| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Neurology, Caritas St. Elizabeth’s Medical Center, Tufts University School of Medicine, Boston, Massachusetts, USA;
Department of Neurobiology and Behaviour, University of California, Irvine, California, USA;
Whitaker Cardiovascular Institute, Boston University School of Medicine, Boston, Massachusetts, USA; and
Department of Anaesthesia, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, USA
1Correspondence: Department of Neurology, Caritas St. Elizabeth’s Medical Center of Boston, Tufts University School of Medicine, 736 Cambridge St., CBR419, Boston. MA, USA. E-mail: henry.querfurth{at}tufts.edu
ABSTRACT
Intracellular deposition of the ß-amyloid (Aß) peptide is an increasingly recognized pathological hallmark associated with neurodegeneration and muscle wasting in Alzheimer’s disease (AD) and inclusion body myositis (IBM), respectively. Previous reports have implicated dysregulation of ß-amyloid precursor protein (ßAPP) expression in IBM. Accumulation of full-length ßAPP, its various proteolytic derivatives including Aß, and phospho-tau into vacuolated inclusions is an early pathogenic event. We previously reported on a statistical tendency favoring fast twitch fiber involvement in IBM, reminiscent of the tissue specific patterns of misfolded protein deposition seen in neurodegenerative diseases. To test this principle, we generated an animal model in which human wild-type (WT) ßAPP expression was limited to postnatal type II skeletal muscle. Hemizygous transgenic mice harboring increased levels of holoßAPP751 and Aß in skeletal muscle fibers became significantly weaker with age compared with nontransgenic littermates and exhibited typical myopathic features. A subpopulation of dissociated muscle fibers from transgenic mice exhibited a 2-fold increase in resting calcium and membrane depolarization compared with nontransgenic littermates. Taken together, these data indicate that overexpression of human ßAPP in fast twitch skeletal muscle of transgenic mice is sufficient for the development of some features characteristic of IBM, including abnormal tau histochemistry. The increase in resting calcium and depolarization are novel findings, suggesting both a mechanism for the weakness and an avenue for therapeutic intervention in IBM.—Moussa, C. E-H., Fu, Q., Kumar, P., Shtifman, A., Lopez, J. R., Allen, P. D., LaFerla, F., Weinberg, D., Magrane, J., Aprahamian, T., Walsh, K., Rosen, K. M., Querfurth, H. W. Transgenic expression of ß-APP in fast-twitch skeletal muscle leads to calcium dyshomeostasis and IBM-like pathology.
Key Words: ß-amyloid • inclusion body myositis • skeletal muscle
HUMAN INCLUSION BODY MYOSITIS (IBM) is the most common muscle disorder affecting the elderly. Despite intense efforts to understand IBM, its etiology remains unknown. Sporadic IBM (sIBM) is an inflammatory condition, but clinical and pathological features also support a primary degenerative cause (1)
. IBM shares several pathological hallmarks with Alzheimer’s disease (AD). Deposits of nonmutant AD-associated proteins ß-amyloid precursor protein (ßAPP), ß-amyloid (Aß), hyperphosphorylated neurofilaments and Tau, ubiquitin, and various shared chaperones and kinases are thought to play pathological roles in the cognitive decline (2)
and muscular failure (3)
that define sporadic AD and IBM, respectively. An interesting distinction between neurodegeneration in AD and IBM involves the accumulation of fibrillar Aß in extracellular brain parenchyma and intracellular myoplasm, respectively (4)
.
Misfolded mutant gene products appear to attack specific brain regions in the various inherited neurodegenerative diseases. For instance, mutant 
-synuclein causes degeneration of the substantia nigra in PD (5
, 6)
, specific sets of striatal neurons are affected in Huntington’s disease by expansion of polyglutamine repeat within the Huntingtin protein (7
, 8)
, and motor neurons are sensitive to mutant superoxide dismutase (SOD) in amytrophic lateral sclerosis (9
10
11
12
13)
. In Alzheimer’s disease, early degeneration preferentially occurs in the entorhinal cortex in association with increasing amyloid load (14
15
16
17)
. In IBM, evidence points to an excess of ßAPP transcripts (18)
and protein (19)
. Induction of ßAPP overexpression by myotubes in culture can recreate some of the hallmarks of IBM (20
, 21)
. Intracellular inclusions bearing AD-associated proteins are relatively sparse, occurring in scattered, non-necrotic angulated myofibers (22)
. These and other data suggest that clinical muscle weakness arises from a more widespread metabolic defect (23)
. In previous work, we suggested that a more generalized mismetabolism of calcium could be one such defect (24)
. Excessive calcium could result in muscle weakness and arise from an exaggerated release or leak involving ryanodine receptors (24)
, reduced sarcoplasmic, or endoplasmic reticulum calcium ATPase (SERCA) reuptake or influx from Aß-forming pores (25)
.
