Sporadic inclusion body myositis (IBM) is the most common age-related muscle disease in humans; however, its etiology is unknown, there are few animal models for this disease, and effective treatments have not been identified. Similarities between pathological findings in Alzheimer's disease brain and IBM skeletal muscle include increased levels of amyloid precursor protein (APP) and amyloid β-protein (Aβ). Moreover, there have been suggestions that elevated levels of free cholesterol might participate in the pathogenesis of Alzheimer's disease and IBM due, in part, to its role in Aβ generation. Here, we tested the hypothesis that rabbits fed cholesterol-enriched diets might faithfully exhibit human-like IBM pathological features. In skeletal muscle of one-third of the female rabbits fed cholesterol-enriched diet but not control diet, we found features of IBM, including vacuolated muscle fibers, increased numbers of mononuclear inflammatory cells, increased intramuscular deposition of Aβ, hyperphosphorylated tau, and increased numbers of muscle fibers immunopositive for ubiquitin. The cholesterol-enriched diet increased mRNA and protein levels of APP, increased the protein levels of βAPP cleaving enzyme, and shifted APP processing in favor of Aβ production. Our study has demonstrated that increased ingestion of high levels of dietary cholesterol can result in pathological features that resemble IBM closely and thus may serve as an important new model with which to study this debilitating disorder.
- amyloid precursor protein
- amyloid beta
- Alzheimer's disease
- skeletal muscle
sporadic inclusion body myositis (IBM) is the most common degenerative muscle disease in people over 50 years of age (8, 20). IBM progresses slowly, affects muscles distally and proximally in lower and upper limbs, and can lead to severe pain and disability. Pathological features of muscle biopsies from IBM patients demonstrate the presence of mononuclear inflammatory cells, vacuolated muscle fibers, intracellular deposition of Congo-red-positive β-amyloid protein, and hyperphosphorylated tau in the form of paired helical filaments (1, 11). Obvious from the above pathological features, IBM exhibits in skeletal muscle many similarities to Alzheimer's disease brain.
Although the etiology and pathogenesis of IBM remains ill-understood, a close link to amyloid precursor protein (APP) and amyloid β-protein (Aβ) is apparent (2, 19). Implicating APP in the pathogenesis of IBM are findings that APP protein levels are increased in IBM patients and that overexpression of APP leads to at least some histopathological features of IBM (18, 26). Recent studies by others and us in nonmuscle cells have demonstrated that increased dietary intake of cholesterol can lead to increased Aβ production and conversely that decreased levels of cholesterol can reduce Aβ generation (10, 13, 14, 22, 23). Further implicating cholesterol in APP processing are findings from IBM muscle fibers that accumulated free cholesterol colocalizes with Aβ, hyperphosphorylated tau, and the intracellular transporter of cholesterol caveolin-1 (16).
The present study demonstrates, for the first time, that rabbits fed cholesterol-enriched diets developed IBM-like pathology, including vacuolated muscle fibers, and increases in inflammation, deposition of Aβ, phosphorylated tau, and ubiquitin in muscle fibers. Cholesterol-enriched diet also increased APP expression and shifted APP processing in favor of Aβ production. Our study demonstrates that increased ingestion of dietary cholesterol results in pathological features that closely resemble IBM. This animal model may speed elucidation of underlying mechanisms for IBM as well as the identification of potential therapeutic interventions against IBM.
MATERIALS AND METHODS
Animals and treatment.
New Zealand White female rabbits (2 yr old, weighing 3–4 kg) were used in the present studies. Animals were randomly assigned to two groups; group 1 was fed normal chow (n = 6), and group 2 was fed chow supplemented with 2% cholesterol (n = 6). Later (12 wk), rabbits were killed with 1-ml intravenous injections of euthasol. At necropsy, animals were perfused with Dulbecco's PBS at 37°C, muscles from forelimb (triceps) were dissected, and separate sections were taken for immunohistochemical, immunoblotting, and RT-PCR experiments. All tissues were frozen immediately on a liquid nitrogen-cooled surface and stored at −80°C until taken for analysis. All animal procedures were carried out in accordance with the U.S. Public Health Service Policy on the Humane Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the University of North Dakota.
