Induction of Protein Conformational Change in Mouse Senile Amyloidosis*

Aggregated amyloid fibrils can induce further polymerization of precursor proteins in vitro, thus providing a possible basis for propagation or transmission in the pathogenesis of amyloidoses. Previously, we postulated that the transmission of amyloid fibrils induces conformational changes of endogenous amyloid protein in mouse senile amyloidosis (Xing, Y., Nakamura, A., Chiba, T., Kogishi, K., Matsushita, T., Fu, L., Guo Z., Hosokawa, M., Mori, M., and Higuchi, K. (2001) Lab. Invest.81, 493–499). To further characterize this transmissibility, we injected amyloid fibrils (AApoAII(C)) of amyloidogenic C type apolipoprotein A-II (APOAIIC) intravenously into 2-month-old SAMR1 mice, which have B type apolipoprotein A-II (APOAIIB), and develop few if any amyloid deposits spontaneously. 10 months after amyloid injection, deposits were detected in the tongue, stomach, intestine, lungs, heart, liver, and kidneys. The intensity of deposition increased thereafter, whereas no amyloid was detected in distilled water-injected SAMR1 mice, even after 20 months. The deposited amyloid was composed of endogenous APOAIIB with a different amyloid fibril conformation. The injection of these amyloid fibrils of APOAIIB (AApoAII(B)) induced earlier and more severe amyloidosis in SAMR1 mice than the injection of AApoAII(C) amyloid fibrils. Thus, AApoAII(C) from amyloidogenic mice could induce a conformational change of less amyloidogenic APOAIIB to a different amyloid fibril structure, which could also induce amyloidosis in the less amyloidogenic strain. These results provide important insights into the pathogenesis of amyloid diseases.

Prion, an abnormal form of the host cellular prion protein (PrP C ), 1 is responsible for the transmissible spongiform encephalopathy, including scrapie of sheep, bovine spongiform encephalopathy, and human Creutzfeldt-Jakob disease (2,3). In transmissible spongiform encephalopathy, prion induces the conformational change of PrP C to the prion form PrP Sc and causes a detectable phenotype or disease in the affected individual. Recent studies with yeast have broadened the definition of prion from proteinaceous infectious agent of transmissible spongiform encephalopathy to infectious proteins or proteinbased genetic elements. Like PrP Sc , prions URE3 and PSI in Saccharomyves cerevisiae can induce the conversion of the cytoplasmic proteins ure2p and sup35p to the prion form. This infectious agent propagates in the cytoplasm and is also transferable to other yeast (4 -6).
The term amyloidosis refers to a group of diverse conditions characterized by the extracellular accumulation of fine amyloid fibrils to which normally innocuous soluble proteins polymerize (7). The nucleation-dependent polymerization model is postulated to explain well the kinetics of amyloid fibrilization and conversion of PrP C to PrP Sc (8,9). This model consists of nucleation and extension phases. Preformed amyloid fibrils accelerate conformational changes of many amyloid precursor proteins and result in rapid extension of amyloid fibrils in vitro (10 -12). Although Alzheimer's disease is not known to be infectious like prion disease (13,14), intracerebral injection of brain homogenate from an Alzheimer's patient into marmoset monkeys induced the formation of amyloid ␤ plaques in these primates (15). Injection of synthetic amyloid-like fibrils or modified silk accelerated amyloid protein A (AA) amyloidosis and showed amyloid-enhancing activity, possibly by serving as a seed on which new AA amyloid fibrils could form (16,17). These results suggested that amyloid ␤ and AA amyloidoses may be transmissible under experimental conditions.
We established the Senescence-Accelerated Mouse (SAM) strains as a good model for senile amyloidosis (18). In mouse senile amyloidosis, apolipoprotein A-II (apoA-II), the second most abundant apolipoprotein of serum high density lipoprotein, polymerizes to form amyloid fibrils and deposits systemically but not in the brain and bone (19,20). Three variants of apoA-II (types A, B, and C) with different amino acid substitutions at four positions (positions 5, 20, 26, and 38 from the N terminus) are present in inbred strains of mice (21). Senescence-Accelerated Mouse-resistant 1 (SAMR1) mice with wild-type B apoA-II (APOAIIB, Pro 5 , and Val 38 ) show few, if any, signs of senile amyloidosis. The R1.P1-Apoa2 C mice are a congenic strain of mice that have the amyloidogenic allele of the apoA-II gene from the Senescence-Accelerated Mouse prone 1 (SAMP1) strain on the genetic background of the SAMR1 strain. This strain with the variant C type apoA-II (APOAIIC, Gln 5 , and Ala 38 ) spontaneously exhibits a high incidence of amyloidosis and severe amyloid deposition with aging (22).
