Inhibitors of ubiquitin-dependent proteolysis can delay programmed cell death of adult intersegmental muscles in the moth Manduca sexta

Authors

  • Ronald J. Bayline,

    Corresponding author
    1. Department of Biology, Washington and Jefferson College, Washington, Pennsylvania
    2. Field of Neurobiology and Behavior, Cornell University, Ithaca, New York
    • Washington and Jefferson College, 60 South Lincoln Street, Washington, PA 15301
    Search for more papers by this author
  • Derek M. Dean,

    1. Field of Genetics and Development, Cornell University, Ithaca, New York
    Current affiliation:
    1. Department of Biology, TBL 110, Williams College, Williamstown, MA 01267
    Search for more papers by this author
  • Ronald Booker

    1. Field of Neurobiology and Behavior, Cornell University, Ithaca, New York
    2. Field of Genetics and Development, Cornell University, Ithaca, New York
    Search for more papers by this author

Abstract

In the moth Manduca sexta, intersegmental muscles (ISMs) undergo rapid programmed cell death (PCD) within 48 hr of adult emergence. ISM PCD involves ubiquitin-dependent proteasomal degradation accompanied by the down-regulation of expression of actin genes and the up-regulation of degradative gene expression such as ubiquitin. Hemin chloride and N-acetyl-leu-leu-norleucinal (ALLN), both inhibitors of proteasomal activity, administered before adult emergence delayed PCD for up to 5 days in ISMs maintained from the larval stage, such as the dorsal internal medial muscle in abdominal segment 4 (DIM-A4). ISMs that developed during metamorphosis from respecified larval muscles such as the DIM-A2 were less dramatically affected. The increase in polyubiquitinated proteins and the decrease in actin mRNA expression accompanying maintained ISM PCD were delayed after inhibitor application. No changes were detected in respecified ISMs. These results reveal a regulatory role for proteasomal activity in an early stage of maintained ISM cell death. Developmental Dynamics 233:445–455, 2005. © 2005 Wiley-Liss, Inc.

INTRODUCTION

During metamorphosis of holometabolous insects, programmed cell death is responsible for the removal of many larval cells and tissues that are no longer necessary for the adult insect. The death of the intersegmental muscles (ISMs) of the moth Manduca sexta is one example of this process. At the time of adult emergence, a set of four segmentally homologous ISMs are found in abdominal segments 2 through 6 (A2–A6) of the moth. The adult moth uses the ISMs to escape from the pupal case and to inflate its wings during the first hours after adult emergence (Finlayson, 1956; Lockshin and Williams, 1965a; Schwartz and Truman, 1983; Truman, 1983). After emergence and wing inflation, the ISMs degenerate and are no longer present within 36–72 hr after adult emergence. Although the degeneration of ISMs occurs after adult emergence, some muscle atrophy that can be characterized by histological examinations in the ISMs occurs before emergence. This early phase of atrophy results in a slow decline in the mass of the ISMs during the final days of adult development. This slow atrophy is not accompanied by a decline in muscle motility or loss of the structural integrity of the muscles (Schwartz and Ruff, 2002). After adult emergence, the ISMs switch to a program of rapid atrophy, which is characterized by a dramatic increase in the rate of ISM regression, the loss of myofibrillar structure, the degeneration of muscle nuclei and the ruffling of the sarcolemma. The ISMs undergo cell death in response to a decline in the circulating titers of the steroid hormone ecdysone (Schwartz and Truman, 1983).

Along with histological signs of atrophy, the ISMs also exhibit changes in gene and protein expression during PCD (Schwartz et al., 1990; Haas et al., 1995; Jones et al., 1995; Muller and Schwartz, 1995). The regression of the ISMs is characterized by the repression of muscle mRNA transcripts coding for structural proteins, including actin and myosin (Schwartz et al., 1990). There is also a dramatic increase in the expression of genes that are involved in the proteasomal pathway, which is responsible for the majority of extralysosomal protein degradation in cells (Rock et al., 1994; Ciechanover, 1994; Ciechanover et al., 2000; Naujokat and Hoffmann, 2002). The regression of the ISMs is associated with an increase in the expression of ubiquitin-conjugating enzymes, a 10-fold increase in the expression of polyubiquitin-conjugated proteins and changes in the subunit composition of the 26S proteasome (Schwartz et al., 1990; Haas et al., 1995; Jones et al., 1995). These changes in the pattern of gene expression are thought to facilitate the rapid removal of the ISMs.

