Main

Amyloidosis is a group of diseases caused by structural disorders of proteins whereby normally soluble proteins are deposited in tissues as highly ordered, insoluble amyloid fibrils made up of β-pleated sheets.1 Several serious human diseases are associated with amyloid fibril deposition such as Alzheimer’s disease, type II diabetes, prion disease and familial amyloid polyneuropathy.2, 3 Several apolipoprotein family proteins have been identified as major components of amyloid fibrils. Apolipoprotein serum amyloid A (ApoSAA), apolipoprotein A-I (apoA-I), apolipoprotein A-II (apoA-II) and apolipoprotein A-IV (apoA-IV) are currently considered to be amyloidogenic apolipoproteins. In mice, apoA-II, the second most abundant apolipoprotein in serum high-density lipoprotein (HDL), associates to form amyloid fibrils (AApoAII) and is deposited in the body, though not in brain, in an age-associated manner.4, 5 In laboratory mice, three major alleles (Apoa2a, Apoa2b and Apoa2c) of the apoA-II gene encode three variants of the apoA-II protein: APOA2A with Pro5 and Met26, APOA2B with Pro5 and Val38 and APOA2C with Gln5 and Ala38, respectively.6, 7 Several genetic analyses have indicated that the Apoa2c allele might markedly accelerate age-associated deposition of AApoAII.8, 9, 10

Previously, we have described prion-like transmission of AApoAII amyloidosis in which intravenous, peripheral and oral injection of AApoAII amyloid fibrils markedly accelerated amyloid deposition in young R1.P1-Apoa2c mice.11, 12 We also observed that young R1.P1-Apoa2c mice showed rapid onset of amyloidosis when they share cages with old R1.P1-Apoa2c mice with severe amyloid deposits. Furthermore, offspring nursed by amyloidosis-induced mothers showed early development of amyloidosis.12, 13 The propagation of amyloidosis among mice probably occurred through the consumption of AApoAII fibrils contained in feces and milk. Injection of AApoAII fibrils also induced amyloidosis in less-amyloidogenic mouse strains with Apoa2a or Apoa2b alleles.13, 14 Injection of various kinds of amyloid fibrils into R1.P1-Apoa2c mice also induced amyloidosis.15 These results demonstrated that a common amyloid fibril structure could serve as a seed for amyloid fibril formation (cross-seeding) in vivo. Prion-like transmission was also reported in mouse inflammation-associated amyloid A (AA) amyloidosis.16, 17 Many factors such as aging, primary sequence and mutation of amyloid proteins, genetic background of patients, as well as epigenetic factors, including inflammation, food composition and rearing conditions, may influence fibril formation and deposition in tissues. Transmission of amyloid fibrils from the environment may influence fibril formation as an important epigenetic factor.18

The R1.P1-Apoa2c strain has been a valuable model system for the investigation of genetic and epigenetic factors in amyloid deposition. However, at least 2–3 months are required to evaluate the degree of amyloid deposition even after the injection of AApoAII fibrils. In addition, it has been shown that the efficiency with which amyloid fibrils induce amyloidosis is significantly decreased by cross-seeding compared to self-seeding by homologous amyloid fibrils.19, 20, 21 In fact, the rate of fibril formation is dependent on the concentration of amyloid protein both in vitro and in vivo. To develop an advanced model for investigating the mechanism of amyloidosis, we established a transgenic mouse strain that overexpresses amyloidogenic Apoa2c mRNA.

In this study of mApoa2c transgenic mice, we found that treatment with a very small quantity of amyloid fibrils led to greater amyloid deposition and higher levels of plasma apoA-II compared to non-transgenic mice.

Materials and methods

Establishment of Apoa2c Transgenic Mouse

cDNA of the mouse Apoa2c gene was isolated by reverse transcription-PCR (RT–PCR) of messenger RNA extracted from R1.P1-Apoa2c mouse liver. The pmApoa2c-CAGGS vector DNA containing mouse Apoa2c cDNA inserted between cytomegalovirus (CMV) immediate early gene (CMV-IE) enhancer/chicken β-actin promoter and rabbit β-globin polyA signal22 was microinjected into fertilized eggs of BDF2 mice. Then, the eggs were transplanted into the uteri of pseudo-pregnant ICR mice. Transgenic founders were identified by PCR using primers for Apoa2-1 (5′-AGGAATTCCATCATGAAGCTGCTCGC-3′) and rβGlb-R (5′-TAGCCAGAAGTCAGATGCTC-3′). These primers can distinguish positive and negative mice by amplifying a 495 bp fragment of the transgene (Figure 1a). Since BDF2 mice have a less amyloidogenic Apoa2a allele, the founder mice were crossed with R1.P1-Apoa2c, a congenic strain of mice with Apoa2c from the senescence-accelerated mouse prone strain (SAMP1) on the genetic background of the senescence-accelerated mouse-resistant strain (SAMR1), to make transgenic mice homozygous for the Apoa2c allele (Apoa2c Tg+/+, Apoa2c/c). To minimize the possible effects of the genetic background, transgenic mice were backcrossed with R1.P1-Apoa2c mice for six generations. The Apoa2c allele was identified using allele c-specific primers of A2/acc-F (5′-AAGAGACAGGCGGACGGACA-3′) and A2/acc-R (5′-GAGGTCTTGGCCTTCTCCAC-3′). Seven transgene-positive founders were obtained, but only one founder (I-6) could pass the transgene to the next generation. Thus, all studies were in a line derived from founder I-6.

