Extraction of samples containing polysorbate for HPLC assay

[lu_pharm]

Hi, friend,

Our samples contain very high Tween 80, and the HPLC result showed interference of peaks of tween 80 in the samples after extraction (we used liquid-liquid extraction). WE used C18 column with ACN/water as mobile phase. We tried different organic solvents to extract samples, but did not solve problem. Anyone has the experience with the extraction of samples containing Tween 80? We are looking for a solution. If it can not be solved, we may need to use some reagent to digest tween 80. Or change a expensive high putiry tween 80 since the interference peaks could come from impurity in the tween 80, but we are not sure. Thanks.


Ze

[Uwe Neue]

Depending on the nature of your analytes, you could potentially work out a solid-phase extraction procedure to get rid of the remaining Tween 80. If your analytes are ionizable compounds, you could use an ion-exchanger, for example.

[HW Mueller]

Some manufacturers (Calbiochem, maybe Pierce, I have seen others which I don´t remember) give extensive info on getting rid of detergents. If the molecular size is quite different from your analyte you can also try size exclusion chrom., ultrafiltration, or dialysis. A multistep chrom. might be a solution as well (I am thinking of using at least two different normal sized columns in series).

[JM]

Apart from the above suggestions , tween 80 components can be saparated from analyte by using ACN/water gradients and can be eluted at end of the run. I have analysed one drug in pure tween 80 using C18 column with ACN/water gradient.



JM

[lu_pharm]

Thank you for all your reply. We are running HPLC for a nonpolr drug using ACN/water gradient. As everyone know, Tween 80 is a mixture and showed multiple peaks. The residual tween 80 in samples showed a peak at very end (nonpolar portion). It is not very big but big enough to interfere the analysis (same retention time as our analyte.

[lu_pharm]

Thank you for all your reply. We are running HPLC for a nonpolr drug using ACN/water gradient. As everyone know, Tween 80 is a mixture and showed multiple peaks. The residual tween 80 in samples showed a peak at very end (nonpolar portion). It is not very big but big enough to interfere the analysis (same retention time as our analyte.

[SIELC_Tech]

Lu_Pharm

what is your drug? Does it have any ionizable groups, because even "non-polar drug" might have some ionizable groups which you can use for your benefits (steroids with acidic or basic groups, gemfibrozil and ibuprophen) to extend (ion-exchange) or reduce retention (ion-exclusion) while keeping the same amount of ACN

[SIELC_Tech]

Sorry forgot to add the link. So if you have ionizable group you can use it to slow down you drug and then elute you TWEEN at 80% ACN:

http://www.sielc.com/compound_209.html (this is 50 mm columnon 150 or 250 you will have enough room to "squeeze in" Tween.

(although THERE IS NO Tween in this application but it shows how you can use ionic properties if you have them)

[lu_pharm]

Thank you for all your reply. We are running HPLC for a nonpolr drug using ACN/water gradient. As everyone know, Tween 80 is a mixture and showed multiple peaks. The residual tween 80 in samples showed a peak at very end (nonpolar portion). It is not very big but big enough to interfere the analysis (same retention time as our analyte.

[lu_pharm]

The drug is paclitaxel from samples in release study with tween 80 is in release medium. No ionic group in paclitaxel.

[HW Mueller]

Assuming that you optimized your chromatography and still get this "last" Tween peak to overlap, you may just pass this paclitaxel/Tween peak onto a second, different column.
It would always help to give some hints as to the structure, mol wt, etc., who knows what paclitaxel is.

[SIELC_Tech]

Dear Hans,

I want to tell you a little secret, but do not tell anybody. There is a great invention called Google-it will give you almost any structure. Wikipedia (wikipedia.com) will give you a lot of structures and information on chemical compounds.

Accordinng to Wikipedia paclitaxel (Taxol) is:

http://en.wikipedia.org/wiki/Taxol

Another source is Merck Index and Aldrich.

regards,

Vlad

[Uwe Neue]

Here are a few suggestions that have a chance of solving the overlap problem:

1. Replace the acetonitrile with methanol. You will need significantly more methanol to get about the same retention.

2. Use a column with an embedded polar group instead of a standard C18. The resolution between paclitaxel and the Tween peak will change.

3. Throw a dash of THF into the mobile phase.

[HW Mueller]

Vlad, have you not followed the discussions on how people can save time of those who want to help them??

[PJ8]

Uwe Neue:1. Replace the acetonitrile with methanol. You will need significantly more methanol to get about the same retention.

I have some data on Isoelutropic strength mobile phases (estimated from a database of lots of compounds, I think from drylab.)

It suggests that 40% MeOH is approximately equal to 33 % MeCN which is approximately equal to 23 % THF.

Of course the exact % differs for each analyte, but this has helped me narrow down my where to start trial and error method development when changing from MeCN to MeOH in the past.

Cheers, Pat.