Why is recombinant EPO differently charged?

June 8, 2007 8:35 AM

What accounts for the charge difference in recombinant human erythropoietin (EPO) compared to endogenous material?

I wonder if changes to recombinant EPO would make it harder to detect doping. As I understand it, doping with EPO by athletes is detected using isoelectric focusing, relying on a charge difference between endogenous and recombinant EPO.

The answer probably lies in one of these papers below, but I am no longer affiliated with a research institution or university so it's harder to get primary sources.

Wide L, Bengtsson C. (1990) Molecular charge heterogeneity of human serum erythropoietin. Br J Haematol 76:121–7.

Wide L, Bengtsson C, Berglund B, Ekblom B. (1995) Detection in blood and urine of recombinant
erythropoietin administered to healthy men. Med Sci Sports Exer 27:1569–76

This letter and response (PDF) from 2003 hints that glycosylation might account for the difference, but the situation is complicated by differences between serum EPO and that found in urine.

posted by exogenous to science & nature (5 comments total)

As far as I know, previously EPO was made using blood material of hamsters somewhere in the process. So, its use could be discovered because hamster material isn't really a natural part of human blood.

Right now, the more expensive brands of EPO have gotten rid off the hamsters, and use only human material. So, its illegal use by athletes cannot be discovered anymore.

The latest figures I know, tell me that 80% of all the EPO produced isn't used by kidney patients, sold to athletes.
posted by ijsbrand at 8:48 AM on June 8, 2007

This is a pretty good description of the problems with testing as of 2005.
I also found this article which is pretty illuminating.
posted by exogenous at 9:11 AM on June 8, 2007

Some detail on the Lasne test.
posted by caddis at 9:31 AM on June 8, 2007

and some more (pdf)
posted by caddis at 9:39 AM on June 8, 2007

You're right, the difference in charge between native human Epo and recombinant Epo comes from differences in their glycosylation (specifically, different numbers of a charged sugar called sialic acid).

It is quite common to find that recombinant human proteins made in non-human cells have somewhat different glycosylation patterns even though the protein (i.e. amino acid) sequences are identical. The "machinery" that glycosylates the protein is a little bit different in different species, and it makes slightly different oligosaccharide structures which are eventually attached to the protein. These structures are not controlled by the gene for Epo (or whatever), so they are characteristic of the host species, not the original human protein.

So if the recombinant product is made in big tanks of, say, hamster cells, it will have a different "hamster-style" glycosylation pattern that will show up on isoelectric focusing or capillary electrophoresis.

Some of the testing confusion seems to arise from antibodies which detect Epo plus other proteins, giving false positives. A better method, which it sorta sounds like that Australian company is working on, would be to combine antibodies (to fish out Epo plus whatever other proteins it binds) with IEF or CE to give a so-called "2 dimensional" method. The antibodies retain Epo and let you get rid of most of the uninteresting stuff, and the IEF or CE differentiates between Epo and the other cross-reacting proteins.

There are also differences between Epo in blood/serum and in urine, but again a 2D method would allow you, in principle, to measure either form. (Again, from the articles linked above, the differences seem to be in the glycosylation pattern rather than in the protein sequence.)
posted by Quietgal at 12:02 PM on June 8, 2007 [1 favorite has favorites]