Figure 1
figure 1

The structures of the pCAGGS-mApoa2c vector and its insertion site. (a) The structure of the pCAGGS-mApoa2c vector, Apoa2c cDNA was inserted between CMV-IE enhancer/chicken β-actin promoter and rabbit β-globin Poly A site. pCAGGS-mApoa2c vector was microinjected into fertilized eggs. It was inserted in chromosome 13 associated with the deletion of genomic DNA (4789 bp) and insertion of IAP and LINE elements. Adjoining genes (Pfkp and Adarb2) are 490 and 1100 kbp away from the integrated site. Two copies of pCAGGS-mApoa2c were integrated in tandem in the transgenic mice. Arrow, the transcription start codon; R180, R230, F2381 and 13-L1-F primers used for chromosomal localization of the transgene; Apoa2-1, rβGlb-R, 13F and R112 primers used for identification of the transgenic mice; cβAct-F and Apoa2-2 primers used for detection of the apoA-II mRNA transcribed from transgene. (b) Identification of the negative (Tg−/−), heterozygous (Tg+/−) and homozygous (Tg+/+) mice for the transgene by PCR. Specific primer pairs (13F and R112, Figure 1a) recognizing inserted transgene and pairs (13F and 13R, Figure 1a) recognizing wild-type genomic sequence were used to distinguish Tg−/−, Tg+/− and Tg+/+ mice.

Full size image

Mice were maintained under specific pathogen free (SPF) conditions at 24±2°C with a light-controlled regimen (12 h light/dark cycle) in the Division of Laboratory Animal Research, Department of Life Science, Research Center for Human and Environment Science, Shinshu University. A commercial diet containing 5.3% fat (MF; Oriental Yeast, Tokyo, Japan) and tap water were available ad libitum. All experiments were performed with the consent of the Animal Care and Use Committee of Shinshu University School of Medicine.

Chromosome Localization of the Integrated Apoa2c cDNA Vector

The location of the inserted Apoa2c cDNA vector in the chromosome was determined by chromosome walking according to the protocol of the DNA Walking SpeedUp Kit (Seegene Inc., Seoul, Korea). We used two PCRs to amplify genomic DNA upstream of the inserted transgene in the transgenic mice. In the first PCR, we used the primers of DW-ACP2 (5′-ACP-TGGTC-3′) and R203 (5′-CCATTGACGTCAATGGAAAGTCC-3′). Cycling conditions were as follows: one cycle at 94°C for 30 s; 30 cycles at 94°C for 40 s, 55°C for 40 s, 72°C for 2 min; and one cycle at 72°C for 7 min. The second PCR was performed on 2 μl of PCR product from the first round of PCR with the second set of primers, DW-ACP-N (5′-ACPN-GGTC-3′) and R180 (5′-CTATTGGCGTTACTATGGGAACATACG-3′) (Figure 1a). Cycling conditions for the second round of PCR were as follows: one cycle at 94°C for 30 s; 30 cycles at 94°C for 40 s, 60°C for 40 s, 72°C for 2 min; and one cycle at 72°C for 7 min. PCR products were resolved on 2% agarose gels. Next, we used the same method to amplify sequences downstream of the inserted cDNA using primers F2381 (5′-GCCTAATGAGTGAGCTAACTCAC-3′) and 13-L1-F (5′-CACCTGAAAAAATGTTCAACATCCTTAG-3′). We determined their location in the chromosome by comparing their sequences with the draft sequence of the mouse genome.

We also performed PCRs using the primers of 13F (5′-GCAAGCAGTGTCCAGTTAGG-3′) and R112 (5′-CCAGGCGGGCCATTTACCGTAAGT-3′) and with the primers of 13F and 13R (5′-CAGGCTCATACTACTTGCAG-3′) to distinguish homozygous, heterozygous and negative mice (Figure 1b).

