Muscle specific expression of human amyloid precursor protein (hAPP) induces an age-dependent reduction in climbing and flying activity
In order to study the role of human APP (hAPP) on the development and function of skeletal muscles, we took advantage of a transgenic fly line that expresses hAPP under the control of the Upstream Activating Sequence (UAS) [24
]. These flies were crossed to a line that expresses the Gal4 transcription factor under the control of the muscle transcription factor Dmef
], which expresses in all skeletal muscles and a few circadian neurons within the brain [26
We employed several controls in each of these studies, including nontransgenic w1118 flies and transgenic flies expressing bacterial β-galactosidase (LacZ) under the control of Dmef-Gal4. The expression of LacZ both validated the anticipated expression of hAPP in muscles (data not shown) and controlled any effects caused by the competition of Gal4 promoter for general transcriptional machinery within the cells. Ectopic hAPP expressing flies (■) eclosed in normal numbers and displayed comparable longevity to the control lines (▲) suggesting that ectopic hAPP (■) was non-toxic (Figure ).
Figure 1 hAPP-expressing flies display age-dependent defects in climbing and flying. A. Climbing assays with wild type and LacZ expressing flies were performed every week. N = 136 and 85 for w1118 and LacZ expressing flies respectively. Mean +/- SEM. B. In order (more ...)
To assess the effects of ectopic hAPP on behavior, flies were tested for their ability to climb (Figure ). Both nontransgenic (w1118 ●) and LacZ-expressing (■) flies displayed comparable levels of climbing activity during the one-month testing period. Transgenic flies expressing hAPP exhibited wild-type levels of climbing activity during the first two weeks of adulthood, but it declined during the subsequent weeks (Figure ) so that by the end of the fourth week, only about 20% of the transgenic flies could climb. In contrast, 75% of 1-month old wild-type w1118 control animals could climb. The climbing defect observed in transgenic flies was not due to a loss in their negative-geotropism, but rather, an apparent loss of strength that caused them to fall from the test cylinder before they could reach the test mark.
To evaluate flying behavior, animals were dropped into an oil-coated 500 ml graduated cylinder and then scored for their ability to remain near the top. Approximately 55% of the hAPP-expressing flies (dark gray) fell to the bottom 200 ml of the cylinder (poor fliers), while only about 12% of the controls (light gray) fell that distance (Figure ). Conversely, about 70% of the controls (light gray) remained within the top 200 ml (normal fliers), while only 20% of the transgenic flies (dark gray) were in this category. Taken together, the climbing and flying assays suggest that ectopic expression of hAPP results in an age-dependent defect in motor ability.
Electrophysiological analysis of the neuromuscular junction
Defects in motor function could reflect aberrations within the central nervous system, at the neuromuscular synapse, or within the muscle itself. To narrow our focus, we performed electrophysiological analyses of control and hAPP-expressing adult Drosophila. Intracellular glass recording microelectrodes were placed in the thoracic dorsal longitudinal flight muscle (DLFM) and tergotrochanteral motor muscle (TTM) of 1- and 3-week old adults, and the motor neurons were stimulated via the giant fibers using tungsten electrodes placed in the brain or directly in the thoracic ganglion. Both w1118 control and hAPP-expressing animals displayed normal DFLM muscle responses comprised of an evoked junction potential and a muscle action potential when repeatedly stimulated at 100Hz (Figure ). There was no significant difference in amplitude size between animals recorded at 1 week versus animals recorded after 3 weeks (Student T-test, P < 0.05, Table ). This suggests that motorneuron activation reliably results in muscle action potentials that should be sufficient to trigger a behavioral outcome mediated by muscle contraction. Consequently, the decay of climbing behavior in 3 week old hAPP expressing flies is not the result of a defect in motorneuron function and thus we focused on potential hAPP effects on the muscle.
Figure 2 Electrophysiological analysis of wild-type and hAPP-expressing flight muscles. Sample traces of intracellular recordings of muscle responses from the DLFM following electrical stimulation of 1 week (A) and 3 week (B) old wild-type w1118 flies and 1 week (more ...)
Amplitude sizes of recordings from the DLFM and the TTM
Anatomical analysis of muscles
To evaluate the effects of hAPP on muscle development and function, we performed both light and electron microscopy on the indirect flight muscles of control and hAPP-expressing flies. In Figure we examined coronal (3 week old animals) and sagittal (4 week old animals) sections of the thorax to examine both the flight and leg muscles. At the light level, hematoxylin and eosin staining revealed well-developed muscles in both control (3A and 3C) and transgenic animals (3B and 3D), consistent with their ability to eclose, walk, climb, and fly as adults (Figure ). Fiber number and diameter were comparable in wild-type and transgenic animals. These data support the hypothesis that ectopic expression of hAPP does not negatively impact myogenesis.
Figure 3 Light microscopy of wild-type and hAPP-expressing flight muscles. Age-matched transverse sections from three week old w1118 (A) and hAPP (B) expressing flies and age-matched sagittal sections from four week old w1118 (C) and hAPP (D) were conducted. Asterisks (more ...)