In two previous reports, transgenic mice that overexpress the C-terminal 100 amino acid (C100) fragment of ßAPP displayed some of the pathology associated with IBM (26
, 27)
. More recently we have shown in humans and in a transgenic mouse line that expresses a non-IBM related mutant ß-APP under the control of a general muscle creatine kinase promoter that type II fibers are to a modest degree more vulnerable to pathological changes (28)
.
To study the hypothesis that ßAPP gene expression, when confined to a specific muscle fiber type, can reproduce both the disease phenotype and defect in calcium homeostasis in vivo, we generated a transgenic mouse in which ßAPP production is restricted to fast-twitch fibers through control by a myosin light chain (MLC) 1/3 promoter/enhancer (29)
. These mice develop myopathological and clinical features resembling those associated with IBM, including colocalization of immunoreactivities to ßAPP and Aß-sequence-containing peptides, and ultrastructural and histopathological changes that indicate myodegeneration and skeletal muscle weakness. In addition, an alteration in muscle membrane potential and dysregulation of calcium homeostasis was discovered.
MATERIALS AND METHODS
Generation of transgenic mice
The complete ßAPP 751 open reading frame (in Bluescript, gift of C. Abraham) was cloned into pMEX-M2-myc3 expression vector downstream of the 800 bp 5' flanking sequence of the myosin light chain (MLC) 1/3 promoter (SacI restriction site) and upstream of the Simian virus 40 polyadenylation sequence and the MLC enhancer flanking the 3' end (gift of Drs C. Neville and N. Rosenthal, Massachusetts General Hospital). The entire expression cassette containing the promoter, the cDNA, and the enhancer can be isolated as a single fragment using one of the infrequent Srf I cutting enzymes engineered on either side of the cassette (Fig. 1
A).
|
Pronuclear stage zygotes were harvested from B6/C3F1 mice (Charles River Laboratories, Cambridge. MA, USA). The linearized expression cassette was injected into the male pronucleus and between 25 and 30 embryos each were implanted per pseudopregnant CD1 foster mother. After delivery of litters, distal tail samples were collected from each offspring and the DNA was extracted in Hot Shot reagent (2.5 mM NaOH, 0.2 mM disodium EDTA) and neutralized in 40 mM Tris-HCl. Transgenic mice were identified using a polymerase chain reaction (PCR) -based assay. To avoid the amplification of endogenous mouse DNA, we designed the sense primer from the rat MLC 1/3 promoter (5'-GCG TGT GTC AAG GTT CTA TTA GGC-3') and antisense strand primer (5'-ACA TCC GCC GTA AAA GAA TG-3') from the KPI domain of human ßAPP 751. Both PCR-derived cDNAs generated from transgenic and control mice genomic DNAs and total genomic DNA digested with EcoR1, were electrophoresed and transferred to a charged nylon membrane. Southern blots were hybridized with a probe generated using the 1.2 kb PCR product derived from the MLC-APP751 cDNA construct (Fig. 1A
) as template and random priming with -32P dATP (Perkin Elmer Life Sciences, Boston, MA, USA).
