Amyloidosis-inducing activity of blood cells in mouse AApoAII amyloidosis.

Mouse senile amyloidosis is a disorder in which apolipoprotein A-II (APOA2) deposits as amyloid fibrils (AApoAII) in many organs. We previously reported that AApoAII amyloidosis can be transmitted by feces, milk, saliva and muscle originating from mice with amyloid deposition. In this study, the ability of blood components to transmit amyloidosis was evaluated in our model system. Blood samples were collected from SAMR1.SAMP1-Apoa2c amyloid-laden or amyloidosis-negative mice. The samples were fractionated into plasma, white blood cell (WBC) and red blood cell (RBC) fractions. Portions of each were further separated into soluble and insoluble fractions. These fractions were then injected into recipient mice to determine amyloidosis-induction activities (AIA). The WBC and RBC fractions from amyloid-laden mice but not from amyloidosis-negative mice induced AApoAII amyloid deposition in the recipients. The AIA of WBC fraction could be attributed to AApoAII amyloid fibrils because amyloid fibril-like materials and APOA2 antiserum-reactive proteins were observed in the insoluble fraction of the blood cells. Unexpectedly, the plasma of AApoAII amyloidosis-negative as well as amyloid-laden mice showed AIA, suggesting the presence of substances in mouse plasma other than AApoAII fibrils that could induce amyloid deposition. These results indicated that AApoAII amyloidosis could be transmitted across tissues and between individuals through blood cells.

Introduction

Amyloidosis consists of a group of structural disorders in which normally soluble proteins undergo conformational changes leading to their deposition in tissues as abnormally ordered, insoluble amyloid fibrils [3, 30]. Over 30 different amyloid proteins have been found in human and animal amyloidoses, including amyloid light chain amyloidosis, familial amyloid polyneuropathy and reactive amyloid A amyloidosis [40]. An intriguing feature of the amyloidoses is that they are transmissible. The transmissibility was first demonstrated in prion disease. The transmissible nature of infectious prion protein (PrPSc) has been demonstrated at 4 levels: (1) direct molecular transmission (self-propagation), (2) spreading within tissue, (3) spreading across tissue and (4) from organism to organism [4]. Subsequently, prion-like transmission was reported in other amyloidoses, such as inflammation-associated AA amyloidosis, Alzheimer’s disease and progressive neurodegenerative diseases in mice and other animal species [4, 21, 28, 32, 46].

In order to reduce the prevalence of amyloidoses, it is essential to elucidate the mechanisms of development and progression. Previous studies have demonstrated the importance of blood and its components, including monocytes, platelets, and plasma, as agents for progression and transmission of the diseases. Prion infectivity can reside in the blood of sheep and humans [4, 19]. Furthermore, prions were transmitted by animal blood transfusion prior to the clinical onset of the disease in deer [20, 29]. In mouse amyloidosis models, AA amyloidosis could be transmitted by peripheral blood monocytes [41]. Also, intraperitoneally injected amyloid β (Aβ) fibrils could be detected in blood monocytes in a transgenic mouse model of Alzheimer’s disease [5]. Supposedly, these cells transport Aβ seeds from the periphery to the brain where they induce cerebral Aβ amyloidosis. Blood platelets express the amyloid precursor protein (APP) and enzymatic machinery to process APP proteins into Aβ peptides. Furthermore, platelets can transform soluble Aβ into fibrillar structures, absorb Aβ fibrils and accumulate in amyloid deposits in cerebral vessels [10]. Recently, transmissible spongiform encephalopathy-associated prion protein (PrPTSE) has been demonstrated in extracellular vesicle preparations containing exosomes from the plasma of mice infected with mouse-adapted variant Creutzfeldt-Jakob disease [37]. Still, much remains to be done to understand the molecular mechanism of propagation and transmission of amyloidosis.

