Cysteine / Cystine estimation in BSA (long(winded))

Artem Evdokimov AEVDOKIMOZ at
Sun Oct 21 10:34:48 EST 2001

Dear Josh,

It seems that you are confused. Let's see if I can make sense of what you

> The problem is, the absorbances at 412 nm for the DTNB don't match up, ie
> peak absorbance builds up gradually to fraction 6 of 12, but doesn't
> to zero like the readings for Bradford method.

Does this mean that you added DTNB and THEN ran the size exclusion column ?
No wonder that your DTNB peak is delayed from your protein peak - DTNB is a
small molecule which will come out of the sizing column MUCH LATER than the
large protein. Does it help ?

Alternatively, if you did not load DTNB on the column, but rather tested all
the fractions that came out, then the higher activity in later fractions is
due to the fact that your beta-ME came out in the small molecule fraction
too, so DTNB found lots of free SH to react with. Does that help ?


Ellman 's test is simple. You measure the concentration of protein, as
accurately as you can. Then you do the test with an aliquot of your protein,
repeat several times at different concentrations, and normalize using a
curve- or a line- fit. That's it. You're running the sizing column
presumably to get rid of the excess reducing agent (otherwise the test would
be meaningless).

So, in conclusion:

1) you take your protein and reduce it with b-ME.
2) you run it down a sizing (in this case, essentially, desalting) column,
in order to get rid of excess b-ME
3) use Bradford's test to determine which fractions contain protein, and how
much protein is there
40 use DTNB to determine the concentration of reducing groups, and then
relate that to protein conc. thus obtaining the number of SH- groups per
protein molecule.

Hopefully, it helps. Read the original paper, mentioned in the URL above.


"Josh" <jj_barnett at> wrote in message
news:3bd2e42c.0 at
> Hi there. Sorry for the low-brow question, but I am struggling with a prac
> write-up at uni. Basically;
> 1) std curve of BSA in Guanidine HCl - Tris HCl by Bradford method -
> Absorbance at 595 nm
> 2) run reduced ( by beta-M.EtOH ) BSA through sephadex column, collect 12
> fractions, add DTNB and record absorbance at 412 nm
> The question the lecturer has set is;
> " Calculate the concentration of sulfhydryl group per unit mass of protein
> before and after reduction. Estimate how many moles of cystine are present
> in a mole of BSA."
> We are given the eqn,
> A 412 = E(DTNB).Cc
> E(DTNB) = 13 700 L/(
> Cc = concentration of cysteine reisudes
> OK, so I have my std curve, I also have set of readings by Bradford method
> from column fractions ie
> 1    0
> 2    0
> 3     1.10
> 4    0
> so the protein all came out in fraction 3, right?
> The thing I am trying to get my head around is why the -SH groups that
> with the DTNB elute more gradually than the whole protein? That is, how do
> relate the concentration of cystine in BSA to the concentration of protein
> that came out of the column?
> Or are my results just shite?
> Note: previously in this prac we had to determine experimentally the molar
> absorptivity of tryptophan and tyrosine in 6M Guanidine HCl -Tris HCl, and
> then work out the number of tyrosine and tryptophan residues per molecule
> BSA. I did this after much headache, but this prac has got me baffled!
> Thanks for any pointers...:)

More information about the Proteins mailing list