PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of hmgLink to Publisher's site
 
Hum Mol Genet. 2009 December 1; 18(23): 4467–4477.
Published online 2009 August 19. doi:  10.1093/hmg/ddp399
PMCID: PMC2773263

Ku70 regulates Bax-mediated pathogenesis in laminin-α2-deficient human muscle cells and mouse models of congenital muscular dystrophy

Abstract

The severely debilitating disease Congenital Muscular Dystrophy Type 1A (MDC1A) is caused by mutations in the gene encoding laminin-α2. Bax-mediated muscle cell death is a significant contributor to the severe neuromuscular pathology seen in the Lama2-null mouse model of MDC1A. To extend our understanding of pathogenesis due to laminin-α2-deficiency, we have now analyzed molecular mechanisms of Bax regulation in normal and laminin-α2-deficient muscles and cells, including myogenic cells obtained from patients with a clinical diagnosis of MDC1A. In mouse myogenic cells, we found that, as in non-muscle cells, Bax co-immunoprecipitated with the multifunctional protein Ku70. In addition, cell permeable pentapeptides designed from Ku70, termed Bax-inhibiting peptides (BIPs), inhibited staurosporine-induced Bax translocation and cell death in mouse myogenic cells. We also found that acetylation of Ku70, which can inhibit binding to Bax and can be an indicator of increased susceptibility to cell death, was more abundant in Lama2-null than in normal mouse muscles. Furthermore, myotubes formed in culture from human laminin-α2-deficient patient myoblasts produced high levels of activated caspase-3 when grown on poly-l-lysine, but not when grown on a laminin-α2-containing substrate or when treated with BIPs. Finally, cytoplasmic Ku70 in human laminin-α2-deficient myotubes was both reduced in amount and more highly acetylated than in normal myotubes. Increased susceptibility to cell death thus appears to be an intrinsic property of human laminin-α2-deficient myotubes. These results identify Ku70 as a regulator of Bax-mediated pathogenesis and a therapeutic target in laminin-α2-deficiency.

INTRODUCTION

Aberrant activation of a mitochondrial cell death pathway is a significant contributor to pathogenesis in mouse models of several neuromuscular diseases including laminin-α2-deficiency (Congenital Muscular Dystrophy Type 1A; MDC1A), Oculopharyngeal Muscular Dystrophy (OPMD), Collagen VI-deficiency (Bethlem or Ullrich Congenital Muscular Dystrophy; BCMD, UCMD) and Spinal Muscular Atrophy (SMA) (18). In previous studies, for example, we showed that inhibiting the mitochondrial cell death pathway through genetic or pharmacological interventions decreases pathology and more than doubles lifespan in laminin-α2-deficient (Lama2−/−) mice which are a model for human MDC1A (4,5,9). Because anti-cell death strategies ameliorate pathology in models of several skeletal muscle diseases, it may be that similar molecular mechanisms underlie excessive activation of cell death pathways in pathogenesis. If so, a single therapeutic approach could be of benefit in multiple diseases.

Further development of anti-cell death therapeutic strategies for neuromuscular diseases will benefit from additional understanding of cell death pathways in diseased skeletal muscles. From previous studies which showed that inactivation of the pro-apoptotic protein Bax is sufficient to significantly decrease pathology in mouse models of MDC1A and SMA (4,8), it appeared that Bax is a key promoter of aberrant cell death in at least these two disease models. The mechanisms that lead to Bax activation and cell death in neuromuscular diseases such as laminin-α2-deficiency remain, however, incompletely understood. In this work, therefore, we have examined molecular mechanisms that regulate Bax and cell death in normal and laminin-α2-deficient skeletal muscles and cells.

The most common form of MDC1A is caused by mutation of the human LAMA2 (mouse Lama2) gene that encodes laminin-α2 (1011). Loss of laminin-α2 function in this autosomal recessive disease results in severe neuromuscular dysfunction. Partial laminin-α2-deficiency is found in some other diseases, including Muscle–Eye–Brain disease, Fukuyama Congenital Muscular Dystrophy, and Walker-Warburg Syndrome, though these diseases are caused by primary mutations in genes other than LAMA2 (2). In skeletal muscles, laminin-α2 assembles with laminin-β1 and -γ1 to form laminin-211. Heterotrimeric laminins that include laminin-α2 have been termed merosins, and MDC1A has thus also been known as merosin-deficient congenital muscular dystrophy. Laminin-α2 has binding partners in both the extracellular matrix—entactin/nidogen, which in turn binds to collagen IV—and on the plasma membrane—α-dystroglycan and α7-integrin (1). Laminin-α2-deficiency causes aberrant activation of Bax-mediated muscle cell death (4,5,1213).

Recent work in non-skeletal muscle cells has shown that interaction of Bax with the Ku70 protein is an additional mechanism to regulate cell death (1417). Under normal conditions, Ku70 binds to Bax, and Bax is thereby retained in the cytoplasm in an inactive form. In response to a pro-death signal, however, acetylation of one or more critical lysines on Ku70 occurs and this modification inhibits Ku70 binding and frees Bax for translocation to mitochondria (18). Ku70 is acetylated by CBP and PCAF and deacetylated by SIRT1, SIRT3 and probably other deacetylases (14,18,19). Ku70 is a multifunctional protein that is found in both nuclei, where it participates in DNA break repair, and in the cytoplasm (15). Ku70 also appears to act as a Bax deubiquitinase (20). To determine if Ku70 regulates Bax-mediated pathogenesis in laminin-α2-deficiency, we have now analyzed the Ku70/Bax pathway in normal and laminin-α2-deficient myogenic cells and tissues. Taken together, our results identify Ku70 as a regulator of Bax-mediated pathogenesis in laminin-α2-deficiency and the Bax/Ku70 pathway as a therapeutic target in laminin-α2-deficiency.

RESULTS

To analyze the role of Ku70 in normal and diseased skeletal muscles, we studied systems with complementary experimental advantages, including cells of the mouse C2C12 myogenic cell line; primary cells derived from muscles of wild-type and genetically-modified mice; muscle tissues from Lama2−/− mice; and human myogenic cells obtained both from healthy individuals and from three patients with a clinical diagnosis of MDC1A (Table 1 and Materials and Methods). Use of the mouse-derived materials circumvented the experimental limits imposed by the limited mitotic capability of the human cells and the lack of human muscle biopsy material. We first used immunoblots to demonstrate that Ku70 was expressed in both the cytoplasm and nuclei of cultured primary mouse myoblasts (Fig. 1A). Less Ku70 appeared to be in the cytoplasm in Bax-null than in wild-type cells, a difference which is consistent with the possibility that the cytoplasmic Ku70 was more rapidly degraded or translocated to the nucleus when binding to Bax was not possible.

