Biotech Values

[urche]

Re: Plant produced insulin

Thanks for the info on SemBioSys---a player in transgenic proteins that will be interesting to watch.

I would expect Insulin to be one of the simpler proteins to create, being only 51 amino acids. Nonetheless, these have to be arranged in two chains with 2 disulfide bridges and the A chain has another bridge to itself. How important these are to function I don't know. But, it may not be as easy as creating the linear amino acid sequence.

Look at this amazing link if you are interested in the complexity of the insulin molecule:
http://www.cryst.bbk.ac.uk/pps97/course/section11/insulin.html

The picture link in prior post, BTW is the crystalline complex structure of insulin, as produced historically from pancreas extract.

More relevant, here is a description of the insulin monomer, but the link gives the pictures which speak thousands of words:

The insulin monomer is a compact globular structure with a hydrophobic core. Although the surface residues are primarily polar, there are two hydrophobic surfaces on each side of the molecule which are buried during the formation of dimers and 2-zinc hexamers. In the insulin fold, the A chain is a compact unit around which the B chain is wrapped.

The A chain consists of two anti-parallel stretches of imperfect alpha helices (A2 Ile - A8 Thr and A13 Leu - A19 Tyr) which are joined by a turn from A9 Ser to A12 Ser, stabilized by the A6-A11 disulphide. The A chain lies in a plane in which the N and C terminii are brought to the same side, bringing A2 Ile and A19 Tyr into van der Waals contact (see diagram).


The B chain consists of an alpha-helix (B9 Ser - B19 Cys) from which both N and C terminii residues extend. The glycine residues at B20 and B23 allow the chain to fold back on itself in an approximate V- shape, and this brings the C terminal residues B24 Phe and B26 Tyr into van der waals contact with B15 Leu and B11 Leu of the alpha-helix (see diagram).

The insulin fold is formed when interchain disulphides at A7 and A20 form interchain disulphides with the B chain cysteines at B7 and B19 respectively. The (A7-B7) disulphide is fully exposed on the surface of the molecule, whereas the (A20-B19) disulphide is part of the hydrophobic core (see next diagram). Burial of the intrachain (A6 - A11) disulphide and the non-polar side chains of A16 Leu, B11 Leu, B15 Leu, A2 Ile and B24 Phe provides the hydrophobic interior stabilising the fold. The N termini residues of the B chain are folded across and run anti-parallel to the turn in the A chain, giving rise to hydrogen bonding between A11 Leu and B4 Gln, A7 Cys and B5 His, and between A19 Tyr and B25 Phe. Further stability is also provided by a salt bridge between the polypeptide chains at A11 Cys and B4 Gln, between the A7 carbonyl oxygen and the B5 His side chain, and between the A19 carbonyl oxygen and the B25 back-bone nitrogen. Further stability is also provided by a salt bridge between B29 Lys and A4 Glu and between the positively charged B22 Arg side chain and the negatively charged A21 terminal carboxyl group (in molecule 1 only).

My point is that this is a relatively simple molecule and to have it produced in a plant, which lacks the natural enzymatic steps to allow human style folding is a daunting undertaking. Despite the claim in the PR ("The study demonstrates bio-equivalence of purified insulin from Arabidopsis seeds when compared with commercial insulin products in an animal model.)" put me down as still in the dubious if not incredulous category.

BTW, this is all relevant to GTCB in that, IMO, the goat's two main claims to fame are:
1) the ability to cost-effectively produce therapeutic proteins;
2) the production of these proteins in a mammalian in vivo system facilitates natural folding and thereby lessens chances of immunogenicity and diminished efficacy.

Urche