Reverse phase C18 column will probably be able to differentiate the tetramer from the dimer/monomer if only you are able to run the RP-C18 HPLC at the conditions where these tetrameric/dimeric /monomeric states are present. But it will not be able to differentiate the between the NAD bound or unbound states simply because the large differences of the molecular weight and hence the hydrophobicity of the protein multimers compared to the NAD.

RP for protein samples are often ran at pH 2 (due to the presence of the additive, TFA) and at this pH most of the protein residues protonate, which unfolds the protein structure. If you are trying HPLC at this condition then due to the protein unfolding the NAD should come out from the binding pocket and elute with the flow through (before initiating the gradient for B). So you can run two experiments at this condition among which one will have NAD in flow through (for the protein bound with NAD) and another will not have NAD in the HPLC flow through. Though it's not a direct proof of the NAD bound protein conformation. Here another problem is the instability of NADH at this pH. So after the RP HPLC while you will try the MS to confirm the molecular weight you will end up at a confusing situation due to decomposition of the NADH.

The RP-HPLC is not the best way to try this experiment. I shall try the size exclusion chromatography or ESI-MS or NMR to confirm the bound and unbound structures of the protein.

If I were to answer the questions you listed, my preffered chromatography method would be size exclusion chromatography (prefferably with multi angle light scattering in line, so you can measure the molar masses of the eluting peaks regardless of their elution position) and multiwavelength UV detector (diode array) so you can take UV/Vis spectra for slices of the eluting peaks. The presence of the nucleotide will be obvious from elevated absorbance at 260nm in comparison to regular protein for which A280/A260 is 2 (i.e. twice as much absorbance at 280nm rather than at 260nm). I had performed such experiment and even computed that there was ~ 0.5 nucleotide per a monomer of a protein (i.e. it was partially saturated with the nucleotide). IF the binding is of low affinity, including teh nucleotide in the runnign buffer with ensure that the nucleotide remains associated iwth the polypeptidde.

Perhaps better would be to use a more sensitive techniques like Fluorescence spectroscopy to monitor the bindings.