Mutant Huntingtin Expression Induces Mitochondrial Calcium Handling Defects in Clonal Striatal Cells

Huntington disease (HD) is caused by a pathological elongation of CAG repeats in the huntingtin protein gene and is characterized by atrophy and neuronal loss primarily in the striatum. Mitochondrial dysfunction and impaired Ca2+ homeostasis in HD have been suggested previously. Here, we elucidate the effects of Ca2+ on mitochondria from the wild type (STHdhQ7/Q7) and mutant (STHdhQ111/Q111) huntingtin-expressing cells of striatal origin. When treated with increasing Ca2+ concentrations, mitochondria from mutant huntingtin-expressing cells showed enhanced sensitivity to Ca2+, as they were more sensitive to Ca2+-induced decreases in state 3 respiration and ΔΨm, than mitochondria from wild type cells. Further, mutant huntingtin-expressing cells had a reduced mitochondrial Ca2+ uptake capacity in comparison with wild type cells. Decreases in state 3 respiration were associated with increased mitochondrial membrane permeability. The ΔΨm defect was attenuated in the presence of ADP and the decreases in Ca2+ uptake capacity were abolished in the presence of Permeability Transition Pore (PTP) inhibitors. These findings clearly indicate that mutant huntingtin-expressing cells have mitochondrial Ca2+ handling defects that result in respiratory deficits and that the increased sensitivity to Ca2+ induced mitochondrial permeabilization maybe a contributing mechanism to the mitochondrial dysfunction in HD.

Huntington disease (HD) is caused by a pathological elongation of CAG repeats in the huntingtin protein gene and is characterized by atrophy and neuronal loss primarily in the striatum. Mitochondrial dysfunction and impaired Ca 2؉ homeostasis in HD have been suggested previously. Here, we elucidate the effects of Ca 2؉ on mitochondria from the wild type (STHdh Q7/Q7 ) and mutant (STHdh Q111/Q111 ) huntingtin-expressing cells of striatal origin. When treated with increasing Ca 2؉ concentrations, mitochondria from mutant huntingtin-expressing cells showed enhanced sensitivity to Ca 2؉ , as they were more sensitive to Ca 2؉ -induced decreases in state 3 respiration and ⌬⌿m, than mitochondria from wild type cells. Further, mutant huntingtinexpressing cells had a reduced mitochondrial Ca 2؉ uptake capacity in comparison with wild type cells. Decreases in state 3 respiration were associated with increased mitochondrial membrane permeability. The ⌬⌿m defect was attenuated in the presence of ADP and the decreases in Ca 2؉ uptake capacity were abolished in the presence of Permeability Transition Pore (PTP) inhibitors. These findings clearly indicate that mutant huntingtin-expressing cells have mitochondrial Ca 2؉ handling defects that result in respiratory deficits and that the increased sensitivity to Ca 2؉ induced mitochondrial permeabilization maybe a contributing mechanism to the mitochondrial dysfunction in HD.
Huntington disease (HD) 2 is a neurodegenerative disease that is inherited in an autosomal dominant manner. It belongs to a family of CAG expansion diseases and is caused by the pathological elongation of the CAG repeats in exon one of the huntingtin protein gene (1). Symptoms and disease progression are caused by dysfunction and loss of neu-rons starting in the striatum (specifically medium spiny neurons), but progressing to cortex and to a lesser extent to other brain regions in the later stages of the disease (2). Disease is caused by the toxic gain of function of mutant protein but some loss of function may also contribute to the pathogenesis (for review see Ref. 3). The toxic gain of function of mutant huntingtin has not been clearly defined, but there are findings suggesting that mutant huntingtin causes transcriptional dysregulation (4), ubiquitin-proteasome system dysfunction (5), Ca 2ϩ homeostasis dysfunction (6,7), and mitochondrial dysfunction (7)(8)(9)(10).
Mitochondrial dysfunction in HD has been suggested primarily by the studies showing impairment of mitochondrial complexes (II, III, and IV) specifically in the striatum in the late stages HD patients (8 -10). Administration of the mitochondrial complex II inhibitor 3-nitropropionic (3-NP) in both rodents and nonhuman primates resulted in symptoms and neuropathology that resemble HD (11,12). Further, studies have shown impairment of mitochondrial Ca 2ϩ buffering in HD lymphoblast cell line and brain mitochondria from the fulllength mutant huntingtin transgenic mice (YAC72) (7,13).
Striatum, the primary region to get affected in HD is highly innervated by cortical glutaminergic projections (2). Previously it has been demonstrated that mitochondrial dysfunction can lead to neuronal sensitization to glutamate leading to excitotoxic cellular dysfunction and cell death (14,15). Hence, even though HD is not a classic mitochondrial disease (16), elucidation of mitochondrial dysfunction mechanisms would likely provide important insight in HD pathogenesis.
To study the effects of mutant huntingtin on mitochondrial function, conditionally immortalized cells of striatal origin that express endogenous, comparable levels of either wild type (STHdh Q7/Q7 ) or mutant (STHdh Q111/Q111 ) huntingtin were used (17). These cell lines are prepared from wild type (Hdh Q7/Q7 ) and mutant huntingtin knock-in mice (Hdh Q111/Q111 ) (17) and therefore the STHdh Q111/Q111 cell line is a genetically accurate cell model of HD. In our previous study (18), we investigated the effects of mutant huntingtin on mitochondrial electron transport chain complexes using STHdh Q7/Q7 and STHdh Q111/Q111 cell lines. Given the fact that the metabolic thresholds and enzyme activities of electron transport chain complexes were not different between the two cell lines, it is likely that the mitochondrial complex deficits are a later event in the course of HD pathogenesis, indeed in low grade HD cases no deficits in the enzyme activities of electron transport chain complexes were observed (19).
