Discussion:
amino acids by DABS HPLC
(too old to reply)
rdd
2007-06-19 13:22:48 UTC
Permalink
Anyone use dabsylation/HPLC for determination of aa's? If so can you
comment on success of method, column used, column life, pitfalls, etc?

Thx.
Shankar Bhattacharyya
2007-07-22 14:18:02 UTC
Permalink
Post by rdd
Anyone use dabsylation/HPLC for determination of aa's? If so can
you comment on success of method, column used, column life,
pitfalls, etc?
We use it. It has advantages and disadvantages, as always. Any decent
C-18 column will do. Most vendors provide conditions you can start
from.

Advantages:
The derivative seems quite stable on an analytical scale. The reaction
is easy to do. All you need is a UV/VIS detector. Pierce has a little
instruction pamphlet for how to do the derivatization. You can
download that from www.piercenet.com.

Disadvantages:
You mix an aqueous solution of your sample (buffered) with an acetone
solution of the reagent, evaporate to dryness, take up in ethanol-
water, heat and inject. You need a good understanding of how much free
amino group you have in the matrix or else you could use up your
reagent. Given that you have a need for evaporation of the sample this
may not be sufficiently quick, once you find out you have a problem.
The same time problem arises if you do not have sufficient
concentration of analyte. Our solution is to make a 10-20X dilution of
the sample in water, dilute 10X, 100X and 1000X in buffer and then
analyze all three diluted solutions, hoping that we have provided a
sufficient range of dilution to cover the needs. If you have any
meaningful number of samples that is a loony solution in terms of
management.
It may be worth mentioning that in any pre-column derivatization it is
essential to have sufficient reagent to cover the needs of all
compounds which can react with the reagent. In post column work all
you need is sufficient to cover what is coming off the column at any
given moment.
The reagent is not all that stable. It needs some care in storage.

Caution: You will see a huge peak for unused reagent (presumably). You
will see a smaller peak for hydrolyzed reagent (presumably). Run
blanks. Otherwise you will find yourself wondering why one or the
other does not behave well on a calibration curve. Also, make sure you
get both compounds off in the run, or you could have them messing up
later runs.

Derivatized amino acids will run before and after these peaks, so you
need to pay attntion. Not that that is anything new but we all have a
tendency to see a nice large peak and say, "Hey that's a great
response." I have not done that in at least a couple of days but I
expect I will soon enough.

We do very little amino acid work (basically for glutamate in foods)
but my preference would be to do chromatography as the amino acids and
then do post-column derivatization with a reagent which does not
itself have signal under the detection conditions. Ortho-
phthallaldehyde comes to mind. Pre-column derivatization is more
common but the methods say to inject OPA derives within a minute or
two of derivatizations. I'm darned if I know why - never gave it much
thought, and I am not inclined to worry about whether that is a
reasonable concern.

There are other fluorescence probes, too.

I think most people nowadays do reverse-phase separations but this is
not really my turf, so take that with a pinch of salt. Dionex has very
good ion-exchange columns, so I am tempted to do ion chromatography
with some variety of post-column derivatization. Amperometric or
conductivity detection may be sufficient at food levels, which would
simplify my life. With that solution or some post-column solution, the
overhead of pre-column (in fact pre-hplc) derivatization presented by
DABSylation goes away. We seldom have the luxury of scheduling an
instrument for tomorrow and not finishing with the instrument that
day. I could live with in-line pre-column derivatization but that
requires me to mess with my hplc systems more than I want to mess with
them.

Unless what you are trying to do is limited in scope, you may be
better off finding a good amino acid lab to do it. I can recommend one
in the US, if you like, where I send my more complicated stuff. That's
all they do.

- Shankar
David Stone
2007-07-23 13:16:39 UTC
Permalink
In article <***@127.0.0.1>,
Shankar Bhattacharyya <***@worldnet.att.net> wrote:

[snip]
Post by Shankar Bhattacharyya
We do very little amino acid work (basically for glutamate in foods)
but my preference would be to do chromatography as the amino acids and
then do post-column derivatization with a reagent which does not
itself have signal under the detection conditions. Ortho-
phthallaldehyde comes to mind. Pre-column derivatization is more
common but the methods say to inject OPA derives within a minute or
two of derivatizations. I'm darned if I know why - never gave it much
thought, and I am not inclined to worry about whether that is a
reasonable concern.
Probably because the reaction between aldehyde and amine group
is easily reversible unless you take additional steps. IIRC, mild
acid hydrolysis will cause the OPA to drop off, amongst over things.
Shankar Bhattacharyya
2007-08-11 17:10:25 UTC
Permalink
Post by David Stone
[snip]
Post by Shankar Bhattacharyya
We do very little amino acid work (basically for glutamate in
foods) but my preference would be to do chromatography as the
amino acids and then do post-column derivatization with a reagent
which does not itself have signal under the detection conditions.
Ortho- phthallaldehyde comes to mind. Pre-column derivatization is
more common but the methods say to inject OPA derives within a
minute or two of derivatizations. I'm darned if I know why - never
gave it much thought, and I am not inclined to worry about whether
that is a reasonable concern.
Probably because the reaction between aldehyde and amine group
is easily reversible unless you take additional steps. IIRC, mild
acid hydrolysis will cause the OPA to drop off, amongst over
things.
Sorry for this very asynchronous exchange but I am here somewhat
erratically. I suspect that the explanation is in the "amongst other
things" you consider above.

While that sort of reaction may be reasonably easily reversible, what
I don't see is why it should, in effect, reverse itself in the same
reaction mixture within minutes. There is not enough time involved for
any large scale change in the local chemical environment, so we appear
to be seeing some odd loss of analyte. I don't see any obvious reason
why the derivatives would destroy themselves rapidly in some way and
reversal of the reaction seems unreasonable without some change in the
environment.

I don't know enough about the chemistry of the derivatives to be
entitled to an opinion on the subject but here it is anyway. Given the
fluorescence there is obviously some change in the photochemical
behaviour and that is the only possibility I see for a mechanism. The
methods talk about injecting within a window quite short compared to
the run time for the hplc work. One has to assume that no serious loss
of analyte is happening on the column.

All in all, I am somewhat confused by this. I am a geek (but rusty)
organic chemist by training and an analytical chemist by function. I
no longer see any reason to overthink this and have generally accepted
that this is the constraint with OPA derivatives. It still annoys me,
since it is a simple reaction to do.

- Shankar

Loading...