Since the initial report of the increase in the expression of the components of the proteasomal pathway before the death of the ISMs, a similar pattern has been observed in several other systems. Increased expression of ubiquitin is observed in dying mouse spinal motor neurons (Mazurkiewicz, 1991), lymphocytes (Delic et al., 1993), as well the cell deaths associated with diseases such as amyotrophic lateral sclerosis (Bergmann, 1993) and Alzheimer's disease (Mori et al., 1987). This positive association between the expression and cell death is consistent with a constitutive role for the proteasomal activity in the rapid removal of the cells and tissue during cell death (Muller and Schwartz, 1995). However, the role of the ubiquitin proteasomal pathway is not necessarily restricted to the bulk degradation of proteins. Proteasomal activity is known to be involved in regulating several cellular processes, including the cell cycle, cellular transformation, and signal transduction (Naujokat and Hoffmann, 2002). There is also evidence that ubiquitin-dependent proteolytic activity is involved in regulating the level of both pro- and antiapoptotic molecules and that the inhibition of proteasomal activity can both promote or inhibit the onset of cell death, depending on the system under study (Hirsch et al., 1998; Stefanelli et al., 1998; Dallaporta et al., 2000; Wing et al., 2002; Agapite et al., 2002). Thus, it appears that proteasomal activity can play at least two roles in regulating cell death. First, ubiquination and proteasomal activity could potentially play a key role in regulating the activation of the cell death program. Once cell death is initiated, proteasomal activity may be involved in mediating the cellular degeneration that results in the final removal of the doomed cell or tissue. In the absence of direct experimental evidence, it is not possible to define the functional role(s) of ubiquination and proteasomal activity in cell death.

Here, we present the results of a series of experiments designed to define the role of proteasomal activity in the death of the ISMs that occurs after adult emergence. The administration of inhibitors of proteasomal activity before adult emergence blocked the onset of ISM rapid atrophy for up to 4 days. The application of the inhibitors not only delayed cellular degeneration but also blocked the changes in the pattern of gene expression that accompanied the death of the ISMs, including the dramatic increase in the level of ubiquitin expression.

RESULTS

Slow and Rapid Atrophy of the ISMs

Segments A2 through A6 of larval Manduca sexta each contain four, paired intersegmental muscles, or ISMs, the ventral internal medial (VIM), ventral internal lateral (VIL), dorsal internal medial (DIM), and dorsal internal lateral (DIL) muscles. During metamorphosis, the fates of the ISMs varies in a segment specific pattern (Bayline et al., 1998). In segments A4 to A6, all of the larval ISMs are maintained throughout adult development, whereas in segment A3, only two muscles are maintained. In contrast, the remaining two ISMs in A3 and all four ISMs in A2 undergo a process referred to as respecification. The respecification of the ISMs involves the regression of the larval muscle during the early stages of adult development followed by a period of regrowth accompanied by the acquisition of adult specific characteristics (Bayline et al., 1998). The growth and differentiation of the respecified ISMs is not completed until the time of adult emergence.

Around the time of adult emergence the ISMs undergo a period of regression and their removal is completed 2 to 3 days after adult emergence (Schwartz and Truman, 1983; Figs. 1, 2A,B). The pattern of regression observed depends upon the developmental history of the muscle. For the maintained ISMs, the regression occurs in two phases (Schwartz, 1992). In the 3 days before adult emergence, the maintained ISMs undergo a process referred to as slow atrophy. During the period of slow atrophy, the diameter of the maintained ISMs declined by 15% to 20% (Fig. 1A). During the period of slow atrophy, the muscles remained contractile and a histological analysis revealed the regression was not associated with the cellular degeneration of the muscles. The myofibrillar organization and sarcolemma remained intact, and no signs of degenerating muscle nuclei were observed (Fig. 1B). In contrast to the maintained ISMs during the final days of adult development, the respecified ISMs of A2 and A3 exhibited a steady increase in diameter and completed the terminal stages of adult differentiation (Fig. 1A; Bayline et al., 1998).

Figure 1.

Sizes of dorsal internal medial (DIM) muscles in segments A2 and A4 during normal development. A: Total muscle size for the DIM-A4 (solid lines) and DIM-A2 (dashed lines). B: Percentage of fibers undergoing rapid atrophy as defined by the breakdown of myofibrillar structure and the sarcolemma and the appearance of degenerating nuclei. C: Number of fibers in the DIM-A4 muscle. Values are mean ± SEM for six or more muscles from three or more animals for each data point.

Figure 2.

A–C: Whole abdomens showing the abdominal musculature of untreated controls at AE+0 (A), AE+48 (B), and an AE+48 animal that was treated with hemin at AE-48 (C). In each panel, the dorsal internal medial (DIM) -A2 (left arrow) and DIM-A4 (right arrow) are indicated. A: Before adult emergence, each segment contains four sets of intersegmental muscles (ISMs). B: After adult emergence, the ISMs undergo rapid atrophy and, within 48 hr, all that remains of the ISMs are a few thin degenerating fibers. C: In animals treated with hemin 2 to 3 days before the expected time of adult emergence resulted in a delay in the onset of rapid atrophy. In this abdomen isolated from an animal treated at stage AE-48, there is little or no evidence of muscle regression.