We also performed transmission electron microscopy on the DLFM from 3-week old wild-type animals, which revealed well-developed Z-lines, M-lines, contractile apparatus and mitochondria (Figure ). Neither the muscle fibers themselves nor the internal membrane systems were swollen or disrupted. Anatomically, the DLFM of 3-week old transgenic animals were also grossly normal, and displayed well-defined sarcomeres (Figure ). As in the control muscles, dense rows of mitochondria were sandwiched between the bundles of contractile elements perpendicular to the Z bands. At the subcellular level, the organization of the TTM muscles differs from that of the DLFM both in terms of sarcomeric structure and the abundance of internal membrane systems of the T-tubule and sarcoplasmic reticulum (SR) (Figure and ). At the electron microscopic level, 3 week old wild-type and hAPP transgenic TTM muscles were grossly indistinguishable (Figure and ).
Figure 4 Transmission electron microscopy of wild-type and hAPP-expressing flight muscle. Dorsal longitudinal flight muscles (DLFM) from control (w1118, A) and hAPP (B) transgenic flies at three weeks of age. TTM muscles from control (w1118, C) and hAPP (D) transgenic (more ...)
Environmental control of muscle pathogenesis in hAPP-expressing flies
We performed many replicates of the climbing assay and consistently observed comparable age-dependent defects in transgenic animals. At one point in the study however, we switched from glass to polypropylene plastic vials for rearing the adults. While longevity was unchanged under these different rearing conditions (data not shown), the severity of the hAPP-induced climbing defect of flies cultured in the plastic vials was significantly reduced over the course of one month (Figure ). In order to verify that the rearing container impacted climbing behavior, cohorts of wild-type and hAPP-expressing animals were separated into glass or plastic vials and then subjected to the climbing assay. Wild-type animals displayed statistically improved climbing activity in plastic vials (♦) relative to glass vials (▲) (85% vs: 56% at week four; p < 0.05 ) (Figure ). A much more dramatic effect was observed for the hAPP-expressing animals. At 4 weeks 65% of plastic reared animals (●) could climb versus 25% for glass-reared animals (■) (p < 0.05). The benefits of plastic versus glass vials were restricted to the leg muscles and did not extend to the flight muscles (Figure ), since we did not observe any differences in flying activity between these two populations. These data suggest that only muscles that are actively challenged are the ones affected by the type of rearing substrate.
Figure 5 Effect of vial type on climbing behavior in wild-type and hAPP transgenic flies. A. The climbing ability of wild type and hAPP expressing flies reared in glass or plastic vials was determined. Percentage represents the portion of flies displaying normal (more ...)
One possible mechanism for the differential effects of the vial composition on the development of muscle weakness is that the smoothness of glass relative to plastic requires more muscle strength to climb. To test this hypothesis, we siliconized plastic vials to make their surfaces smoother, and then tested new groups of control and transgenic animals. Survival was comparable in all three types of vials (data not shown). Wild type control flies climbed equally well in all three types of vials until week 3. After that, there was a small but statistically significant decline in climbing ability for flies reared in glass vials (▲) relative to coated (■) and uncoated plastic vials (●) (Figure ). However, the effects of vial type were much more dramatic in the hAPP transgenic flies (Figure ). Transgenic flies reared in siliconized plastic vials (■) displayed a rapid loss of climbing activity beginning even after the first week relative to those reared in glass (▲) or uncoated plastic vials (●). Between three and four weeks, there was a dramatic decline in climbing for hAPP transgenic flies reared in glass vials. These data suggest that the environmental factors (vial material) can have a significant impact on APP-induced pathogenesis.
Coexpression of human Parkin rescues a hAPP-mediated climbing defect
Mutations in the ubiquitin E3 ligase Parkin is the primary cause of Autosomal Recessive Juvenile Parkinsonism [27
]. Loss of Parkin function endangers some cells, most notably midbrain dopaminergic neurons. However, in vitro
studies with muscle cells have demonstrated that expression of ectopic Parkin can protect muscle from the toxic effects of accumulation of amyloid peptides [28
]. To determine if this protein could also protect skeletal muscles in vivo
from hAPP-induced damage, we generated transgenic flies that express human Parkin and/or hAPP in the muscles, and then monitored the ability of the adults to climb (Figure ). Flies expressing hAPP (▲) demonstrated a significant decline in climbing ability beginning in the third week. Co-expression of Parkin (♦) rescued this defect and allowed the flies to climb at wild type levels (Figure ).
Figure 6 Ectopic Parkin rescues hAPP-induced climbing defects in transgenic flies. Climbing assays with transgenic flies expressing hAPP alone, hParkin alone, and hAPP plus hParkin in glass vials were conducted each week. N = 54 for all genotypes. Mean +/- SEM. (more ...)
