Inhibition of Calpain Cleavage of Huntingtin Reduces Toxicity ACCUMULATION OF CALPAIN/CASPASE FRAGMENTS IN THE NUCLEUS*

Huntington’s disease (HD) is a neurodegenerative dis-order caused by a polyglutamine (polyQ) tract expansion near the N terminus of huntingtin (Htt). Proteolytic processing of mutant Htt and abnormal calcium signal-ing may play a critical role in disease progression and pathogenesis. Recent work indicates that calpains may participate in the increased and/or altered patterns of Htt proteolysis leading to the selective toxicity observed in HD striatum. Here, we identify two calpain cleavage sites in Htt and show that mutation of these sites ren-ders the polyQ expanded Htt less susceptible to proteolysis and aggregation, resulting in decreased toxicity in an in vitro cell culture model. In addition, we found that calpain- and caspase-derived Htt fragments preferentially accumulate in the nucleus without the requirement of further cleavage into smaller fragments. Calpain family members, calpain-1, -5, -7, and -10, have increased levels or are activated in HD tissue culture and transgenic mouse models, suggesting they may play a key role in Htt proteolysis and disease pathology. In-terestingly, calpain-1, -5, -7, and -10 localize to the cytoplasm

Huntington's disease (HD) 1 is an autosomal-dominant neurodegenerative disease caused by a CAG expansion in the huntingtin gene (htt) (1), and is characterized by involuntary movements, personality changes, dementia, and early death (2).
Huntingtin protein (Htt) is expressed throughout the central nervous system as well as non-neuronal cells, yet causes selective neuronal death in the striatum and cortex (3)(4)(5)(6). The selective neurodegeneration found in HD may be linked to the expression of particular receptors and proteases in the affected striatal and cortical cells, but the molecular details of these events are not well characterized. One hypothesis that has received a great deal of attention is that production of Nterminal Htt fragments plays a key role in disease pathogenesis (7)(8)(9)(10). Our recent studies have shown that calpain cleaves Htt at multiple sites to produce small N-terminal fragments (11). In addition, we have demonstrated that the active form of calpain is substantially increased in the caudate of HD patients suggesting that calpain-derived Htt fragments may contribute to the etiology of HD (11).
Recently, new insights have been gained as to how polyQexpanded Htt may alter Ca 2ϩ homeostasis at early stages of the disease, and these findings may be directly linked to excitotoxicity and the activation of the Ca 2ϩ -dependent proteases, the calpains (12)(13)(14). N-Methyl-D-aspartate receptor, subtype 2B, and metabotropic glutamate receptors class 1/5 are enriched in the striatum and play a role in the dysregulation of Ca 2ϩ in HD. The polyQ-expanded form of Htt sensitizes N-methyl-Daspartate receptors in the brain, leading to enhanced Ca 2ϩ influx following receptor stimulation (12,15,16). The polyQexpanded form of Htt sensitizes endoplasmic reticulum type 1 inositol (1,4,5)-trisphosphate receptor also leading to enhanced Ca 2ϩ release following metabotropic glutamate receptors class 1/5 activation (14). Furthermore, the disease form of Htt has been implicated in dysfunctional mitochondrial Ca 2ϩ buffering (17,18).
We have shown that both calpains and caspases cleave Htt (9 -11, 19 -22). Far more is known about the proteolytic processing of Htt with caspases in vitro and in vivo (9, 10, 19 -22). This is not surprising, because Htt was the first caspase substrate identified in a neurodegenerative disease, and the mechanism of action of the caspase family members has been studied extensively in the apoptotic cell death process (9,23). However, little is understood about the role of calpain family members in cell death and neurodegeneration. Calpains were discovered over 40 years ago, and at least 15 distinct human family calpain members have been identified (24 -26). At least 6 of the 9 family members found in the brain have not been characterized (25).
