SELECTIVE RECOVERY OF ANALYTES IN HUMAN PLASMA?

[swheelz]

Dear All

I am carrying out reverse-phase HPLC on human plasma (with dual channel UV detection) to analyse concentrations of retinol, tocopherols and carotenoids.

Everything was going swimmingly for the first 3 batches. Then, without warning or changes in the setup, I suddenly lost most of alpha-tocopherol on the 292nm channel and all of the carotenoids on 452nm. The retinol was fine, the tocopherol acetate internal standard was fine, but the alpha-tocopherol was extremely blunted and the carotenoids (alpha- and beta-carotene, beta-cryptoxanthin, lutein and lycopene) were missing completely. I ran reference standards (fresh) and got the same result.

I am told that this is impossible. I have put through large doses of beta-carotene reference standards on manual injection and do get a tiny peak (well, a small hump anyway). The same happens to a much lesser effect with alpha-tocopherol. We have changed the column (to a guard column to eliminate the possibility of sample retention) and changed the detector bulb to no effect.

I am using a Spherisorb analytical column with Kontron tertiary pump, a degasser and Chromeleon software. Isocratic solvents: methanol, acetonitrile and chloroform with methanol primer.

Suggestions very welcome please!

Best Wishes,

Simon

[Bruce Hamilton]

My guesses would be that either one of your solvents ( extraction, sample prep, mobile phase ) has changed, or that you detector settings have changed. Another possibility, of many, is that you have inadvertently contaiminated your mobile phase with the analytes.

I'm assuming the IS and other peaks still have the same retention time on the column, and have similar repsonse as before. If that's the case, I'd look closely at your sample preparation, and the reagents involved.

I'd also consisder environmental contamination, such as light or heat - but your retinoids are OK, and they would probably fall over as well.

Good luck,

Bruce Hamilton

[swheelz]

I really don't think it is contamination / degradation. It literally changed overnight and both the reference standards and the plasma samples were affected equally, even though they were prepared separately.

The only common item was the tocopherol acetate which was mixed before the first batch and had been used for 3 batches without problem. No one else uses my solutions or has used the machine to mess with the settings or methods.

All preps were done in semi-darkness using amber and cool solvents. Besides retinol is the most affected by degradation and that one's fine!

So, that leaves your suggestion about detector settings. Other than choosing the wavelengths, what sort of other detector settings might cause this effect?

Thanks for your comments.

Simon

[HW Mueller]

Could it be that some of your analytes moved out of the wavelength window?

[swheelz]

The stone in the chloroform jar was corroded for some strange reason (no one has put water in it before you ask!). I've changed the column (Spherisorb ODS2), the stone and the chloroform and things are a lot better, but still not happy about the b-carotene.....

Cleaned the flow cell this morning - no change.

Will post the solution when found.