Abigail S. Hackam

Caspase Cleavage of Gene Products Associated with Triplet Expansion Disorders Generates Truncated Fragments Containing the Polyglutamine Tract

The neurodegenerative diseases Huntington disease, dentatorubropallidoluysian atrophy, spinocerebellar atrophy type 3, and spinal bulbar muscular atrophy are caused by expansion of a polyglutamine tract within their respective gene products. There is increasing evidence that generation of truncated proteins containing an expanded polyglutamine tract may be a key step in the pathogenesis of these disorders. We now report that, similar to huntingtin, atrophin-1, ataxin-3, and the androgen receptor are cleaved in apoptotic extracts. Furthermore, each of these proteins is cleaved by one or more purified caspases, cysteine proteases involved in apoptotic death. The CAG length does not modulate susceptibility to cleavage of any of the full-length proteins. Our results suggest that by generation of truncated polyglutamine-containing proteins, caspase cleavage may represent a common step in the pathogenesis of each of these neurodegenerative diseases.

Kennedy's disease: caspase cleavage of the androgen receptor is a crucial event in cytotoxicity.

Abstract : X‐linked spinal and bulbar muscular atrophy (SBMA), Kennedys disease, is a degenerative disease of the motor neurons that is associated with an increase in the number of CAG repeats encoding a polyglutamine stretch within the androgen receptor (AR). Recent work has demonstrated that the gene products associated with open reading frame triplet repeat expansions may be substrates for the cysteine protease cell death executioners, the caspases. However, the role that caspase cleavage plays in the cytotoxicity associated with expression of the disease‐associated alleles is unknown. Here, we report the first conclusive evidence that caspase cleavage is a critical step in cytotoxicity ; the expression of the AR with an expanded polyglutamine stretch enhances its ability to induce apoptosis when compared with the normal AR. The AR is cleaved by a caspase‐3 subfamily protease at Asp146, and this cleavage is increased during apoptosis. Cleavage of the AR at Asp146 is critical for the induction of apoptosis by AR, as mutation of the cleavage site blocks the ability of the AR to induce cell death. Further, mutation of the caspase cleavage site at Asp146 blocks the ability of the SBMA AR to form perinuclear aggregates. These studies define a fundamental role for caspase cleavage in the induction of neural cell death by proteins displaying expanded polyglutamine tracts, and therefore suggest a strategy that may be useful to treat neurodegenrative diseases associated with polyglutamine repeat expansions.

Huntingtin interacting protein 1 induces apoptosis via a novel caspase-dependent death effector domain.

Huntington disease is a devastating neurodegenerative disease caused by the expansion of a polymorphic glutamine tract in huntingtin. The huntingtin interacting protein (HIP-1) was identified by its altered interaction with mutant huntingtin. However, the function of HIP-1 was not known. In this study, we identify HIP-1 as a proapoptotic protein. Overexpression of HIP-1 resulted in rapid caspase 3-dependent cell death. Bioinformatics analyses identified a novel domain in HIP-1 with homology to death effector domains (DEDs) present in proteins involved in apoptosis. Expression of the HIP-1 DED alone resulted in cell death indistinguishable from HIP-1, indicating that the DED is responsible for HIP-1 toxicity. Furthermore, substitution of a conserved hydrophobic phenylalanine residue within the HIP-1 DED at position 398 eliminated HIP-1 toxicity entirely. HIP-1 activity was found to be independent of the DED-containing caspase 8 but was significantly inhibited by the antiapoptotic protein Bcl-xL, implicating the intrinsic pathway of apoptosis in HIP-1-induced cell death. Co-expression of a normal huntingtin fragment capable of binding HIP-1 significantly reduced cell death. Our data identify HIP-1 as a novel proapoptotic mediator and suggest that HIP-1 may be a molecular accomplice in the pathogenesis of Huntington disease.

Cleavage of Atrophin-1 at Caspase Site Aspartic Acid 109 Modulates Cytotoxicity

Dentatorubropallidoluysian atrophy (DRPLA) is one of eight autosomal dominant neurodegenerative disorders characterized by an abnormal CAG repeat expansion which results in the expression of a protein with a polyglutamine stretch of excessive length. We have reported recently that four of the gene products (huntingtin, atrophin-1 (DRPLA), ataxin-3, and androgen receptor) associated with these open reading frame triplet repeat expansions are substrates for the cysteine protease cell death executioners, the caspases. This led us to hypothesize that caspase cleavage of these proteins may represent a common step in the pathogenesis of each of these four neurodegenerative diseases. Here we present evidence that caspase cleavage of atrophin-1 modulates cytotoxicity and aggregate formation. Cleavage of atrophin-1 at Asp109 by caspases is critical for cytotoxicity because a mutant atrophin-1 that is resistant to caspase cleavage is associated with significantly decreased toxicity. Further, the altered cellular localization within the nucleus and aggregate formation associated with the expanded form of atrophin-1 are completely suppressed by mutation of the caspase cleavage site at Asp109. These results provide support for the toxic fragment hypothesis whereby cleavage of atrophin-1 by caspases may be an important step in the pathogenesis of DRPLA. Therefore, inhibiting caspase cleavage of the polyglutamine-containing proteins may be a feasible therapeutic strategy to prevent cell death.