The first calpains discovered,and m-calpain (calpain-1, calpain-2, and calpain-4 (small regulatory subunit)), have been extensively studied and are distinguished by their differences in Ca 2ϩ dependence, with -calpain requiring micromolar Ca 2ϩ for activation whereas m-calpain requires millimolar levels (27)(28)(29)(30). Autoproteolysis following Ca 2ϩ addition converts the catalytic subunit of -calpain from a 80-kDa protein to a 76-kDa protein, whereas activation of the m-calpain catalytic subunit results in 20 amino acids being removed from the 80-kDa protein N terminus. The regulatory subunit is common to both calpains and requires sequential cleavage events to convert it from the inactive 28-kDa precursor into the active 21-kDa form upon Ca 2ϩ addition. Calpains are activated by many apoptotic and necrotic stimuli, particularly those that alter Ca 2ϩ levels in the cell and their activation often lies upstream of caspase activation (31)(32)(33)(34)(35)(36)(37)(38).
Calpain-5, -7, -10, -12, and -13 are expressed in the brain, but their physiological or pathological roles have not been studied. Calpain-5 is expressed ubiquitously and may have a role in sex determination (25). Calpain-10 has been identified as a type 2 diabetes susceptibility gene (39) and calpain-13 is a homologue of a Drosophila small optic lobe gene (40). Calpain-10 was found to be partially localized to the nucleus, and nuclear localization was enhanced upon ionomycin treatment (41).
As part of our ongoing efforts to understand the proteolytic cleavage events in HD, we investigated the role calpain cleavage plays in polyQ-expanded Htt cellular toxicity and evaluated whether any of the calpain family members are expressed or activated during altered Ca 2ϩ homeostasis in HD cell culture and transgenic mouse models. We found that calpains cleave Htt at amino acids 469 and 536. The sites of cleavage are clustered between amino acids 437-537, an area that overlaps with caspase cleavage sites in Htt (amino acids 513-586). Reduced cellular toxicity and Htt aggregation were observed when the calpain-resistant form of Htt was expressed in a cell culture model with altered Ca 2ϩ homeostasis. The calpain/ caspase-derived Htt fragments were found to preferentially accumulate in the nuclear compartment without the requirement of further proteolysis. Finally, we found calpain family members calpain-1, -5, -7, and -10 levels are increased and/or activated in HD tissue culture and transgenic mouse models.
Cloning of Epitope-tagged Calpain-5 cDNA-RNA from 293T cells was extracted using TRIzol reagent (Invitrogen). Oligo(dT)-primed cDNA was generated using a Superscript cDNA kit (Invitrogen) according to the manufacturer's instructions. Calpain-5 cDNA was amplified using primers CAPN5 F, 5Ј-GGGCAGCAGCCACCATGTTCTCG-3Ј, and CAPN5-FLAG R 5Ј-CTTATCGTCGTCATCCTTGTAATCGA-CAGCCATGAGGGAGGTGCT-3Ј. The reverse primer contains codons encoding a FLAG epitope. Reaction mixtures contained 25 pmol of each primer, 0.5 mM dNTPs, 10ϫ Pfu Turbo polymerase buffer, 5 l of Me 2 SO (Sigma), and 2.5 units of Pfu Turbo DNA polymerase (Stratagene). Total reaction volume was 50 l. Amplification parameters were: denaturation at 94°C for 1 min, 35 cycles at 94°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 2 min. The PCR product was purified using a QIAquick PCR purification kit (Qiagen) and cloned into a pcDNA3.1/V5-His vector using a TOPO TA Expression kit (Invitrogen).
Cell Cultures-Superfect reagent (Qiagen) was used for transient transfections of human embryonic kidney 293T cells with Htt constructs. Thapsigargin (500 nM, 2.5 M, or 10 M; Calbiochem) or Me 2 SO (0.05-0.25%; Sigma) was added 36 h following transfection, and cells were collected at 48 or 60 h post-transfection for analysis.
Toxicity Measurements-Caspase activity was measured using the ApoAlert Caspase-3 Fluorescent Assay Kit (Clontech, Palo Alto, CA). Briefly, cells were lysed in Clontech cell lysis buffer for 10 min on ice. Following centrifugation to remove cell debris, lysate was added to reaction mixture containing 20 mM Pipes, pH 7.2, 100 mM NaCl, 1% Chaps, 10% sucrose, 10 mM dithiothreitol, and 100 M DEVD-7-amino-4-trifluoromethyl coumarin. Reactions were incubated at 37°C and read at an excitation of 400 nm and an emission of 505 nm for a 1-h period. The rate of change in fluorescence within the linear range was normalized to protein levels to determine relative DEVD-ase activity. Protein levels were determined using the BCA assay (Pierce) for all experiments.