Mouse senile amyloidosis is a disorder in which apolipoprotein A-II (APOA2) of plasma high-density lipoprotein deposits as amyloid fibrils (AApoAII) in many organs [18, 42]. In inbred strains of mice, 6 alleles (a, b, c, d, e and f) of the APOA2 gene (Apoa2) have been found [24]. AApoAII amyloidosis is observed frequently in aged mice of various strains with these alleles. During aging, strains with the Apoa2 allele in particular exhibit a high incidence of severe spontaneous systemic amyloid deposits [12, 15, 17, 39].

SAMR1.SAMP1-Apoa2 (R1.P1-Apoa2) mice constitute a congenic strain with the amyloidogenic Apoa2 allele of the SAMP1 strain on the genetic background of SAMR1 mice. This strain exhibits a high incidence of AApoAII amyloidosis, and it has provided a useful system to study the pathogenesis of amyloidosis [16]. We have shown that AApoAII amyloid fibrils act as seeds that induce conversion of the native APOA2 to fibrillar AApoAII assemblies in vitro [33, 38]. In addition, intravenous injection of a small amount of an AApoAII amyloid fibril fraction extracted from various organs of amyloid-laden mice into young R1.P1-Apoa2 mice readily induces AApoAII amyloidosis [13, 26, 47]. Transmission of AApoAII amyloidosis through feces and from mother to babies through milk has also been demonstrated in R1.P1-Apoa2 mice [25, 43]. Thus, mouse AApoAII amyloidosis could be transmitted through a seeding mechanism similar to that of infectious prion diseases and other amyloidoses [36, 43].

In this study, we examined whether amyloid deposition could be induced by blood cells of mice with AApoAII amyloidosis using the R1.P1-Apoa2 mouse model. We demonstrated amyloidosis-inducing activities (AIA) in the white blood cell (WBC) and red blood cell (RBC) fractions of AApoAII amyloid-laden mice.

Materials and Methods

Animals

We used R1.P1-Apoa2 mice that were developed in our laboratory. Apoa2 knockout mice (Apoa2) were purchased from Jackson Laboratories. A SAMR1-Apoa2 (R1-Apoa2) congenic strain was established by introducing the Apoa2 allele into the SAMR1 strain by a standard backcross procedure in our laboratory. We confirmed that APOA2 protein was absent in R1-Apoa2 mice by Western blotting analysis (Supplementary Fig. S1). Mice were maintained at the Division of Animal Research, Research Center for Support of Advanced Science, Shinshu University under specific pathogen-free conditions at 24 ± 2°C with a light-controlled regimen (12-h light/dark cycle). A commercial diet (MF; Oriental Yeast, Tokyo, Japan) and tap water were provided ad libitum. Only female mice were used in this study to avoid AA amyloidosis or other adverse impacts caused by fighting or other aggressive behavior among male mice reared in the same cage. All experiments were performed with the approval of the Committee for Animal Experiments of Shinshu University (Approval No: 250023 and 280014).

Preparation of donor mice with and without AApoAII amyloid deposition

The AApoAII amyloid fibril fraction was an aqueous suspension of fibrils isolated from the livers of 13-month-old R1.P1-Apoa2 mice as described previously [43] (Supplementary Fig. S2). Amyloid fibrils were resuspended at a concentration of 1.0 mg/mL in distilled water (DW). Sonicated amyloid fibrils (100 μg) were then injected into the tail vein of 2-month-old R1.P1-Apoa2 mice to induce AApoAII amyloidosis [13]. Two, 4, 7 and 10 months after the injection, 3 mice were sacrificed by cardiac puncture under isoflurane anesthesia and blood was collected in tubes containing heparin. Three R1.P1-Apoa2 mice without AApoAII injection were sacrificed at 12 months of age. Blood was collected from these mice and used as an AApoAII amyloid-free control. Three R1-Apoa2 mice without AApoAII injection were sacrificed at 2 months of age. Blood collected from these mice was used as APOA2-free material.