Analysis of the Expression of apoA-II in the Transgenic Mice

Total RNA was extracted from mouse organs using QIAamp RNA Mini Kits (Qiagen Inc., Valencia, CA, USA) according to manufacturer’s protocol. First strand cDNA was synthesized using ReverTra Ace-á kit (Toyobo, Ltd, Osaka, Japan) and used in PCR. We used primers Apoa2-1 (5′-AGGAATTCCATCATGAAGCTGCTCGC-3′) and Apoa2-2 (5′-AGGAATTCCTCACTTAGCCGCAGGAGC-3′) for RT–PCR of the apoA-II mRNA transcribed from endogenous apoA-II gene and transgene. We used primers cβAct-F (5′-ACTGACCGCGTTACTCCCAC-3′) and Apoa2-2 for the apoA-II mRNA transcribed from transgene (Figure 1a). We carried out RT–PCR for the housekeeping gene β-actin using primers mβAct-F (5′-ACAATGAGCTGCGTGTGGCC-3′) and mβAct-R (5′-CCTCGTAGATGGGCACAGTG-3′).

Following perfusion with PBS, organs were removed from mice and maintained at −80°C until proteins could be extracted for western blot analysis of apoA-II protein.23 Samples were loaded on Tris-Tricine/SDS–16.5% polyacrylamide gels electrophoresis (PAGE) as follows: 50 μg of protein extracted from the testis, liver, kidney, heart, lungs, spleen, tongue and stomach, 100 μg from brain and intestine and 0.1 μl of the plasma.24 After electrophoresis at 15 mA for 6 h, proteins were transferred to a polyvinylidene difluoride membrane using a semidry western blot apparatus (Nihon Eido, Tokyo, Japan) at 150 mA for 1.5 h. The membrane was then reacted with primary antibody solution, either with polyclonal rabbit anti-mouse apoA-I (diluted 1:2000) or apoA-II antisera (diluted 1:4000) in 5% skim milk in PBS containing 0.1% Tween 20 (T-PBS) for 1 h at room temperature with gentle shaking.25 Membranes were incubated for 1 h with horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (Daiichi Pure Chemicals, Tokyo, Japan) solution (1:500). ApoA-I and apoA-II were detected by the enhanced chemiluminescence (ECL) method (Amersham International, Buckinghamshire, UK) and quantitated using a densitometric image analyzer with NIH Images Ver. 1.61.

Plasma Lipids, Lipoprotein Quantity and HDL Particle Size

Plasma was collected from mice that had fasted overnight. Plasma total cholesterol levels were determined by an enzymatic procedure (cholesterol C test, Wako Pure Chemical Industries, Osaka, Japan). HDL cholesterol was estimated according to a modified heparin–manganese precipitation procedure (HDL-cholesterol Test, Wako). To determine the HDL particle size, plasma (3 μl) pre-stained for lipids by Sudan Black B was electrophoresed on a nondenaturing PAGE gel containing a 5–15% linear polyacrylamide gradient.26 Electrophoresis was carried out at 25 mA for 2 h. The distribution of apoA-II protein among the HDL species was determined by western blot analysis of 0.1 μl plasma separated by native PAGE. Cholesterol profile in plasma lipoproteins was analyzed by a dual detection high performance liquid chromatography (HPLC) system with two tandem TSKgel LipopropakXL columns (300 Å∼7.8 mm) according to the method of Usui et al27 (Liposearch System, Skylight Biotech Inc, Akita, Japan).

Isolation of AApoAII Fibrils and Induction of AApoAII Amyloidosis

AApoAII amyloid fibrils were isolated from the liver of an 18-month-old R1.P1-Apoa2c mouse with severe amyloidosis as described by Pras et al.28 The isolated amyloid fibril fraction was further purified by ultracentrifugation as described previously.11 Pellets after ultracentrifugation were resuspended in distilled water and kept at −70°C. For induction of AApoAII amyloidosis, a single dose of 10−8, 10−5, 10−3, 10−1 and 1 μg of sonicated AApoAII amyloid fibrils13 were injected into the tail vein of 2-month-old transgene positive and negative mice. After 1, 2 and 3 months, the treated mice were killed by cardiac puncture under diethyl ether anesthesia, and amyloid deposition was determined.

Detection of Amyloid Deposition

Deposition of amyloid fibrils was identified by polarizing microscopy using Congo Red-stained sections in which green birefringence indicates the presence of amyloid.29 Amyloid fibril protein, AApoAII and AA proteins were identified immunohistochemically using the avidin–biotin HRP complex method with specific antiserum against mouse AApoAII or AA.5 The intensity of the AApoAII amyloid deposition was determined semi-quantitatively using the amyloid index (AI) parameter. The AI parameter represents the average degree of AApoAII deposition, graded 0 to 4, in the seven organs examined (liver, heart, spleen, tongue, stomach, intestine and skin) in Congo Red stained sections.13 Tissues were examined by two independent observers who were blinded to the experimental protocol.