In previous studies, we found that intravenous injection of AApoAII(C) fibrils isolated from livers of old R1.P1-Apoa2 C mice markedly accelerated amyloid deposition in young R1.P1-Apoa2 C mice (23). The in vitro experiment also demonstrated that the extension of AApoAII(C) proceeds by the association of APOAIIC to the ends of existing fibrils (24). The nucleation-dependent polymerization model addressed these phenomena. We fed young R1.P1-Apoa2 C mice with AApoAII(C) fibrils or reared the young mice with old severe amyloid-deposited R1.P1-Apoa2 C mice in the same cage. All of the treated mice developed amyloid deposits (1). These results suggested that the oral transmission of amyloid fibrils results in the fibril formation of amyloid proteins in AApoAII amyloidosis and presented a possible pathogenesis of amyloidosis. In this study, we investigated whether exogenous AApoAII fibrils can induce amyloid deposition in less amyloidogenic SAMR1 mice. Our results revealed that exogenous amyloid fibrils can change the less amyloidogenic wild-type apoA-II monomer to amyloid fibrils; these new amyloid fibrils can more easily induce de novo amyloid deposition in SAMR1 mice. The results of this experiment provide further evidence for the hypothesis of transmission in mouse senile amyloidosis.

EXPERIMENTAL PROCEDURES
Animals-R1.P1-Apoa2 C and SAMR1 mice were raised in the Institute for Frontier Medical Science of Kyoto University under conventional conditions at 24 Ϯ 2°C with a light-controlled regimen (12-h light/dark cycle). A commercial diet (CE-2, Nihon CLEA, Tokyo, Japan) and tap water were available ad libitum. Mice were killed by cardiac puncture under diethyl ether anesthesia. All of the experiments were performed with the consent of the Animal Care and Use Committee of both Kyoto University Graduate School of Medicine and Shinshu University School of Medicine.
Isolation and Injection of Amyloid Fibrils-AApoAII(C) amyloid fibril fractions were isolated from the livers of 18 -21-month-old R1.P1-Apoa2 C mice using Pras' method with some modification (25,26). Purified AApoAII(C) fibrils were suspended in distilled water (DW) at a concentration of 1.0 mg/ml and sonicated five times on ice for 30 s at 30-s intervals using a microtip-equipped Astrason ultrasonic processor W-380 (Heat System-Ultrasonics Inc., Farmingdale, NY) (23). A single dose of 0.1 ml of sonicated AApoAII(C) fraction (0.1 mg) was immediately injected into the tail veins of 2-month-old SAMR1 mice. An equal volume of DW was injected to SAMR1 and R1.P1-Apoa2 C mice of the same age. After 1-20 months, the treated mice were sacrificed and their organs were subjected to pathologic or biochemical analysis as described below. The amyloid fractions from the tongue and lungs of each animal were used for biochemical and electron microscopy studies. The amyloid fraction from the severely deposited tongue of one SAMR1 mouse was injected into an additional group of 2-month-old SAMR1 mice. After 1-12 months, these mice were killed and amyloid deposition was determined. We also injected 0.1 mg of sonicated AApoAII(C) into six 18-month-old SAMR1 mice. These mice were killed for pathologic examination 3 months after injection.