APP has been the focus of intensive investigation for its possible role in human diseases, most notably Alzheimer's disease. Numerous studies have demonstrated that ectopic expression of APP, or its proteolytic products, most notably Aβ42
, can trigger synapse loss and neuron death in both in vitro
and in vivo
]. In fly, ectopic expression of Aβ42
results in age-dependent neuron loss [29
]. Interestingly, males were more severely impacted than females, although this may reflect differences in driver expression rather than differential gender sensitivity.
These same proteins have been proposed to accumulate within aggregates in patients with other diseases as well, such as inclusion body myositis [30
], although this observation has been controversial [17
]. Nevertheless, experimental studies have demonstrated that ectopic expression of either APP or Aβ42
is sufficient to induce muscle cell death both in vitro
] and in transgenic mouse models [22
]. In the nematode Caenorhabditis elegans
, expression of Aβ42
results in protein aggregations within body wall muscles that result in paralysis and reduced longevity [32
The transgenic fly model described in the present study complements and extends some of the data obtained with mouse and worm models, as well as some of the features of s-IBM. In s-IBM patients and animal models designed to simulate the disorder, individuals produce muscles that appear to be morphologically and physiologically normal, but develop progressive age-dependent muscle weakness in mid-to-late adulthood [12
] (Figure and ). This loss of muscle strength is not accompanied by detectable changes in neuromuscular synapse activity, suggesting that the defects arise within the muscles themselves. Indeed, we did not observe any changes in following frequency or spike amplitude in the muscles of our hAPP-expressing animals (Figure ). These data would argue against a negative influence of hAPP expression in the small subset of circadian neurons in the brain that also express MEF2 [26
The fly model described in this study does differ from both s-IBM and other hAPP transgenic mouse and worm models in that they did not produce protein aggregates within the sarcoplasm that can be detected with either Congo Red staining (data not shown) or ultrastructural analysis (Figure ). It is possible that the level of hAPP expression was below the threshold required for macro-aggregation, or alternatively, that flies have intracellular mechanisms that limit aggregate formation. It has been noted that flies do not make inclusion bodies in the skeletal muscles with other aggregation prone proteins, like the polyglutamine-rich Huntington protein [35
]. Nevertheless, the present fly model may provide insight into the molecular mechanisms that mediate hAPP induced pathology.
Several mechanisms have been proposed for the toxic effects of APP on skeletal muscle function including: defective regulation of ryanodine receptor-dependent sarcoplasmic Ca2+
], CD8+ cytotoxic T cell invasion [37
], an autophagic mechanism [38
], and myostatin activity [39
The observation that ectopic expression of the ubiquitin E3 ligase Parkin can rescue the hAPP-associated defects in climbing in our transgenic model agrees well with in vitro
results from other groups [16
]. At least two possible Parkin associated mechanisms can ameliorate ectopic APP-induced behavioral defects. In mammalian muscle, Parkin prevents hAPP-induced muscle degeneration by inhibiting accumulation of toxic Aβs and also protects cells from mitochondrial-specific toxins like rotenone and carbonyl cyanide 3-chlorophenylhydrazone, but not from other toxins like calcium ionophore A23187 or H2
]. Loss of function mutations in the parkin gene result in age-dependent defects in flight muscle maintenance in fly and are also associated with mitochondrial defects [42
]. Interestingly, these defects can be prevented by increased mitochondrial fission [43
]. Taken together, these data support the hypothesis that Parkin may reduce mitochondrial oxidative stress or maintain respiratory functions to help preserve mitochondrial integrity.
We have demonstrated that environmental factors, such as rearing surface (glass versus plastic), have a dramatic effect on the timing and severity of hAPP-induced pathogenesis, suggesting a possible interplay between environmental and genetic factors. To our knowledge, this is the first report of any behavioral activities that can be tied to the material used for rearing flies. The simple manipulation of changing the rearing vessel may represent a valuable tool for genetic screens in fly designed to identify work-associated genes in fly muscle.
Only limited data exist on the role of exercise in the progression of myopathies in human [44
]. Interestingly, Arnardottir et al.
] have suggested that moderate exercise may retard the symptoms of s-IBM. Rearing flies in plastic vials greatly reduced the timing and severity of symptoms relative to rearing in glass (Figure ). Since the only difference between animals reared in glass versus plastic was the nature of the vessel, we speculated that surface properties accounted for the observed effects. Glass vials have a smoother surface than plastic and therefore it is presumably harder for the fly to climb. This extra work would increase the mechanical stress on the muscle fibers and might contribute to damage or generation of reactive oxygen species (ROS) by mitochondria. We tried to test this hypothesis directly by setting up devices that would force the animal to walk more, such as placing the vials on a slowly moving rocking table. Unfortunately, these efforts to voluntarily increase motor activity did not appear to alter animal behavior, so instead we changed the surface properties of the plastic vials by siliconizing them. This subtle manipulation had a profound effect on the time course and severity of the hAPP-induced abnormal activities. This effect was specific to the muscles that were forced to work, since this treatment did not alter flying behavior.