Skeletal muscle was homogenized mechanically in tissue protein extraction reagent extraction buffer (purchased from Pierce and contains a proprietary detergent in 25 mM bicine and 150 mM sodium chloride at a pH of 7.6.) at a ratio of 1:20 (wt/vol) in the presence of a protease inhibitor cocktail (Sigma) and phosphatase inhibitors (5 mmol/l sodium fluoride and 50 μmol/l sodium orthovanadate). The detergent-soluble fraction was isolated by centrifugation at 100,000 g for 1 h at 4°C. Protein concentration was determined by Bradford assay. Equal amounts of protein (20 μg) from detergent-soluble fractions were resolved by SDS-PAGE under reducing conditions, transferred to polyvinylidene difluoride membranes, and subjected to immunoblotting. APP was probed with mouse monoclonal NH2-terminal APP antibody (1:1,000; Chemicon). C99 fragment of APP was detected using 6E10 monoclonal antibody (1:500; Zymed) or rabbit polyclonal COOH-terminal APP (1:500; Sigma). C83 fragment and APP intracellular domain (AICD) were detected with rabbit polyclonal COOH-terminal APP (1:500, Sigma). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control for all of the immunoblotting experiments.
To detect Aβ, 200 mg of skeletal muscle tissue were mechanically homogenized and suspended in 70% formic acid for 3 h on ice. Formic acid was then evaporated with a Speed Vac, and samples were suspended in 1× sample buffer. Aβ was resolved using 16.5% Tris-Tricine SDS gels, transferred to nitrocellulose membranes, and probed for Aβ with 6E10 (1:500).
Immunohistochemical and histological studies.
Immunohistochemical studies were performed on 10-μm-thick frozen skeletal muscle sections. Aβ was probed with 4G8 (1:500) mouse monoclonal antibody (Zymed). Phosphorylated tau of paired helical fragments was detected using SMI-31 (1:500) mouse monoclonal antibody. Mononuclear inflammatory cells were probed with CD11b (1:500) monoclonal antibody. Ubiquitin was detected using a mouse anti-ubiquitin antibody (1:100; Santa Cruz). Sections were developed with diaminobenzidine substrate using the avidin-biotin horseradish peroxidase system (Vector Laboratories) and counterstained with hematoxylin. Omission of the primary antibody served as a negative control (data not shown). Engel-Gomori trichrome staining was performed to visualize vacuolated muscle fibers as described previously (12). Congo red staining was used for the detection of intramuscular amyloid deposition as described previously (17).
Total RNA was isolated from skeletal muscle using TRIzol reagent (Invitrogen, Carlsbad, CA). First-strand cDNA was synthesized using oligo(dT)12–18 primer according to manufacturer's instructions of the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). The following primers were used: βAPP forward primer 5′-TCTACCCTGAACTGCAG-3′ and βAPP reverse primer 5′-CTGCATGTCTCTTTGGC-3′; GAPDH forward primer 5′-TCTCTCAAGATTGTCAGCAACG-3′ and GAPDH reverse primer 5′-CTTCACAAAGTGGTCATTGAGG-3′. The expected sizes of the PCR products were 250 bp for APP and 454 bp for GAPDH. Thirty cycles of PCR products were separated with electrophoresis in 2% agarose gel, gels were stained with ethidium bromide, and gels were photographed under ultraviolet light.
Visualizing free cholesterol and immunofluorescent staining.
Filipin staining for cholesterol was performed essentially as described by others (16). Briefly, a stock solution containing 2.5 mg/ml filipin complex (Sigma, St. Louis, MO) was prepared in dimethyl sulfoxide and stored in the dark at −20°C. Formalin-fixed frozen muscle sections (10 μm) were first immunostained for Aβ or phosphorylated tau and were then incubated for 1 h at room temperature in the dark with filipin diluted 1:75 (vol/vol) in PBS. Subsequently, sections were rinsed in PBS, covered with cover slips, and examined under fluorescence microscopy. For double fluorescent staining, muscle sections were first stained for Aβ with 4G8 (1:500) or for phosphorylated tau with SMI-31 (1:500) mouse monoclonal antibody and then stained with filipin for cholesterol. Images were collected using a Leica fluorescent microscope.
