|Home | About | Journals | Submit | Contact Us | Français|
Levels of full-length huntingtin (FL htt) influence organ and body weight, independent of polyglutamine length. The growth hormone-insulin like growth factor-1 (GH-IGF-1) axis is well established as a regulator of organ growth and body weight. In this study, we investigate the involvement of the IGF-1 pathway in mediating the effect of htt on body weight. IGF-1 expression was examined in transgenic mouse lines expressing different levels of FL wild-type (WT) htt (YAC18 mice), FL mutant htt (YAC128 and BACHD mice) and truncated mutant htt (shortstop mice). We demonstrate that htt influences body weight by modulating the IGF-1 pathway. Plasma IGF-1 levels correlate with body weight and htt levels in the transgenic YAC mice expressing human htt. The effect of htt on IGF-1 expression is independent of CAG size. No effect on body weight is observed in transgenic YAC mice expressing a truncated N-terminal htt fragment (shortstop), indicating that FL htt is required for the modulation of IGF-1 expression. Treatment with 17β-estradiol (17β-ED) lowers the levels of circulating IGF-1 in mammals. Treatment of YAC128 with 17β-ED, but not placebo, reduces plasma IGF-1 levels and decreases the body weight of YAC128 animals to WT levels. Furthermore, given the ubiquitous expression of IGF-1 within the central nervous system, we also examined the impact of FL htt levels on IGF-1 expression in different regions of the brain, including the striatum, cerebellum of YAC18, YAC128 and littermate WT mice. We demonstrate that the levels of FL htt influence IGF-1 expression in striatal tissues. Our data identify a novel function for FL htt in influencing IGF-1 expression.
Huntington disease (HD) is an autosomal-dominant neurodegenerative disorder characterized by an age-dependent loss of motor coordination, cognitive impairment and psychiatric disturbances (1). In addition to these core neurological symptoms, a number of other abnormalities are observed in HD patients including weight loss, skeletal muscle wasting, osteoporosis and testicular degeneration (2).
Huntingtin (htt), a polyglutamine tract-containing protein that, in the mutant form, causes HD, is involved in a number of cellular functions including intracellular trafficking, transcriptional regulation, cell survival and neuroprotection (3,4). The disease results when the length of the polyglutamine tract in htt exceeds 35 glutamines, which confers a property whose consequences lead to pathogenesis during a normal human life-span (toxic gain-of-function; 5). There is also evidence that loss of normal htt function during the disease process, as a result of reduced htt protein levels or disrupted activity, may contribute to the features of HD (3,4). For example, htt has been shown to promote the expression of BDNF (6) and this activity is impaired in the presence of mutant htt (7), resulting in reduced BDNF expression in HD (6).
Not all functions of htt are disrupted in the presence of the expanded polyglutamine tract. This is underscored by the observation that mutant htt can rescue mice with a targeted disruption of the HD gene from embryonic lethality (8,9), and suggests that many of the essential functions of htt are maintained in the presence of the polyglutamine expansion. Indeed, another example of normal htt function that is unaffected by polyglutamine expansion is the influence of full-length (FL) htt levels on body weight (10). Increased expression of FL wild-type (WT) or mutant htt in mice is associated with a dose-dependent increase in body weight, an effect that is not accounted for by increased food consumption (10).
Insulin-like growth factor 1 (IGF-1) plays an important role in organ growth and body weight regulation (11). In this study, we aimed to investigate the potential involvement of the IGF-1 pathway in influencing the effect of FL htt on body weight. We demonstrate using htt YAC and BAC transgenic mice that htt may mediate its polyglutamine length-independent effect on body weight by influencing the expression of IGF-1. Given the ubiquitous expression of IGF-1 in the central nervous system (CNS), we also examined the impact of FL htt levels on IGF-1 expression in different regions of the brain including the striatum, the region most affected in HD (12). We demonstrate that FL htt levels modulate IGF-1 expression in the striatum, but not in the cortex or cerebellum. Our study identifies a novel biological function for htt with implications for the weight loss observed in patients with HD.
We have previously shown that FL htt transgenic YAC mice have increased body weight represented by an increase in both fat mass and fat-free (lean) mass (10; Fig. 1A), and that this effect was greater with increased FL htt expression levels and is independent of polyglutamine length (Fig. 1B). The increase in lean mass is reflected in increased weights of several peripheral organs, including the heart, liver, kidney and spleen (10). The lack of correlation between the extent of organ weight gain and organ-specific htt levels suggested that the effect of FL htt on organ and body weight gain might be mediated by a systemically circulating growth factor.