Figure 1.
Ku70 expression in normal and diseased muscle cells and tissue. (A) In primary cultures of mouse myoblasts, Ku70 was found in both nuclear and cytoplasmic fractions, with relatively less Ku70 in the cytoplasm of Bax-null than in wild-type cells. Primary ...
Table 1.
Clinical data on laminin-α2-deficient and normal human myoblast donors

We next used combined immunoprecipitation and immunoblotting to demonstrate that Ku70 and Bax interact in mouse skeletal muscle tissue (Fig. 1B). Extracts of 6- or 12-week-old mouse soleus muscles were immunoprecipitated with an antibody specific for Bax. Subsequently, the immunoprecipitated proteins were subjected to SDS-PAGE and immunoblotting with antibodies specific for Ku70 or Bax. Ku70 was found with Bax in the precipitates, showing that Ku70 interacted with Bax either most likely by direct binding, as found in other cell types, or indirectly as part of a multi-subunit complex.

To determine if laminin-α2-deficiency altered the interaction of Bax and Ku70, we compared the amounts of Ku70 that co-immunoprecipitated with Bax from wild-type and Lama2−/− mouse muscle extracts. We found a consistent decrease in the amount of Ku70 that was co-immunoprecipitated with Bax from extracts of Lama−/− soleus muscle compared with wild-type (Fig. 1B and C). This result suggests that a smaller percentage of Bax is in a complex with Ku70 in Lama−/− than in wild-type muscles. Because dissociation of Ku70 from Bax is associated with increased cell death, this result is consistent with the high levels of cell death previously found in Lama−/− mouse muscles (5).

We next used Bax-inhibiting peptides (BIPs) to examine the role of Ku70 in muscle cell death. BIPs are simple pentapeptides with a sequence from the region of Ku70 that binds to Bax. Previous work in non-muscle cells has shown that BIPs can bind to Bax and can inhibit apoptotic cell death by inhibiting Bax activation and translocation to mitochondria (15,16,21). We synthesized BIPs based on mouse and human Ku70 sequences either as the simple pentapeptide or with a novel modified structure that included fluorescein isothiocyanate (FITC, to demonstrate cell permeability) linked by beta-alanine to the BIP pentapeptide sequences. Thus, the mouse-specific BIP (mBIP) was the simple pentapeptide VPTLK and the fluorescent version (fl-mBIP) was FITC-ßA-VPTLK, whereas the human-specific BIP (hBIP) was PMLKE and the fluorescent version (fl-hBIP) was FITC-ßA-PMLKE. A pentapeptide, IPMIK that does not have Bax-inhibiting activity was similarly prepared with or without the FITC tag for use as a control. FITC-tagged BIPs were used for the short-term experiments described in Figures 2 and and3,3, whereas untagged BIPs were used for the longer duration experiments described in Figures 6 and and77.

Figure 2.
Mouse Bax-inhibiting peptide (mBIP) inhibited staurosporine-induced nuclear abnormalities in C2C12 cells. (A) After 6 h treatment with 0.1 mm staurosporine (STS), a large percentage of the nuclei in cultures of C2C12 myoblasts showed abnormal morphology ...
Figure 3.
In myogenic cells, mBIP inhibited staurosporine-induced translocation of Bax. (A) Myoblasts of the mouse C2C12 myogenic cell line were left untreated (None) or treated with 0.1 mm staurosporine (+STS) for 6 h either alone in or combination with 200 µ ...
Figure 6.
Caspase-3 activation in myotubes formed from human laminin-α2-deficient myoblasts was inhibited by hBIP or growth on laminin-α2-containing substrate. (AC) Normal human myoblasts of line 363/07 (A) and two independent lines of ...
Figure 7.
Caspase-3 enzymatic activity was increased in myotubes formed from human MDC1A myoblasts, but this increase was inhibited by culture with hBIP or growth on laminin-α2-containing substrate. Quantitative caspase-3 enzyme assays were performed on ...

Our initial experiments showed that treatment with fl-mBIP, when compared with treatment with control peptide, significantly decreased the percentage of C2C12 myoblasts with abnormally-shaped nuclei (i.e. condensed, fragmented or with an irregular or blebbed perimeter) that were induced by staurosporine (Fig. 2). Staurosporine is a non-selective kinase inhibitor that induces Bax-mediated apoptotic cell death (cf. 22). We previously showed that induction by staurosporine of such nuclear abnormalities correlates with the induction of Annexin V staining and subsequent cell death (5). In our earlier study, we showed that myoblasts from Bax-null mice were more resistant to staurosporine than myoblasts from Bax+/+ wild-type mice (22). Consistent with that study, we found that nuclear abnormalities induced by staurosporine in C2C12 myoblasts, which are Bax-positive, was likely Bax-dependent, because staurosporine treatment induced a significantly higher percentage of abnormal nuclei in C2C12 myoblasts than in Bax-null primary myoblasts (Fig. 2C). All cells in fl-mBIP-treated cultures showed cytoplasmic FITC fluorescence, indicating that the peptides entered cells (not shown). Treatment with fl-mBIP also inhibited staurosporine-induced abnormalities in C2C12 myotubes (not shown).

Staurosporine-induced translocation of Bax was also inhibited by fl-mBIP treatment (Fig. 3A). We determined the subcellular distribution of Bax in C2C12 myoblasts that were left untreated or treated with 0.1 mm staurosporine for 6 h alone in or combination with either mBIP or the control peptide. At the end of treatment, cells were divided into two fractions, one enriched for cytoplasmic proteins and a second membrane fraction that was enriched for mitochondria. The fractions were analyzed by SDS-PAGE and quantitative immunoblotting for Bax. In addition, immunoblots for ATP synthase (mitochondrial marker) and lactate dehydrogenase (LDH, cytoplasmic marker) confirmed successful cell fractionation (Fig. 3B). The percentage of Bax in the mitochondrial-enriched fraction was significantly increased upon staurosporine treatment, consistent with the expected translocation of Bax from the cytoplasm to mitochondria upon induction of cell death (Fig. 3A). This staurosporine-induced translocation of Bax was prevented by treatment with mBIP but not with the control peptide (Fig. 3A).