The ISMs are used around the time of adult emergence to propel the adult from the pupal cuticle and to aid in the inflation of the wings (Finlayson, 1956; Lockshin and Williams, 1965a; Schwartz and Truman, 1983; Truman, 1983). Within a few hours of adult emergence, all of the abdominal ISMs initiate cell death and undergo a period of rapid atrophy that results in their complete removal within 36 to 48 hr (Figs. 1A, 2A,B). For the maintained ISMs, rapid atrophy is defined by an eightfold increase in the rate of regression. Unlike slow atrophy, the rapid atrophy of the maintained ISMs was accompanied by the cellular degeneration of the muscle fibers. For example, as early as AE+12, 41% DIM-A4 muscles examined contained fibers that possessed degenerating nuclei, disorganized myofibrils, and a ruffled sarcolemma (Fig. 1B). By AE+24, the average number of fibers counted in DIM-A4 had decreased by approximately 30% and all of the remaining fibers exhibited signs of the cellular degeneration that typically accompanies rapid atrophy (Fig. 1B,C). By AE+36, all that remained of the degenerating DIM-A4 was a diffuse mat in which individual fibers could no longer be distinguished and the removal of the muscle was complete by AE+48. The cellular events of rapid atrophy for the respecified ISMs were similar to those observed for the maintained ISMs (Fig. 1B). The only difference was that the respecified ISMs regressed at a slightly slower rate compared with the maintained ISMs.

Inhibitors of Proteasome Proteolytic Activity Delay the Death of the ISMs

Both the slow and rapid atrophy of ISMs observed around the time of adult emergence are correlated with an increase in the expression of the components of the ubiquitin–proteasome pathway and of proteasomal activity (Schwartz et al., 1990; Haas et al., 1995; Jones et al., 1995). This correlation raises the possibility that proteasomal activity plays a key role in regulating and/or mediating the cellular degeneration that accompanies the regression and final removal of the ISMs observed around the time of adult emergence (Schwartz and Truman, 1983; Schwartz and Ruff, 2002). As a direct test of this possibility, we examined the effect of administering two inhibitors of proteasomal activity, hemin chloride and N-acetyl-leu-leu-norleucinal (ALLN), on the fates of the ISMs. Both of these inhibitors have been shown previously to inhibit proteasome-dependent protein degradation in vitro (Haas and Rose, 1981; Vierstra and Sullivan, 1988; Rock et al., 1994). The application of the inhibitors during the final 5 days of adult development appeared to have little or no effect on the fate of the ISMs before adult emergence. For example, in animals treated with 23 μl/g body weight of hemin at stage AE-48 the size of both the maintained (Fig. 3A) and respecified ISMs (Fig. 3B) at the time of adult emergence (AE+0) were similar to the ISMs in control animals. Even in animals treated as early as AE-96, no observable changes in the size of the ISMs at AE+0 were observed. Similar results were seen after application of 10 μl/g body weight of ALLN. Thus, it appears that neither the slow atrophy of the maintained ISMs nor the terminal differentiation of the respecified ISMs, were altered in late stage developing adults treated with the proteasomal inhibitors.

Figure 3.

Hemin injections at AE-48 delay the onset of rapid atrophy in the intersegmental muscles (ISMs). A: Change in total muscle diameter for the dorsal internal medial (DIM) -A4 during both normal development and after hemin injection. B: Change in total muscle diameter for the DIM-A2 during both normal development and after hemin injection.

Proteasomal inhibitors had a much more dramatic effect on the rapid atrophy that occurred in the ISMs after adult emergence. The maintained ISMs were most sensitive to the application of the inhibitors. For example, after a 23 μl/g body weight dose of hemin 2 days before adult emergence, intact maintained ISMs are still visible in the abdomen 2 days after adult emergence (Fig. 2C). In animals treated with hemin between AE-48 and AE-72, the rate of ISM regression after adult emergence slowed dramatically; as a result, the final removal of the maintained ISMs was delayed by as much as 96 hr (Fig. 3A). The effect on the respecified muscles was less dramatic. At AE+48, little to no DIM-A2 muscle is observed after hemin treatments (Figs. 2C, 3B). However, the rate of respecified ISM regression is slower after hemin treatments than in untreated animals. As a result, in the treated animals, the time required for the complete removal of the respecified ISMs often increased by as much as 48 hr compared with controls (Fig. 3B).

