Methods for recombinant peptide production

#The present invention relates generally to improved methods for the production of recombinant peptides from bacterial cells.

#Although bioactive peptides can be produced chemically by a variety of synthesis strategies, recombinant production of peptides, including those in the 5-50 amino acid size range, offers the potential for large scale production at reasonable cost. However, expression of very short polypeptide chains can sometimes be problematic in microbial systems, including in bacterial cells such as Escherichia coli. This is true even when the peptide sequence is expressed as part of a fusion protein. As part of a fusion protein, peptides may be directed to specific cellular compartments, i.e. cytoplasm, periplasm, or media, with the goal of achieving high expression yield and avoiding cellular degradative processes.

#Preparation of a peptide from a fusion protein in pure form requires that the peptide be released and recovered from the fusion protein by some mechanism and then obtained by isolation or purification. Methods for cleaving fusion proteins have been identified. Each method recognizes a chemical or enzymatic cleavage site that links the carrier protein to the desired protein or peptide [Forsberg et al., I. J. Protein Chem. 11, 201-211, (1992)]. Chemical cleavage reagents in general recognize single or paired amino acid residues which may occur at multiple sites along the primary sequence, and therefore may be of limited utility for release of large peptides or protein domains which contain multiple internal recognition sites. However, recognition sites for chemical cleavage can be useful at the junction of short peptides and carrier proteins. Chemical cleavage reagents include cyanogen bromide, which cleaves at methionine residues [Piers et al., Gene, 134, 7, (1993)], N-chloro succinimide [Forsberg et al., Biofactors 2, 105-112, (1989)] or BNPS-skatole [Knott et al., Eur. J. Biochem. 174, 405-410, (1988); Dykes et al., Eur. J. Biochem. 174, 411-416, (1988)] which cleaves at tryptophan residues, dilute acid which cleaves aspartyl-prolyl bonds [Gram et al., Bio/Technology 12, 1017-1023, (1994); Marcus, Int. J. Peptide Protein Res., 25, 542-546, (1985)], and hydroxylamine which cleaves asparagine-glycine bonds at pH 9.0 [Moks et al., Bio/Technology 5, 379-382, (1987)].

#1. Bacterial Expression Vector Construct pING4702

#A bacterial expression vector which would encode a peptide fusion protein, was constructed. This vector contains a sequence for a gene encoding subunit D of a human osteogenic protein (“Bone D”) (see, amino acids 23 through 161 of SEQ ID NOS: 1 and 2), linked to a sequence encoding a linking sequence that includes the dipeptide Asp-Pro and a sequence encoding a peptide derived from the sequence of BPI (SEQ ID NO: 3). This vector construct, pING4702, was prepared in several steps as described below.

#First, two synthetic oligonucleotides were synthesized that encode a BPI-derived peptide, an Asp-Pro dipeptide and appropriate restriction enzyme recognition sites for cloning. The oligonucleotides encoding this sequence were: 1 5′-GATCCACCGAAAGTGGGTTGGCTGATCCAGCTGTTCCACAAAAAGTAAAGC-3′ (SEQ ID NO: 4) 5′-TCGAGCTTTACTTTTTGTGGAACAGCTGGATCAGCCAACCCACTTTCGGTG-3′ (SEQ ID NO: 5)