In our experience, there have been scattered problems in terms of instrumental analysis with respect to the identification of many carotenoids: LC-MS provides important results in terms of mass, but it usually detects protonation, dehydration, sodium adducts, etc. Thus, the slight difference in the position of a functional group can lead to incorrect results. It is better to use MS/MS if possible and confirm peak retention times, fragment ions, and absorption spectra matching standards at least. If only literature values are used, they may sometimes not match, so it is better to use a standard. If standards are not available, a mixture of the compounds can be obtained from the recombinant E.coli and compared.

The best and safest approach would be to isolate and purify each compound after culture and measure it by high-resolution MS, NMR, or CD spectroscopy. These can be compared to literature values such as Corotenoids and Carotenoid Handbook.Really, mis-identification of carotenoid structures is quite frequent...

Good luck.

Yasuhiro Nishida

Of course it is best to have pure reference standards. In the real world, it is understood that this is not always possible. For a qualitative result, you may be able to make an educated ID using a properly developed HPLC method with MS and DAD detection modes, in-line (compound has chromophores that can be detected AND it must also ionize well under the analysis conditions used). MS/MS would add more information, and another orthogonal dimension of analysis. The quality of the HPLC method is the key as the results will only be as good as the method used.

• Please note that MS (MS/MS) detection is NOT 'universal' and due to differences in instruments, ionization, solution chemistry, mobile phase and/or detector settings, some compounds may not be detected at all (similar to UV/VIS as neither detection mode will 'see' everything).

We try to use as many different orthogonal methods of analysis as possible (combine detection methods which use different chemical or physical properties) with or without standards to come up with a proposed ID. That is what leads to a good QUALITATIVE ID.