In the present study, we examined the effects of Ca 2ϩ on mitochondria from STHdh Q7/Q7 and STHdh Q111/Q111 cells. Isolated mitochondria were treated with increasing Ca 2ϩ concentrations and mitochondrial function was assessed using different assays. We determined that mutant huntingtin-expressing cells have decreased Ca 2ϩ uptake capacity, and exhibit increased sensitivity to Ca 2ϩ -induced decreases in respiration and ⌬⌿m. The ⌬⌿m defect was attenuated in the presence of ADP and the decrease in Ca 2ϩ uptake capacity was abolished in the presence of Permeability Transition Pore (PTP) inhibitors. This study clearly demonstrates that mitochondrial Ca 2ϩ buffering capacity in STHdh Q111/Q111 cells is compromised, and suggests increased sensitivity to Ca 2ϩ -induced mitochondrial permeabilization as a mechanism of mitochondrial dysfunction in HD.

EXPERIMENTAL PROCEDURES
Materials-All chemicals were from Sigma-Aldrich unless otherwise noted. All buffers used in experiments with crude mitochondrial preparations were prepared in water (Sigma, catalogue no. 95305) that is standardized for Ca 2ϩ content ([Ca 2ϩ ]Յ 0.000001%).
Cell Culture-In this study, conditionally immortalized striatal progenitor cell lines: STHdh Q7/Q7 cell line expressing endogenous wild type huntingtin and the homozygous mutant STHdh Q111/Q111 cell line expressing comparable levels of mutant huntingtin with 111 glutamines were used. Cell lines were prepared from wild type mice and homozygous Hdh Q111/Q111 knock-in mice and were described previously (17). Culturing conditions were the same as described in our previous study (18).
Isolation of Mitochondria-Cells were grown on 150-mm plates until Ϸ 80 -90% confluency, washed twice with cavitation buffer (250 mM sucrose, 5 mM HEPES, 3 mM MgCl 2 , 1 mM EGTA, pH 7.3 corrected with 5 M KOH) and scraped into cavitation buffer using soft rubber scrapers. Cells were opened using N 2 cavitation for 5 min at 250 psi on ice, and samples were additionally homogenized with 1 stroke in a glass Dounce homogenizer. Homogenates were centrifuged at 7000 ϫ g for 10 min at 4°C. Supernatants were aspirated, and pellets were resuspended in cavitation buffer and used as crude mitochondrial preparations. Protein concentrations in crude mitochondrial preparations were determined using the bicinchoninic acid assay (Pierce), and aliquots were then prepared that contained the indicated protein content for each measurement. Aliquots were centrifuged at 7000 ϫ g for 10 min and kept on ice in cavitation buffer until use in each assay.
Measurement of Mitochondrial Respiration-Respiration rates were measured using an oxygraph (Hansatech Instruments) as described previously (18). Crude mitochondrial preparations (0.5-mg aliquots) were resuspended in respiration buffer (130 mM KCl, 20 mM HEPES, 2 mM MgCl 2 , 2 mM EGTA, 2 mM potassium phosphate (KH 2 PO 4 /K 2 HPO 4 , 1:1.78), 1% essentially fatty acids free bovine serum albumin, pH 7.2 adjusted with 5 M KOH) to a final concentration of 1 mg/ml. The mitochondrial suspension (0.5 ml volume) was placed in the respiratory chamber and allowed to equilibrate for 2 min. Respiratory substrate (glutamate (10 mM) plus malate (10 mM) or succinate (5 mM) with rotenone (10 M)) was then added and state 4 respiration was measured for 2 min, ADP (1.5 mM) was then added, and state 3 respiration was measured for a further 2-4 min. Rates were normalized to citrate synthase activity in the same samples. Citrate synthase activity was determined as previously described (18).
Ca 2ϩ Titration Experiments-Respiration buffers containing specific free Ca 2ϩ concentrations (Ca 2ϩ -EGTA respiration buffers) were prepared on the day of the experiment. To calculate the amount of total Ca 2ϩ that was needed to achieve the appropriate free Ca 2ϩ concentration in the respiration buffer that contained 2 mM EGTA we used MaxChelator software (20). Each Ca 2ϩ -EGTA respiration buffer was prepared separately by diluting each specific 100ϫ CaCl 2 stock in the respiration buffer and correcting its pH to 7.2 using 0.1 M KOH in the respiration buffer. CaCl 2 stocks were prepared from CaCl 2 ⅐2H 2 O (minimum 99%), that was dried overnight and stored in a desiccation chamber until use. Crude mitochondrial preparations were dissolved in prepared Ca 2ϩ -EGTA respiration buffers, and respiration rates were measured as described above. The period between buffer addition to the mitochondrial preparation and initiation of state 3 was ϳ5 min.
Free Ca 2ϩ concentrations in the Ca 2ϩ -EGTA buffers were checked using a calibrated Ca 2ϩ electrode on the day of the experiment. Measured concentrations were averaged and presented on the X-axis of Ca 2ϩ titration experiments graphs. Actually concentrations were always slightly higher than those calculated by software.
Cytochrome c and NADH Respiration Experiments-Respiration experiments were performed as described above. State 3 respiration was measured for 2 min prior to the addition of cytochrome c (30 M), and respiration was monitored for another 2 min. This was followed by the addition of NADH (5 mM), and respiration was monitored for an additional 2 min.
Determination of Mitochondrial Ca 2ϩ Uptake Capacity-Ca 2ϩ uptake capacities were measured using a Ca 2ϩ electrode (World Precision Instruments). Crude mitochondrial preparation was resuspended in Ca 2ϩ uptake buffer (130 mM KCl, 20 mM HEPES, 2 mM MgCl 2 , 2 mM potassium phosphate (KH 2 PO 4 /K 2 HPO 4 , 1:1.78), 1% bovine serum albumin, pH 7.2 adjusted with 5 M KOH) and placed in the oxygraph respiratory chamber. The respiratory chamber was thermostatted at 37°C, and its contents were constantly mixed with an electromagnetic stirrer bar. Glutamate (10 mM) and malate (10 mM) were added as respiratory substrates. Ca 2ϩ and reference electrodes were added to the chamber from the top. Starting volume of the reaction was 2 ml. The chamber was kept open during an experiment. Ca 2ϩ additions were performed using fine tubing and a Hamilton syringe. 5, 10, and 20 mM CaCl 2 stocks were used to make 10, 20, 40, or 80 nmol of Ca 2ϩ additions. The Ca 2ϩ electrode measures extramitochondrial Ca 2ϩ and increases in the signal present as downward deflections on the traces.