Transmission Electron Microscopy

Aliquots of 5 μl of amyloid fibril fraction isolated from liver of homozygous mouse treated by AApoAII were diluted with 45 μl distilled water, and 20 μl aliquots of the diluted fraction were applied to 400-mesh collodion-coated copper grids (Nissin EM Co. Ltd, Tokyo, Japan) for 1 min and subjected to negative staining with 20 μl of 1% phosphotungstic acid (pH 7.0) for 1 min. The negative stained samples were observed with an electron microscope (1200 EX; JEOL, Tokyo, Japan) operated at 80 kV.

Statistical Analysis

A Statview software package (Abacus Concepts, Berkeley, CA, USA) was used to analyze the data. All data are presented as the mean±s.d. Student’s t test was used for all data except for AI. The Mann–Whitney U-test was used to analyze the AI of AApoAII deposition.

Results

Establishment of the mApoa2c Transgenic Mouse Strain

To study the mechanism of amyloidogenesis and to generate a novel model mouse strain with improved sensitivity to amyloidogenic materials in the environment, the amyloidogenic Apoa2c gene was introduced into the mice. A 320 bp Apoa2c cDNA fragment was amplified and inserted into the vector between the CMV-IE enhancer/chicken β-actin promoter and the rabbit β-globin polyadenylation site. The construct was linearized with SalI and PstI, then microinjected into fertilized eggs. Transgenic offspring were screened by PCR, using primers specific for the transgene spanning Apoa2c gene and rabbit β-globin polyadenylation site (Figure 1a). One line of the founder mice was bred successfully for further studies.

We used the DNA walking PCR procedure with primers annealed to the 5′ and 3′ end of the pCAGGS-mApoa2c vector to localize the integrated chromosomal site. DNA sequencing of the genome surrounding the integrated vector revealed that it integrated near the centromere of chromosome 13 (Figure 1a). In association with integration of the vector, intracisternal A particles (IAP) and long interspersed nuclear elements (LINE) were integrated together in tandem and a 4784 bp genomic segment was deleted. Adjoining genes (Pfkp; phosphofructokinase platelet and Adarb2; adenosine deaminase, RNA-specific, B2) are 490 and 1100 kbp away from the integrated site. The results indicated that homozygous Tg+/+ mice had four additional copies and heterozygous Tg+/– mice had two extra copies of the Apoa2c gene in addition to the two endogenous copies (Figure 1a). We designed primers that recognized deleted chromosomal DNA and integrated vector sequences to distinguish homozygous (Tg+/+), heterozygous (Tg+/–) and wild-type (Tg–/–) mice for the transgene. PCR with these primers clearly identified the Tg–/–, Tg+/– and Tg+/+ mice (Figure 1b). When we mated male and female heterozygous transgenic mice, the birth rates of Tg–/–, Tg+/−, and Tg+/+ mice were 25.7, 52.8 and 21.5%, respectively. All three strains with different transgenes developed normally and there were no marked pathological abnormalities.

Expression of Apoa2c in Tissues of mApoa2c Transgenic Mice

In Tg+/− and Tg+/+ mice, high-level expression of Apoa2c mRNA transcribed from the transgene (transgene Apoa2c mRNA) was found in the testis, liver, brain, kidney, heart, lungs, spleen, stomach and intestine by RT–PCR using transgene specific primers (Figure 2a). In contrast, transgene Apoa2c mRNA was not detected in any of the tissues from transgene negative (Tg−/−) littermates. The total expression of Apoa2c mRNA was determined by RT–PCR with primers recognizing both endogenous and transgene Apoa2c mRNA. High-level expression of Apoa2c mRNA was detected in all tissues of Tg+/− and Tg+/+ mice. However for Tg−/− mice, expression was only detected in the liver, brain, lung and stomach (Figure 2b). The amount of apoA-II protein in tissues was analyzed by western blot analysis (Figure 2d). In Tg−/− mice, apoA-II protein was not detected in brain, stomach or intestine. In Tg+/+ mice, apoA-II was elevated in all tissues except intestine. Furthermore, Tg+/+ mice showed higher expression than Tg+/− mice.

Figure 2
figure 2

Expression of Apoa2c in tissues. (a) The expression of transgene-mediated Apoa2c mRNA was detected by RT–PCR using a specific primers pair (cβAct-F and Apoa2-2; Figure 1a). The expression of Apoa2c mRNA from the transgenic gene was observed in all tissues examined; testis, liver, brain, kidney, heart, lungs, spleen, stomach and intestine in Tg+/− and Tg+/+ mice but not in Tg−/− mice. (b) The expression of total Apoa2c mRNA originated both from endogenous and transgenic Apoa2c genes were detected by RT–PCR using the primer pair (Apoa2-1 and Apoa2-2; Figure 1a). Endogenous expression of Apoa2c mRNA was detected in the liver, brain, lungs and stomach in Tg−/− mice. (c) The expression of housekeeping gene β-actin was examined as a control. (d) ApoA-II protein was analyzed by western blot analysis of the homogenate of each tissue. Endogenous apoA-II protein was detected only in the testis, liver, heart, lungs, spleen and tongue in Tg−/− mice. However, apoA-II was detected in all tissues except intestine in Tg+/+ mice.