Organs were then taken from mice and fixed in 10% neutral buffered formalin, embedded in paraffin and cut into serial 4-μm sections. Amyloid deposition was evaluated in Congo red-stained sections by polarizing microscopy. The degree of AApoAII deposition in 7 organs (heart, liver, spleen, stomach, small intestine, tongue and abdominal skin) was determined using an amyloid score (AS) that was graded from 0 to 4 as described previously [44]: grade 0, no AApoAII deposition; grade 1, a minute amount; grade 2, small amounts; grade 3, a moderate amount and grade 4, extensive AApoAII deposition. The degree of AApoAII deposition in mice was expressed as the amyloid index (AI), which was the average of the AS grades in 7 organs. The identity of the deposited amyloid fibril proteins in organs was verified by immunohistochemical analyses using an avidin-biotin horseradish peroxidase complex method with specific antisera against mouse AApoAII and mouse AA [14].

Separation and processing of plasma, WBC and RBC fractions

Plasma was isolated from whole blood by centrifugation for 20 min at 3,000 × g at 4°C. Blood cell fractions were then separated into the WBC and RBC fractions through use of Lymphocyte M (Mouse) (Cedarlane, Burlington NC) according to the manufacturer’s instructions. The cells in the fractions were enumerated with a cell counting chamber (Erma, Tokyo, Japan). After 3 PBS washing steps, the WBC and RBC fractions did not contain plasma and WBCs were not found in the RBC fractions but some amounts of RBCs were found in the WBC fractions (Supplementary Fig. S3). Although the majority of WBCs in the WBC fraction consisted of lymphocytes and monocytes with only traces of granulocytes but considerable amounts of RBCs (~50%) were contained. Separated blood components were stored at −70°C until use.

For Western blotting and electron microscopic analyses, WBC, RBC and plasma fractions collected from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice were pooled. The protein content of the fractions was determined with the BCA assay (Pierce BCA Protein Assay Kit, Thermo Scientific, Tokyo, Japan). Parts of the pooled components were denatured (1.0 mg/mL) in a solution of 6 M guanidine hydrochloride (GdnHCl), 0.1 M Tris-HCl (pH. 10.0), 50 mM dithiothreitol for 24 h at room temperature. Denatured samples were dialyzed against DW to remove GdnHCl and subjected to freeze-drying (FD-5N, EYELA Co, Tokyo Japan). The freeze-dried samples, corresponding to 20 µL volumes of the original RBC, WBC or plasma fractions, were dissolved in 100 µL PBS just prior to use.

Portions of the pooled RBC and WBC fractions from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice were homogenized in PBS using a Power Masher (Nippi Co, Tokyo, Japan). Homogenates were then subjected to centrifugal fractionation. Homogenates were first centrifuged at 3,000 × g for 5 min at 4°C. Supernatants were transferred to new tubes and the sediment was resuspended in PBS (3KG-P fraction). The supernatants were centrifuged at 100,000 × g for 1 h at 4°C. The supernatants were transferred to new tubes (Sol fraction) and the sediments were resuspended in PBS (100KG-P fraction).

Determination of the AIA of blood components

Samples consisted of WBC, RBC or plasmas fractions (20 µL per fraction) (Table 1), or processed blood components (3KG-P, 100KG-P, Sol and GdnHCl treated fractions) were diluted with PBS to 100 µL, and injected into the tail veins of 3 two-month-old R1.P1-Apoa2 recipient mice. Two months after injection, the recipient mice were sacrificed, and the degree of amyloid deposition (AI) was measured using the methods detailed above. The AIA of each sample was determined as the mean AI value across the 3 recipient mice.

Table 1.