Semi-quantitative RT-PCR for Calpain cDNAs-RNA from transfected 293Tcells were extracted using TRIzol reagent (Invitrogen). Oligo(dT)-primed cDNA was generated using a Superscript cDNA kit (Invitrogen) according to the manufacturer's instructions. The cDNA mixture was then diluted in water to a volume of 100 l. cDNAs were amplified in a semi-quantitative PCR (45) using the following primers: R ϭ A or G, and Y ϭ C or T. Reaction mixtures contained 2.5 l of cDNA, 25 pmol of each primer, 0.5 mM dNTPs, 10ϫ Taq polymerase buffer, and 2.5 units of TaqDNA polymerase (Roche Applied Science) in a total reaction volume of 25 l. Amplification parameters were 30 cycles of denaturation at 94°C for 30 s, annealing at 60°C for 30 s, and extension at 68°C for 1 min. PCR products were analyzed on 1% agarose gels.

Identification of Two Calpain Cleavage Sites in Htt-We
have previously shown that calpains cleave Htt to produce three N-terminal cleavage products (11). In those studies, we found that incubation of in vitro translated pRcCMV-Htt15Q-(1-1212) with calpain-1 (-calpain) or -2 (m-calpain) produced 62-, 67-, and 72-kDa N-terminal products, two of which (67and 72-kDa fragments) are in close proximity to the caspasesensitive region of Htt (70 -80 kDa, Fig. 1A). Both in cell culture and in vitro translation experiments, the major Htt calpain cleavage product appears at 72/82 kDa (15Q versus 138Q) and lies between the caspase-3 sites at amino acids 513 and 552 (Fig. 1, A and B). The exact site of cleavage in Htt and whether calpain cleavage modulates cellular toxicity of the polyQ-expanded form of Htt were not evaluated in our previous work (11). Therefore, we determined the site of calpain cleavage in Htt using deletion analysis and analyzed the cytotoxicity of these constructs in a cell culture model.
Deletion of predicted calpain recognition sequences was used to identify calpain cleavage sites in Htt. Although the primary amino acid sequences of calpain substrate cleavage sites have similar properties (i.e. small hydrophobic amino acids in P2 position), calpains also recognize the three-dimensional structure of their protein targets (26). This distinguishes them from the caspases, which only recognize conserved primary amino acid sequences. First, we evaluated whether deleting a Ser-Ser-Ser motif at amino acids 535-537 of Htt, an identified calpain cleavage site in protein kinase C (26), would alter proteolytic cleavage of Htt by calpains. Western blot analysis of Htt15Q-(1-1212) when expressed in 293T cells produces one prominent Htt cleavage product at 72 kDa along with three minor products at 67, 70, and 75 kDa (Fig. 1C). We found that Htt15Q-(1-1212) ⌬535-537, when expressed in 293T cells, was resistant to calpain cleavage and eliminated production of the 72-kDa calpain-derived Htt fragment (Fig. 1, B and C). Confirming our results, in vitro translation of Htt15Q-(1-1212) ⌬535-537 treated with recombinant calpain eliminated the production of the 72-kDa fragment.
Next, we identified the calpain cleavage site in Htt that produces the 67-kDa fragment. We deleted a Leu-Thr-Ala motif at amino acids 468 -470 of Htt, because this site is an identified calpain cleavage site in caspase-12 (46). We found that Htt15Q-(1-1212) ⌬468 -470, when expressed in 293T cells, was resistant to calpain cleavage and eliminated production of the 67-kDa calpain-derived Htt fragment (Fig. 1, B and C). To validate the specificity of our deletion analysis, we produced deletions in Htt at other potential calpain cleavage sites, which were only four amino acids from the identified calpain cleavage sites (amino acid 469 and 536). We found these deletions had no effect on calpain proteolysis of Htt. Deletion of a  Fig. 1A (sequence noted in red) did not influence the production of the 72-kDa calpain-derived Htt fragment. Therefore, deletion of other potential calpain cleavage sites immediately adjacent to the identified Leu-Thr-Ala site (amino acids 468 -470) or Ser-Ser-Ser (amino acids 535-537) did not affect calpain cleavage of Htt and further supports the specificity of these deletion mutants and the ability to successfully utilize them in studying calpain proteolysis and toxicity of Htt. Finally, we evaluated whether deletion of these calpain cleavage sites would affect caspase cleavage of Htt since these sites are in close proximity to one another (Fig. 1A). Treatment of these calpain-resistant constructs Htt15Q-(1-1212) ⌬468 -470, Htt15Q-(1-1212) ⌬535-537 and Htt15Q-(1-1212) ⌬468 -470 ⌬535-537 with recombinant caspase (data not shown) and production of caspasederived Htt fragments when these calpain-resistant constructs are expressed in culture (see Fig. 4) demonstrated they were still susceptible to caspase cleavage.
