Trying to do post-column derivitization for cycloserine and some amino acid analogs, all primary aliphatic amines. All react well at 40C to yield great fluorescence using 340 excitation and 455 emission per a published method for cycloserine in biological fluids. However, a related compound-cycloserine dimer (see below) does not react. Even when temperature is increased to 60C, the suspect “oxyamine “ compounds don’t appear to react. Following is a structure for oxyamine dimer: 2 beta aminoxy D-alanines combine to form 2 amine bond linkages and a resulting ring structure with extended oxyamines.

H2N-0-CH2-"di-amide bond ring"-CH2-O-NH2

Does the oxyamine structure prevent reaction with OPA. Is there a way around this, or is there a better reagent that would react with all these amine compounds using a post-column system for fluorescence.

The problem might be connected to your reaction buffer system. For this reaction to work, the analytes need to be in the free base form but the oxygen leakage will significantly lower the basicity of this compound. Are you sure that your buffer pH is optimum for this reaction?

Alternatively, this compound might have a very long retention time since it should be di-derivatized. Have you checked for late eluting analytes?

Dear Anonymous;

I have no experience with the specific molecules you are working with. However, I have developed many method using derivative formation of primary and secondary amines. My first impression is that perhaps OPA is not the best choice in this case, the reactions can be very pH dependent and the compounds formed are not very stable.

What I would do in your case is to first try a simpler precolumn approach with FMOC. This reagent almost never fails and works well with fluorescence detectors, and the derivatives are stable for at least several hours or days. Some people argues that the formation of a byproduct, FMOC-OH, is a problem. I really do not find this to be the case since it is easily separated from other products.

Another reagent to try is Fluorescamine. This reacts very fast, forms highly fluorescent products, and it has the special characteristic of the reagent not being fluorescent by itself. This has allowed me in some cases to do the analysis without separation.

I hope these ideas help you. Good luck.

Benjamin

Chris,

The post-column reaction is optimized using a commercially prepared OPA reagent with a high pH maintained in a borate buffer. More aggressive gradients have been run to confirm absence of any late eluters. We're sure of the dimer peak identity in our elution profile based on its formation in degradation profiles of a parent compound and the degradation profiles of the dimer itself. The post-column reaction won't alter that profile, just enhance the sensitivity of the peaks. We do know from stability studies that the dimer slowly degrades in pH 10-11 carbonate buffer to its primary amine(s)by clipping the peptide bonds. Wondering if the high pH post-column buffer may be detrimental to derivative formation even though the cleaved compounds are still primary amines in their own right.

Ben,

Thanks for the inputs. We are trying to use existing HPLC conditions since we have things optimized for resolution and detection of all compounds in a fast 15 minute gradient. Pre-column derivitization requires more method dev. and having to optimize new chromatography. We simply need to enhance sensitivity, so post-column is our preferred choice. I will try to find commercially prepared post-column reagent alternatives to OPA.

Anonymous,

As been mentioned above, it looks like your problem might be due to the instability of the derivative. It's not very well known but this reaction will work with any mercaptan. Surprisingly, the poorest stability is achieved with the recommended reagent: mercaptoethanol. Mercaptoacetic acid, mercaptopropionic acid and mercaptoethanesulfonate all form much more stable adducts. You might want to try one of these in the place of mercaptoethanol.