Total serum cholesterol and high-density lipoprotein (HDL) were measured in venous blood collected from ear veins of rabbits. Lipid levels were measured by standard techniques with an Olympus AU640 clinical analyzer.
All data were expressed as means and SE. Statistical significance was determined by a two-tailed Student's t-test. P < 0.05 was considered to be statistically significant.
Rabbits fed cholesterol-enriched diets exhibited increased serum levels of cholesterol.
Rabbits fed cholesterol-enriched diets have been used extensively to model cardiovascular disorders, including atherosclerosis and more recently to model sporadic Alzheimer's disease (24). Because of findings that cholesterol might play a role in the pathogenesis of sporadic IBM (3), we tested the extent to which increased dietary cholesterol in rabbits induces IBM-like pathology in skeletal muscle and might serve to model this disorder. Female rabbits fed a standard diet enriched with 2% (wt/wt) cholesterol for 12 wk exhibited 10-fold increases in total serum cholesterol concentrations (mean ± SE) from 64 ± 7 mg/dl in controls (n = 6) to 644 ± 106 mg/dl in cholesterol-fed rabbits (n = 6, P < 0.05). Serum levels of HDLs decreased from 34 ± 4 mg/dl in controls to 14 ± 2 mg/dl in cholesterol-fed rabbits (n = 6, P < 0.05).
Rabbits fed cholesterol-enriched diets exhibited pathological features in skeletal muscle of IBM.
Pathological features commonly associated with IBM in skeletal muscle include increased numbers of intramyofibril vacuoles, increased inflammation, infiltration in skeletal muscle fibers of inflammatory cells, increased amyloidosis, hyperphosphorylation of tau, and increased ubiquitin. Accordingly, these features should be present if rabbits fed a cholesterol-enriched diet model faithfully the human disorder. In Engel-Gomori trichrome-stained triceps muscle samples, it was clear that the cholesterol-enriched diet led to gross changes in muscle morphology and induced the formation of intramyofibril vacuoles in two out of the six rabbits (Fig. 1). In contrast, none of these features was present in muscle of control rabbits (Fig. 1). Relatively few mononuclear inflammatory cells were present in control samples (Fig. 2A), as identified immunohistochemically with anti-CD11b antibody, but large numbers of mononuclear inflammation cells were observed in triceps muscle taken from the same two rabbits fed cholesterol-enriched diet (Fig. 2B); obvious as well were findings that mononuclear cells were especially enriched near dysmorphic muscle fibers (Fig. 2B). Congo red staining of intracellularly deposited Aβ was absent in control rabbits (Fig. 2C), but positive staining for Aβ was readily apparent with Congo red (Fig. 2D) and anti-4G8 antibody (Fig. 2F) in rabbits fed cholesterol-enriched diet. In Fig. 2D as in Fig. 2F, numerous inflammatory cells were apparent surrounding the muscle cells. Intracellular deposition of hyperphosphorylated tau as demonstrated immunohistochemically with the anti-SMI-31 monoclonal antibody was at background levels in control rabbits (Fig. 2G) but was increased markedly in muscle from the same two rabbits fed cholesterol-enriched diet (Fig. 2H). Ubiquitin-positive muscle fibers were largely absent in control tissues (Fig. 2I) but were increased in the same two cholesterol-fed rabbits (Fig. 2J). Using adjacent sections, we found that increased expression levels of ubiquitin and hyperphosphorylated tau of paired helical filaments colocalized around intramyofibrillar vacuoles of varying sizes (Fig. 3).
Free cholesterol colocalized with intramuscular deposition of Aβ.
Enrichment of diet with cholesterol can lead to increased accumulation of free cholesterol intracellularly (15, 21). Accordingly, we determined next on a cellular basis using double fluorescent staining the extent to which free cholesterol colocalized with Aβ and phosphorylated tau. As expected, elevated levels of free cholesterol as determined using filipin staining were observed in skeletal muscle from rabbits fed cholesterol-enriched diet (Fig. 4A). When sections were stained with the 4G8 antibody against Aβ, we saw a pattern very similar to that observed with filipin staining, and, when colocalization studies were performed, it was obvious that free cholesterol indeed colocalized with intracellular deposition of Aβ (Fig. 4A), a finding consistent with that of others using muscle biopsies from sporadic IBM patients (16). Similar studies were then performed using the SMI-31 antibody against hyperphosphorylated tau, and, although high levels of free cholesterol as detected with filipin and hyperphosphorylated tau as detected with SMI-31 were observed, colocalization was not apparent (Fig. 4B).