Cell numbers contained in 20 µL WBCs and RBCs suspensions injected into recipient 2-month-old R1.P1-Apoa2 mice

Donor group	Injection
	WBCs (×107)	RBCs (×107)
4 M R1.P1-Apoa2c + AApoAII	0.46 ± 0.10	6.15 ± 1.48
6 M R1.P1-Apoa2c + AApoAII	0.49 ± 0.06	6.53 ± 0.65
9 M R1.P1-Apoa2c + AApoAII	0.49 ± 0.08	4.64 ± 0.70
12 M R1.P1-Apoa2c + AApoAII	0.43 ± 0.09	3.40 ± 0.38
12 M R1.P1-Apoa2c	0.46 ± 0.05	5.22 ± 0.24
2 M SAMR1.Apoa2-/-	0.50 ± 0.07	6.42 ± 1.08

WBCs and RBCs were isolated from 3 mice in each group and cell numbers were calculated. WBCs and RBCs samples (20 µL) were diluted with PBS to 100 µL and injected into the tail vein of 3 recipient R1.P1-Apoa2 mice.

Western blotting and ultrastructural analyses

Western blotting analysis was performed using Tris-Tricine/SDS-16.5% polyacrylamide gels (SDS-PAGE). Proteins in the WBC fraction separated with SDS-PAGE were transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, Hercules, CA). The membrane was incubated with polyclonal rabbit anti-mouse APOA2 antiserum [14] and reacted proteins were detected with the enhanced chemiluminescence method (Amersham Biosciences, Buckinghamshire, England) with X-ray film (Amersham Biosciences). The negatively stained 3KG-P, 100KG-P and Sol fractions of pooled WBC fraction from 9-month-old amyloid-laden R1.P1-Apoa2 mice were observed with a JEM-1400 electron microscope (JEOL, Tokyo, Japan) with an acceleration voltage of 80 kV [46].

Statistical analysis

Statistical analysis was performed with SPSS (Abacus Concepts, Berkley, CA) and R software (The R Development Core Team, Vienna University of Economics and Business, Vienna, Austria). For non-parametric statistical analysis of AI of recipient mice injected with each sample, we applied the Kruskal-Wallis test followed by the Steel-Dwass test for adjusting multiple comparisons.

Results

AApoAII amyloid deposition in donor mice

In order to prepare blood that could harbor AIA, AApoAII amyloidosis was induced in 2-month-old R1.P1-Apoa2 mice by administration of AApoAII amyloid fibrils. The mice were sacrificed at 4, 6, 9 and 12 months of age. The degree of AApoAII amyloid deposition increased with age in each organ. Accordingly, the AI of the mice increased from 1.56 (mean of 3 mice) at the age of 4 months to 2.76 (6 months), 2.95 (9 months) and 3.09 (12 months) (Fig. 1). Amyloid deposition was not observed in any organs in 12-month-old R1.P1-Apoa2 or 2-month-old R1-Apoa2 mice that were not administered AApoAII fibrils. Using immunohistochemical staining with anti-mouse AApoAII and AA antisera, we confirmed that amyloid deposition was AApoAII and not AA, which is a major amyloid deposit associated with inflammation. These results indicated that these mice could be used as blood donors to evaluate their AIA.

Fig. 1.

Age-dependent increase in AApoAII amyloid deposition in donor mice. (A) AI of R1.P1-Apoa2 mice that were injected with AApoAII amyloid fibrils at 2 months of age, and euthanized at the ages of 4, 6, 9 and 12 months (Circle), R1.P1-Apoa2 mice without AApoAII injection (inverted triangle), and R1-Apoa2 mice without AApoAII injection (triangle). Each symbol represents the AI of an individual mouse. Figures in parentheses indicate the numbers of amyloid-laden mice/numbers of mice examined. (B) Representative polarized light microscopic images of Congo red staining (upper panels) and bright-field light microscopic images of immunostaining with anti-APOA2 antiserum (lower panels). These images show the livers of 4 and 9-month-old R1.P1-Apoa2 mice with induction of amyloidosis and 12-month-old R1.P1-Apoa2 without induction. Scale bars = 100 μm.

