I think I know why my protein results do not match but, would like some help from the community to increase my confidence or at least lead me in the right direction.

Background

Note* - Technically I am talking about peptides and not intact proteins but use the word protein mostly in my explanation.

I performed protein analysis on a natural product extract using the Pierce™ BCA Protein Assay Kit, Catalog number: 23225. The calibration curve was generated using the albumin standard in the kit. My BCA result = 16% of the material being protein. However, the CofA from the vendor reported Proteins (N X 6.25) = 71%. That's 55% difference in results!

Another result that may be of value is, over 90% of the 16% protein measured, passed through a 3KD (kilo-dalton) molecular weight cut-off filter. 

What I think is happening

I assume the vendor is using the Kjeldahl (or similar) method to determine protein based on total nitrogen since it is reported as (N x 6.25). This natural product extract is chewed up pretty good so most of the nitrogen is not part of a polymerized protein. BCA turns color in the presence of cysteine, tyrosine, and tryptophan in the amino acid sequence. Also, the peptide backbone contributes to color formation. This is important because polymerized protein will activate more color than free amino acids. 

The calibration curve standard (albumin) is different than your material so your result could be biased. I thought about this but, I don't think it would be biased by a magnitude of 55%. Since the peptide backbone creates a color change this minimizes variability due to amino acid sequence differences.

Conclusion

The difference I am seeing is due to polymerized protein versus total nitrogen. Total nitrogen is assumed to have been protein once upon a time before the extraction process.

What's happening?

Thank you in advance for your help!

-Dave