Full size image

Plasma Apolipoproteins and Lipoproteins of mApoa2c Transgenic Mice

Plasma was collected from 2-month-old Tg−/−, Tg+/− and Tg+/+ mice after fasting overnight. For reference, plasma was collected from C57BL/6J with Apoa2a and SAMR1 with Apoa2b. The concentrations of apoA-I and apoA-II proteins were determined by western blot analysis (Figure 3a). The concentrations of apoA-II were lower in Apoa2a (C57BL/6J) and Apoa2c (Tg−/−) mice. The Apoa2c transgene increased the concentration of apoA-II in Tg+/− and Tg+/+ mice, while apoA-I concentrations were similar between strains. Quantitation of apoA-I and apoA-II levels revealed that the concentration of apoA-II in Tg+/+ mice, was 1.57- and 1.26-fold greater than in Tg−/− and Tg+/− mice, respectively. In contrast, significant changes were not observed in apoA-I concentration (Figure 3b). As a result, the ratio of apoA-II/apoA-I was much higher in Tg+/+ mice. Specifically, the ratio of apoA-II/apoA-I for Tg+/+ mice was 1.65- and 1.36-fold higher than Tg−/− and Tg+/− mice, respectively (Figure 3c). Plasma total cholesterol and HDL-cholesterol levels were also determined (Figure 3d). Total and HDL-cholesterol levels were significantly higher in Tg+/− mice compared to Tg−/− mice and further increased in Tg+/+ mice to the levels of SAMR1 mice. To determine the effect of additional Apoa2c expression on lipoprotein particles, we determined lipoprotein particle size by non-denatured PAGE and by gel permeation HPLC analysis. The predominant form observed on non-denaturing PAGE was HDL3. The sizes of HDL3 particles in C57BL/6J and Tg−/− mice were smaller than those in SAMR1 mice (Figure 3e). Higher concentrations of apoA-II and higher ratios of apoA-II to apoA-I in Tg+/− and Tg+/+ mice did not change the particle sizes of HDL3. The amount of apoA-II was increased in all HDL3, HDL2 and HDL1 particles in Tg+/− and Tg+/+ mice. HPLC analysis confirmed that the quantity of HDL increased but the size of HDL did not change in Tg+/+ mice, but the HDL particle size remained smaller than in SAMR1 mice (Figure 3f).

Figure 3
figure 3

The plasma concentrations of apolipoproteins and cholesterols in the transgenic mice. (a) ApoA-I and apoA-II were detected with specific antisera after SDS–PAGE and western blot of 0.1 μl of mouse plasma. (b) The plasma concentrations of apoA-I and apoA-II were quantitated using a densitometric image analyzer with NIH Images. The number of mice used were: 3, 3, 9, 13 and 9 for C57BL/6J, SAMR1, Tg−/−, Tg+/− and Tg+/+, respectively. (c) The plasma apoA-II/apoA-I ratio was calculated from the plasma concentrations (mg/dl) of apoA-I and apoA-II. (d) The concentrations of total and HDL cholesterol were determined by an enzymatic procedure. (e) HDL particle size was analyzed by native 5–15% PAGE of 3 μl mouse plasma prestained for neutral lipids using Sudan Black B. SAMR1 with Apoa2b allele and C57BL/6J with Apoa2a allele were used as a control for comparison with the transgenic mice (Tg−/−, Tg+/−, and Tg+/+). The distribution of apoA-II protein among the HDL species (HDL1, HDL2 and HDL3) was determined by western blot analysis of 0.1 μl plasma separated by native PAGE. (f) Cholesterol profiles in plasma lipoproteins were analyzed by a dual detection HPLC system. Overexpression of Apoa2c mRNA did not change the particle size of HDL but increased concentration in Tg+/+ mice. Values are means±s.d. (*P<0.05, **P<0.01).