Hemin was more effective in slowing the regression of the maintained ISMs than ALLN. In animals treated with hemin between AE-12 and AE-96, the maintained DIM-A4 regressed by only 25% to 40% of its size at adult emergence by AE+48 (Fig. 4A). In contrast, in animals treated with 10 μl/mg body weight of ALLN between AE-96 and AE-12 by 2 days after emergence, the DIM-A4 had regressed by 45% to 90% (Fig. 4A). For both hemin and ALLN, there is a sharp decline in the ability of the inhibitors to slow the rate of ISM regression after adult emergence when administered between AE-12 and AE+0 (Fig. 4A). Larger dosages of hemin (up to 45 μl/g body weight) and ALLN (up to 20 μl/g body weight) did not produce a qualitative change in the pattern of muscle regression. For example, animals treated with 45 μl/g body weight of hemin at AE-12, the maintained ISMs still regressed by more than 50% of their AE+0 size by AE+48 (data not shown). In animals treated at the time of emergence, there was little or no evidence that the inhibitors slowed the rate of the maintained ISMs regression. The proteasomal inhibitors were less-effective at blocking the rapid regression of the respecified ISMs, regardless of the stage and dosage of application of the inhibitors (Fig. 4B). Based on these results, dosages of 23 μl/g body weight of hemin and 10 μl/g body weight of ALLN were used for the remaining experiments.

Figure 4.

A,B: The extent to which the inhibitors slowed the regression of the respecified dorsal internal medial (DIM) -A4 (A) and maintained DIM-A2 (B) was dependent upon the time of treatment. Animals were treated with hemin chloride (squares) or N-acetyl-leu-leu-norleucinal (ALLN, circles) at the times indicated, and the muscles were examined 2 days after adult emergence (AE+48). Upper and lower dashed lines represent the size of the muscles in control animals at AE+0 and AE+48, respectively. n = 4 or more muscles for each point.

Although hemin and ALLN applications revealed different patterns of regression in muscle size, both were effective at delaying the appearance of cytological aspects of rapid atrophy in the maintained muscles. In animals treated with hemin at AE-48 and examined at AE+48, it was clearly evident that the cellular degeneration was reduced in the maintained ISMs. Whereas in control animals, all muscles were absent at AE+48, the muscles in the treated animals contained similar numbers of intact fibers as the muscles at the time of adult emergence (Fig. 5A–C). Even as late as stage AE+96 in some of the treated animals, it was not unusual to find maintained ISMs that exhibited few if any signs of the onset of rapid atrophy. Although ALLN was less effective than hemin at slowing the regression in the size of DIM-A4 after adult emergence, the histological signs of rapid atrophy were also absent in a majority of muscle fibers examined at AE+48 (Fig. 5A–C). For the respecified muscles, the cytological aspects of rapid atrophy were not as dramatically affected by application of hemin or ALLN. For example, by AE+48, all respecified muscles in the treated animals showed signs of the onset of rapid atrophy, as did the control muscle (Fig. 5D–F).

Figure 5.

A–F: The application of proteasomal inhibitors delays the onset of rapid atrophy of the maintained dorsal internal medial (DIM) -A4 (A–C) but not the respecified DIM-A2 (D–F) examined 2 days after adult emergence (AE+48). For hemin and N-acetyl-leu-leu-norleucinal (ALLN) treatments, injections of the inhibitors were given between AE-72 and AE-48. A,D: Regression in total muscle size. B,E: Percentage of intact (i.e., nondegenerating) fibers. C,F: The number of intact fibers counted in the DIM-A4 muscle of treated and control animals. Values are mean ± SEM for six or more muscles for each data point.

Inhibitors of Proteasomal Activity Block the Molecular Correlates of ISM Cell Death

Proteasomal activity potentially could play several different roles in the death of the abdominal ISMs after emergence. As has been observed in several different systems, proteasomal activity might play a role in either the initiation or the execution of the cell death program. By using inhibitors of proteasomal activity, it is possible to determine whether proteasomal activity is necessary for different aspects of the cell death pathway. For example, proteasomal inhibitors applied before the PCD may block the initiation of the PCD program and prevent all aspects of PCD. Alternatively, ISM cell death program might be activated in the presence of inhibitors, but an early step(s) in the cell death program such as the changes of the pattern of gene expression that define rapid atrophy may be prevented. Finally, the inhibitors could simply block the proteolytic activity responsible for mediating the global atrophy of the ISMs observed during programmed cell death. In an effort to distinguish between these possibilities, we examined the effect of the inhibitors on the pattern of protein and gene expression that accompanies the regression and death of the ISMs around the time of adult emergence. We examined the expression of ubiquitin protein using antibodies specific to both the free and conjugated forms of the protein (Fahrbach and Schwartz, 1994). Consistent with previous results, maintained ISMs isolated before adult emergence express relatively low levels of ubiquitin (Fig. 6A). However, within a few hours of adult emergence, there is a dramatic increase in anti-ubiquitin staining, with the staining reaching maximal levels in all the fibers within 24 hr of emergence (Fig. 6B). A similar pattern of anti-ubiquitin staining was observed for the respecified ISMs (Fig. 6D,E), as well as for another respecified muscle that degenerates after adult emergence, the tergosternal muscle (TS; Fig. 6B). Treating developing adults with proteasomal inhibitors delayed the expected increase in ubiquitin expression in the maintained ISMs after adult emergence. In animals treated with the inhibitors at stages AE-72 and AE-48, the expected increase in ubiquitin expression was delayed in the maintained ISMs. In the treated animals examined at stage AE+24, the intensity of the anti-ubiquitin staining in the maintained ISMs was similar to the level of staining observed in muscles isolated from animals before adult emergence and in muscles that are not destined to die (AE-32; Fig. 6C). ALLN treatments at the same stages produced similar results produced similar results (data not shown). The first signs of an increase in the expression of ubiquitin protein in the maintained ISMs of treated animals were observed at stage AE+48. However, the anti-ubiquitin staining was not uniform. In the DIM-A4s isolated from stage AE+48 treated animals, only 20% the fibers showed an increased level of anti-ubiquitin staining (data not shown). Over the next 48 hr of adult life, treated animals exhibited a steady increase in the percentage of DIM-A4 fibers with enhanced levels of anti-ubiquitin staining. In contrast to the maintained muscles, the application of the inhibitors failed to delay the increase in ubiquitin protein expression in the respecified ISMs at AE+24 (Fig. 6F) or at any later stages examined.