Immunocytochemical and histological analysis of skeletal muscle
Immunohistochemistry was performed on 10 micron-thick muscle sections. ßAPP 751 was probed with 22C11 (1:800) mouse monoclonal antibody (mAb) (Chemicon International; Temecula, CA, USA). Aß was probed with 6E10 (1:600) mouse mAb (Signet; Dedham, MA, USA). Specific anti-Aß1–42 antibodies included a rabbit polyclonal (1:60) (Chemicon International) and a mouse monoclonal (21F12; 1:60) (gift of Dr. Dennis Selkoe, Brigham and Women’s Hospital. Boston) to probe for Aß1–42 immunoreactivity. Total tau was probed with tau-5 (1:500) mouse mAb (Biosource International, Inc. Camarillo, CA, USA). Immunolabeled specimens were immunoperoxidase stained using a VECTASTAIN avidin-biotin complex (ABC) system (Vector Laboratories, Inc.; Burlingame, CA, USA) and counterstained with hematoxylin. Further histological staining was performed, including routine hematoxylin and eosin (H&E), modified Gomori trichrome, ATPase at pH 4.6, and thioflavin S. Serially sectioned muscles were stained with (1:40) myosin heavy chain (fast fiber specific) antibody (Ab) (Vector Laboratories) and counterstained with hematoxylin. mAb against mouse LCA (CD45, PharMingen, Inc.; San Diego, CA, USA) was used (1:100) as a marker for cellular inflammation.
Immunoblot analysis
Tissues were snap-frozen in liquid nitrogen, powdered using a tissue grind pestle (Kontes; Vineland, NJ, USA), homogenized 1:5 (w:v) in homogenization buffer (150 mM NaCl, 15 mM EDTA, 10 mM Tris·Cl, pH 7.4 and protease cocktail inhibitor; Roche Diagnostics; Mannheim, Germany), and centrifuged at 20,000 g for 20 min at 4°C. To probe for ßAPP, the supernatant was dried by speed vacuum and the pellet was resuspended in 1:5 (w:v) Laemmli sample buffer (2.5% SDS) and analyzed by Western blot using mouse monoclonal 22C11 Ab. To probe for Aß, 400 mg of skeletal muscle tissue was snap-frozen in liquid nitrogen, mechanically homogenized, and suspended in 70% formic acid for 2–3 h on ice. Formic acid was evaporated using speed vacuum and the tissue homogenate was washed (3x) in 0.1 M Tris/HCl (pH 7.4), 2 mM EDTA, and 0.5% SDS in the presence of protease cocktail inhibitor. Aß was immunoprecipitated in 300 µl 1 x STEN buffer (50 mM Tris (pH 7.6), 150 mM NaCl, 2 mM EDTA, 0.2% Nonidet P-40, 0.2% BSA, 20 mM PMSF and protease inhibitor cocktail) using (1:100) rabbit polyclonal R1282 Ab (gift of Dr. Dennis Selkoe, Brigham and Women’s Hospital; Boston, MA, USA) and analyzed by Western blot using 6E10 mouse mAb.
Maximal Force Generation Strength Test
Isometric grip strength was evaluated using a Shimpo digital force gauge (Shimpo Instruments, FGV-1; Itasca, Japan). Prior to testing, the gauge was zeroed and set horizontally in recording mode to measure maximal strength in gram force (g-force). The animal was trained to grasp with either fore- or hind-paw a T-bar or triangular grasping ring, respectively. Using one hand, the animal is grasped about three-fourths of the way up toward the base of the tail and steadily pulled (
1 in/s) away from the ring/T-bar in line with the transducer axis until the grip is broken and maximal force is digitally recorded. Typically 15 recordings were taken for each animal with an intertrial interval long enough to record the data and zero the gauge meter for next trial. All statistical analyses were performed using GraphPad Prism version 4 (GraphPad Software, Inc.; San Diego, CA, USA).
In vivo electromyography (EMG) recordings
Electrically grounded animals were gently restrained while a concentric needle (30G, DCF25, Medtronic, Inc., MN, USA) was inserted into the hamstring to measure individual spontaneous motor unit action potentials (MUAP) using Medtronic’s keypointTM. MUAP quantification was performed through decomposition analysis (30)
. Each MUAP was reviewed blinded to the animal transgene status and the cursors reset manually. Duplicates were detected and removed. Criteria for MUAP selection were rise time: 
0.8 ms, amplitude: 
50 µV.
Membrane potential and resting [Ca2+]i recording
Flexor digitorum brevis (FDB) muscles were dissected out and fibers were enzymatically dissociated. Dissociated fibers were plated onto plastic dishes for microelectrode recording and myoplasmic resting [Ca2+]i and the plasma membrane potentials were recorded simultaneously using double-barreled Ca2+-selective microelectrodes as described previously (24)
.
