Chromatography Forum

dear all,
can anybody suggest some general remedies to extract the drugs which is bound by protiene.

with regard
pliva

Enzymatically or chemically hydrolize the protein?

There are a trillion methods if binding is ionic, or van der Waal´s type, etc., like chaotropic salts (KSCN ......) other chaotropes. Probably the most common method is MeOH or ACN extraction. Precipitation of the proteins with trichloroacetic most likely will not work (unless drugs are quite water soluble). If it is "covalently" bound maybe Kostas´s method or specific reagents for a specific bond (there is lots of information on some of the latter in the novabiochem catalog).

The general approach is to use a strong acid, such as phosphoric acid or so. Contrary to the comment by HW, this is the most commonly used procedure for drugs in plasma. We have been successful also with strong base, ammonia or NaOH. The addition of a lot of acetonitrile often works also, but this has occasionally been questioned. The combination of an acid with acetonitrile is like to work very well.

If precipitation with acid, without making sure that the drug did not coprecipitate, is the most common method than I am worried about our medication (actually I am generally extremely worried about medicine anyway......have seen too much).

HW: I did not talk about a coprecipitation, which is commonly very rare at the low concentrations that most drugs are used at. I was talking about protein binding. I also was not talking about a tiny touch of acid. Common procedures are designed to do the opposite what you are trying to do: they are designed to denature the protein. After having seen many, many different SPE procedures in our labs and others that start with an acidification of the plasma sample, and having never seen a problem that was attributable to residual analyte bound to plasma after treatment with acid, I think that your objection is fiction.

Grown up with acid and acetonitrile precipitation of plasma proteins, I was very suspicious to denaturate with phosphoric acid, followed by SPE. But I completely agree with Uwe: we did some experiments with drugs, 80%-90% protein binding, and found recoveries of approximately 100%.

Regards Bert

Uwe, Bert,
as long as you make sure that you have adequate recovery everything is fine, of course. I am not saying that acid precipitation can´t work, I just wanted to point out that it may not work if the compound has low solubility in water, has a high affinity to protein (natural or denatured), and if protein precipitation is too rapid (a low concentration of analyte doesn´t help in the latter case). Incidentally, I have had very good results in precleaning via partial precipitation of proteins with Na2SO4 and a bit of methanol, followed by a restricted access column. I also used precipitation with trichloroacetic acid (of course, I have done thousands of anals with MeOH, or ACN precipitation/extractions which can almost always be made to work).
Thus, I don´t see how acid precipitation is the "opposite of what I am trying to do". Again, I am advocating that for some substances you need to make sure that they are soluble in the aqu. phase when you precipitate proteins. Denaturing, especially in such cases, does not necessarily remove the analyte from the protein.
Here is an example: Clin Chem, 34, 1816 (1988), as discussed in J Chrom B, 678, 137 (1996), analyte: cortisol.
Anyway, is there anybody out there who doesn´t check his recovery?

Cortison sticks to things via hydrophobic interaction. Therefore it needs some amount of organic solvent to break this interaction. This non-specific interaction does not require a specific protein binding.

Uwe, if one has what you call specific binding (its nature is probably the same as "hydrophobic interaction", maybe some ionic interaction thrown in) which one, maybe, destroys via denaturing of the protein then this substrate will immediatly attach via "hydrophobic interaction" (nonspecific van der Waals, London....) to the protein and/or vessel wall if it has low solubility in the aqu. phase.

Now an example of where adsorption SPE has not even been considered as the primary pre-cleaning (extraction) step by the people working in that field:
To estimate ouabain in blood or cerebrospinal fluid one uses about 0.5 liter of these liquids and extracts them with MeOH, etc. Now ouabain is reasonably soluble in H2O, but at an extremely low concentration if present at all (in a live person).

Incidentally, there are some people that believe that cortisol has its carrier protein (specific binding), I don´t see why this should be necessary, though. Also, cortison is not cortisol, though in the present context it´s similar.