To observe the effects of PTP inhibition on Ca 2ϩ uptake capacity we used cyclosporine A (1 M) plus ADP (50 M) plus oligomycin (2 g/ml). The PTP inhibitors were added to the respiratory chamber prior to Ca 2ϩ additions (21).
To calculate Ca 2ϩ uptake capacity, we counted number of Ca 2ϩ additions until the addition after which no uptake was observed (trace horizontal). The number of additions was multiplied by the nmol of Ca 2ϩ per addition, and normalized to protein content.
Mitochondrial Membrane Potential (⌬⌿m) Determination in Live Cells-Mitochondrial membrane potential was estimated using the specific mitochondrial probes: Mitotracker Red (CM-H 2 TMRos) and tetramethylrhodamine ethyl ester (TMRE) (Molecular Probes) (22)(23)(24)(25)(26). Cells were grown on poly-L-lysine-coated plates and cultured for 3 days. The cells were then loaded for 30 min with CM-H 2 TMRos in KRH-glucose, washed, and allowed to equilibrate for 15 min. Cell plates were then mounted in a chamber on the stage of a confocal laser scanning microscope (Leica model TCS SP2). Quantitative measurements of CM-H 2 TMRos fluorescence were performed by confocal microscopy (Leica model TCS SP2), using a 40ϫ water immersion lens. CM-H 2 TMRos fluorescence images were obtained by excitation at 563 nm, reflection off a dichroic mirror with a cut-off wavelength at 564 nm, and longpass emission filtering at 590 nm. For TMRE experiments, cells were loaded with TMRE (100 nM) for 45 min in KRH-glucose, and then were mounted on the stage for confocal microscopy. TMRE fluorescence was detected exciting with a 561 nm He-Ne laser line very heavily attenuated (10% laser power), and the emission was collected at Ͼ563 nm (22). Signal from control cells and cells treated with different stimuli were compared using identical settings for laser power, confocal thickness, and detector sensitivity for each dye and separate experiment (23,25,26). The images were analyzed with LCS Leica confocal software and recorded as the mean Mitotracker Red or TMRE fluorescence signal per live cell.
Measurement of Mitochondrial Membrane Potential (⌬⌿m) in Mitochondrial Preparations-⌬⌿m was measured using 5,5Ј,6,6Ј-tetrachloro-1,1Ј,3,3Ј-tetraethylbenzimidazolylcarbocyanine iodide (JC-1; Molecular Probes) according to a published protocol with modifications (27). Modifications were made so that method could be used with isolated mitochondria. To measure ⌬⌿m at different Ca 2ϩ concentrations, crude mitochondrial preparation was aliquoted into the wells of 96-well plate (50 g/well). The plate was centrifuged at 3220 ϫ g for 10 min at 4°C, and supernatants were carefully aspirated. Ca-EGTA respiration buffers with 0, 0.4, or 0.6 M (software calculated) free Ca 2ϩ or Ca 2ϩ uptake buffer with 150, 500, or 1000 M Ca 2ϩ each supplemented with glutamate (10 mM), malate (10 mM), and with (for state 3) or without (for state 4) ADP (1.5 mM) were added to separate wells in duplicates (50 l/well) and incubated at 37°C for 10 min. Supernatants were carefully aspirated, and the same buffers but containing JC-1 (5 g/ml) were added to the wells (50 l/well). Wells in which FCCP (20 M) was also added were considered as positive controls. The plate was incubated for 30 min, at 37°C in dark, supernatants were aspirated, and fluorescence was read at 485/528 nm and 530/ 590 nm. The ratio between the fluorescence was used to describe ⌬⌿m as published previously (27).
Measurement of Mitochondrial H 2 O 2 Production-To determine mitochondrial ROS production we used an Amplex Red (Molecular Probes) assay (28). Crude mitochondrial preparations were aliquoted (100 g/well) and pelleted onto a 96-well plate as described for measuring ⌬⌿m. Mitochondrial pellets were covered with Ca-EGTA respiration buffers with 0, 0.4, or 0.6 M (software calculated) free Ca 2ϩ or Ca 2ϩ uptake buffer with 150, 500, or 1000 M Ca 2ϩ each supplemented with glutamate (10 mM), malate (10 mM), Amplex Red (50 M), horseradish peroxidase (0.01 units/ml or 0.1 units/ml), and with (for state 3) or without (for state 4) ADP (1.5 mM). Plate was read in the kinetic mode for 30 min at excitation/emission wavelengths 530/590 nm at 37°C. Rates of H 2 O 2 production were determined using a standard curve.
Statistical Analysis-Results were analyzed using ANOVA, Student's t test, or paired t test as indicated. Differences were considered significant if p Յ 0.05.

Effects of Ca 2ϩ on Respiration in Mitochondria from STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant) Cells-It
has been shown previously that at the free concentrations higher than 1 M, Ca 2ϩ causes strong inhibition of oxidative phosphorylation (29). To determine the effects of Ca 2ϩ on oxidative phosphorylation in mitochondria isolated from the cells expressing endogenous levels of wild type (STHdh Q7/Q7 ) or mutant (STHdh Q111/Q111 ) huntingtin, we measured state 4 and state 3 respiration rates in respiration buffers containing increasing free M Ca 2ϩ concentrations. In these experiments, EGTA-based respiration buffer was used for 0 M Ca 2ϩ and Ca 2ϩ -EGTA respiration buffers were prepared as described under "Experimental Procedures." At 0 M Ca 2ϩ , we observed no differences in the state 4 or state 3 respiration rates between wild type and mutant cells when glutamate plus malate (complex I substrate) or succinate (complex II substrate) were used as substrates (Fig. 1A). As described earlier (29), with increasing free M Ca 2ϩ concentrations decreases in the state 3 rates were observed (Fig. 1B). However, this decrease was more pronounced in the mitochondria from the mutant huntingtin-expressing cells, reaching significance at lower Ca 2ϩ concentrations than in the wild type (Fig. 1B). State 4 rates increased with increasing Ca 2ϩ concentrations, reaching significance only in the mutant at the highest Ca 2ϩ concentration used (Fig. 1B). To describe overall changes in the respiration rates, we calculated Respiratory Control Ratios (RCRs) at the different Ca 2ϩ concentrations. RCR was calculated as the ratio between state 3 and state 4 rates. A decrease in RCR was observed with increasing Ca 2ϩ concentrations and was more pronounced in mitochondria from mutant cells, reaching significance at the lower Ca 2ϩ concentrations than in the wild type cells (Fig. 1C). These results indicate that mitochondria from STHdh Q111/Q111 (mutant) cells are more sensitive to Ca 2ϩ -induced changes in oxidative phosphorylation than mitochondria from STHdh Q7/Q7 (wild type) cells.