The calpain cleavage site at amino acid 536 of Htt was confirmed by producing a neoepitope antibody to this specific Htt calpain cleavage product (Fig. 1D). When Htt15Q-(1-1212) is cleaved with low levels of calpain-2 (0 -0.3 U/l), the neoHtt 536 antibody detected increasing amounts of the N-terminal 536 amino acid Htt fragment being produced. A Htt stop construct (Htt15Q-(1-536)) which expresses Htt protein from amino acids 1 to 536 was used as a positive control and ran at the same molecular weight as the identified cleavage product.
We were not able to identify the calpain cleavage site responsible for producing the 62-kDa N-terminal Htt fragment. A Val-Leu-Ser motif at amino acids 436 to 438 and a Ser-Arg-Lys motif at amino acids 438 to 440 were two of the potential calpain cleavage sites investigated and neither of these deletions eliminated production of the 62-kDa calpain-derived Htt fragment (data not shown). Since the 62-kDa N-terminal cleav-age product represents a minor calpain cleavage product of Htt in cell culture (often not detected), identification of this site was not pursued further.
Calpain-resistant Htt Mutants Reduce Cellular Toxicity in an in Vitro Cell Culture Model-In our previous work we evaluated the proteolysis and toxicity of Htt in tissue culture models under conditions that activated caspase-dependent cell death (21). It is not clear whether this pathway of cell death activation is relevant in vivo since recent evidence suggests that altered Ca 2ϩ homeostasis plays an important role in HD pathogenesis and progression (47). Therefore, we established a simple cell culture model to evaluate the cleavage of Htt by both caspases and calpains under conditions that modulate intracellular Ca 2ϩ levels (11). 293T cells overexpressing wild-type, caspase-resistant, calpain-resistant or calpain/caspase-resistant Htt were treated with thapsigargin, a pro-apoptotic agent that increases intracellular Ca 2ϩ levels through inhibition of the endoplasmic reticulum Ca 2ϩ /Mg 2ϩ -ATPase (Figs. 1-4,  and 6).
Because we established cell culture conditions that modulate Ca 2ϩ homeostasis and mapped the calpain cleavage sites in

Calpain-resistant Htt Mutants Reduce Htt Proteolysis and
Aggregation in an in Vitro Cell Culture Model-Currently, three proteolytic cleavage pathways have been reported for Htt in vivo that may influence cellular toxicity and aggregation. These include cleavage of Htt by caspases (9,21), calpains (11,48), and an unknown aspartic endopeptidase (49). Whether these proteolytic pathways act independently or sequentially is currently not known. Because we generated calpain-resistant Htt constructs, we tested whether cleavage at these Htt calpain sites was required for aggregation and the production of smaller N-terminal fragments. Calpain-resistant Htt138Q-(1-1212) ⌬468 -470 ⌬535-537 showed a decrease in proteolysis relative to Htt138Q-(1-1212) (Fig. 3B). Elimination of calpain-sensitive region(s) (⌬468 -470 ⌬535-537) from Htt138Q-(1-1212) not only eliminated production of the specific calpain-derived cleavage products but led to a reduction in further proteolysis of Htt (4.2-fold decrease by densitometry). Most notably, a reduction in the amount of lower molecular weight N-terminal Htt cleavage products was observed when the calpain-resistant form of Htt138Q-(1-1212) was expressed, which may represent the aspartic endopeptidase-cleaved N-terminal Htt product described previously (Fig. 3B) (49).