Electron microscopy
Muscle tissue was fixed in (1:4, v:v) 4% paraformaldehyde-picric acid solution and 25% glutaraldehyde overnight, then washed 3 x in 0.1M cacodylate buffer and osmicated in 1% osmium tetroxide/1.5% potassium ferrocyanide for 3 h, followed by another 3 x wash in distilled water. Samples were next treated with 1% uranyl acetate in maleate buffer for 1 h, washed 3 x in maleate buffer (pH 5.2), then exposed to a graded cold ethanol series up to 100% and ending with a propylene oxide treatment. Samples were embedded in pure plastic and incubated at 60°C for 1–2 days. Blocks were sectioned on a Leica ultracut microtome at 95 nm, picked up onto 100 nm formvar-coated copper grids, and analyzed using a Philips Technai Spirit transmission electron microscope.
Creatine kinase and muscle area measurements
To measure creatine kinase levels, blood was collected from the tail artery and immediately analyzed by an enzymatic rate method using the SYNCHRON LX20 system according to the manufacturer’s protocol (Beckman Coulter, Inc., Fullerton, CA, USA). Whole muscle area was measured by manually tracing the perimeter of mid belly of cross sections using Spot program (Version 3.4, Diagnostic Instruments, Inc., Sterling Heights, MI, USA).
RESULTS
Identification of transgenic mice
Transgene integration was analyzed in total DNA extracted from distal tail samples both by PCR and Southern blot hybridization (Fig. 1B, C
). Transgenic founders were backcrossed to B6/C3F1 mice (Jackson Laboratories; Bar Harbor, ME, USA). One female founder gave rise to a line consisting of at least 5 transgenic males nos. 27 and 28 (DOB 6/1/2002), nos. 10 and 15 (DOB 7/20/03) and no. 19 (DOB 7/25/03). One male founder gave rise to at least one transgenic female no. 24 (DOB 6/12/2002). Transgenic 10, 15, and 19 were backcrossed with WT B6/C3F1. Transgenic mice and control littermates were aged for up to 2.5 years.
Transgene-derived expression of ßAPP in MLC-ßAPP mice
Steady-state levels of transgene-derived human ßAPP were determined by Western blot analysis on total protein extracted from skeletal muscle of transgenic mice. Human ßAPP was detected in skeletal muscle extracts from transgenic mice (nos. 24, 27, and 28) aged 2.5 years (Fig. 1D
). We detected 4- to 5-fold more ßAPP in skeletal muscle homogenates of transgenic mice compared with nontransgenic littermates. The ß-amyloid peptide derived from ß- and 
-secretase mediated ßAPP proteolysis was also detected in total protein extracted from skeletal muscle of transgenic mice (nos. 24, 27, and 28) (Fig. 1E
) compared with nontransgenic littermates. Age-dependent analysis of transgene expression showed a steady presence of ßAPP in transgenic mice as early as 3 months of age compared with age-matched nontransgenic littermates (Fig. 1F
). ßAPP levels remain constant throughout development. Monomeric Aß, on the other hand, accumulated in an age-dependent manner, first detected at 3 months of age. Oligomeric Aß was also increasingly evident in an age-dependent manner (Fig. 1G
). No noticeable difference in ßAPP expression in the brain, liver, kidney, and heart (Fig. 1H
) was detected between transgenic and nontransgenic littermates.