Increased APP expression and Aβ production in rabbits fed cholesterol-enriched diet.
Others and we have shown in nonmuscle cells that high levels of cholesterol increase Aβ production, and cholesterol depletion decreases Aβ generation (10, 13, 14, 22–24). Here, we examined the extent to which rabbits fed cholesterol-enriched diet exhibit altered APP processing and levels of Aβ in skeletal muscle. Immunoblot images and RT-PCR agarose gel images from two control rabbits and two cholesterol-fed rabbits that exhibited extensive pathological features of IBM were shown in Fig. 5. Immunoblot images and RT-PCR agarose gel images from the other four control and four cholesterol-fed rabbits that did not exhibit extensive IBM pathological features were shown in Supplementary Fig. 1 (Supplemental data for this article is available at the American Journal of Physiology: Regulatory, Integrative and Comparative Physiology website.). Summarized and statistically analyzed data from all six control and six cholesterol-fed rabbits were shown in Fig. 5. We found that cholesterol-enriched diet significantly increased levels of APP protein (n = 6, P < 0.01) (Fig. 5A). Cholesterol-enriched diet also significantly increased levels of APP mRNA (n = 6, P < 0.01) (Fig. 5B). In beginning to study possible altered APP processing in these animals, we examined C99 and C83 fragments of APP using an antibody that recognizes the COOH terminal of APP. C99, or CTF-β, is the product of cleavage of βAPP by β-secretase (BACE-1), and C83, or CTF-α, is the product of the nonamyloidogenic α-secretase pathway. We found that cholesterol-enriched diet significantly increased levels of C99 fragments (n = 6, P < 0.01) but did not change levels of C83 fragments (n = 6, P = 0.12) (Fig. 5C). Using 6E10 antibody that recognizes the C99 fragment of APP and 4G8 antibody that recognizes the C83 fragment of APP, we observed similar molecular masses for the fragments and similar changes in the levels of these fragments as identified by the COOH terminal APP antibody in cholesterol-fed rabbit compared with control rabbits (data not shown). Cholesterol-enriched diet also significantly increased the generation of the intracellular domain (AICD) of APP (n = 6, P < 0.01) (Fig. 5C). Next, we examined the protein levels of βAPP cleaving enzyme (BACE-1), which is responsible for the generation of C99 fragments. We found that cholesterol-enriched diet increased protein levels of BACE-1 by ∼60% (n = 6, P = 0.09) (Fig. 5D). Last, we examined the extent to which cholesterol-enriched diet affects Aβ generation. In formic acid extractions of rabbit skeletal muscle from the two animals that exhibited extensive pathological features of IBM, we found marked increases in Aβ generation (Fig. 5E); levels were 0.04 ± 0.01 in controls and 0.30 ± 0.12 (relative units) in cholesterol-fed rabbits (n = 2). Together, these data indicate that cholesterol-enriched diet increases levels of APP and alters its processing to favor Aβ production.
Although the etiology of sporadic IBM is not known, several factors have been implicated in its pathogenesis, including Aβ, free cholesterol, protein misfolding and aggregation, and endoplasmic reticulum stress (1, 3, 4). Increased APP and Aβ may be particularly important factors because intramuscular deposition of Aβ is a pathological hallmark of IBM (5), APP gene transfer to cultured muscle cells induces an IBM phenotype (6, 7), and transgenic mouse models overexpressing APP induce a subset of histological features of IBM (18, 26). An increased accumulation of free intracellular cholesterol has been proposed to have a pathological role in sporadic IBM in part because increased levels of free cholesterol have been shown to increase Aβ production in nonmuscle cells and colocalize with intramuscular depositions of Aβ (10, 16, 22, 23). However, the pathogenesis of sporadic IBM remains ill-understood, animal models for this disease have not been well-characterized, and the role of intracellular accumulation of free cholesterol in sporadic IBM has not been determined.
