Most studies of experimental amyloid A protein (AA) amyloidosis in mice have been performed in type A mice with BALB/c as the prototype. In these mice the products of two genes, SAA1 and SAA2, are the major apo-SAA isoforms on high density lipoprotein (HDL). Of these two isoforms, that differ at nine amino acids, only apo-SAA2 is rapidly cleared and deposited as amyloid fibrils. No mouse strain has ever been shown to be completely resistant to amyloid induction. We have found the CE/J mouse strain to be exceedingly resistant to amyloidogenesis. Data indicate that this resistance is not due to a lack of apo-SAA synthesis but rather resides in the unique apo-SAA isoform in this strain. CE/J mice have a single major apo-SAA isoform (pI 6.15) the product of a single gene. This is a hybrid molecule with features of both apo-SAA1 and apo-SAA2, differing from the latter at only six amino acids. When CD studies were performed to explore the structural relationship of this isoform to apo-SAA1 and apo-SAA2, we found that when bound to heparan sulfate proteoglycan the CE/J pI 6.15 isoform fails to undergo the beta-sheet folding typical for apo-SAA2. This evidence suggests that the folding effect of heparan sulfate proteoglycan on apo-SAA2 is important in amyloid formation.