Ca 2ϩ Uptake Capacity in Mitochondria from STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant) Cells-Several studies have suggested that there is reduced mitochondrial Ca 2ϩ buffering capacity in HD. Panov et al. (7,13) demonstrated diminished Ca 2ϩ uptake capacity in mitochondria from HD lymphoblast cell lines, and brain mitochondria from the full-length mutant huntingtin-overexpressing mice (YAC72) (7), whereas others demonstrated diminished Ca 2ϩ uptake in muscle mitochondria from R6/2 mice (30). To comprehensively describe the effects of Ca 2ϩ on mitochondria in our model, we determined mitochondrial Ca 2ϩ uptake capacity in STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells. For these experiments we used a Ca 2ϩ -sensitive electrode, as described under "Experimental Procedures." To determine, mitochondrial Ca 2ϩ uptake, isolated mitochondria (1.5 mg/2 ml) were placed in a 37°C thermostatted chamber and challenged with 10-nmol Ca 2ϩ pulses every 3 min. Representative traces are shown in Fig. 2C. Ca 2ϩ uptake capacity was calculated as described under "Experimental Procedures." We observed that mitochondria from STHdh Q111/Q111 (mutant) cells have significantly diminished Ca 2ϩ uptake capacity compared with mitochondria from STHdh Q7/Q7 (wild type) cells ( Fig. 2A). To determine the "initial uptake" rates we calculated the average of the rates after the second, third, and fourth additions of Ca 2ϩ and determined that the "initial uptake" rates were significantly diminished in the mitochondria from the mutant cells (Fig. 2B). These results indicate that mitochondria from STHdh Q111/Q111 (mutant) cells have a Ca 2ϩ -buffering defect, as they can take up less Ca 2ϩ than the mitochondria from wild type cells. Because ER contamination of the mitochondrial preparation was a possibility, we confirmed the mitochondrial nature of the Ca 2ϩ uptake in our mitochondrial preparations, as the addition of uncoupler (FCCP) caused release of Ca 2ϩ , and pretreatment of the cells with thapsigargin (which blocks the Ca 2ϩ uptake pump of the ER Ref. 31) did not produce any change in the Ca 2ϩ uptake capacity (Fig. 2D). Effects of Ca 2؉ on respiration in mitochondria from STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells. A, state 4 (st4) and state 3 (st3) respiration rates measured in crude mitochondrial preparations in EGTA-based respiration buffer. Crude mitochondria from wild type and mutant cells were incubated with either glutamate plus malate (glu.ϩmal.) or succinate with rotenone (succ.ϩrot.) as respiratory substrates, and respiration rates were determined as described under "Experimental Procedures." Rates were normalized to citrate synthase activity measured in the same samples. No significant differences were observed. B, respiration rates measured in the presence of increasing free Ca 2ϩ concentrations. Crude mitochondria were resuspended in Ca-EGTA buffers with the indicated free Ca 2ϩ concentrations, and state 4 and state 3 rates were determined as described under "Experimental Procedures." Glutamate plus malate was used as the respiratory substrate. Results are expressed as percentage of the state 3 rates at 0 Ca 2ϩ for each of the cell lines. Significant decreases in state 3 rates at the low M free Ca 2ϩ concentrations were observed, with the decreases occurring at the lower Ca 2ϩ concentrations in mitochondria from the mutant cells compared with wild type mitochondria. State 4 rates showed a trend toward increasing with increasing free Ca 2ϩ concentrations, with significance being reached only at the highest Ca 2ϩ concentration used in the mutant cells. C, RCRs in the presence of different free Ca 2ϩ concentrations were calculated as ratios between state 3 and state 4 rates presented in B. The decrease in RCR reached significance at lower Ca 2ϩ concentrations in mutant cells than in the wild type group. All data are mean Ϯ S.E. of 3-4 independent experiments. For statistical analyses ANOVA followed by the Tukey post test (*, p Ͻ 0.05) was used.

Analysis of Mitochondrial Membrane Integrity Before and
After Ca 2ϩ Addition in STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant) Cells-Ca 2ϩ overload of mitochondria results in increased mitochondrial membrane permeability (32). To further study the cause of differences between mitochondria from wild type and mutant cells in their sensitivity to Ca 2ϩ , we wanted to determine if the decrease in respiration observed in the presence of free M Ca 2ϩ concentrations was associated with increased permeability of the mitochondrial membrane. First, we analyzed the integrity of mitochondrial membrane in the basal conditions. NADH is the substrate for mitochondrial complex I. However, the inner mitochondrial membrane is not permeable to exogenous NADH (33). When pyruvate plus malate was added to provide reduced adenine dinucleotides (NADH, FADH 2 ) inside the mitochondria we observed significant state 3 rates upon ADP addition (Fig. 3A). However, when NADH was used as the respiratory substrate, we did not observe induction of state 3 respiration in mitochondria from any of the cell lines (Fig. 3A). Representative traces are shown (Fig. 3A). This indicates good integrity of mitochondrial inner membrane in basal conditions in mitochondria from both cell lines.