Sequences in the Amino Termini of GABA ρ and GABAA Subunits Specify Their Selective Interaction In Vitro

Abstract: Molecular cloning has revealed that there are six classes of subunits capable of forming GABA‐gated chloride channel receptors. GABAA receptors are composed of α, β, γ, δ, and ε/χ subunits, whereas GABAC receptors appear to contain ρ subunits. However, retinal cells exhibiting GABAC responses express α, β, and ρ subunits, raising the possibility that GABAC receptors may be a mixture of subunit classes. Using in vitro translated protein, we determined that human GABAA receptor subunits α1, α5, and β1 did not coimmunoprecipitate with full‐length ρ1, ρ2, or the N‐terminal domain of ρ1 that contains signals for ρ‐subunit interaction. To explore the molecular mechanism underlying these apparently exclusive combinations, chimeric subunits were created and tested for interaction with the wild‐type subunits. Transfer of the N terminus of β1 to ρ1 created a β1ρ1 chimera that coimmunoprecipitated with the α1 subunit but not with the ρ2 subunit. Furthermore, exchanging the N terminus of the ρ1 subunit with the corresponding region of β1 produced a ρ1β1 chimera that interfered with ρ1 receptor expression in Xenopus oocytes, whereas the full‐length β1 subunit had no effect. Together, these results indicate that sequences in the N termini direct assembly of ρ subunits and GABAA subunits into GABAC and GABAA receptors, respectively.

The fatal attraction of polyglutamine-containing proteins

Eight human neurodegenerative disorders are caused by the expansion of a CAG trinucleotide encoding polyglutamine within specific genes, including Huntington disease (HD), dentatorubralpallidoluysian atrophy (DRPLA), spinobulbar muscular atrophy (SBMA), and spinocerebellar ataxia (SCA) types 1, 2, 3, 6, and 7 (1-7). The polyglutamine tract in the proteins responsible for each of these diseases is polymorphic in the normal population; expansion of the polyglutamine tract beyond a threshold characteristic for each disease results in the onset of a progressive and fatal neurodegeneration (1, 4, 8, 9). Each disease has a specific and different neuropathy. However, because each illness is caused by a similar mutation, these diseases may share a similar basic mechanism for their pathogenesis despite obvious differences in their symptomology and pathology. Several recent reports have highlighted a basic feature that may unite many of these polyglutamine diseases. Intracellular inclusions or aggregates have been found in neurons of affected

Down with mutant GATA1

Hematological neoplasias frequently involve somatically derived chromosomal translocations. Pediatric acute megakaryoblastic leukemia (AMKL) has been associated with translocation of chromosomes 1 and 22, resulting in the expression of an oncogenic fusion protein. AMKL also frequently occurs in individuals with trisomy 21, which is also known as Down syndrome (DS). Interestingly, not only is the risk for AMKL much higher in DS patients than in the general population, but it occurs without chromosomal translocation. Therefore, induction of malignant transformation in Down syndrome-related AMKL (DSAMKL) may involve a unique mechanism. A recent report by Wechsler et al. describes a novel mutation that leads to AMKL, that occurs only in the context of Down syndrome. DS-AMKL is characterized by the overgrowth of abnormal blast cells that express various erythroidand megakaryocytic-specific genes, such as GATA1. The GATA1 gene has an important function in hematopoesis as a tissue-specific transcription factor that regulates expression of globins and other erythroid genes. Previous studies have also demonstrated an essential role for GATA1 in promoting and inhibiting megakaryocyte and erythroid differentiation. Therefore, GATA1 is a reasonable candidate gene for involvement in AMKL. To determine whether GATA1 contributes to DS-AMKL pathogenesis, a panel of bone-marrow cells was analyzed from 75 individuals with various myeloid leukemias, including several classifications of acute myelogenous leukemia (AML),