Figure 6.

A–F: Inhibition of proteasomal activity alters the pattern of ubiquitin protein expression demonstrated by anti-polyubiquitin staining in 10-μm cross-sections of dorsal internal medial (DIM) -A4 (A–C, asterisks) and DIM-A2 (D-F, asterisks). A: On the day before adult emergence AE-32, the level of anti-polyubiquitin staining in the cytoplasm of DIM-A4 is low. The superficial external muscle (Ext), which does not undergo degeneration, shows a similar level of anti-polyubiquitin staining. B: A degenerating DIM-A4 isolated from a stage AE+24 control animal. The muscle fibers all show signs of rapid atrophy including the loss of myofibrillar organization. There is also dramatic increase in the intensity of anti-polyubiquitin labeling in the cytoplasm, while the level of anti-polyubiquitin staining in the external muscle remains low. Another muscle that degenerates, the tergosternal muscle (TS), also shows a high level of anti-polyubiquitin staining. C: DIM-A4 muscle isolated at stage AE+24 from an animal treated with hemin at AE-48. The muscle shows no signs of cellular degeneration. The intensity of the anti-polyubiquitin staining is similar to that observed in muscle isolated before emergence and to that of the underlying external muscle, which does not undergo cell death. D: DIM-A2 at AE-32. At this stage, the developing DIM-A2 is still differentiating and is not undergoing atrophy. At this stage, anti-polyubiquitin staining is at a low level. E: By stage AE+24, the fibers of DIM-A2 all show signs of the onset of cell death. At this stage, there is also a dramatic increase in the intensity of anti-polyubiquitin staining. The level of anti-polyubiquitin staining remains low in the underlying external muscle. F: Anti-ubiquitin staining at stage AE+24 isolated from an animal treated with hemin at stage AE-48. The application of the inhibitor fails to block the dramatic increase in anti-polyubiquitin staining in the DIM-A2 that occurs after adult emergence. Scale bar = 100 μm in C (applies to A–F).

The rapid atrophy of the maintained ISMs is also preceded by a dramatic decline in the level of expression of muscle structural proteins, including actin and myosin heavy chain mRNAs (Schwartz et al., 1993). The decline in the expression of actin and myosin heavy chain mRNAs begins the day before adult emergence (AE-24) and reaches minimal levels by within a few hours of adult emergence (Schwartz et al., 1993). In the DIM-A4s isolated from animals treated with proteasomal inhibitors between stages AE-72 and AE-48, the decline in the expression of actin mRNA was delayed by as much as 48 hr (Fig. 7A). In the muscles isolated from stage AE+24 treated animals, the level of actin mRNA expression was similar to that observed on AE-24. In two of the four DIM-A4s isolated from stage AE+48 treated animals, the level of actin mRNA expression was also similar to pre-emergence levels. In contrast to the maintained ISMs, the administration of the inhibitors failed to block the decline in actin mRNA expression in the respecified DIM-A2 muscle (Fig. 7B). Similar to controls, actin mRNA expression in the treated DIM-A2 muscle declined to undetectable levels by stage AE+3.

Figure 7.

The inhibition of proteasomal activity blocks the expected decline in the expression of actin mRNA in the intersegmental muscles (ISMs) after adult emergence. A,B: The top panels show actin transcript levels, whereas the bottom panels shows the ethidium bromide-stained 18s RNA as a loading control. A: Actin mRNA expression in the maintained ISMs isolated from segments A4-A6. Maintained ISMs isolated from control animals exhibit a sharp drop of actin mRNA within 3 hr of adult emergence (AE+3). In the maintained ISMs isolated from animals treated with the proteasomal inhibitors between stages AE-72 and AE-48, this drop in actin mRNA expression was blocked for at least 24 hr after emergence. B: The respecified ISMs found in A2 also show a dramatic drop in actin mRNA expression shortly after adult emergence. However, in contrast to the maintained ISMs, the application of the inhibitors fails to block the down-regulation of actin mRNA expression in the respecified DIM-A2 muscles. Each time point was analyzed by examining RNA preparations from at least four animals.