Full size image

AApoAII Amyloid Deposition in mApoa2c Transgenic Mice

To induce amyloidosis, sonicated AApoAII amyloid fibrils were injected intravenously into 2-month-old Tg−/−, Tg+/− and Tg+/+ mice at doses of 10−8, 10−5, 10−3, 10−1, and 1 μg. Amyloid deposition in tissues was determined by Congo Red staining and immunohistochemistry after 2 months. In Tg+/+ mice, injection of 1 μg AApoAII amyloid fibril induced severe amyloid deposition that was observed by green birefringence under polarizing microscopy in the spleen (Figure 4a and d), liver (Figure 4b and e), tongue (Figure 4c and f), testis, kidney, heart, stomach, intestine and skin after 2 months, while no amyloid deposition was observed in the brain. The amyloid deposition stained positively with anti-AApoAII antiserum (Figure 4g–i) but negatively with anti-AA antiserum immunohistochemically (data not shown). To analyze amyloid deposition in tissues, we isolated amyloid fibrils from the liver of Tg+/+ mice treated with AApoAII amyloid fibrils. Western blot analysis identified the protein as AApoAII (Figure 5a). We observed negatively stained amyloid fibrils by transmission electron microscopy (Figure 5b), the fibrils were rigid, nonbranching and ∼10 nm in diameter. We compared the levels of amyloid deposition in the six major organs of spleen, heart, liver, tongue, stomach and intestine using a grading system.13 In transgenic positive mice, significantly higher amyloid depositions were observed in all organs except for the stomach as compared with Tg−/− mice (Figure 6). No amyloid deposition was observed in the brain in any strain. To quantitate the relative sensitivities of mApoa2c transgenic mice to induction of amyloidosis by amyloid fibrils, the AApoAII amyloid fibrils were serially diluted in distilled water and injected into the tail vein of Tg−/−, Tg+/− and Tg+/+ mice at doses ranging from 10−8 to 1 μg. Mice were killed 2 months after the injections. The degree of amyloid deposition in the whole body was compared using the AI in the seven organs (Figure 7a). Small amyloid deposits were found in four of the eight Tg+/+ mice after injection of 10−8 μg AApoAII fibrils; no deposition was detected in Tg−/− and Tg+/− mice. We found amyloid deposition in Tg+/− and Tg+/+ mice and no deposition in Tg−/− mice after injection of 10−5 μg fibrils. When 10−3 μg, 10−1 μg and 1 μg AApoAII fibrils were injected, the degree of amyloid deposition increased in a dose-dependent manner in all Tg−/−, Tg+/− and Tg+/+ mice. However, AI values were significantly different between the strains and always in the following order: Tg−/−<Tg+/−<Tg+/+. Amyloid deposits were observed only in the tongue, spleen, stomach and small intestine at low quantities of injected AApoAII fibrils (10−8 μg in Tg+/+ and 10−3 μg in Tg−/− mice) (Figure 7b). To characterize the rate at which deposition occurred in mApoa2c transgenic mice (Tg+/+), AI was determined 1, 2 and 3 months after injection of 1 μg amyloid fibrils (Figure 7c). The amyloid deposition was observed in all three strains 1 month after injection. The extent of amyloid deposition (AI) further increased 2 and 3 months after injection. AI values were significantly different between strains and always in the following order: Tg−/−<Tg+/−<Tg+/+. Amyloid deposition was found systemically in the liver, heart, lungs, spleen, tongue, stomach, intestine and skin in Tg+/+ mice (Figure 7d). Amyloid deposition was found only in the tongue, heart and stomach in Tg−/− mice 1 month after injection, but found in the liver, heart, lungs, spleen, tongue, stomach, intestine and skin 3 months after injection. The distribution profiles of amyloid in the organs did not show any difference in all Tg−/−, Tg+/− and Tg+/+ mice. Namely, some particular organs showed earlier deposition, always in this order: tongue, intestine and stomach, spleen and heart, liver.

Figure 4
figure 4

Heavy AApoAII amyloid deposition in tissues of transgenic mice. Amyloid deposition was observed in (a), (d), (g) spleen, (b), (e), (h) liver, and (c), (f), (i) tongue from Tg+/+ mice obtained 2 months after injection of 1 μg of AApoAII. (a, b and c), amyloid deposition was identified by red color in Congo stained sections. (d, e and f), heavy amyloid fibril deposition was observed by green birefringence in polarizing microscopy. (g, h and i), AApoAII amyloid deposition was confirmed immunohistochemically with anti-AApoAII antiserum. Optical magnification, × 100 (c, f and i), and × 200 (a, b, d, e, g, and h).

Full size image
Figure 5
figure 5

The identification of amyloid fibrils deposited in tissue. (a) Amyloid fibrils were extracted as water-suspended fraction from the liver of Tg+/+ mice 2 months after 1 μg AApoAII injection and were separated by SDS–PAGE. The gel was stained with Coomassie Brilliant Blue R-250. Molecular weight marker (lane 1), amyloid fibrils of the liver (lane 2). Western blot analysis of amyloid fibrils, detected with anti-AApoAII antiserum, two bands showed monomer and dimer (lane 3). (b) Transmission electron microscopic images showed the negatively stained amyloid fibrils isolated as a water-suspended fraction of the liver of Tg+/+ mice 2 months after AApoAII injection. Amyloid fibrils were rigid, nonbranching, ∼10 nm in diameter and 439 nm in mean length. The scale bar indicates 200 nm.