Having established that Ku70 is expressed in myogenic cells and that it appeared to regulate Bax-dependent muscle cell death (based on the inhibition of induced nuclear abnormalities and Bax translocation by mBIP), we performed further experiments to determine if Ku70 regulation of Bax might be abnormal in laminin-α2-deficient muscle cells. For these studies, we further examined muscle tissues from Lama2−/− compared with wild-type mice; and we analyzed cultures of myogenic cells obtained from patients with a clinical diagnosis of MDC1A compared with cells from healthy donors. Details of the human myoblast donors are presented in Table 1 and in the Materials and Methods.

In mouse muscles, we found that significantly more acetylated Ku70 was present in Lama2−/− than in wild-type muscles (Fig. 4). We used soleus muscles for these assays because the time course and severity of pathogenesis is known (4,5) and this muscle has a mixture of slow and fast fiber types as is the case with human muscles. We prepared extracts of soleus muscles from Lama2−/− and wild-type mice and carried out immunoprecipitation with an antibody specific for Ku70. The immunoprecipitates were analyzed by SDS-PAGE and immunoblotting with antibodies specific for Ku70 or acetyl-lysine. Quantitative immunoblots showed that acetylated Ku70 was ~4X more abundant in Lama2−/− muscles than in wild-type muscles, whereas the amount of total Ku70 was similar in Lama2−/− and wild-type muscles (Fig. 4). A previous study of cultured non-muscle cells demonstrated that Ku70 contains eight lysines that can be acetylated and that acetylation of K539, K542, or to a lesser extent K553, appears to inhibit Ku70 binding to Bax and is associated with increased cell death (18). Though the acetyl-lysine antibody we used was not specific for the lysine residues on Ku70 that are critical for Bax binding, the increased acetylation of Ku70 in Lama2−/− muscles is nonetheless consistent with the aberrantly high level of Bax-dependent cell death in the Lama2−/− mouse model of MDC1A (cf. 4, 5).

Figure 4.
Increased Ku70 acetylation in Lama2−/− mouse muscle. Extracts of soleus muscles were prepared from wild-type (WT) and Lama2−/− mice and immunoprecipitated (IP) with an antibody specific for Ku70. The immunoprecipitates ...

To determine if the Ku70/Bax pathway might contribute to pathogenesis in human, as well as mouse, laminin-α2-deficient muscle cells, we compared cultures of proliferating and differentiated myogenic cells that were obtained from three independent patients with a clinical diagnosis of MDC1A to similar cultures of myoblasts obtained from healthy human donors (Materials and Methods; Table 1). Our initial characterization confirmed that the patient cells were myogenic and had the expected laminin-α2-deficient phenotype (Fig. 5). More than 90% of the cells in low density, proliferating cultures of both the wild-type and laminin-α2-deficient cells were desmin-positive myoblasts (Fig. 5A and D). When grown on a poly-l-lysine substrate, both the normal human and the patients' laminin-α2-deficient myoblasts were capable of forming multinucleate myotubes that expressed myosin heavy chain within 3–4 days after switching to low serum differentiation medium (Fig. 5B and E). Finally, laminin-α2 was expressed by myotubes formed from wild-type human myoblasts, but, as would be expected for patients with MDC1A, laminin-α2 was not detectable in myotubes formed from each of the three patient lines (Fig. 5C and F and not shown). Though the clinical observations, including early onset and apparently complete lack of laminin-α2 protein, support the MDC1A diagnosis, we currently have mutation information for only one of the three lines (see Materials and Methods) and causality still needs to be definitively established. Thus, we will refer to the human cells as laminin-α2-deficient, though we expect that they are from bona fide MDC1A patients.

Figure 5.
Myotubes formed from myoblasts obtained from donors with a diagnosis of MDC1A did not express laminin-α2. Myoblasts obtained from humans with normal laminin-α2 expression and from patients with a clinical diagnosis of MDC1A were cultured ...

Though human laminin-α2-deficient myoblasts formed myotubes when cultured on poly-l-lysine, these myotubes exhibited several abnormalities indicative of instability and increased susceptibility to cell death. First, we noted that many of the laminin-α2-deficient myotubes, but not normal myotubes, became rounded and began to detach from the dish after 4 or 5 days in differentiation medium. Second, immunoprecipitation and immunoblotting analyses of Ku70 expression and acetylation similar to those in Figure 4 showed that the ratio of acetylated Ku70 to total Ku70 was 2.8 ± 0.55 (Ave. ± SE, n = 3) times higher in cultures of laminin-α2-deficient myotubes than in cultures of normal myotubes. This increase in acetylated Ku70 in the laminin-α2-deficient human myotubes was similar in magnitude to that found in the Lama2−/− mouse muscles (Fig. 4) Third, Annexin V bound to many laminin-α2-deficient myotubes, but not to normal myotubes, which is consistent with ongoing cell death of laminin-α2-deficient myotubes (not shown). Finally, laminin-α2-deficient myotubes, but not normal myotubes, showed high levels of activated caspase-3 (a marker for incipient cell death) at 4–5 days after they were formed (Figs 6 and and7).7). After 5 days in differentiation medium, for example, we found that many of the myotubes formed from human laminin-α2-deficient myoblasts showed intense immunostaining for activated caspase-3 (Fig. 6B and C), whereas myotubes formed from wild-type myoblasts did not (Fig. 6A). Myotubes formed from each of the three available lines of patient myoblasts (which were obtained from three individual donors) showed similar patterns of intense immunostaining for activated caspase-3 when grown on poly-l-lysine (Fig. 6B and C, and not shown). Quantitative measurements of caspase-3 enzyme activity also demonstrated that caspase-3 enzymatic activity was significantly higher in differentiated laminin-α2-deficient than in normal cultures (Fig. 7). (N.B., because Lama2−/− mouse myoblasts did not adhere to poly-l-lysine, we could not carry out a similar experiment with primary cultures of mouse myogenic cells.)