DISCUSSION

Inhibitors of Ubiquitin-Dependent Proteolysis Slow the Rate of ISM Regression

The atrophy and final removal of the abdominal ISMs occurs in two steps. During the final 3 days of adult development, the maintained ISMs undergo a program of slow atrophy, which results in a 20% decline in the diameter and a 40% decline in the total mass of the muscle (Schwartz and Truman, 1983; Schwartz and Ruff, 2002). After adult emergence, all the ISMs initiate a program of rapid atrophy, which is characterized by a dramatic increase in the rate of regression with the final removal of the muscles being complete within 36 to 72 hr. Both slow and rapid atrophy are characterized by the increased expression of ubiquitin protein and enhanced levels of proteasomal activity (Schwartz, 1992; Schwartz et al., 1993; Haas et al., 1995). Despite the increased level of proteasomal activity observed during the period of slow atrophy, the application of the proteasomal inhibitors failed to slow the regression of the maintained ISMs that occurs during the final days of adult development. Although these results are consistent with proteasomal activity having little or no role in mediating the slow atrophy of the maintained ISMs, our results are not definitive. One possible explanation is that, in the treated animals, the level of residual proteasomal activity was sufficient to allow slow atrophy to proceed at a normal rate. In contrast to the maintained ISMs, the respecified ISMs are growing and undergoing their terminal differentiation during the final days of adult development. The application of the inhibitors during this period of growth had no noticeable affect on the fate of the respecified ISMs before adult emergence.

Within hours of adult emergence, the ISMs initiate cell death and the removal of the muscle is complete within 36 to 48 hr after emergence. In animals that were treated with the proteasomal inhibitors between stages AE-96 to AE-12, the rate the ISMs regressed slowed to the point that 2 to 5 extra days were required for the complete removal of the muscles. During the first 3 to 4 days of adult life, the maintained ISMs in the treated animals regressed at a rate similar to that observed during the period of slow atrophy that occurs before emergence. This finding raised the possibility that administering the inhibitors during the last days of adult development resulted in the process of slow atrophy simply continuing into the early part of adult life. However, a cellular analysis revealed that is not the case. Slow atrophy is characterized by the regression of the maintained ISMs, resulting from a decline in the average diameter of the individual fibers in the absence of any signs of the cellular degeneration. Thus, by the time of adult emergence, slow atrophy results in a 20% decline in the diameter of the maintained ISMs, while there is no sign of the cellular degeneration of the individual muscle fibers and their numbers were stable. In contrast, rapid atrophy is characterized by a dramatic increase in the rate of ISM regression accompanied by the cellular degeneration and eventual loss of individual muscle fibers. In control animals, all of the fibers of the maintained ISMs exhibited signs of cellular degeneration within 24 hr of adult emergence. Our examination of maintained ISMs isolated from treated animals revealed that the regression of the ISMs that occurs resulted from the regression and complete removal of a limited number of muscle fibers. Even as late as stage AE+96 in the animals treated before emergence, we found ISMs that showed few if any signs of cellular degeneration and that were still contractile.

Proteasomal Activity Is Required to Initiate Rapid Muscle Atrophy

Our results are consistent with proteasomal activity playing a role in regulating an early step in the ISM cell death program. The onset of rapid atrophy is accompanied by changes in the pattern of protein expression that are thought to support the rapid degeneration of the ISMs. These changes include an increase in the expression of the components of the proteasomal pathway and a decline in the expression of the muscle structural proteins actin and myosin. The application of the proteasomal inhibitors delayed the expected changes in the expression of both ubiquitin protein and actin mRNA expression. In the treated animals, the expected changes in the pattern of protein expression in the maintained were delayed by up to 48 hr. Whereas application of the inhibitors during the final days of adult development also slowed the rate that the respecified ISMs regressed after adult emergence, we failed to detect an effect of the inhibitors on the pattern of protein expression that accompanies the regression of the respecified ISMs. One possibility is that any delays in the expected changes in the pattern of expression of ubiquitin and actin were of short duration and went undetected. Alternatively, the differences in the response to the inhibitors could represent differences in how the process of cell death is regulated in the respecified and maintained ISMs.

As the time of adult emergence approaches, there is a dramatic decline in the ability of the inhibitors to block the cellular degeneration of the maintained ISMs. When the inhibitors were applied at or just after adult emergence, they failed to delay the timing of the onset of the cellular degeneration of the fibers. Thus, the ability of the inhibitors to delay the death of the fibers of the maintained ISMs is lost many hours before the first signs of the onset of cellular degeneration. In the animals treated with the inhibitors at the time of emergence, the finding that the final removal of the maintained ISMs occurs does not necessarily mean that proteasomal activity does not play a critical role in mediating the cellular degeneration of the fibers. Given the dramatic increase in the expression of the components of the proteasomal pathway, the doses of inhibitors used may not have been sufficient to yield a significant reduction in the level of enzymatic activity.