It has been suggested that damage of the outer mitochondrial membrane results in the activation of alternative respiratory pathway in the presence of exogenous NADH and cytochrome c (34 -37). In this pathway, NADH is oxidized at the outer mitochondrial membrane leading to reduction of exogenous cytochrome c. As described previously, if outer membrane is being compromised, cytochrome c will translocate to the complex IV and stimulate respiration (35). To assess mitochondrial membrane integrity, we measured the effects of NADH and cytochrome c on state 3 respiration in the absence FIGURE 2. Ca 2؉ uptake capacity measured in mitochondria from STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells. A, Ca 2ϩ uptake capacities. Ca 2ϩ uptake was measured using a Ca 2ϩ -sensitive electrode as described under "Experimental Procedures." Crude mitochondria (1.5 mg in 2 ml) were given 10-nmol Ca 2ϩ pulses every 3 min, until uptake could not be observed anymore. Ca 2ϩ uptake capacities were calculated by multiplying the number of pulses by 10 nmol and normalizing it to mg of protein. Significantly lower Ca 2ϩ uptake capacity in mitochondria from mutant cells as compared with mitochondria from wild type cells was observed. B, initial Ca 2ϩ uptake rates. Initial uptake rates were calculated by averaging rates of uptake after the second, third and fourth additions of Ca 2ϩ . Initial Ca 2ϩ uptake rates were significantly lower in mitochondria from mutant cells compared with wild type cells. Data are mean Ϯ S.E. of three independent experiments. Student's t test (*, p Ͻ 0.05) was used for statistical analyses. C, representative traces of Ca 2ϩ electrode recordings (Ca 2ϩ uptake is reflected by an upward deflection of the trace). Small arrows indicate Ca 2ϩ additions. Large arrows indicate the last pulses included in the calculations. D, control experiment that demonstrate the mitochondrial nature of Ca 2ϩ uptake measured by Ca 2ϩ -sensitive electrode in the crude mitochondrial preparation. Mitochondria from mutant cells were used in these control experiments. Addition of FCCP caused release of Ca 2ϩ from mitochondria as evident by increase in [Ca 2ϩ ] in the reaction buffer (downward deflection of the trace after FCCP). Thapsigargin (1 M) pretreatment of the cells for 30 min did not affect Ca 2ϩ uptake capacity.
or presence of Ca 2ϩ . For these experiments, we chose Ca 2ϩ concentration that caused a significant decrease in state 3 rate in the mitochondria from both cell lines. Experiments were carried out as described under "Experimental Procedures." In the absence of Ca 2ϩ , neither cytochrome c alone nor cytochrome c plus NADH had any effect on respiration (Fig. 3B). This indicated good integrity of the outer mitochondrial membrane in both cell lines. However in the presence of Ca 2ϩ , when the state 3 respiration was decreased, cytochrome c plus NADH caused a significant increase in state 3 (Fig. 3B). Alamethicin, forms pores in the membrane, was used as a positive control for the method (not shown). 3 Cytochrome c alone did not affect state 3 respiration in the presence of Ca 2ϩ (Fig. 3B), suggesting that no substantial loss of cytochrome c is causing the decrease in state 3 rate. These results suggest that mitochondria from both wild type and mutant cells do have good membrane integrity in the absence of Ca 2ϩ . Further, the decrease in state 3 rates in the presence of M Ca 2ϩ concentrations is associated with the increased permeability of mitochondrial membrane but not substantial loss of cytochrome c.
Differential Effects of Ca 2ϩ on Mitochondrial Membrane Potential (⌬⌿m) in STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant) Cells-To determine the effects of Ca 2ϩ deregulation on ⌬⌿m, the ratiometric dye JC-1 was used (38). In these experiments, isolated mitochondria were incubated with increasing Ca 2ϩ concentrations, keeping the same [mitochondrial mass/Ca 2ϩ buffer volume] ratio as in the respiration experiments. Ca 2ϩ concentrations used were: 0 M (EGTA-based respiration buffer), two low M concentrations: 0.4 M, 0.6 M (software calculated) that correspond to 1.1 M, 2.2 M (Ca 2ϩ electrode determined) in Fig. 1B, at which decreases in state 3 respiration was observed, and 3 high M concentrations: 150, 500, and 1000 M, where 150 M corresponds approximately to the Ca 2ϩ concentration at which we no longer observed Ca 2ϩ uptake by the wild type mitochondria (Fig. 2). ⌬⌿m was also measured in Ca 2ϩ uptake buffer (Cab), which is respiration buffer without EGTA, and contains ϳ10 M Ca 2ϩ as determined by using the Ca 2ϩ electrode. ⌬⌿m was determined as described under "Experimental Procedures." FCCP was used to induce maximal decrease of ⌬⌿m, as the positive control for the assay (Fig. 4). In FIGURE 3. Analysis of mitochondrial membrane integrity before and after Ca 2؉ addition to mitochondria from STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells. A, integrity of the mitochondria in wild type and mutant cells. Representative respiration traces are shown to demonstrate the inability of exogenous NADH (5 mM) (inner mitochondrial membrane impermeable) to act as a respiratory substrate in mitochondria from both wild type and mutant cells. When pyruvate (10 mM) plus malate (10 mM) (pyrϩmal) were used as respiratory substrates, a significant increase in O 2 consumption was observed upon ADP addition, while this was not the case when NADH was used. Numbers on the traces indicate O 2 consumption rates. B, effects of cytochrome c (cyt c) or cytochrome c plus NADH (cyt cϩNADH) on respiration before and after Ca 2ϩ addition. Crude mitochondria were resuspended in respiration buffer without or with Ca 2ϩ (0.6 M free Ca 2ϩ as calculated by software), and respiration was measured as described under "Experimental Procedures." Upon induction of state 3 for 2 min (state 3), cytochrome c was added, and the rate measured for 2 min (state3ϩcyt c) and then NADH was added and the rate measured for additional 2 min (state3ϩcyt cϩNADH). In the absence of Ca 2ϩ neither cytochrome c nor cytochrome c plus NADH affected state 3 respiration. However, in the presence of Ca 2ϩ , where significant decreases in state 3 rates were observed, cytochrome c plus NADH addition significantly increased respiration. Cytochrome c alone did not affect state 3 rates in the presence of Ca 2ϩ . Data are mean Ϯ S.E., n ϭ 4 independent experiments. For statistical analyses, we used ANOVA with Tukey post test (*, p Ͻ 0.05, compared with state 3 -Ca 2ϩ ; #, p Ͻ 0.05, compared with state 3 ϩ Ca 2ϩ ). the state 4 condition (ADP not added), we observed a Ca 2ϩ concentration-dependent decrease of ⌬⌿m. This decrease was significant in the mutant at as low as 0.4 M, but in the wild type only at 500 M Ca 2ϩ (Fig. 4A). The significant difference between the two could be observed at 0.6, Cab (ϳ10 M), 150, and 500 M Ca 2ϩ , and was about 80% at 0.6 M and 65% at 150 M. However, in the state 3 condition (ADP added) the significant difference in ⌬⌿m could be observed only starting at 150 M and was about 30% (Fig. 4B). These results show that ⌬⌿m is significantly more sensitive to Ca 2ϩ in STHdh Q111/Q111 (mutant) compared with STHdh Q7/Q7 (wild type) cells. Also, this difference can be strongly attenuated by the presence of ADP.

