Preparation of 1mg/ml BSA

[TytoAlba95]

I had to prepare 1mg/ml BSA for Total protein estimation by Lowry method. I was suggested to add 1ml distilled water to 1mg of BSA in a 2 ml ependoff tube, instead of adding distilled water upto the 1ml mark. The explanation that was given to me was, that, 1mg is a very small amount and it doesn't affect the total volume, much if we add that to 1ml of dH2O. Also as BSA crystals doesn't dissolve easily, and forms froth in the process of dissolving, it becomes more difficult to add water till the1ml mark, as the miniscus isn't formed.

So my question, is the rationale correct? (It sounds okay to me)
Should I be doing this for all solutions where the amount of solute is very less (<1g?)?

 

[Ygggdrasil]

It depends on how accurate you need to be. For most applications (e.g. preparing protein standards for common biochemical assays), the method you describe is fine. It is very difficult to measure out 1mg, so there will be a lot of error associated with that step, and using the marks on the Eppendorf tubes to measure out 1mL is also not very accurate, so adding 1mL of water to 1mg of BSA is probably not going to be much more accurate and reproducible than dissolving 1 mg of BSA to 1mL of total solution.

If you need more accuracy, I'd probably use an orthogonal method (e.g. UV/vis spectrophotometry) to measure the concentration of the solution you create and use that measured concentration as the real concentration rather than assuming it is 1 mg/mL.

 

[TytoAlba95]

Thank you.

 

[epenguin]

Common error or deformation amongst students and beginners (I guess induced by brainwashing and didactic terrorism) to do with high precision a procedure step when other steps before or after have intrinsically considerably less precision, thus wasted time and effort.

When you measure protein concentration by the Lowry method, you are dependent on the assumption that the extinction coefficient of the coloured product formed from your protein is the same as that formed your albumin standard (reflecting rough similarity(?) of amino acid compositions especially aromatic aa's between one and the other). Good luck with that. If you're measuring, I don't know, say total albumin levels in circulating blood, that may be good enough. And if you're only interested in the variation of these levels according to something else, well even a systematic error would not affect your conclusions very much. Yes it will practically always be a lot better than just order of magnitude, but I have known a case where estimating protein concentration of a purified protein of which only small quantities were available give the conclusion that there were four binding sites for certain ligands per protein molecule.Later better investigations some years later showed that instead there were six sites per molecule.