Have a bioactive fraction - how to go about ?

[Charmchrome]

Hi,
I saw this forum on Linked in and was hoping you could help. I have an awesome bioactive fraction from a plant source that I have extracted using methanol. When I ran the dried methanol extract through TLC with acetone:hexane - the band gave me great activity - checked using bioautography. I need to find out what is in the band. I have a GC, but my technical staff says it has a polar column and wont let me run my sample. I am neither allowed to run it in my HPLC. Basically I am working with an unknown...I want to know what it might be. I have budget but my skill in this field is low. Pl advise.

[Peter Apps]

If this stuff is as good as you say it is at whatever it does (and how you tested it already gives a clue) then then the best advice I can give is that you need to keep discussions about it off the internet until you have secured your intellectual property rights over it. Depending on circumstances the organisation you work for will probably want to have some significant input into that process.

Bear in mind that there are lots of things that kill bugs in silico, but also kill humans just as effectively - these are not promising leads as wonder drugs !

Good luck.

Peter

[lmh]

If your organisation has no interest, then you are doomed, because isolating and identifying natural products isn't a trivial task.

Peter is right about bioassays and things that kill bugs. It is very easy to find yourself purifying a random chemical that washes out of filters (or something like that) and happens to make bugs do something exciting. At the biological level, you need to do blank extractions without the plant to make sure that the same place on the tlc plate doesn't have the activity!

If this is a crude methanol extract it will also probably contain hundreds or thousands of chemicals in widely differing abundances, and even after purification on tlc you might have tens or hundreds of chemicals in the spot. Yes, ideally you might like to do some GC-MS on it, but think what you're going to do when you get multiple peaks? You may need to do multiple steps of pre-purification and bioassay to get to a pure compound (pure means pure, not just that it is now so weak that only a single peak is visible...). The good thing about GC-MS is that you might get a library-match for identity, but libraries aren't perfect, so you would then need to go back and buy/synthesise the matched chemical and check it has the activity. But if it does, you haven't found a novel chemical (although you may have found a novel use for it). If you don't get a good match, you will then be trying to identify the chemical from spectral information. And, of course, it might be something that doesn't work in GC anyway. In fact it almost certainly won't, without derivatisation, because you're already drying your tlc-plate, so it can't be all that volatile??

HPLC won't help you at all, because it will just be a spot on a column instead of a spot on a tlc. LC-MS might, with the best of luck, get you some limited information (e.g. empirical formula from accurate mass), but it can be a very large struggle to identify anything reliably by LC-MS alone.

Which leaves you with purifying enough to identify by NMR.

Natural products are fascinating and important, but you will need extensive technical support if this is to have a good outcome. Find a collaborator with a genuine interest and access to the right resources.