Proteasomal Activity May Act Independently of Hormonal and Neural Control of Rapid Atrophy

The application of inhibitors of proteasomal activity is just one of several treatments that are known to delay the final removal of the maintained ISMs. The maintained ISMs become committed to initiate cell death when the circulating titer of ecdysteroids falls below a critical point, some 12 to 18 hr before adult emergence (Schwartz and Truman, 1983; Schwartz et al., 1990, 1993). Triggering a precocious drop in the circulating titer of ecdysteroid accelerates the rate of ISM regression, while preventing the decline in the ecdysone titer through the administration of exogenous hormone slows the rate of ISM regression. Manipulating the titer of the ecdysteroids also influences that pattern of protein expression within the maintained ISMs. Thus, it is plausible that the application of proteasomal inhibitors might slow ISM regression by either delaying the decline in the ecdysteroid titer or interfering with the ability of the ISMs to respond to the falling titer of ecdysteroids. There is no evidence supporting a role of the proteasomal inhibitors blocking or slowing the decline in the titer of ecdysteroids. In addition to triggering the rapid atrophy of the ISMs, the decline in the titer of ecdysteroids during the final stages of adult development is responsible for triggering several other developmental events around the time of adult emergence. These events include the development of wing pigmentation and triggering the series of behaviors that allow the adult to emerge from the old pupal cuticle (Schwartz and Truman, 1983). These and other ecdysteroid-regulated events still occurred in the animals treated with the inhibitors. However, it is possible that the inhibitors act by disrupting a tissue-specific response to the declining titer of ecdysteroids. For example, the inhibitors could interfere with the proper regulation of the expression ecdysteroid receptors or a transcription factor required for the ISMs to respond to the falling titer of ecdysteroids.

Enhanced motor activity can also slow the rate of ISM regression after adult emergence motor activity (Truman, 1983; Lockshin and Williams; 1965b, c; Lockshin, 1971). Manduca pupate underground and use the large abdominal ISMs to help it dig to the surface after adult emergence. If freshly emerged adults are forced to dig for an extended period of time, the complete removal of the ISMs can be delayed for as much 96 hr. The direct electrical stimulation of the muscles and the application of pharmacological agents that promoted muscle activity also slow the rate of ISMs regression after adult emergence (Lockshin and Williams, 1965c; Lockshin, 1971). However, it does not appear that the inhibitors slow ISM regression by directly enhancing motor activity. The activity-dependent rescue of the ISMs requires that the ISMs be functionally innervated (Lockshin, 1971). Both denervation and the addition of atropine blocked the activity-dependent decline in the rate of ISM regression (Lockshin and Williams, 1965c, Lockshin, 1971). The ability of the proteasomal inhibitors does not appear to require that the ISMs be functionally innervated. The response of the ISMs to the application of the inhibitors was unchanged after their denervation (Bayline, unpublished results).

In summary, we have demonstrated that inhibitors of proteasomal activity can delay the onset of ISM cell death that occurs after adult emergence. The decision to initiate cell death appears to be made at the level of the individual muscle fibers. The muscle fibers are most sensitive to the inhibitors during a 24-hr period, with the response declining around the time the muscle fibers become committed to undergo cell death. The inhibitors appear to block an early step in the cell death program of the maintained ISMs. This conclusion is based on the ability of the inhibitors to block several of the changes in protein and RNA expression that precede the first signs of the cellular degeneration of the fibers of the maintained ISMs.

EXPERIMENTAL PROCEDURES

Animals

Individual larvae were raised at 27°C under a 16:8 hr light:dark cycle on an artificial diet (Bell and Joachim, 1976). Approximately 21 days after hatching, larvae underwent pupation, and approximately 20 days later, the adults emerged. The animals were staged relative to the time of adult emergence. Over the last 4 days before adult emergence, the animals were staged according to changes in the wing pigmentation and pupal cuticle degradation as previously described by Schwartz and Truman (1983). Thus, an animal that displayed the wing pigmentation characteristic of an animal that would undergo adult emergence in 48 hr was staged as AE-48 (adult emergence minus 48 hr). After emergence, the animals were staged as the hours after adult emergence. For example, adults that had emerged 48 hr previously were staged as AE+48.