Calpain and Caspase Cleavage Products of Htt Preferentially Localize to the Nuclear Compartment-The presence of polyQexpanded fragments in the nucleus has been correlated with cellular cytotoxicity in HD, as well as other polyQ diseases (10, 20, 50 -55). Cellular toxicity of N-terminal Htt fragments can be enhanced by addition of a nuclear localization signal and decreased by addition of a nuclear export signal (NES) (53). Indeed, Htt contains a naturally occurring NES, suggesting full-length Htt may shuttle back and forth between the cytosol and nucleus (56). The majority of Htt is cytosolic (Ͼ95%) suggesting the importance of the NES signal in maintaining the subcellular distribution of Htt (57,58). Many studies have also suggested Htt must be truncated to 50 kDa or less for Nterminal fragments to preferentially accumulate in the nucleus, implying caspase/calpain-derived fragments may not redistribute to the nucleus unless further truncated (10,20,52). These conclusions were based mainly on immunofluorescence studies evaluating Htt aggregates and not subcellular fractionation. Clearly, it is of interest to determine whether caspase/ calpain-derived fragments can redistribute from the cytoplasm to the nucleus. In addition, a small percentage of full-length Htt exists in the nucleus, and therefore it is also important to establish whether cleavage can occur in this subcellular compartment.
We have recently produced and characterized Htt antibodies that specifically recognize the caspase cleavage products of Htt at amino acid 513 and 552 (22). Using the neoHtt 513 antibody, the caspase-3-derived Htt cleavage product at amino acid 513 was identified primarily in the nuclear fraction (asterisk, Fig.  4A, and data not shown). Utilizing the neoHtt 552 antibody, similar results were observed for the caspase-2/3-derived Htt cleavage product at amino acid 552 by subcellular fractionation experiments (arrowheads, Fig. 4C). Complete fractionation was confirmed using antibodies to the cytoplasmic protein, ␤-tubulin, and the nuclear protein, poly(ADP-ribose)polymerase (PARP) (Fig. 4A, lower panels). Immunofluorescence studies using the neoHtt 552 antibody also demonstrate the nuclear localization of the Htt cleavage product at amino acid 552 along with accumulation of this fragment into perinuclear aggregates (Fig. 4D). Our studies do not address whether cleavage of Htt occurs in the nucleus or cytoplasm.
Most Calpain Family Members Localize to and Are Activated in the Nuclear Compartment-Although proteolysis of cytoplasmic and nuclear substrates by caspase family members has been extensively studied, our knowledge of calpain cleavage events is limited. Three of the fifteen calpain family members identified, calpain-1, calpain-2, and calpain-4 (small regulatory subunit), have been well characterized. Three other calpain family members, calpain-5, calpain-7, and calpain-10, are highly expressed in the brain tissue (41) 2 and therefore may play a role in Ca 2ϩ -mediated HD pathogenesis. In addition, mRNA levels of calpains-5, -7, and -10 are significantly altered in the R6/2 HD mouse brains relative control tissue, 2 making them promising candidates for further investigation.
To further understand which calpain and caspase isoforms cleave Htt and in what subcellular compartment cleavage could occur, Western blot analysis of cytosolic/nuclear fractions from 293T cells were probed with relevant calpain and caspase antibodies to determine subcellular localization of the active and inactive forms of these proteases (Fig. 5, n ϭ 3). Western analysis showed that both inactive (arrow) and active (arrowhead) forms of calpain-2 as well as the calpain regulatory subunit were primarily localized to the cytoplasmic fraction (Fig. 5A). Calpain-7 and -10 showed the inactive form in both nuclear and cytoplasmic compartments and the active form exclusively in the nuclear compartment (Fig. 5A). Both inactive and active forms of calpain-5 were found in the cytosol and nucleus (Fig. 5A).
Caspases included in these studies were those previously identified as cleaving Htt, as well as caspase-12, which is activated following induction of endoplasmic reticulum stress (46). The inactive form of caspase-2 was localized to the nucleus and cytosol, whereas the active form was limited to the cytosolic compartment (Fig. 5B). The majority of inactive forms of caspase-3, caspase-6, and caspase-7 was localized to the cytosolic fraction (Fig. 5B). We found less activation of caspases in the thapsigargin treatment model utilized in these studies relative to the tamoxifen model used to activate caspases in previous studies (21).