The WBC and RBC fractions from AApoAII amyloid-laden mice had AIA

Both the WBC and RBC fractions collected from the AApoAII amyloid-laden mice showed AIA. It was expected that the WBC and RBC fractions collected from older mice with more severe AApoAII amyloid deposition would have higher AIA. Contrary to this expectation, the mean AIA values of the WBC fractions collected from mice at the ages of 4 (0.43), 6 (0.57), and 9 months (0.84) were not statistically different. AIA values of the WBC fractions collected from 12-month-old mice (0.35) were significantly lower than those of the WBC fractions from 9-month-old mice (Figs. 2A and C). AIA values of the WBC fractions from AApoAII amyloidosis-negative R1.P1-Apoa2 and R1-Apoa2−/− mice were virtually undetectable, with mean values of 0.02 and 0.00, respectively. Injection of PBS to recipient mice did not induce amyloid deposition (AIA = 0). The AIA values in the WBC fractions of amyloid-laden mice were significantly higher than those in mice without amyloid deposition.

Fig. 2.

Comparison of AIA values of the WBC (A) and RBC (B) fractions collected from R1.P1-Apoa2 mice (n = 3 in each point) with varying severities of AApoAII amyloidosis. Mice were induced amyloidosis at the age of 2 months and were euthanized at the ages of 4, 6, 9 and 12 months. Also, in the absence of induction, 12-month-old R1.P1-Apoa2 mice (control) and 2-month-old R1-Apoa2 (Apoa2) mice were examined (n = 3 in each group). PBS: PBS-injected mice (n=3). The AIA of each donor’s cells was measured and (circle) indicates the mean AIA of recipient mice. Bar indicates mean AIA of 3 donor mice. Significant differences between amyloidosis-positive and -negative mouse groups are indicated (#, P<0.05; ##, P<0.01; ###, P<0.001. Kruskal-Wallis test followed by the Steel-Dwass test). (C) Representative polarized light microscopic images after Congo red staining of the intestines and tongues of recipient mice that were administered the WBC fraction collected from the following donor animals: 4- and 9-month-old R1.P1-Apoa2 mice with induction of amyloidosis and 12-month-old R1.P1-Apoa2 mice without induction. Scale bars = 100 μm.

Mean AIA values of RBC fraction collected from AApoAII amyloid-laden mice at the ages of 4, 6, 9 and 12 months were 0.33, 0.27, 0.45 and 0.35, respectively (Fig. 2B). AIA values of the RBC fraction collected from amyloidosis-negative R1.P1-Apoa2 and R1-Apoa2−/− mice were virtually undetectable, with mean values of 0.00 and 0.05, respectively.

Plasmas from both AApoAII amyloid-laden and amyloidosis-negative mice had AIA

The mean AIA values of plasmas collected from AApoAII amyloid-laden mice at the ages of 4, 6, 9 and 12 months were 0.40, 0.48, 0.67 and 0.49, respectively. There was no significant difference between these values (Figs. 3A and B). Plasma collected from AApoAII amyloidosis-negative R1.P1-Apoa2 and R1-Apoa2−/− mice showed similar AIA values with mean values of 0.46 and 0.66, respectively.

Fig. 3.

Comparison of AIA values of plasmas collected from donor R1.P1-Apoa2 mice (n=3 in each point) with varying severities of AApoAII amyloidosis. Mice were induced amyloidosis at the age of 2 months and were euthanized at the ages of 4, 6, 9 and 12 months. Also, in the absence of induction, 12-month-old R1.P1-Apoa2 mice (control) and 2-month-old R1-Apoa2 (Apoa2) mice were examined (n = 3 in each group). PBS: PBS-injected mice (n=3). The AIA of each donor’s plasma was measured and (circle) indicate the mean AIA of recipient mice. Bar indicates the mean AIA of donor mice. (B) Representative polarized light microscopic images after Congo red staining of the intestines and tongues of recipient mice that were administered plasma collected from the following donor animals: 4- and 9- month-old R1.P1-Apoa2 mice with induction of amyloidosis and 12-month-old R1.P1-Apoa2 mice without induction. Scale bars = 100 μm.

