High-density lipoproteins and their major protein, apolipoprotein A-I (apoA-I), remove excess cellular cholesterol and protect against atherosclerosis. However, in acquired amyloidosis, non-variant full-length apoA-I deposits as fibrils in arteries contributing to atherosclerosis. In hereditary amyloidosis (AApoAI), a potentially fatal disease, N-terminal fragments of variant apoA-I deposit in vital organs and damage them. There is no cure for apoA-I amyloidosis and its structural basis is unknown. Previously, AApoAI mutations were mapped on the crystal structure of the human C-terminally truncated Δ(185-243)apoA-I. The results suggested that the mutation-induced destabilization of the lipid-free protein initiates β-aggregation. Our biophysical studies showed that amyloidogenic mutations G26R, W50R, F71Y and L170P did not necessarily destabilize the native structure, prompting us to search for additional triggers of apoA-I misfolding. We mapped residue segments predicted to promote β-aggregation (termed amyloid hot spots) on the atomic structure of ∆(185-243)apoA-I. The results suggested that perturbed packing of these hot spots, particularly residues 14-22, triggers amyloidosis. This enabled us to propose the first molecular mechanism of apoA-I misfolding. To explore a potential mechanism, we combined structural, stability, dynamics and functional studies of several amyloidogenic mutants and a non-amyloidogenic control, L159R. All mutants reduced structural protection of the segment 14-22, supporting our hypothesis that increased dynamics of this segment triggers AApoAI. The non-amyloidogenic mutant showed helical unfolding near the mutation site indicating susceptibility to proteolysis. We propose that the major factors that make apoA-I amyloidogenic are reduced protection of the major amyloidogenic segments combined with the structural integrity of the four-helix bundle to facilitate protein aggregation. Together, our results suggest that the fate of apoA-I in vivo depends on the balance between its misfolding, proteolysis, and protective protein-lipid interactions. Our structural and bioinformatics analysis of other members of the apolipoprotein family (A-II, A-IV, A-V, B, C-I, C-II, C-III, E, SAA) showed that apolipoproteins’ propensity to form amyloid is rooted in the proteins’ hydrophobicity, which is key to the lipid binding ability. The overlap of functional and pathologic interfaces suggests competition between normal protein function and misfolding. Therefore, increasing apolipoprotein retention on the lipid surface provides a potential therapeutic strategy against amyloidosis.