Hemin and ALLN Injections

To determine the role of the ubiquitin proteolytic pathway in the PCD of the Manduca ISMs, animals were given a single injection of hemin chloride (Sigma) or of ALLN (Sigma). The stages injected ranged from AE-96 to within 4 hr after adult emergence. The hemin solution was prepared by adding 32.6 mg of hemin to 100 μl of 0.1 N NaOH. The solution was vortexed and raised to 1 ml with dH2O, and the solution was vortexed again. The hemin solution was then centrifuged, and the supernatant was collected. CO2-anesthetized pharate adults were given an injection of 12.5–45 μl of hemin solution per gram body weight into the abdominal cavity using a syringe. The stages injected ranged from AE-96 to AE-4. The treated animals were then killed between AE+0 and AE+168 hr (0–7 days) after adult emergence. ALLN was dissolved 10 mg/ml in dimethyl sulfoxide (DMSO) and stored at −20°C until use. For experiments, 2 parts stock was diluted into 3 parts DMSO or EB saline and between 5 and 20 μl of solution/g body weight was injected as described above. As a control, animals were injected with solutions prepared as above but without the inhibitors. The experimental animals were killed between AE+0 and AE+168 hr (0–7 days) after treatment.

Muscle Measurements

Abdomens were removed from staged animals, cut along the dorsal midline, stretched, and pinned to Sylgard-lined dishes (Dow–Corning) and rinsed in insect saline (Ephrussi and Beadle, 1936). After the removal of the gut, the tissue was fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS: 0.01 M phosphate buffer, pH 7.4, 137 mM NaCl, 2.7 mM KCl) overnight at room temperature. After fixation, the ISMs were rinsed in PBS then covered with insect saline and the diameter of each fiber of each of the ISMs was measured near its midpoint using a stereomicroscope equipped with an ocular micrometer. The sum of the diameters of all fibers was used as an estimate of the total diameter of the muscle.

Whole-Mount Histology

Fragments of the fixed muscle were surgically removed from the abdomens. The tissue was stained for 7 min in a solution of cresyl violet acetate (Sigma), prepared by diluting a saturated solution of the dye (in 50% ethanol) with 3 volumes of 50% ethanol. The stained muscle fragments were dehydrated through an ethanol series, cleared in methyl salicylate, and mounted in Permount (Fisher) for viewing on a compound microscope using interference contrast optics.

Sectioning

Tissue from both control and treated animals was prepared for sectioning as described in Fahrbach and Schwartz (1994). Briefly, dorsal internal medial muscles from segments A2 and A4 (DIM-A2 and DIM-A4) were fixed in either 4% paraformaldehyde in PBS or alcoholic Bouin's fixative (1 g of picric acid; 70% ethanol; 25% formalin; 5% glacial acetic acid) for 24 hr at room temperature. The Bouin's-fixed tissue was rinsed repeatedly in 70% ethanol saturated with LiCO3 to remove the excess picric acid, whereas the paraformaldehyde-fixed tissue was rinsed in PBS. After dehydration through a graded ethanol series, the tissue was cleared in xylene and embedded in TissuePrep (Fisher). Ten-micrometer cross-sections of the muscle fibers were cut and affixed to poly-L-lysine (0.1 mg/ml) -coated slides.

Immunocytochemistry

The expression of ubiquitin was monitored using an antibody that was specific to the conjugated form of the protein (kindly provided by Dr. A.L. Haas; Haas and Bright, 1985; Riley et al., 1988). The immunocytochemistry was performed using a method similar to that outlined in Fahrbach and Schwartz (1994). Briefly, the sectioned tissue was rehydrated and blocked in a solution of PBS + 0.3% Triton X-100 (PBS-T) containing 10% normal goat serum (NGS), 5% nonfat dry milk, 0.5% bovine serum albumin, and for 1 hr. The tissue was then incubated in a 1:3,000 dilution of the anti-ubiquitin antibody in PBS-T with 1% NGS overnight at 4°C. The sections were rinsed and incubated in a 1:200 dilution of biotinylated goat anti-rabbit antibody (Vector) for 4 hr at room temperature. The sections were rinsed in high-salt PBS (PBS with 0.5 M NaCl) and incubated in ABC reagent (Vector Labs) for 2 hr at room temperature. After rinsing, the histochemistry was performed using standard techniques. The sections were dehydrated, cleared in xylene, and mounted under Permount (Fisher).

RNA Isolation and Northern Blots

ISMs from segments A4 to A6 were dissected under saline from a cold-anesthetized staged animal. The tissue from a single individual was homogenized in 1 ml of Tri Reagent (Molecular Research Corporation), and total RNA was isolated using standard protocols. For detection of actin mRNA, a single-stranded DNA oligonucleotide complementary to positions 720-770 of the published cDNA sequence for Manduca actin (Schwartz et al., 1993) was synthesized (Cornell University's BioResource Center). The 3′ end of the oligonucleotide was labeled with Biotin dT (Glen Research Corporation). Prehybridizations and hybridizations were carried out using standard procedures (Sambrook et al., 1989).

Acknowledgements

The authors thank Dr. Carol Miles for critical review of this manuscript. R.J.B. and D.M.D. were supported by the NIH.

Ancillary