The increased level of activated caspase-3 seen in laminin-α2-deficient myotubes was not found, however, when the patients' myogenic cells were either grown on a substrate that contained laminin-α2 or were treated with the human Bax-inhibiting peptide (hBIP) (Figs 6D, E and and7).7). To determine if provision of laminin-α2 could prevent caspase-3 activation, we cultured patient cells on a substrate of human placental laminin, which contains laminin-α2 as a subunit of laminin-211 and laminin-221. Both immunostaining and enzyme assays showed that differentiated laminin-α2-deficient cultures grown on the human placental laminin substrate had approximately the same caspase-3 levels as wild-type cells, which was significantly less than in the laminin-α2-deficient cultures grown on poly-l-lysine (Figs 6D and and7).7). Similarly, differentiated laminin-α2-deficient cultures grown on poly-l-lysine in the absence laminin-α2 but in the presence of hBIP had about the same caspase-3 levels as wild-type cells, which was significantly less than in laminin-α2-deficient cultures grown similarly in the absence of hBIP (Figs 6E and and77).

When we examined Ku70 expression by immunostaining, we found that, when compared with normal myotubes, myotubes formed from the laminin-α2-deficient patient myoblasts had an altered subcellular distribution of total Ku70. Laminin-α2-deficient myotubes showed staining for total Ku70 in all nuclei but showed little staining in the cytoplasm of myotubes, whereas normal myotubes showed staining for total Ku70 in both the cytoplasm and nuclei (Fig. 8). In contrast to the different patterns in myotubes, myoblasts from both normal and laminin-α2-deficient lines showed similar staining patterns with signals in both cytoplasm and nuclei (not shown). Quantitative immunoblots of cytosolic and nuclear fractions purified from differentiated cultures confirmed that differentiated laminin-α2-deficient cells had a significantly smaller percentage of the total Ku70 in the cytoplasm than did normal myotubes (Fig. 9A and B.) Furthermore, Ku70 acetylation patterns were also altered in differentiated laminin-α2-deficient cells (Fig. 9A and C). In the cytosolic fraction, the ratio of acetylated Ku70 to total Ku70 was ~2.5X–4X higher for each of the three laminin-α2-deficient patient lines than in cultures of healthy cells (Fig. 9C). In the nuclear fraction, in contrast, the ratio of acetylated Ku70 to total Ku70 was lower in cultures of each of the three laminin-α2-deficient patient cultures, ranging from 0.49X to 0.74X the ratio in normal cell cultures (Fig. 9C). Thus, cytoplasmic Ku70 in human laminin-α2-deficient myotubes was both reduced in amount and more highly acetylated than in normal myotubes.

Figure 8.
The intracellular distribution of total Ku70 was altered in human laminin-α2-deficient myotubes. Myoblasts from normal lines 363/07 and 2/08 (A, B) and laminin-α2-deficient line 96/04 (Lama2-def; C, D) human donors were cultured on poly- ...
Figure 9.
Acetylation and intracellular distribution of Ku70 was altered in human laminin-α2-deficient myotubes. (A) Differentiated laminin-α2-deficient cultures had less Ku70 in the cytosol than normal cell cultures. Cytosolic (Cyt) and nuclear ...

DISCUSSION

The results of this study identify the multifunctional protein Ku70 as a regulator of Bax-mediated pathogenesis and a possible therapeutic target in laminin-α2-deficiency. Our studies of normal and laminin-α2-deficient human myogenic cells, complemented by assays of normal and Lama2−/− mouse cells and tissues, produced an outline of pathological changes in the Ku70/ Bax pathway. In particular, we found increased acetylation of Ku70 in Lama2−/− mouse muscles, increased acetylation and loss of Ku70 from the cytoplasm of human laminin-α2-deficient myotubes, increased activation of caspase-3 in human laminin-α2-deficient myotubes and inhibition of this caspase-3 activation by hBIP. All of these results are consistent with a mechanism of pathology in which human laminin-α2-deficient myotubes are intrinsically more susceptible to cell death because lack of laminin-α2, perhaps due to aberrant signaling through its unliganded receptors, causes hyperacetylation of Ku70 which in turn makes it more likely that Bax can be activated for mitochondrial translocation and induction of cell death.

MDC1A is a severely debilitating, autosomal recessive disease caused by LAMA2 mutations that compromise function of the laminin-α2 protein. Previously, loss of functional laminin-α2 had been shown to lead to increased cell death in cultures of mouse myogenic cells and in laminin-α2-null mice (5,12). This aberrant induction of cell death was shown to occur through a mechanism that is mediated by activation of the pro-apoptotic protein Bax (4,5,9,1213,23), as shown, for example, by our finding that genetic inactivation of Bax function is sufficient to significantly extend lifespan and ameliorate pathology in laminin-α2-null mice (4,5). In this study, we sought to extend our understanding of Bax-mediated pathogenesis in laminin-α2-deficiencies by identifying mechanisms underlying the aberrant activation of Bax in laminin-α2-deficient model systems.

Activation of the mitochondrial pathway of cell death, as in laminin-α2-deficiency, can occur when Bax undergoes a conformational change and translocates to the mitochondria (reviewed, 24). Though Bax interacts with multiple other members of the Bcl-2 family, including Bak and BH3-only proteins, it also appears to interact with additional non-Bcl-2 family proteins (24). Among these Bax-interacting proteins is the multifunctional protein Ku70. Previous studies had shown that Ku70 interacts with Bax to regulate cell death in some types of non-muscle cells such as neural cells (16), though the possible function of Ku70 in normal and diseased skeletal muscle had not previously been studied. In the nucleus, Ku70 is a component of the protein complex that repairs DNA double strand breaks, but a large fraction of Ku70 is also found in the cytoplasm where Ku70 binding to Bax appears to inhibit Bax activation and translocation (16,18). This Bax inhibitory function is regulated by post-translational acetylation of specific Ku70 lysines which prevent binding of Ku70 to Bax (18), thereby freeing Bax to undergo the activating conformational change and translocation to mitochondria in response to pro-apoptotic signals. Because Bax-mediated cell death is a significant contributor to pathology in laminin-α2-deficiency, these previous findings in non-muscle cells prompted us to examine the role of Ku70 in normal and diseased skeletal muscle.

Our finding that BIPs can inhibit signs of induced and spontaneous cell death provides strong evidence that Ku70 interaction with Bax regulates cell death in normal and in laminin-α2-deficient skeletal muscle cells. The BIPs were designed to bind to and prevent activation of Bax even under pro-cell death conditions when binding of Ku70 to Bax is inhibited by acetylation (15,16,21). Here we found that appropriate species-specific BIPs inhibited both the staurosporine-induced cell death in normal mouse myogenic cells and the aberrant activation of caspase-3 that occurs in human myotubes formed from laminin-α2-deficient patient myoblasts in culture.