Effects of Ca 2ϩ on H 2 O 2 Production in Mitochondria from STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant)
Cells-Increased ROS production in mitochondria is usually associated with perturbations of electron transfer in the oxidative phosphorylation process (39) and has been described in conditions of increased mitochondrial membrane permeability (40). To assess the effects of Ca 2ϩ on ROS production in the mitochondria from wild type and mutant cells the Amplex Red assay was used (28). This assay measures H 2 O 2 production and was carried out as described under "Experimental Procedures." Isolated mitochondria were incubated with the increasing Ca 2ϩ concentrations, and H 2 O 2 production was measured in the course of incubation. The Ca 2ϩ concentrations used were as described for the ⌬⌿m experiments (Fig. 4). In state 4, we observed Ca 2ϩ concentration dependent increases in H 2 O 2 that reached significance at 1000 M in wild type and at 500 and 1000 M in the mutants (Fig. 5A). No differences at any of the data points were observed when wild type and mutants were compared (Fig.  5A). In the state 3 condition (ADP added), we observed increases in H 2 O 2 production with Ca 2ϩ that reached significance only in the mutant at 1000 M Ca 2ϩ (Fig. 5B). The trend of increased H 2 O 2 production with increasing Ca 2ϩ concentrations was similar to what was observed for the ⌬⌿m change. At all Ca 2ϩ concentrations, mutant mitochondria displayed a greater increase in H 2 O 2 production than wild type (Fig. 5B). This increase was statistically significant at 0.4, 0.6, and 500 M (Fig. 5B). In these experiments, rotenone treated mitochondria were used as an assay positive control. As expected, in the wild type, rotenone treatment caused significant increases in H 2 O 2 production. Interestingly, mutant mitochondria generated significantly less H 2 O 2 upon rotenone treatment when compared with the wild type.
Effects of PTP Inhibitors on Mitochondrial Ca 2ϩ Uptake Capacity in STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant) Cells-Diminished Ca 2ϩ uptake capacity in the mutant could be explained by a lower threshold for PTP opening in the mutant cells. To test this hypothesis, Ca 2ϩ uptake capacity in the presence of PTP inhibitors was measured. For these experiments, cyclosporine A plus ADP plus oligomycin was used, as this combination has been shown to be very efficient in inhibiting PTP in brain mitochondria (21). Addition of PTP inhibitors increased mitochondrial Ca 2ϩ uptake capacity in both cell lines. This increase was greater and statistically significant in the mutant mitochondria (ANOVA, Tukey post test; p Ͻ 0.01, n ϭ 5 for mutant and n ϭ 8 for mutant plus PTP inhibitors group) (Fig. 6). In the presence of PTP inhibitors, mitochondrial Ca 2ϩ uptake capacity reduction was strongly attenuated in the mutant mitochondria, as the difference between the wild type and mutant mitochondria no longer reached statistical significance (Fig. 6). These results suggest that STHdh Q111/Q111 (mutant) cells exhibit a lower threshold for PTP opening when compared with STHdh Q7/Q7 (wild type) cells.
Effects of Ca 2ϩ on ⌬⌿m in STHdh Q7/Q7 (Wild Type) and STHdh Q111/Q111 (Mutant) Cells in Situ-Prior to evaluating the effects of Ca 2ϩ on ⌬⌿m in situ with CM-H 2 TMRos and TMRE, control studies were carried out to validate the use of these fluorescence dyes. These preliminary studies demonstrated that treatment of both wild type or mutant cells with the mitochondrial uncoupler FCCP resulted in a robust decrease in CM-H 2 TMRos and TMRE fluorescence, indicating a loss of ⌬⌿m (Fig. 7, A and B). To determine the effects of changes in cytosolic Ca 2ϩ levels on ⌬⌿m, cells were treated with a low concentration of the ionophore 4-BrA23187 (1 nM), which does not negatively impact cell viability (41). After treatment with the ionophore for 5 min, 4 mol of Ca 2ϩ additions were made approximately every 5 min, to obtain the increasing concentrations of Ca 2ϩ in the medium (2, 4, 6 mM). With increasing Ca 2ϩ concentrations a decrease in ⌬⌿m was observed as measured with both CM-H 2 TMRos (Fig. 7C) and TMRE (Fig. 7D), and it was significantly more pronounced in mutant cells (Fig. 7, C  and D). The difference in ⌬⌿m between wild type and mutant cells was significant starting at 2 mM Ca 2ϩ in the media (Fig. 7,  C and D). This was in accordance with the results obtained in mitochondrial preparations as shown in Fig. 4.