Characterization of amyloidosis-inducing factors in mouse blood cells and plasma

GdnHCl treatment of the WBC, RBC and plasma fractions collected from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice completely abolished their AIA (Table 2).

Table 2.

Amyloidosis-inducing activities (AIA) of denatured WBCs, RBCs and plasma samples

Injection	No. of mice (n)	Mean AIA	No. of positive mice (positive/total)
WBCs	3	0	0/3
RBCs	6	0	0/6
Plasma	6	0	0/6

The denatured samples containing the same amount of protein in 20 µL of the original pooled sample were injected into recipient R1.P1-Apoa2 mice.

The WBC and RBC fractions of 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice were fractionated into insoluble (3KG-P), 100,000 x g sediment (100KG-P) and soluble (Sol) fractions. AIA was detected in the 3KG-P fraction from the WBC and RBC fractions, but not in the soluble fractions from either WBC or RBC fraction (Fig. 4, Supplementary Table S1).

Fig. 4.

AIA of 3 fractions (3KG-P, 100KG-P and Sol) prepared from the pooled WBC and RBC fractions of 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice. Each bar shows the mean AIA of recipient mice in the group.

Amyloid fibril-like structures were observed by electron microscopy in the 3KG-P fraction of the WBC fraction from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice (Fig. 5A). The fibrils from WBC fraction differed in appearance from those in the liver (Supplementary Fig. S2), exhibiting much thinner, smaller and noodle-like fibrils resembling the fibrils from milk [25], muscle [36] and feces [32, 46]. In contrast, such structures were not observed in either the 100kG-P or the Sol fractions from the amyloid-laden mice or any fraction from amyloidosis-negative R1-Apoa2 mice.

Fig. 5.

A transmission electron microscopic image (A) and Western blotting images (B and C) of the 3 fractions of WBC fractions. (A) Amyloid fibril-like structures in the 3KG-P fraction of WBC fraction from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 mice are indicated by arrows. Scale bar = 100 nm. (B and C) Western blotting images with anti-APOA2 antiserum of the 3 fractions of WBC fraction collected from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 (B) and 2-month-old R1-Apoa2 mice (C). 0.5 µg AApoAII in (B and C) and 1 µL plasma in (B) of 2-month-old R1.P1-Apoa2 mouse were used as positive references. The specific bands at ~7 kDa (monomer) and ~15 kDa (dimer) in 3KG-P are indicated by white circles in (B).

We conducted Western blotting analyses of the 3 fractions of WBC fractions from 9-month-old AApoAII amyloid-laden R1.P1-Apoa2 and amyloidosis-negative R1-Apoa2 mice using anti-mouse APOA2 antiserum. The study revealed specific bands of ~7 kDa (monomers) and ~15 kDa (dimers) in the 3KG-P fraction and a band of ~15 kDa in the 100KG-P fraction of WBC fraction isolated from the amyloid-laden mice (Fig. 5B), but not in the 3KG-P fraction from amyloidosis-negative mice (Fig. 5C).

Discussion

In this paper, the ability of blood components to transmit amyloidosis (AIA) was evaluated with the mouse AApoAII amyloidosis model. The study showed that the WBC and RBC fractions of AApoAII amyloid-laden mice harbored AIA, and accordingly could induce AApoAII amyloidosis when transfused into recipient mice. These findings reinforce the notion that blood can transmit amyloidosis.

