Bioproduction of human enzymes in transgenic tobacco.

Transgenic plants have significant potential in the bioproduction of complex human therapeutic proteins due to ease of genetic manipulation, lack of potential contamination with human pathogens, conservation of eukaryotic cell machinery mediating protein modification, and low cost of biomass production. Tobacco has been used as our initial transgenic system because Agrobacterium-mediated transformation is highly efficient, prolific seed production greatly facilitates biomass scale-up, and development of new "health-positive" uses for tobacco has significant regional support. We have targeted bioproduction of complex recombinant human proteins with commercial potential as human pharmaceuticals. Human protein C (hPC), a highly processed serum protease of the coagulation/anticoagulation cascade, was produced at low levels in transgenic tobacco leaves. Analogous to its processing in mammalian systems, tobacco-synthesized hPC appears to undergo multiple proteolytic cleavages, disulfide bond formation, and N-linked glycosylation. Although tobacco-derived hPC has not yet been tested for all posttranslational modifications or for enzymatic (anticlotting) activity, these results are promising and suggest considerable conservation of protein processing machinery between plants and animals. CropTech researchers have also produced the human lysosomal enzyme glucocerebrosidase (hGC) in transgenic tobacco. This glycoprotein has significant commercial potential as replacement therapy in patients with Gaucher's disease. Regular intravenous administration of modified glucocerebrosidase, derived from human placentae or CHO cells, has proven highly effective in reducing disease manifestations in patients with Gaucher's disease. However, the enzyme is expensive (dubbed the "world's most expensive drug" by the media), making it a dramatic model for evaluating the potential of plants to provide a safe, low-cost source of bioactive human enzymes. Transgenic tobacco plants were generated that contained the human glucocerebrosidase cDNA under the control of an inducible plant promoter. hGC expression was demonstrated in plant extracts by enzyme activity assay and immunologic cross-reactivity with anti-hGC antibodies. Tobacco-synthesized hGC comigrates with human placental-derived hGC during electrophoretic separations, is glycosylated, and, most significantly, is enzymatically active. Although expression levels vary depending on transformant and induction protocol, hGC production of > 1 mg/g fresh weight of leaf tissue has been attained in crude extracts. Our studies provide strong support for the utilization of tobacco for high-level production of active hGC for purification and eventual therapeutic use at potentially much reduced costs. Furthermore, this technology should be directly adaptable to the production of a variety of other complex human proteins of biologic and pharmaceutical interest.

Structure and regulation of the glpFK operon encoding glycerol diffusion facilitator and glycerol kinase of Escherichia coli K-12.

The glpFK operon maps near minute 88 on the linkage map of Escherichia coli K-12 with glpF promoter proximal. The glpF gene encodes a cytoplasmic membrane protein which facilitates the diffusion of glycerol into the cell. The glpK gene encodes glycerol kinase. In the present work, the nucleotide sequence of the 5'-end of the operon, including the control region, the glpF gene, and part of the glpK gene, was determined. The facilitator was predicted to contain 281 amino acids with a calculated molecular weight of 29,780. It is a highly hydrophobic protein with a minimum of six potential transmembrane alpha helices. The transcription start site for the glpFK operon was located 71 base pairs upstream from the proposed translation start codon for glpF. Preceding the transcription start site were sequences similar to the -10 and -35 consensus sequences for bacterial promoters. Binding sites for the cAMP-cAMP receptor protein (CRP) complex and the glp repressor were identified by DNase I footprinting. The region protected by the cAMP.CRP complex contained tandem sequences resembling the consensus sequence for CRP binding. The CRP sites were centered at 37.5 and 60.5 base pairs upstream of the start of transcription. The glp repressor protected an extensive area (-89 to -7 relative to the start point of transcription), sufficient for the binding of four repressor tetramers. Two additional binding sites for the repressor were identified within the glpK coding region. The DNA containing these two operators synergistically increased the apparent affinity of glp repressor for DNA fragments containing the four operators in the promoter region of the glpFK operon. With this study, a total of 13 operators for the glp regulon have been characterized. Comparison of these operators revealed the consensus 5'-WATGTTCGWT-3' for the operator half-site (W = A or T). The relative affinity of the glp repressor for the various glp operators was assessed in vivo using a promoter-probe vector. The relative apparent affinity of the control regions for glp repressor was glpFK greater than glpD greater than glpACB greater than glpTQ. The degree of catabolite repression for each of the operons was assessed using a similar system. In this case, the relative sensitivity of the glp operons to catabolite repression was glpTQ greater than glpFK greater than glpACB greater than glpD.

Purification and characterization of the repressor for the sn-glycerol 3-phosphate regulon of Escherichia coli K12.

The glpR gene encoding the repressor for the sn-glycerol 3-phosphate regulon of Escherichia coli was cloned downstream from the strong pL promoter of bacteriophage lambda. This allowed overproduction of the repressor upon thermal induction of a cryptic lambda lysogen harboring the cI857 gene. The repressor was purified 40-fold to homogeneity from an induced strain. The purification scheme utilized polyethyleneimine and ammonium sulfate fractionation, followed by phosphocellulose and DEAE-Sephadex chromatography. Purification was monitored by measuring the binding of radiolabeled inducer (sn-glycerol 3-phosphate) to the repressor. The purified repressor migrated as a single band exhibiting a subunit molecular weight of 30,000 assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The molecular weight of the repressor under nondenaturing conditions was 100,000-130,000 suggesting the repressor is a tetramer under native conditions. Interaction of the repressor with sn-glycerol 3-phosphate was studied using flow dialysis. Scatchard analysis of the data indicated four binding sites/repressor tetramer and a dissociation constant of 31 microM. Interaction of the repressor with DNA was studied using band-shift electrophoresis. The repressor specifically bound DNA fragments containing the control regions for the glpD, glpK, and glpT-A genes. Binding of DNA by the repressor was diminished in the presence of sn-glycerol 3-phosphate.