DISCUSSION
In this study, we provide evidence for the first time that mitochondrial Ca 2ϩ handling defects in cells of striatal origin that express endogenous levels of mutant huntingtin result in impairment of respiration, which could contribute to neuronal dysfunction and death in HD. Treatment of isolated mitochondria from mutant cells with increasing Ca 2ϩ concentrations, resulted in a significant decrease in state 3 respiration at lower Ca 2ϩ than mitochondria from wild type cells. Further, the Ca 2ϩdependent decrease of ⌬⌿m was significantly greater in the mutant cells compared with the wild type cells. However, the ⌬⌿m defect was markedly attenuated in the presence of ADP. Additionally the mitochondrial Ca 2ϩ uptake capacity in mutant cells was significantly lower than what was observed in mitochondria from wild type cells, which was completely abolished by the presence of PTP inhibitors. Taken together these data demonstrate that the presence of mutant huntingtin at physiologically relevant levels results in impaired Ca 2ϩ handling by FIGURE 5. Effects of Ca 2؉ on H 2 O 2 production in mitochondria from STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells. A, H 2 O 2 production in state 4. H 2 O 2 production was evaluated using the Amplex Red assay as described under "Experimental Procedures." Ca 2ϩ -induced increases in H 2 O 2 production and significance was reached at 1000 M Ca 2ϩ in wild type and at 500, 1000 M Ca 2ϩ in mutant. Cab indicates measurements done in Ca 2ϩ uptake buffer (respiration buffer without EGTA) Results are mean Ϯ S.E., n ϭ 5 independent experiments. Statistical analyses were done using ANOVA followed by Tukey post test (*, p Ͻ 0.05, compared with 0 M Ca 2ϩ for each cell line). B, H 2 O 2 production in state 3. Ca 2ϩ -induced increases in H 2 O 2 production, and significance was reached only in mutant cells at 1000 M Ca 2ϩ . When compared with wild type cells, H 2 O 2 production upon Ca 2ϩ treatment was trending higher in mutant cells, reaching significance at 0.4, 0.6, and 500 M Ca 2ϩ points when a paired t test analysis was done. Rotenone (Rot) addition was used in the assay as a positive control. Interestingly, in mutant cells H 2 O 2 production was significantly lower upon rotenone treatment compared with wild type. Results are mean Ϯ S.E., n ϭ 4 independent experiments. Statistical analyses were done using ANOVA followed by Tukey post test (*, p Ͻ 0.05, compared with 0 M Ca 2ϩ for each cell line), paired t test (#, p Ͻ 0.05) to compare the two cell lines at different data points, and Student's t test (&, p Ͻ 0.05) for rotenone treatment groups. FIGURE 6. Effects of PTP inhibitors on mitochondrial Ca 2؉ uptake capacity in STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells. Mitochondrial Ca 2ϩ uptake was measured as described under "Experimental Procedures." A combination of cyclosporine A plus ADP plus oligomycin was used to inhibit PTP and was added to the mitochondrial suspension (1 mg of protein in a 2-ml volume) before Ca 2ϩ additions. Additions of 20, 40, or 80 nmol of Ca 2ϩ were used in separate experiments but the regimen of Ca 2ϩ additions was always the same for both cell types in a given experiment. Our results indicate that PTP inhibitors addition caused an increase in Ca 2ϩ uptake capacity and this was significant for the mutant cells but not for the wild type cells. In the absence of PTP inhibitors, mutant cells had a significantly decreased Ca 2ϩ uptake capacity compared with the wild type cells. In the presence of cyclosporine A plus ADP plus oligomycin the difference between wild type and mutant was no longer significant. Data are mean Ϯ S.E., n ϭ 5 -9 independent measurements per group. Statistical analysis was done using ANOVA followed by Tukey post test (* and #, p Ͻ 0.05).
mitochondria which negatively impacts their function and hence likely impairs proper neuronal function.
Excitotoxicity has been suggested as a key mechanism that is responsible for neurodegeneration in HD (42), and dysfunction at the level of the mitochondria could be a mediator of this toxicity. Mitochondrial dysfunction could result in an increase in the sensitivity of neurons to glutamate, leading to Ca 2ϩ induced cellular dysfunction and eventually cell death (the role of mitochondria in excitotoxicity is discussed in Refs. [43][44][45]. It has been shown previously that decreases in oxidative phosphorylation and state 3 rates are early events in excitotoxicity, and occur prior to the commitment to cell death (46). Moreover, it has been shown that free M Ca 2ϩ concentrations cause significant decreases in oxidative phosphorylation in isolated mitochondria (29). To determine if mutant huntingtin alters mitochondrial response to Ca 2ϩ , we measured respiration rates in the presence of increasing Ca 2ϩ concentrations in mitochon-dria from wild type (STHdh Q7/Q7 ) and mutant huntingtin (STHdh Q111/Q111 )-expressing cells. We observed that the decrease in the state 3 rate (ADP phosphorylation rate) occurred at significantly lower Ca 2ϩ concentrations in mutant cells compared with wild type cells, suggesting an increased sensitivity to Ca 2ϩ .
Elevated mitochondrial Ca 2ϩ levels are usually associated with the opening of PTP. However, the mechanism of the oxidative phosphorylation decrease due to elevated Ca 2ϩ levels is not quite clear (discussed in Ref. 46). It has been suggested that permeabilization of the outer mitochondrial membrane leads to activation of an alternative respiratory pathway, which utilizes exogenous NADH oxidation on the outer mitochondrial membrane and translocation of subsequently reduced cytochrome c to the inner membrane where it feeds into complex IV (34,35). In our study, addition of NADH and cytochrome c caused an increase in the state 3 rates in the presence of Ca 2ϩ , but not in the absence, and cytochrome c alone did not increase state 3. These results indicate the Ca 2ϩ -induced decrease in oxidative phosphorylation is associated with increased permeabilization of the mitochondrial membrane but not with substantial loss of cytochrome c.
In this study respiration rates were measured in mitochondria isolated from STHdh Q7/Q7 (wild type) and STHdh Q111/Q111 (mutant) cells in KCl-based EGTA containing respiration buffer, and no significant differences were observed. In our previous study, respiration rates were measured in digitonin-permeabilized cells, in a sucrose-based buffer without EGTA, and significant decreases in the state 3 rates were observed in STHdh Q111/Q111 (mutant) cells (18). In the light of the current findings, it is likely that the experimental conditions used in our previous study were permissive for the mutant huntingtin-dependent mitochondrial defects. These conditions resulted in the deficits observed in state 3 rates for the mutant cells. Indeed it is likely that the presence of free Ca 2ϩ in the buffers (e.g. from the sucrose) likely resulted in an increase in mitochondrial membrane permeability to a greater extent in the mutant cells than the wild type cells, which caused the observed differences in respiration.