Detection of Amyloid Deposition-Half of each organ was fixed in 10% neutral buffered formalin, embedded in paraffin, and cut into serial 4-m sections. The other half was stored frozen for biochemical analysis. The amyloid was identified by green birefringence in Congo Redstained sections under polarizing microscopy. Amyloid deposits were also immunohistochemically stained by the horseradish peroxidaselabeled streptavidin-biotin method (DAKO, Glostrup, Denmark) using anti-apoA-II antiserum as the primary antibody (23) and 3,3 diaminobenzidine as the chromogen. The intensity of the AApoAII amyloid deposition was determined semi-quantatively using the amyloid index (AI) as a parameter. The AI is the average of the degree of AApoAII deposition graded from 0 to 4 in the seven organs examined (liver, spleen, skin, heart, stomach, small intestine, and tongue) in sections stained with Congo Red after immunohistochemical confirmation of the AApoAII deposition (18). The amount of AApoAII deposited in each organ was categorized into the following five grades: grade 0 with no AApoAII found; grade 1 with a minute amount of AApoAII deposits; grade 2 with small amounts of AApoAII deposits only in the periportal areas of the liver, in the perifollicular regions of the spleen, in the interstitial tissues covering Ͻ10% of the areas of both ventricles, atrio-ventriclar septum, and both atria of the heart, only in the glandular portion and squamous glandular junction of the stomach and in Ͻ30% papillary layer of the dermis of the skin; grade 3 with a moderate amount of AApoAII deposits in Ͻ30% of the area of the lobules of the liver, in Ͻ30% of the area of the red pulp of the spleen, in the interstitial tissues covering 10 -30% of the area of the heart muscles, in Ͻ50% lamina propria and submucosa of the squamous epithelia of the stomach, and in 30 -80% of the area of the papillary layer of the dermis of the skin; and grade 4 with extensive AApoAII deposits in 30 -80% of the area of the lobules of the liver, in 30 -80% of the area of the red pulp of the spleen, in the interstitial tissues covering Ͼ30% of the area of the heart muscles, in Ͼ50% lamina propria and submucosa of the squamous epithelia of the stomach, and in almost all parts of the papillary layer of the dermis and around the hair follicles and sebaceous glands of the skin. The grading in the small intestine and tongue was described previously (23). Two observers who had no information regarding the examined tissue graded and averaged the AI independently for each mouse.
Electron Microscopy-10 l of purified amyloid fractions (0.4 mg/ml) and 10 l of 2% phosphotungstic acid (pH 7.0) were mixed. Half of the carbon-coated plastic grid (Ouken, Tokyo, Japan) was immersed in this mixture for 1 s. The negatively stained samples were observed with a JEOL 1200 EX electron microscope (JEOL, Tokyo, Japan) operated at 80 kV. We measured amyloid fibrils if both ends were visible on the prints. Absolute breadth and length were calculated using the calibration bar on the photographs.
Biochemical Analysis-Amyloid fractions were purified from the tongue and lungs of a 17-month-old AApoAII(C)-treated SAMR1 mouse with severe amyloid deposition. The isolated amyloid fraction was applied to Tris-Tricine SDS-PAGE gels (16.5% (w/v)) (27) and stained with Coomassie Brilliant Blue R-250 (ICN Biomedicals, Aurora, Ohio). Proteins were blotted onto immunoblot polyvinylidene difluoride membranes (Bio-Rad). ApoA-II on the blots was detected with anti-apoA-II antiserum (1:4000), horseradish peroxidase-linked anti-rabbit IgG antibody (1:3000, New England Biolabs, Beverly, Massachusetts), and the ECL Western blotting analysis system (Amersham Biosciences). To examine the type of apoA-II in the amyloid deposits, a part of each purified amyloid fraction was subjected to preparative two-dimensional PAGE with the Multiphore II system (Amersham Biosciences) essentially as described in the manufacturer's manual. The fraction was soaked into Immobiline Drystrips (pH 3.0, 10 NL, 13 cm, Amersham Biosciences) overnight, and then the strips were processed by isoelectric focusing electrophoresis in the presence of 8 M urea, 0.5% Triton X-100, and 9.7 mM dithiothreitol. After equilibration, the strips were processed by 16.5% Tris-Tricine SDS-PAGE. Separated proteins were blotted onto a polyvinylidene difluoride membrane and immuno-detected as described above. The apoA-II spot located by immunoblotting of a separate gel was excised and digested with 200 l of 25 ng/l pyroglutamine aminopeptidase deblocking buffer (Roche Diagnostics) for 24 h in a 30°C-water bath. The membrane then was air-dried and processed by an automatic peptide sequencer (Applied Biosystems 477 A and 120 A, Foster City, CA).
Statistic Analysis-We used the StatView software package (Abacus Concepts, Berkley, CA) to analyze the data. Because the AI is a nonlinear index, the AI of different groups of mice was compared using the nonparametric Mann-Whitney-U test. The average lengths of amyloid fibrils were compared with the Student's t test.

