL PAPER M. Terashima á Y. Murai á M. Kawamura S. Nakanishi á T. Stoltz á L. Chen á W. Drohan R. L. Rodriguez á S. Katoh Production of functional human a 1 -antitrypsin by plant cell culture Received: 18 January 1999 / Received revision: 26 April 1999 / Accepted: 1 May 1999 Abstract Recombinant human a 1 -antitrypsin (rAAT) was expressed and secreted from transgenic rice cell sus- pension cultures in its biologically active form. This was accomplished by transforming rice callus tissues with an expression vector, p3D-AAT, containing the cDNA for mature human AAT protein. Regulated expression and secretion of rAAT from this vector was achieved using the promoter, signal peptide, and terminator from a rice a-amylase gene Amy3D. The Amy3D gene of rice is tightly controlled by simple sugars such as sucrose. It was possible, therefore, to induce the expression of the rAAT by removing sucrose from the cultured media or by al- lowing the rice suspension cells to deplete sucrose catabolically. Although transgenic rice cell produced a heterogeneous population of the rAAT molecules, they had the same N-terminal amino acids as those found in serum-derived (native) AAT from humans. This result indicates that the rice signal peptidase recognizes and cleaves the novel sequence between the Amy3D signal peptide and the ®rst amino acid of the mature human AAT. The highest molecular weight band seen on West- ern blots (AAT top band) was found to have the correct C-terminal amino acid sequence and normal elastase binding activity. Staining with biotin-concanavalin A and avidin horseradish peroxidase con®rmed the glycosylat- ion of the rAAT, albeit to a lesser extent than that observed with native AAT. The rAAT, puri®ed by im- munoanity chromatography, had the same association rate constant for porcine pancreatic elastase as the native AAT. Thermostability studies revealed that the rAAT and native AAT decayed at the same rate, suggesting that the rAAT is correctly folded. The productivity of rice suspension cells expressing rAAT was 4.6±5.7 mg/g dry cell. Taken together, these results support the use of rice cell culture as a promising new expression system for production of biologically active recombinant proteins. Introduction The biotechnology industry employs a number of ex- pression systems to produce analytical and commercial scale quantities of recombinant proteins. These systems exploit a wide range of biological hosts including ge- netically modi®ed microbes (Georgiou 1988; Harashima 1994), insect cells (Luckow and Summers 1988), trans- genic animals (Lubon et al. 1996) and plants (Kusnadi et al. 1997; Mielle 1997). Each system has its own ad- vantages and disadvantages, but to date no recombinant expression system combines the features of low-cost, high-level expression with proper processing and modi- ®cation while retaining structural integrity and biologi- cal activity. In the case of Escherichia coli, possibly the best studied of all the expression systems, lack of glycosylation and the formation of inclusion bodies re- quire expensive and laborious ex vivo manipulation to obtain pure and functional recombinant proteins (Buchner and Rudolph 1991). In response to the demand Appl Microbiol Biotechnol (1999) 52: 516±523 Ó Springer-Verlag 1999 M. Terashima (&) Department of Chemical Engineering, Osaka Prefecture University, 1±1 Gakuen-Cho, Sakai 599±8531, Japan e-mail: terasima@chemeng.osakafu-u.ac.jp Y. Murai á M. Kawamura Department of Synthetic Chemistry and Biological Chemistry, Kyoto University, Kyoto 606±01, Japan S. Nakanishi á R.L. Rodriguez Section of Molecular Biology and Cellular Biology, University of California, Davis, Davis, CA 95616, USA T. Stoltz Applied Phytologics, Inc., 4110 North Freeway Blvd., Sacramento, CA 95834±1219, USA L. Chen Division of Plant Biotechnology, Scripps Research Foundation, La Jolla, CA 92037, USA W. Drohan American Red Cross Holland Laboratory, Rockville, MD 20855, USA S. Katoh Graduate School of Science and Technology, Kobe University, Rokkodai, Nada-ku, Kobe 657±8501, Japan