Next we tested whether transcription and/or translation of calpains was induced by thapsigargin treatment in our cell culture system. Semi-quantitative PCR studies revealed that calpain-5, -7, and -10 transcripts are increased following thapsigargin treatment relative to control (Fig. 6A). Correlating with increased transcription was an increase in calpain-5, -7, and -10 protein levels along with activation of calpain family members (Fig. 6B). These results are consistent with increased calpain message, protein levels, and activation of calpains under altered Ca 2ϩ homeostasis. Next, we assessed whether the increase and activation of calpain family members detected in our simple cell culture model was relevant in vivo by evaluating these proteases in an HD knock-in mouse model.
Increased Proteolysis in an HD Knock-in Mouse Model Correlates with Calpain Activation-To analyze the contribution of calpain activation and proteolysis in HD in vivo, we utilized an HD knock-in mouse model expressing mouse Htt with 150 polyQ repeats (59). In our studies we utilized both heterozygotes Htt 150Q/7Q and homozygotes Htt 150Q/150Q allowing us to analyze gene dosage effects on proteolysis and activation of calpains. The HD transgenic mouse has late-onset behavioral and neuropathological abnormalities consistent with HD (59). Behavioral deficits are found in gait, cage activity, and rotarod analysis. The neuropathological abnormalities described are nuclear accumulation of Htt and enhanced glial fibrillary acidic protein immunoreactivity in the striatum. We found increased calpain activity and Htt fragmentation in the cortex and striatum of this HD transgenic model (Fig. 7). Of particular significance is the gene dosage effect seen with a neo-epitope antibody to the calpain-derived cleavage product of spectrin (44) and a calpain-5 antibody in the striatum, with homozygous mice, Htt 150Q/150Q , having the highest levels relative to heterozygotes, Htt 150Q/7Q , or controls (Fig. 7B). A 2.2-fold increase in the spectrin cleavage product and a 4.8-fold increase in calpain-5 were found in the striatum of homozygous Htt 150Q/ 150Q relative to control mice as determined by densitometry. Distinct calpain family members were also altered in the cortex of the homozygotes Htt 150Q/150Q and heterozygotes Htt 150Q/7Q when compared with age-matched controls (Fig. 7C). Comparison of calpain-1 (1.5-fold), -7 (1.5-fold), and -10 (1.7-fold) expression in the cortex showed a gene dosage effect, with the homozygotes, Htt 150Q/150Q , having the highest levels when compared with the heterozygotes, Htt 150Q/7Q , or controls. Analysis was carried out at 11 months of age when these mice have impaired motor performance (59). DISCUSSION We used numerous approaches to examine the role calpains play in the proteolysis of Htt in HD cell culture and transgenic mouse models. One pathological mechanism proposed for HD is that production of toxic fragments of the polyQ-expanded form of Htt contributes to HD pathology and progression (7)(8)(9)(10). To continue our investigation of the cleavage pathways involved in Htt proteolysis, we used deletion mutagenesis to identify two calpain sites in Htt at amino acids 469 and 536. These calpain sites are clustered between amino acids 468 and 537, an area that overlaps with caspase cleavage sites in Htt (amino acids 513-586), suggesting this area of Htt is highly susceptible to proteolytic cleavage (Fig. 1A). We found expression of the calpain-resistant polyQ-expanded Htt reduced cytotoxicity, prote-olysis, and aggregate formation when compared with polyQexpanded Htt during Ca 2ϩ dysregulation. In earlier work, we evaluated the caspase-resistant forms of the polyQ-expanded Htt under conditions that stimulate caspase activation but not necessarily Ca 2ϩ dysregulation (21). In fact, here, we found that the caspase-resistant form of Htt has more calpain cleavage products accumulating under conditions of Ca 2ϩ dysregulation. Ca 2ϩ dysregulation is an early event in HD pathogenesis and therefore Ca 2ϩ -stimulated activation of calpains may be particularly important in Htt proteolysis and represent a more relevant in vitro model.  , 12 h). The cellular lysates were fractionated (C ϭ cytoplasmic; n ϭ nuclear). Inactive forms are indicated by arrows, and active forms indicated by arrowheads (n ϭ 3). Equal loading and verification of fractionation was confirmed by probing Western blots with ␤-tubulin and PARP antibody.

