Soluble periplasmic production of human granulocyte colony-stimulating factor (G-CSF) in Pseudomonas fluorescens.

Cost-effective production of soluble recombinant protein in a bacterial system remains problematic with respect to expression levels and quality of the expressed target protein. These constraints have particular meaning today as "biosimilar" versions of innovator protein drugs are entering the clinic and the marketplace. A high throughput, parallel processing approach to expression strain engineering was used to evaluate soluble expression of human granulocyte colony-stimulating factor (G-CSF) in Pseudomonas fluorescens. The human g-csf gene was optimized for expression in P. fluorescens and cloned into a set of periplasmic expression vectors. These plasmids were transformed into a variety of P. fluorescens host strains each having a unique phenotype, to evaluate soluble expression in a 96-well growth and protein expression format. To identify a strain producing high levels of intact, soluble Met-G-CSF product, more than 150 protease defective host strains from the Pfenex Expression Technology™ toolbox were screened in parallel using biolayer interferometry (BLI) to quantify active G-CSF binding to its receptor. A subset of these strains was screened by LC-MS analysis to assess the quality of the expressed G-CSF protein. A single strain with an antibiotic resistance marker insertion in the pfaI gene was identified that produced>99% Met-GCSF. A host with a complete deletion of the autotransporter-coding gene pfaI from the genome was constructed, and expression of soluble, active Met-GSCF in this strain was observed to be 350mg/L at the 1 liter fermentation scale.

Pharmacokinetics and anticoagulant properties of the factor VIIa-tissue factor inhibitor recombinant Nematode Anticoagulant Protein c2 following subcutaneous administration in man. Dependence on the stoichiometric binding to circulating factor X.

Recombinant Nematode Anticoagulant Protein c2 (rNAPc2) is a potent (K(i) =10 pM), inhibitor of the factor VIIa/tissue factor (fVIIa/TF) complex that requires the prerequisite binding to zymogen or activated factor X (fX). In two double blind, place-bo-controlled, sequential dose-escalation phase I studies, rNAPc2 was found to be safe and well tolerated following single and repeat subcutaneous administrations in healthy human male volunteers at doses ranging from 0.3 to 5 micro g/kg. There was a dose-dependent elevation of the prothrombin time reaching almost 4-fold above the baseline value in the highest dose group that directly correlated with rNAPc2 plasma concentration. In contrast, there was little or no effect on the activated partial thromboplastin time, thrombin time or template bleeding time. The pharmacokinetic behavior of rNAPc2 revealed a dose-independent and prolonged elimination half-life (t(1/2)beta) with a mean of >50 hours. A high affinity interaction between rNAPc2 and plasma fX was shown to be essential for the prolonged t(1/2)beta in man using crossed immunoelectrophoresis and was confirmed by exploiting the considerably weaker interaction between rNAPc2 and bovine fX which resulted in an attenuated t(1/2)beta of approximately 1.5 hours in calves. The accumulated data suggests that rNAPc2 is safe and well tolerated following repeat subcutaneous administrations at doses up to 5 micro g/kg in healthy volunteers. In addition, the in vivo fate of rNAPc2 in man appears to be governed by its high affinity interaction with circulating fX. This data supports the continued development of this novel anticoagulant for the prevention and treatment of acute thrombotic disorders.