IBM-like histopathology in MLC-ßAPP mice
Serially sectioned tibialis anterior dissected from transgenic mouse no. 10 showed intracellular deposits of immunoreactive ßAPP (Fig. 2
A). The same deposits were immunoreactive for antibodies vs. Aß (Fig 2C
). The proteinaceous aggregates were shown to correspond with large intramyofiber vacuoles easily identified on Gomori trichrome-developed sections (Fig. 2D
). We confirmed ßAPP immunoreactivity in serially sectioned triceps from transgenic mouse no. 27, where ßAPP deposits (Fig. 2E
) were exclusively localized to fast twitch fibers (Fig. 2F
). Subsarcolemmal inclusions were confirmed by Gomori trichrome staining in a cross section of tibialis anterior of transgenic animal 27 (Fig. 2G
). The paler cytosolic stain suggests relative mitochondrial paucity, such as characteristic of type II fibers, which are expected to bear the transgene. We next probed for immunoreactive Aß deposits (Fig. 2H
) in serially sectioned hamstring from animal no. 15 and confirmed that Aß immunoreactivity was localized to fast-twitch muscle fibers. Thus, Aß-bearing myofibers were ATPase 4.6 negative (Fig. 2I
) and were fast myosin heavy chain positive (Fig. 2J
). We probed for Aß1–42-specific immunoreactivity using either of two antibodies, followed by thioflavin S staining, in cross sections of hamstring from transgenic animals 27 (Fig. 2K
) and 28 (Fig. 2M
). Aß1–42-positive fibers were also thioflavin S positive (Fig 2L, N
), suggesting the presence of beta-sheet structures. ßAPP and Aß immunoreactivity were not widely distributed throughout tissue, but tended to affect clusters of fast-twitch muscle fibers within discrete fascicles. H&E staining of hamstring sections from transgenic mouse no. 27 reveals centralized nuclei (Fig. 2O
) compared with the normal peripheral location of nuclei in nontransgenic control littermates (Fig. 2P
).
|
MLC-ßAPP mice exhibit electromyographic abnormalities and muscle weakness.
Transgenic mice did not display any obvious behavioral or movement disorders, and their body weight and life span were not diminished compared with nontransgenic littermates (data not shown). Transgenic mice with increased levels of ßAPP in skeletal muscle were significantly weaker (84.9±1.3 g-force) as early as 6 months of age (Fig. 3
A) in an isometric forelimb strength test compared with nontransgenic control littermates (124.9±3.5 g-force). Transgene harboring animals remained weak and developed palpable muscle atrophy and waddling gait changes by 2.5 years of age (77.1±4.2 g-force) compared with nontransgenic control littermates (133.9±4.8 g-force). Hind limb isometric grip strength (Fig. 3B
) was also decreased by 6 months (77.4±6 g-force) and up to 2.5 years of age (98.5±4.5 g-force) in transgenic compared with nontransgenic control littermates (140.3±4.8 and 159.9±7.9 g-force, respectively). EMG recordings (Fig. 3)
revealed a greater proportion of smaller MUAP in the transgene-affected mice. These were typically small in amplitude and duration (Fig. 3C
, middle) and, in some instances, polyphasic (Fig. 3C
, bottom). Both the mean amplitude (309.4±23) and mean area (179.6±14) of MUAP (Fig. 3D and E
, respectively) were significantly larger in control than in their transgenic littermates (220.6±18 and 135.8±11, respectively). The frequency distribution of MUAP amplitude in transgenic mice was shifted left (65% <200 µV) relative to control (Fig. 3D
). Similarly, the frequency distribution of MUAP area (Fig. 3E
) was shifted left (75% <100 µV.ms) in transgenic compared with nontransgenic littermates.
|
ßAPP overexpression leads to Ca2+ dyshomeostasis and relative membrane depolarization
A decrease in membrane potential (Vm) in muscle fibers from transgenic mice (–71.8±0.7) was observed compared with nontransgenic (–83.9±0.4) littermates (Fig. 3F
). Moreover, this change was associated with a 2-fold increase in resting intracellular calcium [Ca2+]i (356.8±21.5) compared with nontransgenic (176±10) control littermates (Fig. 3F
). We next classified the transgenic fibers in Fig. 3G
into subpopulations [Tg (1) and Tg (2)] based on a Vm cutoff of –80 mV corresponding to the upper limit of the control values. This split is clearly shown to differentiate the Vm of the two groups. The same Tg (1) subpopulation of fibers from transgenic mice exhibited a mean resting [Ca2+]i (Fig. 3H
) not dissimilar to nontransgenic control littermates. Thus, a smaller population of fibers remains unaffected in MLC-ßAPP mice, whereas the majority of fibers in subpopulation [Tg (2)] exhibited relatively depolarized Vm concordant with increased myoplasmic [Ca2+]i (Fig. 3H
). Comparison recordings were carried out on previously reported muscle creatine kinase (MCK) -ßAPP mice, which express a non-IBM pathogenic Swedish double mutation of ßAPP and without specificity for muscle fiber type (31)
. These mice exhibited a comparable increase in resting [Ca2+]i in all fibers tested (Supplemental Fig. 1). Thus, the distribution of calcium abnormalities into two fiber populations [Tg (1) and (2)] observed in our MLC-ßAPP mice is consistent with the expected chimeric gene expression, type II transgene-bearing fibers having the abnormally raised [Ca2+]i.