AApoAII(C) Injection-induced Amyloid Deposition in SAMR1
Mice-At 1, 2, and 3 months after injection of AApoAII(C) or DW to 2-month-old SAMR1 mice, no amyloid deposition was identified in any tissue. At 10 months after AApoAII(C) injection, we observed mild or moderate amounts of amyloid deposition in the lamina propria of the tongue, the lamina propria and submucosa of the small intestine, the glandular portion and squamous glandular junction of the stomach, and the vessel wall and alveolar septa of the lungs. A slight deposition was also observed in the collecting tubules in the papillae of the kidney, around the central vein in the liver, and in the interstitium of the myocardium. Amyloid deposition detected by green birefringence in Congo Red-stained slides under polarizing microscopy was confirmed by immunohistochemistry to be composed mainly of AApoAII. After 10 months, the intensity of the deposition gradually increased (Fig. 1). At 15 months after treatment, we detected severe amyloid deposition in the tongue (Fig. 2, A and C), stomach (Fig. 2, B and D), small intestine, and lungs. We detected mild deposition in the heart and kidneys and slight deposition in the liver, skin, and spleen. In contrast, we did not observe any amyloid deposition in DW-injected SAMR1 mice, even 20 months after treatment. To study the effect of age at the time of injection on amyloid induction, we also injected AApoAII(C) into 18-month-old SAMR1 mice, but no amyloid deposition was apparent in the tissues 3 months after injection.
R1.P1-Apoa2 C mice were injected with 0.1 ml of DW at 2 months of age. Three months after treatment, mild amyloid deposition was seen in only one mouse. At five months after injection, the AI increased obviously (Fig. 1), and at ten months after treatment, severe amyloid deposition was seen systemically but not in the bone and brain parenchyma.
Characterization of AApoAII from AApoAII(C)-treated SAMR1 Mice-We analyzed amyloid fractions from the tongue and lungs of a 17-month-old AApoAII(C)-treated SAMR1 mouse. The biochemical results indicated no differences between these two organs. We applied the amyloid fractions to 16.5% Tris-Tricine SDS-PAGE gels. One major band with a molecular mass of 6.5 kDa and several faint bands with high molecular masses were detected in the Coomassie Brilliant Blue R-250-stained gel (Fig. 3A). Subsequent immunoblot analysis showed three immunoreactive signals with molecular masses of 6.5, 16.2, and 26.4 kDa (Fig. 3A). Judging by their molecular masses and immunoreactivities, the prominent 6.5-kDa polypeptide and the other larger bands were determined most likely to be the monomeric and oligomeric forms of apoA-II, respectively. Immunoblot analysis after two-dimensional PAGE revealed that the oligomeric forms were almost com-pletely depolymerized to the monomeric form. AApoAII monomers formed four isoforms, one major and three minor forms (data not shown). The major apoA-II membrane spot was cut and subjected to sequencing. The result revealed that the fifth amino acid at the N terminus was proline, which specified APOAIIB (Fig. 3B). No contaminant peaks were detected in the FIG. 1. AApoAII injection-induced amyloid deposition in SAMR1 mice. Two-month-old SAMR1 mice were injected with AApoAII(C) (ϩ) and killed at 1 (n ϭ 5), 2 (n ϭ 4), 3 (n ϭ 5), 10 (n ϭ 6), 12 (n ϭ 3), and 15 (n ϭ 4) months after injection. Control mice were injected with DW (E) and killed at 1 (n ϭ 5), 2 (n ϭ 4), 3 (n ϭ 5), 10 (n ϭ 5), 12 (n ϭ 3), 15 (n ϭ 3), and 20 (n ϭ 3) months after injection. Two-month-old SAMR1 mice were injected with AApoAII(B) fibrils (ϫ) and killed at 1 (n ϭ 5), 3 (n ϭ 5), 7 (n ϭ 3), 10 (n ϭ 3), and 12 (n ϭ 3) months after injection. Two-month-old R1.P1-Apoa2 C were injected with DW (‚) and killed at 1 (n ϭ 3), 2 (n ϭ 3), 3 (n ϭ 3), 5 (n ϭ 3), 7 (n ϭ 3), and 10 (n ϭ 4) months after injection. The intensity of amyloid deposition was determined using AI as a parameter. The AI of each mouse was the average of grades in seven organs.  2) analysis of an amyloid fraction purified from the lungs of an AApoAII(C)-treated SAMR1 mouse. One major 6.5-kDa band was detected in the gel. Three anti-apoA-II-reactive bands were detected in the blot with molecular masses corresponding to the apoA-II monomer (6.5 kDa) and its oligomeric forms (16.2 and 26.4 kDa). B, N-terminal amino acid sequencing of AApoAII from AApoAII(C)-treated SAMR1 mice. Because the enzyme removes the first amino acid, only numbers 2-6 N-terminal amino acids are shown here. The fifth amino acid at the N terminus is proline, the same as in the B type apoA-II monomer.
sequencing. These results show that the major component of the deposited amyloid in AApoAII(C)-treated SAMR1 mice was endogenous APOAIIB and not exogenous AApoAII(C).