In addition to the BIP studies, our work provided additional evidence that Ku70 regulates Bax-mediated cell death in diseased skeletal muscle. In particular, we found that Lama2−/− muscles and human laminin-α2-deficient myotube cultures contained higher levels of acetylated Ku70, particularly in the cytoplasm, than did the corresponding normal muscles or myotube cultures. Furthermore, we found that Ku70 co-immunoprecipitated with Bax from extracts of mouse skeletal muscles, but that less Ku70 co-immunoprecipitated with Bax from Lama2−/− than normal muscles. The combination of a high level of Ku70 acetylation with less binding of Ku70 to Bax would be expected for tissues undergoing high levels of Bax-mediated cell death, which is the case in Lama2−/− mouse muscles (4,5). Thus, the differences that we found between normal and Lama2−/− muscles in Ku70 acetylation and Bax binding are consistent with a role for Ku70 in regulation of Bax-mediated cell death in skeletal muscles.

Our studies of human laminin-α2-deficient myogenic cells in culture are the first to show that these cells recapitulate an important disease phenotype. Previous important studies with laminin-α2-deficient variants of the mouse C2C12 myogenic cell line showed that myotubes were unstable in the absence of laminin-α2 and showed high levels of cell death through the mitochondrial pathway (1213,23). Those studies also identified kinase signaling pathways involved in cell death induced by laminin-α2-deficiency and showed that the high levels of cell death could be inhibited by overexpression of Bcl-2 (which inhibits Bax function) or by differentiation in the presence of laminin-α2. Here we showed that, when grown in the absence of laminin-α2, the human laminin-α2-deficient myoblasts similarly formed unstable myotubes and that these myotubes developed high levels of activated caspase-3 after 4–5 days in differentiation medium. The laminin-α2-deficient myotubes also showed increased levels of acetylated Ku70 and Annexin V staining compared with normal myotubes in culture. Though low levels of activated caspase-3 have been reported to be required for myogenic cell differentiation (25), the high levels of activated caspase-3 in laminin-α2-deficient myotubes were clearly abnormal. A complete absence of laminin-α2, as in the cells from patients with a MDC1A diagnosis that we studied, may be necessary to produce the abnormal myotube phenotypes we observed. A previous study did not find abnormalities in myotubes that were formed from myoblasts of a patient with a different form of congenital muscular dystrophy (not due to LAMA2 mutation) in which laminin-α2 processing was altered but expression continued (26). In addition, the differentiation state of cells may affect outcome, as a study of mitochondria in one line of MDC1A myoblasts did not find abnormalities (27), though myotubes were not examined. The abnormal myotube phenotypes that we identified were consistent over three independent lines of laminin-α2-deficient myoblasts and could be prevented either by growth on a laminin-α2-containing substrate or by treatment with hBIP. Thus, the human laminin-α2-deficient myogenic cell cultures appear to provide a relevant system in which to identify pathological mechanisms and test therapeutic approaches.

Our findings identify Ku70 regulation of Bax as potential therapeutic target for MDC1A and perhaps additional neuromuscular diseases. Genetic and pharmaceutical approaches that inhibit disease-induced mitochondrial abnormalities have been shown to improve outcomes in mouse models of multiple neuromuscular diseases including MDC1A, OPMD, SMA and UCMD (18). Developing therapies based on the Ku70/Bax pathway could include testing BIP function in vivo. One study, for example, has shown that BIPs delivered into the eye can inhibit experimentally-induced retinal ganglion cell death (28), but whether administration through the circulation is possible is not known. In addition, because acetylation of Ku70 is central to regulation of Bax activation, an understanding of the specific acetyltransferases and deacetylases that act on Ku70 in diseased skeletal muscles could provide additional targets for therapeutic development. In non-muscle cells, Ku70 is acetylated by CBP and PCAF and deacetylated by SIRT1, SIRT3 and probably other deacetylases (14,16,18,29). In normal and diseased skeletal muscles, however, it remains to be determined which specific deacetylases and acetyltransferases regulate Ku70 functions and where within the cell they function. Therapeutic strategies that preserve normal mitochondrial function by inhibiting aberrantly activated cell death pathways could have significant therapeutic benefit.

MATERIALS AND METHODS

Mice

Heterozygous Lama2dy-W/+ mice, which carry the targeted dy-W mutation in the Lama2 gene (30), have been maintained in our laboratory for >5 years by breeding with C57BL/6J mice (4,5,9). Mice obtained from crosses of Lama2+/− heterozygotes were genotyped at weaning by PCR (4). Breeding pairs of heterozygous B6.129X1-Baxtm1Sjk mice (31) were obtained from the Jackson Laboratory and mated in our laboratory. Progeny were genotyped (4), and muscle tissue or cells for primary cultures were obtained from tissues of the resulting wild-type or Bax-null littermates at 4–6 weeks of age.

Cells and culture

Myoblasts derived from biopsies obtained from three donors with a clinical diagnosis of MDC1A and from two normal individuals were provided by the Muscle Tissue Culture Collection at the University of Munich. Because of the young age of the MDC1A donors, it was not possible to obtain age-matched normal donors. All experiments with primary human cells were performed with cells at no later than 25 population doublings. All available clinical observations on the donors are shown in Table 1. For MDC1A line 96/04, direct sequencing of LAMA2 gene exons and short stretches of intron flanking sequences was performed (Prevention Genetics, Marshfield, WI, USA). Fifteen polymorphisms were detected including two previously undocumented variants: c.8845-8 T>G and c.1789+10 C>T (nomenclature described in Ref. 32). The c.8845-8 T>G variant at the junction of intron 62 and exon 63 creates a new splice site and is likely causative, whereas the c.1789+10 C>T variant currently has an unknown functional effect. Additional investigations of mRNA and genomic sequences are required to determine if these mutations are causal and to search further for deletions, which can be missed by exon sequencing and are often found in MDC1A (33). We added the 96/04 sequence information to the Leiden Open Variation LAMA2 polymorphism database which was accessed at http://www.dmd.nl/nmdb2/home.php?select_db=LAMA2. Sequences of the LAMA2 genes of the 9/03 and 38/03 lines remain to be investigated.