In previous studies, we demonstrated that AApoAII amyloid fibrils can act as seeds, inducing the conversion of native APOA2 to fibrillar AApoAII assemblies [33, 38]. We also revealed that AApoAII amyloidosis could be induced in young mice by intravenous injection of small amounts of AApoAII amyloid fibril fractions extracted from organs of amyloid-laden mice [13, 26, 47]. In this study, AIA was detected in the insoluble fractions of WBC, RBC and plasma fractions and GdnHCl treatment of the samples abolished the AIA. In addition, amyloid fibril-like structures and anti-APOA2 reactive bands were detected in the insoluble fraction of WBC fraction. These results suggest that AApoAII amyloid fibrils in the WBC and RBC fractions were responsible for their AIA. It seems likely that the injected AApoAII fibrils were taken up by the WBCs and RBCs in circulation, as these cells do not express APOA2 [8] (BIO GPS. http://biogps.org). However, it is known that blood and monocytes transmit prion disease and mouse AA amyloidosis [19, 20, 27, 29, 41]. It was previously shown in mouse AA amyloidosis [34] and transthyretin transgenic mice [6] that monocytes/macrophages phagocytized amyloid fibrils, leading to partial clearance of amyloidosis. It is possible that the phagocytosis of AApoAII amyloid fibrils by monocytes/macrophages contributed to the transmission of amyloidosis when the cells were transfused to recipients. Tissue cells could be damaged by endocytosis of amyloid fibrils in various amyloidoses [22, 35]. Also, Aβ was demonstrated to bind to RBCs [23]. Thus, it is also possible that cell debris and AApoAII amyloid fibrils of damaged cells bound to the WBCs and RBCs that subsequently transmitted amyloidosis. AIA values of the WBC fraction tended to increase until 9 months of age, but decreased significantly in 12-month-old mice (Fig. 2A). The mechanism of the decrease is not clear, but the activity and number of monocyte/macrophages that phagocytize amyloid might decrease with aging along with severe amyloid deposition in the organs of mice [11]. In these experiments, the WBC fractions contained considerable amounts of RBCs, we need further studies to identify which cells are the most responsible for transmission. It remains unclear, however, how the AApoAII amyloid fibrils in or on blood cells interact with precursor APOA2 protein in circulating HDL to convert it to fibrillar AApoAII.

Unexpectedly, AIA was observed in plasma of AApoAII amyloidosis-negative as well as amyloid-laden mice. The identity of the agents with AIA and the mechanisms by which normal plasma induces AApoAII amyloidosis are not apparent. The observation that even plasma of R1-Apoa2 mice, which were deficient in APOA2 production, induced AApoAII amyloidosis in recipient mice indicated that the substance was something other than AApoAII amyloid fibrils. The finding that the AIA of plasma was abolished after GdnHCl treatment suggests that the unidentified substance with AIA in mouse plasma consisted of peptides with an amyloid fibril-like structure or protein aggregates. We have previously reported that injection of fibril fractions extracted from muscles of mice that were free of AApoAII amyloid deposition could induce AApoAII amyloidosis in recipient mice, although these fractions contained no APOA2 [36]. It is known that mouse AApoAII amyloid can be seeded by various heterogeneous amyloid fibrils (cross-seeding) [9, 45]. Many kinds of proteins form amyloid fibril-like structures [7, 31]. There is a possibility that the substances with amyloid fibril-like structures in the plasma cross-seeded with the APOA2. In the mouse AA amyloidosis model, amyloidosis-enhancing activity was demonstrated in the spleen, liver, kidney and heart tissues from normal mice [1, 2]. These observations suggest that substances other than amyloid fibrils could also induce amyloid deposition. Identification of the substances with AIA in plasma awaits further study.

In summary, the data obtained in this study indicated that the WBC and RBC fractions of AApoAII amyloid-laden mice harbored AIA. Still, much remains to be done to fully understand the pathogenesis of AApoAII and other amyloidoses, in particular the molecular mechanism of transmission and propagation of amyloidosis. Mouse AApoAII amyloidosis provides useful guidance in determining the pathogenesis of amyloidosis and developing effective preventive treatments.