Electron Microscopy of Amyloid Aggregation-To analyze the ultrastructure of the AApoAII(B) fibrils, we observed the negatively stained lung and tongue amyloid samples from two AApoAII(C)-treated SAMR1 mice with transmission electron microscopy. Both of the samples exhibited amyloid-characteristic straight and unbranched fibril images with 9-nm mean breadths, lacking any very obvious surface features (Fig. 4A).
AApoAII(C) fibrils were also examined for comparison and were ϳ9 nm in diameter (Fig. 4B). The average length of AApoAII(B) measured was 87.58 Ϯ 10.39 nm (50 fibrils), whereas that of the amyloid fibril AApoAII(C) was 439.28 Ϯ 43.38 nm (80 fibrils). The former fibrils are shorter than the latter fibrils (p Ͻ 0.0001). Compared with AApoAII(B), we could see more obvious helically wound structures in AApoAII(C) fibrils.
AApoAII(B) Injection Induced Amyloid Deposition in SAMR1 Mice-We injected 0.1 mg of AApoAII(B) in solution into 2-month-old SAMR1 mice. Similar to results in AApoAII(C)treated mice, no amyloid deposition was detected at 1 or 3 months after injection. However, at seven months after treatment, we detected severe amyloid deposition in the lamina propria of the tongue and interstitium of the myocardium. We detected moderate amyloid deposition in the glandular portion and squamous glandular junction of the stomach and in the lamina propria and submucosa of the small intestine. Mild amyloid deposits were observed in the extracellular sites around the central veins in the liver. The intensity of the deposition increased after seven months (Fig. 1). The average AI of the AApoAII(B)-treated group at 10 and 12 months after induction was 2.21, whereas that of AApoAII(C)-treated mice at 10 and 12 months after injection was 1.36 (p ϭ 0.0018). We compared the average level of amyloid in five major organs with deposits in tongue, heart, stomach, small intestine, and liver at 10 and 12 months after AApoAII injection (Fig. 5A). In AApoAII(B)-treated 12-and 14-month-old SAMR1 mice, intense amyloid plaques were detected in the heart such that the plaques occupied and destroyed the histologic structure (Fig.  5B). The average AI of the heart was 3.67, which is significantly higher than that of hearts from AApoAII(C)-injected mice of the same age (AI ϭ 1.22, p ϭ 0.0022). Although, in general, the liver did not contain extensive deposits of amyloid, more amyloid deposited in the livers of AApoAII(B)-treated SAMR1 mice (AI ϭ 1.17) than in those of AApoAII(C)-injected mice (AI ϭ 0.11, p ϭ 0.0157).

DISCUSSION
In this study, we found that AApoAII(C) administration causes a conformational change of less amyloidogenic APOAIIB to amyloid fibrils AApoAII(B) in vivo. AApoAII(B) has the structure of an amyloid fibril, and when injected, it can further induce earlier and more severe amyloidosis in SAMR1 mice than AApoAII(C). The histologic results revealed that the heart and liver also contained extensive amyloid deposition in AApoAII(B) fibril-treated SAMR1 mice. We attribute the different effects of these two AApoAII fibrils to the different conformation of the fibrils, although the APOAIIB and APOAIIC monomers differ by only two amino acids. Two observations support this conclusion. First, electron microscopy revealed that the fibrils of AApoAII(B) are shorter and less helically wound than those of AApoAII(C). Because both amyloid fractions were purified by the same method, the structural discrepancy is probably caused by different conformations. Second, an in vitro thioflavine-T binding assay also revealed that AApoAII(B) generates little fluorescence, whereas AApoAII(C) shows high affinity for the fluorescence reagent (data not shown).