Human myoblasts from normal individuals and MDC1A patients were grown in skeletal muscle cell growth medium supplemented with 15% fetal bovine serum, 50 µg/ml fetuin, 1 ng/ml basic fibroblast growth factor, 10 ng/ml epidermal growth factor, 10 µg/ml insulin and 0.4 µg/ml dexamethasone (Promocell, Heidelberg, Germany). Differentiation was induced by replacing the fetal bovine serum with 2% horse serum. Cells of the mouse C2C12 myogenic cell line and primary cultures of mouse myoblasts derived from lower hindlimb muscles were grown and induced to differentiate as previously described (5,22,34,35). Culture dishes for C2C12 cells were coated with gelatin; for Bax+/+ or Bax−/− primary myoblasts with ECL matrix (Entactin–Collagen–Laminin, Upstate Biotechnology, Lake Placid, NY, USA); and for normal and MDC1A human myoblasts with either poly-l-lysine (Sigma-Aldrich, St. Louis, MO, USA) or 1 µg/cm2 human placental laminin (L6274, Sigma-Aldrich).

Antibodies, reagents and assays

Antibodies for activated caspase-3 and mouse acetyl-lysine were from Cell Signaling Technology (Beverly, MA, USA). Goat anti-LDH, anti-Bax mAb 6A7 and goat anti-Ku70 were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit anti-Bax, mouse anti-Ku70 and mouse anti-beta-ATP synthase were from BD Biosciences (San Jose, CA, USA). Anti-desmin mAb was from Sigma-Aldrich (St. Louis, MO, USA). The anti-myosin heavy chain mAb F59 was as previously described (33,36,37). Dr Lydia Sorokin (University of Münster) generously provided the rat anti-laminin-α2 mAb 4H8-2 which reacts with an epitope located within the L4b (previously IVa) globular domain encoded by exons 23–27 in the N-terminal portion of the laminin-α2 molecule (38,39). Staurosporine was from Sigma-Aldrich. Caspase-3 enzyme activity was measured with the CaspACE system (Promega, Madison, WI, USA).

Immunoblots and fractionation

Proteins were transferred to nitrocellulose membranes and signals were detected with Alexa-680-conjugated secondary antibodies with appropriate species specificity. Immunoblots were quantified using the quantification tool in the Odyssey 2.0 software that accompanies the LI-COR Odyssey infrared system (LI-COR Biosciences, Lincoln, NE, USA). Subcellular fractionations were performed by incubating cells in isotonic 0.05% digitonin buffer for 10 min at 4°C followed by centrifugation to separate the cytosolic supernatant from the membrane/mitochondrial/organellar/nuclear pellet as described (40). This fractionation has been validated for use in Bax translocation studies (40). To further purify nuclei, the pellet was resuspended in 10 mm HEPES, 1.5 mm MgCl2, 10 mm KCl, 0.5 mm dithiothreitol, 0.05% NP-40, pH 7.9 for 10 min at 4°C followed by centrifugation at 3000g for 10 min as described (http://www.abcam.com/index.html?pageconfig=resource&rid=11408). This second pellet was then resuspended in 5 mm HEPES, 1.5 mm MgCl2, 0.2 mm EDTA, 0.5 mm dithiothreitol, 26% glycerol (v/v), pH 7.9; and then adjusted to 300 mm NaCl, homogenized on ice in a glass homogenizer, and centrifuged at 24 000g for 20 min at 4°C. The resulting pellet, containing purified nuclei, was used for immunoprecipitation or immunoblotting. When using immunoblots to compare proteins in different cell fractions, we analyzed an equal percentage of the total fractionated material, e.g. 10% of the nuclear and 10% of the cytoplasmic preparation, in each lane.

Immunoprecipitation

The buffer used for cell lysis and Ku70 immunoprecipitation consisted of 20 mm HEPES, pH 7.4, 1% Triton X-100, 200 mm NaCl, 1 mm EDTA, 5 mm sodium pyrophosphate, 20 mm beta-glycerophosphate, 50 mm NaF, 1 mm sodium orthovandate and 1X protease inhibitor cocktail (Calbiochem, San Diego, CA, USA). Bax immunoprecipitation was carried out as previously described (41). Frozen soleus tissue was pulverized under liquid nitrogen in a mortar and pestle and then homogenized using a Polytron in appropriate lysis buffer and sonicated on ice. Insoluble material was removed by centrifugation at 12 000g for 10 min at 4°C. Protein concentration was determined by BCA assay (Bio-Rad Laboratories, Hercules, CA, USA).

Synthesis of BIPs

Sequences of the control peptide, and the mouse and human BIP peptide sequences were adopted from previous studies (15,16). The fl-hBIP and fl-mBIP peptides were labeled at the N-terminus with FITC using beta-alanine as a linker between the BIP peptide and the FITC label. Peptides were prepared by solid state peptide synthesis using an automated peptide synthesizer (ABI Model 433A) and Fmoc (fluorenylmethoxycarbonyl) to block α-amino groups. Coupling of side-chain protected amino acids to a nascent peptide was accomplished by converting the α-carboxyl groups to active benzotriazole esters using the coupling reagent 2-(1H-Benzotriazole-1-yl)-1,1,3,3-teteramethyluronium hexafluorophosphate. The completed peptide was cleaved from the resin and deprotected using a 95% trifluoroacetic acid cocktail containing scavenger molecules (phenol, thioanisole and ethanedithiol). Final purification employed preparative reversed phase HPLC on C18 columns. The quality and identity of the purified peptide was ascertained by mass spectrometry and analytical rpHPLC. Lyophilized peptides were resuspended at 200 mm in dimethylsulfoxide and dilutions from this stock solution were made in culture medium. Initial experiments tested final concentrations of peptides of 200—1 mm for efficacy.

FUNDING

This work was supported by the National Institutes of Health (5R01HL064641, 1U54HD060848); the National Research Initiative Competitive Grant Program of the United States Department of Agriculture (2006-35206-16622); and the Muscular Dystrophy Association of the United States (114981).

ACKNOWLEDGEMENTS

We thank Dr Paul Leavis (Boston Biomedical Research Institute) for peptide synthesis; Dr Sachiko Homma (Boston Biomedical Research Institute) for much helpful advice; Mary Lou Beermann and Maggie Ardelt for technical assistance; Dr Lydia Sorokin (University of Münster) for the laminin-α2 mAb; and Dr Rosário Santos (Centro de Genética Médica J. Magalhães, INSA Porto, Portugal) and Dr Tom Winder (Prevention Genetics Inc.) for analysis and discussion of line 96/04 polymorphisms. In addition, we thank the Muscle Tissue Culture Collection (MTCC) at the University of Munich, particularly Dr Peter Schneiderat and Ira Kaus, for arranging to provide the myoblasts obtained from MDC1A patients. The MTCC is part of the German network on muscular dystrophies (MD-NET, service structure S1, 01GM0601) funded by the German Ministry of Education and Research (BMBF, Bonn, Germany). The MTCC is a partner of Eurobiobank (www.eurobiobank.org) and TREAT-NMD (www.treat-nmd.eu).