A phosphatase mutation implicated in multiple sclerosis

by careful audiometric examination. There were no differences in hearing ability between these nonpenetrant homozygotes and individuals defined as having normal hearing who did not have the DFNB26 haplotype. The appearance of homozygous non-penetrant family members in a severe congenital disease is unusual. Genetic heterogeneity is not a likely explanation since the analysis was performed on a single family, and a rare reversion event is doubtful because the unaffecteds were from five different sibships. To verify the linkage of DFNB26 to 4q31, a second genome-wide linkage analysis was performed using a different set of polymorphic markers. No additional genomic regions were linked to DFNB26, indicating that the initial linkage was accurate. Having confirmed this linkage, Riazuddin et al. investigated the second plausible explanation: the presence of a strongly acting dominant modifier gene segregating in the family. An additional linkage analysis was performed to identify the modifier locus. In this analysis, a dominant inheritance pattern was modeled, the nonpenetrant DFNB26 homozygotes were assigned as affected, deaf family members were assigned as unaffected, and all other individuals were defined as unknown. This analysis linked the non-penetrance trait to chromosome 1q24, at a locus that the authors named DFNM1, for deafness, non-syndromic, modifier 1. A multipoint linkage analysis and a screen for common deafness-associated mitochondrial mutations excluded association with other regions. Haplotype analysis at 1q24 and identification of recombinants refined the DFNM1 locus to a 5.6-cM region delineated by markers D1S2658 and D1S2790. A maximum lod score of 4.31 at u=0 was obtained with D1S2850 within the critical region. Significantly, none of the eight deaf family members carried the DFNM1 haplotype (Fig. 2). Nature has effectively prevented hearing loss in this family, and has provided researchers with a new direction for therapeutics. Future linkage analyses will investigate the exciting possibility that the DFNM1 modifier protects against hearing loss from other genetic or non-genetic sources. Screening for the modifier in affected individuals will be essential for predicting the severity of hearing loss. The identification of the gene product at the DFNM1 locus will provide critical insight into pathways that are perturbed in hearing loss. There are two known candidate genes at or close to the DFNM1 locus. The transcription factor PMX1, which is expressed in the cochlea, maps to the 5.6-cM DFNM1 interval, and DFNM1 itself maps within the 22-cM interval of a previously identified autosomal dominant deafness associated locus DFNA7. The latter Fig. 2. The DFNM1 and DFNB26 loci were identified in a single family. The protective DFNM1 allele is depicted as the green box mapping to chromosome 1q24; the non-protective form of the DFNM1 allele is shown as a red box. The DFNB26 gene is shown as a yellow box at chromosome map location 4q31; and the mutated version is indicated by the red slash. The different combinations of the DFNB26 and DFNM1 alleles account for the hearing phenotypes in the family, as indicated.

A few more pieces of the DM puzzle

The curious puzzle of myotonic dystrophy has recently revealed a few new tantalizing clues. Myotonic dystrophy (abbreviated DM, for dystrophia myotonica) is a muscular disease that is striking in its variability among patients and its seemingly unconnected symptoms. DM has an autosomal dominant pattern of inheritance, and at a prevalence of one in 8000 births, it is the most common form of muscular dystrophy. Primarily a progressive neuromuscular disorder, DM patients frequently exhibit additional characteristic non-muscular symptoms, including ocular cataracts, neuropyschiatric impairments, endocrine abnormalities, and premature balding. Several years ago, the mutation responsible for DM was identified as an expansion of a polymorphic CTG tract within the last intron of the DMPK serine/ threonine protein kinase gene. Due to its position in a non-coding region, the CTG tract in DMPK mRNA is not translated. Therefore, defining the cellular events leading to DM pathology is complicated by the fact that the mutant DMPK gene produces a normal DMPK protein. The heterogeneity of DM implies a complex etiology. Three different pathogenic mechanisms have been proposed. The first implicates reduced DMPK expression, due to a defect in transport and processing of mutant DMPK mRNA. Although lower DMPK mRNA levels have been observed in patient tissue, mice engineered to lack DMPK have only a mild phenotype and no DM-type myotonia or progressive myopathy. The second mechanism involves an alteration of the local chromatin structure by the CTG tract, leading to transcriptional repression of the adjacent transcription factor DMAHP gene (now known as six5) and a corresponding reduction in six5 activity. Third, a transdominant effect of the mutant DMPK mRNA has been suggested since mutant DMPK mRNA interacts with splicing factors and can inhibit myoblast differentiation. The extent that each of these mechanisms contribute to the clinical features in DM is unknown. However, several recent papers investigated the second and third mechanisms and have revealed exciting new clues to the pathogenesis of DM. These papers take advantage of the power of mouse genetics, skillfully using a reductionist approach to test one mechanism at a time. The first report, by Klesert et al., investigated the involvement of six5 in DM by generating mice with a deleted six5 gene. Although this knock-out strategy does not use an expanded CTG, it should appropriately mimic the consequences of reduced six5 expression. A targeting construct was designed to disrupt the six5 gene by removing the first exon of the three-exon gene and replacing it with the LacZ reporter, immediately downstream of the six5 promoter. This insertion not only created a null allele but also permitted tracking of the developmental expression of the six5 promoter by detection of b-galactosidase activity. The absence of six5 expression in the knock-out mice was confirmed by reverse transcription-polymerase chain reaction (RT-PCR). Interestingly, there were no changes in the expression of genes that are normally regulated by six5. However, expression of endogenous DMPK in the homozygous knock-out six5 mice was reduced by 50%, possibly due to an effect of disruption of the adjacent six5 locus. There were no significant differences in DMPK expression between wild-type and heterozygote six5 mice. Although expression of the DMPK and six5 genes were both reduced in six5 animals, histological analyses and electromyography on skeletal muscle from 3and 10-month-old mice revealed no evidence of muscular abnormalities and, particularly, no myotonia. Electrophysiology recordings on acutely isolated muscle fibers from