Ultrastructural changes in the muscle of the MLC-ßAPP mice
Electron microscope analysis of cross sections of hamstring from animal 29 (13 months old) revealed a distortion in Z line clarity, a substantial decrease in mitochondrion number and disrupted junctional triads (Fig 4
A) compared with nontransgenic littermates (Fig. 4B
). A higher magnification view (Fig. 4A
, insert) showed glycogen granule dispersion around abnormal mitochondria. Further examination of cross sections of hamstring from a less affected 6-month-old transgenic animal showed the presence of what may either represent inclusions and/or grossly dilated cisternae (asterisk), as well as tubular aggregates (arrow, Fig. 4C
). Centralized or invading nuclei with early membrane changes were also frequent (Fig. 4D
). A magnified view of a representative vacuolated inclusion, lacking in continuous membrane boundary, from a 13-month transgenic animal revealed the presence of strangely formed multivesicular body (MVB) -like structures and membranous whorls (Fig. 4E
). The content consisted of amorphous material and short 6–10 nm filaments in bundles (Fig. 4F
, insert).
|
Significant hamstring muscle atrophy was noted in transgenic (13.5±1 mm2) compared with nontransgenic (20.1±4.6) littermates (Fig. 4G
). Serum creatine kinase (CK) measurement revealed a 4-fold increase in the serum CK levels in the transgenic (1176.25±379 U/dl) compared with nontransgenic (217.25±116) littermates (Fig. 4H
). This is consistent with 2- to 5-fold increases noted in sIBM patients. Inflammatory cell infiltrates were searched for using anti-CD45 recognizing mouse leukocyte common antigen, but were not found in our transgenic samples (data not shown).
DISCUSSION
The MLC1/3-ßAPP transgenic mouse exhibited many of the clinical and myopathological features characteristic of the human IBM condition. Among these changes, ßAPP and Aß deposition, inclusion formation, centralized nuclei, increased creatine kinase levels, and skeletal muscle weakness with atrophy featured prominently. This is the first demonstration that WT holoßAPP expression in skeletal muscle tissue is sufficient to produce these disease markers in vivo and, as shown in Fig. 1E, G
, to lead to the age-dependent accumulation of Aß monomer and oligomers. The expression of ßAPP, then gradual accumulation of Aß into inclusions in an age-dependent manner, also resemble sIBM, a disorder affecting aged adults, and was orchestrated by the MLC 1/3-promoter/enhancer, which is expressed postnatally (29)
. These results are consistent with culture-based models of ßAPP mismetabolism in muscle (20
, 21)
. Furthermore, the observed increase in resting calcium and relative membrane depolarization in muscle fibers of MLC-ßAPP transgenic mice represents a mechanism relating ßAPP mismetabolism to altered calcium homeostasis and clinical weakness. Abnormalities in calcium metabolism are significant factors in other chronic myopathic conditions, central core disease, and muscular dystrophy (32
33
34
35
36)
. We were able to confirm these physiological findings in the less restricted MCK-ßAPPSwed transgenic mouse of Sugarman et al. (31)
.
WT ßAPP expression was selectively targeted to postnatal fast-twitch skeletal muscle by choosing the MLC 1/3 promoter/enhancer. In previous models, the mutant ßAPP and mutant C99 fragment transgene products were contrived to accelerate Aß genesis (25
, 26
, 31)
. Accordingly, they are removed to a degree from the central ßAPP-sIBM hypothesis (3
, 4)
. Moreover, their pattern of expression was not directed to fiber type or postnatal development as the MLC-ßAPP mouse was.