Full size image
Figure 6
figure 6

Comparison of amyloid deposition in major organs of AApoAII-treated mice. The grade of amyloid deposition in the spleen, heart, liver, tongue, stomach, intestine and brain was determined using Congo red-stained sections.13 The degrees of deposition were significantly different among the Tg−/−, Tg+/− and Tg+/+ mice in all organs except for the stomach (*P<0.05; **P<0.01, Mann–Whitney U-test). No amyloid deposition was observed in the brain.

Full size image
Figure 7
figure 7

Enhanced sensitivity to the induction of amyloidosis by amyloid fibrils in transgenic mice. Amyloid deposition was induced by the injection of AApoAII fibrils. (a) The intensity of AApoAII amyloid deposition was determined semi-quantitatively using the AI parameter, 2 months after female and male Tg−/− (⋄), Tg+/− (○) and Tg+/+ (•) mice received intravenous injections of 10−8, 10−5, 10−3, 10−1 and 1 μg AApoAII amyloid fibrils. The AI parameter represents the average degree of AApoAII deposition in the major seven organs13 (b) Comparison of amyloid deposition in each major organ of 10−8 and 10−3 AApoAII treated Tg−/− and Tg+/+ mice. The first amyloid deposition was observed in the tongue both in Tg−/− and Tg+/+ mice. (c) The intensity of AApoAII amyloid deposition (AI) was determined in female and male Tg−/−, Tg+/− and Tg+/+ mice 1, 2 and 3 months after intravenous injections of 1 μg AApoAII amyloid fibrils. (d) Amyloid deposition in each major organ was compared among Tg−/− and Tg+/+ mice 1 and 3 months after intravenous injections of 1 μg AApoAII amyloid fibrils. Figures in parentheses represent numbers of amyloid positive mice/numbers of mice examined. *P<0.05; **P<0.01, Mann–Whitney U-test.

Full size image

Discussion

In mice, spontaneous senile apoA-II amyloidosis is universally present.30 Severe apoA-II amyloidosis is linked to the Apoa2c allele (Gln5, Ala38). Here, we report the establishment of a new apoA-II transgenic mouse strain that overexpresses the mApoa2c allele. Our long-term goal is to use this model to elucidate metabolic features of the amyloidogenic Apoa2c allele and the pathogenesis of amyloidosis. This strain has much improved sensitivity to induction of amyloidosis by amyloid fibrils.

We have previously showed that the degree of senile apoA-II amyloid deposition in mouse strains varied with the apoA-II allele such that Apoa2c>Apoa2a>Apoa2b.13, 14 Elucidation of the mechanism by which amyloidogenic apoA-IIs (Apoa2c and Apoa2a) induce a low concentration of HDL should shed light on the pathogenesis of amyloidosis (Figure 3). We have previously demonstrated that the mRNA level, synthesis rate, plasma clearance rates and translational efficiencies of apoA-II were similar in SAMP1 (Apoa2c) and SAMR1 (Apoa2b) mice. Impaired secretion was suggested in Apoa2c mice.31, 32 Amyloidogenic apoA-IIs may be retained in the endoplasmic reticulum (ER) by the ER quality control system.33, 34 Overexpression of apoA-II under the CAGGS enhancer/promoter increased plasma apoA-II from 39 mg/dl (Tg−/−) to 61 mg/dl (Tg+/+) resulting in apoA-II concentration 1.57 times higher than in Tg−/− mice. Increased apoA-II was associated with an increased concentration of HDL and total cholesterol. These results were consistent with those observed in C57BL/6J transgenic mice that overexpressed the Apoa2a gene35, 36 and R1.P1-Apoa2c mice that overexpressed the Apoa2b gene.37 Highly expressed apoA-II associated with all species of HDL but mainly with HDL3 particles. The particle size of HDL did not change in mApoa2c transgenic mice, different from mice overexpressing Apoa2b and Apoa2a.

More severe amyloid deposition was observed in the body of Tg+/+ mice compared to Tg−/− mice. Thus, the plasma concentration of amyloid protein contributes to the progress of systemic apoA-II amyloidosis. In our previous study, we reported that SAMP1 mice that were fed a fish oil diet had significantly lower apoA-II concentrations but more AApoAII deposits than those on a butter and safflower oil diet.38 These results suggested altered metabolism or clearance of amyloid protein might be another factor controlling susceptibility towards amyloidosis.39 ApoA-II is synthesized mainly in the liver in mice and secreted to the blood very quickly. Thus, the amount of apoA-II protein in the liver represented by western blot was very low in Tg−/− mice. On the other hand, apoA-II proteins detected in the heart and tongue might be in the blood, whereby perfusion did not wash them away completely. But in Tg+/+ mice, apoA-II mRNA and proteins were expressed in almost all tissues examined. We have previously suggested a possible contribution of apoA-II synthesized in extrahepatic tissues to local amyloid deposition.40 We expected that such extrahepatic synthesis might influence tissue distribution of amyloid deposition after fibril injection. However, notable differences were not observed in tissue distribution of amyloid deposition between Tg+/+ and Tg−/− mice, even though deposition was heavier and earlier in Tg+/+ mice. These results suggest that amyloid protein in blood may be the sole major source of amyloid fibrils in each tissue. In particular, we did not find any amyloid deposits in the brain in spite of the expression of apoA-II mRNA and protein in Tg+/+ mice. We offer three possible explanations: (1) the level of apoA-II protein in the brain was too low to form deposits, (2) the blood–brain barrier prevented the AApoAII fibrils from invading the brain, and (3) other environmental factors such as inflammation were necessary for deposition in the brain.39