Here we demonstrate, for the first time, that rabbits fed a diet enriched in cholesterol exhibit most, if not all, of the featured pathological changes of sporadic IBM, including vacuolated muscle fibers, mononuclear cell inflammation, increased deposition of Aβ, hyperphosphorylation of tau, and increased levels of ubiquitin in muscle fibers. The colocalization of increased free cholesterol with intramuscular deposition of Aβ suggests that cholesterol directly affects APP processing and Aβ production. Although rabbits fed diets enriched in cholesterol may model this disease, other factors that increase the incidence of IBM are apparent because not every rabbit fed the high-cholesterol diet developed extensive IBM-like pathology. The influences of gender and/or age are all among the factors that might affect cholesterol diet-induced IBM-like pathology.
The link between Aβ and IBM in this animal model was strengthened further because we found that high dietary cholesterol increased amyloidogenic mechanisms in rabbit skeletal muscle. We found first that high dietary cholesterol increased mRNA and protein levels of APP. Second, we found increased protein levels of C-99, but not C-83, fragments of APP in skeletal muscle from rabbits fed the cholesterol-enriched diet. Third, we found increased protein levels of BACE-1. Together, these findings suggest a shift of APP processing in favor of Aβ production. It is not yet clear how cholesterol increases APP expression and increases amyloidosis. However, it may be related to our findings that cholesterol-enriched diet increases AICD levels and the findings by others that the intracellular domain of APP (AICD) can translocate to the nucleus and induce the expression of its own precursor APP (27). Future studies investigating the detailed mechanism(s) whereby cholesterol-enriched diet shifts APP processing in favoring of Aβ generation in skeletal muscle are warranted.
In humans, total cholesterol levels ≤200 mg/dl and HDL levels ≥60 mg/dl are considered normal and protective (9). However, significant adverse health risks occur when cholesterol levels are ≥240 mg/dl and HDL levels are ≤40 mg/dl (9). In our model, levels of cholesterol in controls were 64 mg/dl and in cholesterol-fed animals were 10 times higher. Thus the levels of cholesterol measured in the cholesterol-fed rabbits, although highly elevated, were not outside of a range measured in humans (25). Accordingly, this model may be indicative of an environmental/dietary cause of IBM.
Our finding that high dietary cholesterol increases APP and Aβ, and induces IBM-like pathology, is consistent with the idea that increased APP metabolism and amyloidosis play key roles in the pathogenesis of sporadic IBM. Having access to an animal model for this sporadic disease should help with the elucidation of molecular mechanisms underlying this disorder and the testing of promising new therapeutic interventions. This latter point is especially important because no effective treatments are available for patients suffering from IBM.
Perspectives and Significance
Intriguing similarities exist between Alzheimer's disease brain and IBM skeletal muscle; both exhibit similar pathological findings, both are mainly sporadic in nature with no known genetic links or causes, and for both there exits a growing body of literature that suggests strongly that elevated levels of free cholesterol are a major pathogenic contributor. Our present findings raise the intriguing possibility that elevated cholesterol levels in humans might contribute to the pathogenesis of sporadic IBM. However, unclear at present are many issues, including the temporal sequence of events in brain and muscle and the extent to which the genesis of the cholesterol diet-induced effects are gender-related and/or age-dependent. For example, elevated cholesterol levels might first affect brain, and the resulting amyloidosis might lead to upper motor neuron death and degeneration of skeletal muscle. Alternatively, skeletal muscle might be affected first and the resulting amyloidosis might lead to increased amyloid deposition in brain and neuron cell death. Clearly, the amounts of cholesterol necessary to cause these changes and the extent to which the cholesterol-induced effects are gender-related and/or age-dependent are important issues because human males are more prone to IBM than are females, and the disease incidence increases with age. Regardless, establishing a model that robustly mirrors human sporadic IBM pathology will undoubtedly help greatly our understanding of the pathogenesis of this disease as well as facilitate the discovery of effective therapeutic interventions.
This work was supported by National Center for Research Resources Grant P20RR-17699.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society