The reported ultrastructural findings are informative with respect to IBM pathology. Features such as membrane whorls, glycogen granules, incompletely vacuolated deposits containing amorphous material and arrays of small filaments, and mitochondrial cytopathy are all reported in IBM (37)
. Except for the filaments, many of these are seen in other neuromuscular conditions. Additional profiles such as the proliferation and dilatation of sarcoplasmic reticulum networks (e.g., tubular aggregates) are also a nonspecific finding seen in a number of human neuromuscular diseases and clinically normal inbred male mice (38)
. These and/or the inclusion material may account for the uptake of the modified Gomori stain. The abnormal multivesicular structures may be more unique to this model, also noting that mutlivesicular bodies or endosomes are associated with intraneuronal ß-amyloid in AD (17)
.
Several features of human sIBM, however, were not represented in this mouse model, including the lack of endomysial inflammatory infiltrates, severely atrophic angulated myofibers, and dense aggregates of phosphorylated neurofilaments. However, we detected regional increases of ‘thread-like’ immunoreactivity to tau-5 Ab in transgenic compared with nontransgenic littermates (Supplemental Fig. 2
). These changes in tau were accompanied by the appearance of immunoreactivity to the MC1 epitope, an AD conformation-specific tau Ab (39)
, in transgenic compared with nontransgenic littermates. Moreover, an increase in the signal of a high MW tau isoform was noted in the insoluble fraction (Supplemental Fig. 2E). One explanation for these findings is that endogenous tau is partitioned into a less soluble and more aggregate-prone fraction in the presence of the transgene products. More experimental work is needed to fully characterize changes in tau phosphorylation in the MLC-ßAPP mouse. The reason for these variations relative to human IBM is likely due to factors intrinsic to the murine host, as most single and double transgenic models for AD also have similar shortcomings in brain neurofibrillary tangle formation (40
, 41)
. It is noteworthy to mention that some tau isoforms that characterize muscle tau are not microtubule associated and lack certain phospho-epitopes important to AD pathology (e.g., AT8) (42)
.
The MLC-ßAPP mouse model specifying type II fast fiber involvement supports the notion that tissue subtype vulnerability can guide the degenerative phenotype according to the intersection of specific metabolic characteristics with toxic properties particular to the transgene. These results suggest that, to a degree, vulnerability of fast twitch fibers to Aß inclusion formation may contribute to disease in humans (28
, 43)
. We do not, however, prove or claim with this model that human IBM is primarily a disorder of type II fibers, since type I fibers are also affected (28
, 44)
. One possibility for future investigation is that Aß is more prone to aggregation under conditions favoring anaerobic acidic environments (45)
, such as fast twitch myofibers engaged in glycolytic metabolism (46
, 47)
.
ACKNOWLEDGMENTS
The authors thank Dr. Wendy Robinson for assistance with electromyography, Dr. Dennis Selkoe for antibodies R1282 and 21F12, and Dr. Peter Davies for the MC1 Ab. Thanks to Maria Ericsson and Louise Trakimas of the Harvard Medical School electron microscopy facility. Our thanks also to James Happel for assistance with the creatine kinase assays. This work was supported by grants from the Myositis Association and NIH 41373 to H.W.Q.
Received for publication January 20, 2006. Accepted for publication May 8, 2006.
REFERENCES
-synuclein in Parkinson’s disease. Nat. Genet. 18,106-108[CrossRef][Medline]-secretase site mutation develop a myopathy similar to human inclusion body myositis. Am. J. Pathol. 153,1679-1686This article has been cited by other articles:
|
|
 |
|
  X. Chen, O. Ghribi, and J. D. Geiger Rabbits fed cholesterol-enriched diets exhibit pathological features of inclusion body myositis Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2008; 294(3): R829 - R835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Fonte, D. R. Kipp, J. Yerg III, D. Merin, M. Forrestal, E. Wagner, C. M. Roberts, and C. D. Link Suppression of in Vivo -Amyloid Peptide Toxicity by Overexpression of the HSP-16.2 Small Chaperone Protein J. Biol. Chem., January 11, 2008; 283(2): 784 - 791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Weihl, S. E. Miller, P. I. Hanson, and A. Pestronk Transgenic expression of inclusion body myopathy associated mutant p97/VCP causes weakness and ubiquitinated protein inclusions in mice Hum. Mol. Genet., April 15, 2007; 16(8): 919 - 928. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