Conflict of Interest statement. None declared.

REFERENCES

1. Angelin A., Tiepolo T., Sabatelli P., Grumati P., Bergamin N., Golfieri C., Mattioli E., Gualandi F., Ferlini A., Merlini L., et al. Mitochondrial dysfunction in the pathogenesis of Ullrich congenital muscular dystrophy and prospective therapy with cyclosporins. Proc. Natl. Acad. Sci. USA. 2007;104:991–996. [PubMed]
2. Davies J.E., Wang L., Garcia-Oroz L., Cook L.J., Vacher C., O'Donovan D.G., Rubinsztein D.C. Doxycycline attenuates and delays toxicity of the oculopharyngeal muscular dystrophy mutation in transgenic mice. Nat. Med. 2005;11:672–677. [PubMed]
3. Irwin W.A., Bergamin N., Sabatelli P., Reggiani C., Megighian A., Merlini L., Braghetta P., Columbaro M., Volpin D., Bressan G.M., Bernardi P., Bonaldo P. Mitochondrial dysfunction and apoptosis in myopathic mice with collagen VI deficiency. Nat. Genet. 2003;35:367–371. [PubMed]
4. Girgenrath M., Dominov J.A., Kostek C.A., Miller J.B. Inhibition of apoptosis improves outcome in a model of congenital muscular dystrophy. J. Clin. Invest. 2004;114:1635–1639. [PMC free article] [PubMed]
5. Girgenrath M., Beermann M.L., Vishnudas V.K., Homma S., Miller J.B. Pathology is alleviated by doxycycline in a laminin-alpha2-null model of congenital muscular dystrophy. Ann. Neurol. 2008;65:47–56. First published on Dec. 11, 2008 10.1002/ana.21523. [PMC free article] [PubMed]
6. Millay D.P., Sargent M.A., Osinska H., Baines C.P., Barton E.R., Vuagniaux G., Sweeney H.L., Robbins J., Molkentin J.D. Genetic and pharmacologic inhibition of mitochondrial-dependent necrosis attenuates muscular dystrophy. Nat Med. 2008;14:442–447. [PMC free article] [PubMed]
7. Miller J.B., Girgenrath M. Role of apoptosis in neuromuscular diseases and potential usefulness of anti-apoptosis therapy. Trends Mol. Med. 2006;12:279–286. [PubMed]
8. Tsai M.S., Chiu Y.T., Wang S.H., Hsieh-Li H.M., Lian W.C., Li H. Abolishing Bax-dependent apoptosis shows beneficial effects on spinal muscular atrophy model mice. Mol. Ther. 2006;13:1149–1155. [PubMed]
9. Dominov J.A., Kravetz A.J., Ardelt M., Kostek C.A., Beermann M.L., Miller J.B. Muscle-specific BCL2 expression ameliorates muscle disease in laminin-alpha2-deficient, but not dystrophin-deficient, mice. Hum. Mol. Genet. 2005;14:1029–1040. [PubMed]
10. Allamand V., Guicheney P. Merosin-deficient congenital muscular dystrophy, autosomal recessive (MDC1A, MIM#156225, LAMA2 gene coding for alpha2 chain of laminin) Eur. J. Hum. Genet. 2002;10:91–94. [PubMed]
11. Miyagoe-Suzuki Y., Nakagawa M., Takeda I. Merosin and congenital muscular dystrophy. Microsc. Res. Tech. 2000;48:181–191. [PubMed]
12. Vachon P.H., Loechel F., Xu H., Wewer U.M., Engvall E. Merosin and laminin in myogenesis; specific requirement for merosin in myotube stability and survival. J. Cell Biol. 1996;134:1483–1497. [PMC free article] [PubMed]
13. Laprise P., Vallée K., Demers M.J., Bouchard V., Poirier E.M., Vézina A., Reed J.C., Rivard N., Vachon P.H. Merosin (laminin-2/4)-driven survival signaling: complex modulations of Bcl-2 homologs. J. Cell Biochem. 2003;89:1115–1125. [PubMed]
14. Sundaresan N.R., Samant S.A., Pillai V.B., Rajamohan S.B., Gupta M.P. SIRT3 is a stress-responsive deacetylase in cardiomyocytes that protects cells from stress-mediated cell death by deacetylation of Ku70. Mol. Cell. Biol. 2008;28:6384–6401. [PMC free article] [PubMed]
15. Gomez J.A., Gama V., Yoshida T., Sun W., Hayes P., Leskov K., Boothman D., Matsuyama S. Bax-inhibiting peptides derived from Ku70 and cell-penetrating pentapeptides. Biochem. Soc. Trans. 2007;35:797–801. [PubMed]
16. Li Y., Yokota T., Gama V., Yoshida T., Gomez J.A., Ishikawa K., Sasaguri H., Cohen H.Y., Sinclair D.A., Mizusawa H., Matsuyama S. Bax-inhibiting peptide protects cells from polyglutamine toxicity caused by Ku70 acetylation. Cell Death Differ. 2007;14:2058–2067. [PubMed]
17. Mazumder S., Plesca D., Kinter M., Almasan A. Interaction of a cyclin E fragment with Ku70 regulates Bax-mediated apoptosis. Mol. Cell. Biol. 2007;27:3511–3520. [PMC free article] [PubMed]
18. Cohen H.Y., Lavu S., Bitterman K.J., Hekking B., Imahiyerobo T.A., Miller C., Frye R., Ploegh H., Kessler B.M., Sinclair D.A. Acetylation of the C terminus of Ku70 by CBP and PCAF controls Bax-mediated apoptosis. Mol. Cell. 2004;13:627–638. [PubMed]
19. Cohen H.Y., Miller C., Bitterman K.J., Wall N.R., Hekking B., Kessler B., Howitz K.T., Gorospe M., de Cabo R., Sinclair D.A. Calorie restriction promotes mammalian cell survival by inducing the SIRT1 deacetylase. Science. 2004;305:390–392. [PubMed]
20. Amsel A.D., Rathaus M., Kronman N., Cohen H.Y. Regulation of the proapoptotic factor Bax by Ku70-dependent deubiquitylation. Proc. Natl. Acad. Sci. USA. 2008;105:5117–5122. [PubMed]
21. Iriyama T., Kamei Y., Kozuma S., Taketani Y. Bax-inhibiting peptide protects glutamate-induced cerebellar granule cell death by blocking Bax translocation. Neurosci. Lett. 2009;451:11–15. [PubMed]