We determined the apoA-II transgene insertion site. Homozygous Tg+/+ mice had four extra copies of the Apoa2c gene. The insertion of vector DNA into the centromeric region of chromosome 13 eliminated 4784 bp of genomic DNA accompanied with insertion of transposon elements, IAP and LINE. No genes could be identified in the lost DNA, and the neighboring genes may be too far away to be affected by the CAGGS enhancer/promoter insertion elements. Thus, overexpression of apoA-II under the CAGGS enhancer/promoter is likely to be the cause for the severe amyloid deposition in Tg+/+ mice.

Prion diseases are characterized by abnormal deposits of amyloid fibrils as amyloid plaques which induce neurologic damage.41, 42 Such fibrils, which are self-propagating protein conformations occurring both in vitro and in vivo,43, 44, 45, 46, 47, 48, 49, 50 are thus transmissible. When small nuclei or fibril seeds are present, they dramatically accelerate fibril formation by inducing conformational changes in amyloid protein monomers. Thus, in non-prion amyloidosis, the conformational change induced by preexisting amyloid fibrils might make the disease transmissible.3, 51, 52, 53 Here, we induced apoA-II amyloidosis by injection of AApoAII amyloid fibrils in mApoa2c transgenic mice. Without induction, Tg+/+ and Tg+/− mice did not show spontaneous amyloid deposition until 6 months of age (data not shown). Tg+/+ and Tg+/− mice had increased sensitivity to amyloid fibril injection. Amyloid deposition could be detected in Tg+/+ mice injected with 10 fg of amyloid fibrils. On the other hand, amyloid deposition was detected only in Tg−/− mice injected more than 1 ng of amyloid fibrils (Figure 7a). Amyloid deposition was detected in many organs in Tg+/+ mice 1 month after injection of amyloid fibrils (Figure 7d). Thus the period for detection decreased to less than 1 month compared to Tg−/− mice.54

Since fibrilization of several kinds of amyloid proteins is nucleation-dependent, self- and cross-seeding by amyloid fibrils and amyloid fibril-like materials is thought to be a key factor for the progress of amyloidosis. To elucidate the pathogenesis of amyloidosis and to prevent it, it is necessary to improve the current bio-assay system for detection of amyloid fibrils and amyloidogenic materials in the environment. Ideally, the system should have a higher sensitivity and shorter waiting period than in R1.P1-Apoa2c (Tg−/−) mice. The mechanism by which the injected apoA-II fibrils gain access to the tissue remains to be elucidated. We found that the content of apoA-II in HDL particles did not change after incubation of HDL with sonicated AApoAII amyloid fibrils (data not shown). This suggests that typical or fragmented amyloid fibrils do not associate with HDL particles. We believe amyloid fibrils are too large to associate with HDL particles. In mouse apoA-II amyloidosis following induction by amyloid fibrils, the amyloid first deposits in the tongue, stomach and intestine, then extends to other tissues. A difference in access of the fibrils to different tissues may account for the distribution of the induced amyloidosis. We could not exclude the possibility that HDL and other lipoprotein particles might bind to injected fibrils and contribute to the access of fibrils to certain tissues. The new Apoa2c transgenic mouse strain with high sensitivity to exogenous amyloid fibrils should be helpful to elucidate such issues.

Several conditions might contribute to the progress of amyloidosis: (1) structures or mutations of amyloid proteins,18 (2) levels of the amyloid proteins in the blood or local environment in each tissue, (3) transmission of amyloidogenic materials and (4) other environmental conditions such as inflammation, nutrition and aging. In particular, many tissue factors that facilitate the metabolism, aggregation or stability of amyloid fibrils may contribute to a differential tissue distribution of amyloid deposition. Such factors might include the quality control system of proteins in ER, molecular chaperones preventing protein aggregation, proteoglycans increasing fibril stability in the extracellular space and fibril degradation by proteases. Here, we have established a new transgenic mouse strain of an amyloidogenic apoA-II variant (APOA2C), which showed significantly increased sensitivity to induction of amyloidosis associated with an increased concentration of APOA2C. Mouse apoA-II amyloidosis shares many properties with other types of amyloidosis. This line should be a valuable tool to investigate amyloidosis-associated conditions and factors, and should provide important insights into diseases of protein misfolding.