Prions result from a conformational change in a protein, making it infectious (28,29). Previous studies indicate the existence of a species barrier in mammalian and yeast prions (30,31). However, recent work demonstrated that the yeast prions URE3 and PSI facilitate the de novo formation of Rnq1 prion aggregates (32). All of these proteins have similar Gln/ Asn-rich domains but different primary structures. This phenomenon suggests that the appearance of prions is enhanced by heterologous prion aggregates and reveals a general principle of prion transmission. Our results suggest a prion-like transmission of a protein with a specific conformation in the pathogenesis of mouse senile amyloidosis. Although APOAIIB and APOAIIC have only two distinct amino acids, they have clearly different amyloidogenicity. R1.P1-Apoa2 C mice with APOAIIC spontaneously develop amyloidosis as young as 5 months of age, whereas SAMR1 mice with APOAIIB have no amyloid deposition, even at 22 months of age. Like yeast prions, the injection of heterologous AApoAII(C) fibrils may provide a nidus on which the first seeds of the different AApoAIIB can form. This new conformational protein can further induce fibril formation in other SAMR1 mice in an experimental situation and is more effective than AApoAII(C), which suggests more efficient propagation of a homologous seed. The in vitro fibril extension experiment showed that there was no amyloid fibril formation 1 week after incubation of the APOAII(B) monomer from high density lipoprotein with AApoAII(C) (data not shown). This may be because of the long incubation time needed for heterologous nucleation of AApoAII(B) or to other as yet uncovered endogenous factors.
The seeding model has also been proposed to explain how injecting mice with synthetic amyloid-like fibrils or modified silk enhances the formation of AA-amyloid fibrils. Furthermore, it has been reported that in patients with heterozygous familial amyloidosis polyneuropathy, cardiac amyloidosis consists of 46.5-50% wild-type transthyretin (TTR) in addition to mutant TTR (33). After liver transplantation, cardiac amyloidosis can be exacerbated, possibly because of enhanced deposition of wild-type TTR on the template of amyloid fibrils of variant TTR (34).
The heart and liver contain more extensive amyloid deposition in AApoAII(B)-treated than in AApoAII(C)-injected SAMR1 mice. This phenomenon was proposed to relate to the distinct primary structure and conformation of the two amyloid fibrils. Tissues may also provide different environments for protein aggregation (35). A similar observation was made for human hereditary amyloidosis. The first described apolipoprotein A-I (apoA-I) variant Gly 26 -Arg is being associated with peripheral neuropathy, peptic ulcers, and nephrotic syndrome (36). A deletion/insertion mutation in exon 4 of the apoA-I gene could cause amyloid hepatopathy (37). Variant Leu 90 -Pro of apoA-I has been found to be associated with cutaneous amyloid deposition and cardiomyopathy (38). Patients with familial amyloidosis polyneuropathy with the variant TTR Pro 52 and TTR Thr 84 but seldom those with TTR Met 30 have substantial cardiac amyloidosis (33). Further study is needed to elucidate this tissue specificity of amyloid deposition.
The amyloid deposition in R1.P1-Apoa2 C mice increases with age, suggesting the existence of age-related amyloidogenic factors. We injected AApoAII(C) fibrils into 18-month-old SAMR1 mice to determine whether the age at inducement will affect the amyloid formation. No amyloid deposition was detected three months after treatment. Unfortunately, the mice died afterward, and we could not continue the study (data not shown).
We must recognize that amyloid fractions generally contain several minor contaminants including apolipoprotein E (39), amyloid P component, proteoglycans, and ubiquitin (40). Apolipoprotein E, amyloid P component, and proteoglycans are speculated to have "chaperone"-like activity in amyloidogenesis (41). Hence, we cannot rule out the possibility that these components synergistically modulate the amyloidogenesis of APOAIIB. Nevertheless, because a guanidine-hydrochloridetreated amyloid fraction could not induce amyloid fibrils in vivo (23), there is no doubt that AApoAII(C) fibrils are essential for amyloidogenesis.
In conclusion, our study demonstrated that like PrP Sc in prion diseases, exogenous variant AApoAII(C) amyloid fibrils could induce a conformational change in the endogenous wildtype, less amyloidogenic protein APOAIIB. AApoAII(B) fibrils possessing a new conformation can further induce amyloidosis in SAMR1 mice. Our work on the experimental transmission of mouse apoA-II amyloidosis will help to elucidate the pathogenesis of amyloidosis including hereditary human apoA-II amyloidosis (42) as well as that of protein-folding diseases.