22. Nowak J.A., Malowitz J., Girgenrath M., Kostek C.A., Kravetz A.J., Dominov J.A., Miller J.B. Immortalization of myogenic cells does not require loss of p16INK4a, p19ARF, or p53 and is accelerated by inactivation of Bax. BMC Cell Biol. 2004;5:1. [PMC free article] [PubMed]
23. Laprise P., Poirier E.M., Vézina A., Rivard N., Vachon P.H. Merosin–integrin promotion of skeletal myofiber cell survival: differentiation state-distinct involvement of p60Fyn tyrosine kinase and p38alpha stress-activated MAP kinase. J. Cell Physiol. 2002;191:69–81. [PubMed]
24. Er E., Oliver L., Cartron P.F., Juin P., Manon S., Vallette F.M. Mitochondria as the target of the pro-apoptotic protein Bax. Biochim. Biophys. Acta. 2006;1757:1301–1311. [PubMed]
25. Fernando P., Kelly J.F., Balazsi K., Slack R.S., Megeney L.A. Caspase 3 activity is required for skeletal muscle differentiation. Proc. Natl. Acad. Sci. USA. 2002;99:11025–11030. [PubMed]
26. Lattanzi G., Muntoni F., Sabatelli P., Squarzoni S., Maraldi N.M., Cenni V., Villanova M., Columbaro M., Merlini L., Marmiroli S. Unusual laminin alpha2 processing in myoblasts from a patient with a novel variant of congenital muscular dystrophy. Biochem. Biophys. Res. Commun. 2000;277:639–642. [PubMed]
27. Hicks D., Lampe A.K., Laval S.H., Allamand V., Jimenez-Mallebrera C., Walter M.C., Muntoni F., Quijano-Roy S., Richard P., Straub V., Lochmüller H., Bushby K.M. Cyclosporine A treatment for Ullrich congenital muscular dystrophy: a cellular study of mitochondrial dysfunction and its rescue. Brain. 2009;132:147–155. [PubMed]
28. Qin Q., Patil K., Sharma S.C. The role of Bax-inhibiting peptide in retinal ganglion cell apoptosis after optic nerve transection. Neurosci. Lett. 2004;372:17–21. [PubMed]
29. Jeong J., Juhn K., Lee H., Kim S.H., Min B.H., Lee K.M., Cho M.H., Park G.H., Lee K.H. SIRT1 promotes DNA repair activity and deacetylation of Ku70. Exp. Mol. Med. 2007;39:8–13. [PubMed]
30. Kuang W., Xu H., Vachon P.H., Liu L., Loechel F., Wewer U.M., Engvall E. Merosin-deficient congenital muscular dystrophy: partial genetic correction in two mouse models. J. Clin. Invest. 1999;102:844–852. [PMC free article] [PubMed]
31. Knudson C.M., Tung K.S., Tourtellotte W.G., Brown G.A., Korsmeyer S.J. Bax-deficient mice with lymphoid hyperplasia and male germ cell death. Science. 1995;270:96–99. [PubMed]
32. den Dunnen J.T., Antonarakis S.E. Mutation nomenclature extensions and suggestions to describe complex mutations: A discussion. Hum. Mutat. 2000;15:7–12. [PubMed]
33. Oliveira J., Santos R., Soares-Silva I., Jorge P., Vieira E., Oliveira M.E., Moreira A., Coelho T., Ferreira J.C., Fonseca M.J., et al. LAMA2 gene analysis in a cohort of 26 congenital muscular dystrophy patients. Clin. Genet. 2008;74:502–512. [PubMed]
34. Block N.E., Miller J.B. Expression of MRF4, a myogenic helix-loop-helix protein, produces multiple changes in the myogenic program of BC3H-1 cells. Mol. Cell. Biol. 1992;12:2484–2492. [PMC free article] [PubMed]
35. Dominov J.A., Dunn J.J., Miller J.B. Bcl-2 expression identifies an early stage of myogenesis and promotes clonal expansion of muscle cells. J. Cell Biol. 1998;142:537–544. [PMC free article] [PubMed]
36. Miller J.B., Teal S.B., Stockdale F.E. Evolutionarily conserved sequences of striated muscle myosin heavy chain isoforms: Epitope mapping by cDNA expression. J. Biol. Chem. 1989;264:13122–13130. [PubMed]
37. Smith T.H., Block N.E., Rhodes S.J., Konieczny S.F., Miller J.B. A unique pattern of expression of the four muscle regulatory factors distinguishes somitic from embryonic, fetal, and newborn mouse myogenic cells. Development. 1993;117:1125–1133. [PubMed]
38. Allamand V., Sunada Y., Salih M.A., Straub V., Ozo C.O., Al-Turaiki M.H., Akbar M., Kolo T., Colognato H., Zhang X., et al. Mild congenital muscular dystrophy in two patients with an internally deleted laminin alpha2-chain. Hum. Mol. Genet. 1997;6:747–752. [PubMed]
39. Allamand V., Bidou L., Arakawa M., Floquet C., Shiozuka M., Paturneau-Jouas M., Gartioux C., Butler-Browne G.S., Mouly V., Rousset J.P., et al. Drug-induced readthrough of premature stop codons leads to the stabilization of laminin alpha2 chain mRNA in CMD myotubes. J. Gene Med. 2008;10:217–224. [PubMed]
40. Wang J., Wei Q., Wang C.Y., Hill W.D., Hess D.C., Dong Z. Minocycline up-regulates Bcl-2 and protects against cell death in mitochondria. J. Biol. Chem. 2004;279:19948–19954. [PubMed]
41. Yamaguchi H., Paranawithana S.R., Lee M.W., Huang Z., Bhalla K.N., Wang H.G. Epothilone B analogue (BMS-247550)-mediated cytotoxicity through induction of Bax conformational change in human breast cancer cells. Cancer Res. 2002;62:466–471. [PubMed]

Articles from Human Molecular Genetics are provided here courtesy of Oxford University Press