It has been shown that huntingtin associates with the outer mitochondrial membrane (47). We tested the integrity of outer mitochondrial membrane in basal conditions. We found that mitochondria from both cell lines have a good outer mitochondrial membrane integrity as state 3 rates did not increase in the presence of NADH and cytochrome c. The inner mitochondrial membrane was of good integrity as well, because NADH (inner mitochondrial membrane impermeable) did not work as respiratory substrate. Also, state 4 rates were comparable in the two cell lines, indicating similar levels of inner membrane proton leakage. However, it is still possible that changes in the integrity of mitochondrial membrane due to mutant huntingtin are quite subtle and could not be detected with the methods we used.
Mutant huntingtin-expressing cells (STHdh Q111/Q111 ) showed markedly enhanced ⌬⌿m reduction in response to increasing Ca 2ϩ concentrations both in vitro and in situ. Mitochondrial depolarization in response to Ca 2ϩ is caused by Ca 2ϩ uptake itself (partial and reversible depolarization) and by opening of the PTP when Ca 2ϩ uptake capacity is exceeded (complete depolarization) (38). Because in our experiments ⌬⌿m was measured in the population of mitochondria, determined ⌬⌿m values could indicate the portion of mitochondria undergoing PTP associated with complete depolarization or ⌬⌿m levels present in the majority of mitochondria at specific Ca 2ϩ concentrations. As previously described, ADP prevents PTP opening and stabilizes ⌬⌿m by the mechanism that includes binding and stabilization of adenine nucleotide translocator (ANT) in the conformation that prevents PTP opening (48). The differences in ⌬⌿m reduction between wild type and mutant cells were significantly attenuated in the presence of ADP. This suggested that differences in the threshold for PTP opening, significantly contributed to ⌬⌿m differences observed between wild type and mutant cells.
The reduction of ⌬⌿m could be observed at as low as 0.4 and 0.6 M Ca 2ϩ (concentrations at which decreases in oxidative phosphorylation were observed) in the mutant but not in the wild type cells. Because respiration was first monitored in state 4 followed by the induction of state 3, the observed differences in the state 3 rates are likely caused by the differences in the ⌬⌿m before the state 3 was induced. Indeed, no differences in ⌬⌿m were observed at 0.4 and 0.6 M Ca 2ϩ when measured in state 3 conditions. ROS production could contribute to Ca 2ϩ -induced PTP opening (discussed in Ref. 40). As described, determination of released H 2 O 2 is a common and the most reliable measure of mitochondrial ROS production (39). When added to isolated mitochondria, Ca 2ϩ caused a dose-dependent increase in the release of H 2 O 2 . However, no significant difference in released H 2 O 2, between wild type and mutant, was observed at different Ca 2ϩ concentrations (state 4). Therefore, the dramatic differences in ⌬⌿m and Ca 2ϩ uptake capacity observed between wild type and mutant are likely not caused by differences in the ROS production. However, in the presence of ADP (state 3), we observed higher levels of released H 2 O 2 in mutant mitochondria, which reached significance at several data points. Increased ROS in the presence of ADP would likely contribute to more pronounced decrease in state 3 rates observed in mutant mitochondria in the presence of Ca 2ϩ . Unexpectedly, we observed reduced H 2 O 2 release from mutant mitochondria upon rotenone treatment. Our results suggest that ROS production is likely not the mechanism for reduced threshold for PTP opening in the mutant cells. However, there was a trend for mitochondria from mutant cells to exhibit altered ROS homeostasis compared with wild type (modest increase in the presence of Ca 2ϩ and ADP; decrease in the presence of rotenone).
In this article we observed that mitochondria from mutant cells had reduced Ca 2ϩ uptake capacity compared with mitochondria from wild type cells. As suggested in multiple articles, decreases in mitochondrial Ca 2ϩ uptake capacity could be caused by a decreased threshold for PTP opening (7,49). In fact, when mitochondria from mutant cells were treated with PTP inhibitors (cyclosporine A, ADP, oligomycin), the defect in Ca 2ϩ uptake was almost abolished. Decreased mitochondrial Ca 2ϩ uptake has already been reported in other HD models. Panov et al. (7,13) reported decreased Ca 2ϩ uptake in mitochondria from HD lymphoblast cell line, and brain mitochondria from full-length huntingtin transgenic mice (YAC72). Recently, attenuated Ca 2ϩ uptake was reported in muscle mitochondria from R6/2 mice (30). Interestingly, recombinant truncated mutant huntingtin resulted in significant mitochondrial swelling at lower Ca 2ϩ loads than truncated wild type protein when added to isolated mouse liver mitochondria (47). This suggested that the mechanism of mutant huntingtin-induced mitochondrial dysfunction is possibly through its direct effects on mitochondria.
Using multiple mitochondrial functional assays, we demonstrated a mitochondrial Ca 2ϩ handling defect in mutant (STHdh Q111/Q111 ) cells. Although Ca 2ϩ overload could be the cause of the earlier PTP opening in the mutant mitochondria, our data suggest that it is the opening of the PTP at lower Ca 2ϩ concentrations resulting in membrane depolarization rather than Ca 2ϩ overload. ROS could contribute the PTP opening, but it did not correlate with the significant changes in ⌬⌿m observed in the mutant cells, suggesting a different mechanism. Based on these and other findings, we hypothesize that the increased sensitivity of mutant mitochondria to increases in intracellular Ca 2ϩ results in depolarization, and this is an upstream event in the pathological cascade followed by decreases in respiration and Ca 2ϩ buffering capacity. In the light of these current findings, successful PTP inhibition would be beneficial, but, as already discussed in terms of excitotoxicity (38), limiting mitochondrial Ca 2ϩ uptake by manipulating its setpoint (through Ca 2ϩ -uniporter inhibition or Na ϩ -Ca 2ϩ -exchanger activation) is also likely to be beneficial in the treatment of HD.