Rett syndrome : The influence of dysregulated gene slicing

Second only to Down syndrome, Rett syndrome (RTT) is a major genetic cause of mental retardation in females. The disease strikes females only, beginning between the ages of 7 and 18 months, after a seemingly normal development. Regression of developmental milestones is followed by a marked deceleration of growth and deterioration of higher brain functions. A hallmark feature of RTT is loss of purposeful hand use, accompanied by stereotypic ‘hand-wringing’ movements, and loss of speech. Numerous inheritance models for this unusual disease have been proposed, and rejected, including maternal isodisomy, mitochondrial mutation, and genomic rearrangement. The consensus based on analyses of multiple cases is that RTT is an X-linked dominant disease with lethality in hemizygous males. Two recent papers, by Amir et al. and Wan et al., have made significant advances in deciphering the cause of RTT. Aided by exclusion mapping from previous studies refining the RTT gene location, and helped enormously by continuous deposition of sequences from the Human Genome Project, Amir et al. systematically analyzed the genes in a 10-Mb region on chromosome Xq28 for mutations in sporadic and familial RTT patients. Several possible candidate genes were screened based on known function, tissue distribution, and X-inactivation properties, but none contained nucleotide changes exclusively in RTT patients. Surprisingly, mutation analysis of a gene with ubiquitous, rather than brain-specific, expression revealed the gene responsible for RTT. This gene, encoding methyl-CpGbinding protein 2 (abbreviated MECP2), is X-inactivated and maps to the candidate area on Xq28 between the L1CAM and RCP/GCP loci. Amplification of the three coding exons in the MECP2 gene from 21 sporadic and 8 familial classic RTT patients, followed by conformationsensitive gel electrophoresis (CSGE), revealed sequence changes only in patients and not unaffected controls. The mutations found were all nucleotide substitutions or single-base insertions. Comparison of the human MECP2 sequence with the mouse, chicken, and frog MECP2 genes demonstrated that all the nucleotide changes found in affected patients altered amino acids conserved between species, as expected for disease-causing mutations. At this stage, the researchers were extremely fortunate that the fields of human genetics and biochemistry had converged. MECP2 had been cloned earlier and its properties had been analyzed in detail. MeCP2 is a chromosomal protein that binds specifically to the 5-methyl cytosine residues within single CpG dinucleotide pairs (Fig. 1A). The promoter regions of certain genes in mammals are enriched with CpG dinucleotides, and it is their methylation that regulates gene expression by altering chromatin structure, thereby ‘turning off’ the gene. Silencing the expression of a gene or multiple genes by methylation is also the basis for X-chromosome inactivation (XCI) and imprinting. But how could a globally applied mechanism such as gene silencing result in the delayed onset and neuronal specificity observed in RTT? The explanation may reside in the particular genes being silenced by MeCP2, which are presently unknown. MeCP2 expression is essential for development, since mutant mice lacking MECP2 are inviable, similar to the phenotype of human hemizygous males. In contrast, the random pattern of expression of wild-type MeCP2 in the brain of heterozygous females may be sufficient for appropriate neuronal migration and correct formation and organization of brain structures. However, during the post-natal period of brain development, when synaptic connections are established and stabilized, altered MeCP2 activity and the resulting inappropriate expression of MeCP2 target genes may have dire consequences for the development of cognitive functions (Fig. 1A).