FIG. 7. Proteolytic cleavage in
Htt150 knock-in mouse model correlates with calpain activation. Increased calpain expression and activation in the striatum and cortex of HD Htt150 knock-in mouse model. A, Western analysis of full-length Htt from wild-type, heterozygotes Htt 150Q/7Q , and homozygous Htt 150Q/150Q cortical cell lysates (11 months) prior to immunoprecipitation (left panel). Htt was immunoprecipitated with N-terminal Htt BKP1 monoclonal antibody and probed with an antibody that recognizes the expanded polyQ domain of Htt (right panel). The immunoprecipitated Htt fragments shown in the right panel below 75-95 kDa represent calpain/caspase cleavage products, whereas the cleavage product above 60 kDa is likely the aspartyl endopeptidase cleavage product. B, Western analysis of full-length Htt from wild-type, heterozygous Htt 150Q/7Q , and homozygous Htt 150Q/150Q (top panel) mice in striatum. Increased cleavage of spectrin is detected in HD150 knock-in mice using a neo-epitope antibody specific for the calpain cleavage of spectrin (middle panel). Increased calpain-5 expression in the striatum of HD knock-in mice (bottom panel). C, Western analysis of full-length Htt from wild-type, heterozygous Htt 150Q/7Q , and homozygous Htt 150Q/150Q mice (top panel) in cortex. Increased expression and/or cleavage of calpain-1, calpain-7, and calpain-10 is detected in HD150 knock-in mice (bottom panels). Equal loading was verified by probing Western blots with ␤-tubulin antibody.
Currently, three proteolytic cleavage pathways have been reported for Htt in vivo that may influence cellular toxicity and aggregation. These include cleavage of Htt by caspases (9,21), calpains (11,48), and an unknown aspartic endopeptidase (49). Whether these proteolytic pathways are independent of each other or act in a sequential process is currently not known. Analysis of the calpain-resistant Htt constructs demonstrates that cleavage at the calpain sites is required for the production for smaller N-terminal fragments consistent with a sequential cleavage model. We found that both calpains and caspases could cleave Htt independently of each other. Calpain-resistant Htt was cleaved by caspases and conversely caspase-resistant Htt was cleaved by calpains. As noted the caspase-resistant form of Htt produced more calpain cleavage products of Htt (Fig. 2B). In addition, calpain-resistant polyQ-expanded Htt generated more caspase cleavage products (Fig. 4C). Our current studies do not address the order of activation of the protease family members in HD.
Although it is not yet known the relative contribution of calpain, caspases, or other proteases to HD progression, we determined that a number of uncharacterized calpain family members may be increased and activated in HD tissue culture and transgenic mouse models. We found that thapsigargin, an inhibitor of the endoplasmic reticulum Ca 2ϩ /Mg 2ϩ -ATPase, increased transcription and translation of calpain-5, -7, and -10 at in our cell culture model. Paralleling our tissue culture studies we found that the level of calpain-1, -5, -7, and -10 were increased in either striatal or cortical tissue in an HD knock-in mouse model, suggesting these calpain family members may be involved in the etiology of HD.
One particular important finding from our studies is that the proteolytic caspase/calpain-derived Htt cleavage products accumulate in the nucleus without the requirement of further cleavage into smaller fragments. Previous work suggested that truncation of the calpain/caspase Htt fragments is necessary for redistribution and accumulation of Htt in the nucleus (10,20). Although our studies do not address the subcellular site of Htt cleavage, we evaluated the subcellular distribution and activation of calpain and caspases in the cytoplasm and nucleus. We found a number of novel calpain members are localized to the nucleus suggesting that Htt could be cleaved in either compartment.
Our work suggests that calpains play a role in Htt proteolysis and HD pathology. Further work will be directed at understanding whether any of these calpain family members play a critical role in HD pathology and progression, because little is still known about the activation and localization of many of these calpain family members and their particular contribution to HD or neurodegeneration. Although our data suggests calpains are involved in HD pathogenesis, the studies are correlative in nature, and future work will address whether calpains are required for HD pathology and progression.