IGF-1 is the primary mediator of growth hormone action, and its anabolic effects and role as a trophic factor are well established (11,13). We therefore assessed IGF-1 levels in transgenic YAC mice expressing FL htt with 18 (YAC18), or 128 glutamines (YAC128) and their littermate controls at 2 months of age. Consistent with previous studies examining plasma IGF-1 levels in female and male mice on the FVB/N background (14), no significant differences in plasma IGF-1 levels between male and female animals were observed for all genotypes (data not shown). As no gender differences were observed, all the subsequent cohorts of animals employed in this study are of mixed gender. YAC18 and YAC128 animals express significantly higher levels of FL htt compared with WT littermates, with YAC18 (line 212) and YAC128 (line 53) expressing ~1.8 and 1.5 times the levels of FL htt observed in WT mice, respectively (Fig. 1B). Plasma IGF-1 levels were significantly higher in YAC128 compared with WT mice, and plasma IGF-1 levels were significantly higher in YAC18 compared with YAC128 mice (Fig. 1C). Levels of plasma IGF-1 and body weight follow the same pattern as levels of FL htt in these animals, with plasma IGF-1 levels and body weight being significantly higher in YAC18 compared with YAC128 mice, and levels of plasma IGF-1 and body weight being significantly higher in YAC128 compared with WT mice (FL htt levels: YAC18>YAC128>WT; plasma IGF-1 levels: YAC18>YAC128>WT; weight gain: YAC18>YAC128>WT).
To validate these observations in an independent transgenic FL htt mouse model of HD, plasma IGF-1 levels were assessed in BACHD mice expressing FL human htt with 97 glutamines (15) and littermate controls at 2 months of age. Total FL htt levels are significantly higher in BACHD compared with WT littermates, with BACHD mice expressing ~2 times the levels of FL htt observed in WT mice (15). Consistent with the polyglutamine length-independent increases in plasma IGF-1 levels seen in YAC18 and YAC128 mice, plasma IGF-1 levels were significantly higher in BACHD mice compared with WT control (Fig. 1D) and correlated with body weight gain in these animals (Fig. 1E; r2 = 0.288; P = 0.002).
Shortstop mice are transgenic YAC mice expressing an N-terminal htt fragment (amino acids 1–117) with 128 glutamines (16). The expression level of the transgenic htt protein in shortstop mice is similar to that of YAC128 mice (17,18). To evaluate whether over-expression of this N-terminal fragment of htt can modulate IGF-1 expression, we assessed plasma IGF-1 levels in shortstop mice. No difference in plasma levels of IGF-1 is observed between shortstop and WT animals (313.2 ± 43.6 ng/ml for shortstop versus 342.0 ± 37.6 ng/mL for WT; n = 10 for shortstop; 9 for WT; P = 0.140). Moreover, no difference in body weight is observed between shortstop and WT animals (16). These findings along with the findings in the YAC and BAC transgenic mice suggest that FL htt is required for the modulation of IGF-1 expression.
If elevated plasma levels of IGF-1 in YAC128 mice do underlie the increased body weight observed in YAC128 mice, then decreasing plasma IGF-1 levels would be expected to reduce the body weight of YAC128 mice to WT levels. To examine this possibility, we employed 17β-estradiol (17β-ED) which, when administered systemically, lowers the levels of circulating IGF-1 in mammals (19,20). YAC128 animals were treated with 17β-ED starting at 7.5 months of age, and their body weight was monitored on a biweekly-basis until 12 months of age. Control-treatment groups received either 17α-ED, an enantiomer with 200-fold less active as a transactivating hormone (21), or placebo. At 7.5 months of age, placebo-treated YAC128 animals weighed ~30% more than placebo-treated WT animals. Treatment of YAC128 animals with 17β-ED resulted in significantly lower plasma IGF-1 levels compared with placebo- and 17α-ED-treated controls (Fig. 2A). At 12 months of age, the body weight of 17β-ED-treated YAC128 animals was significantly lower than that of placebo- and 17α-ED-treated YAC128s animals and did not differ from that of placebo-treated WT animals (Fig. 2B). As plasma leptin levels per gram of fat in YAC128 and WT animals are not different (data not shown), we assessed plasma levels of leptin in these mice as a measure of fat mass. Plasma leptin levels were significantly higher in placebo-treated YAC128 mice compared with placebo-treated WT mice (Fig. 2C), consistent with significantly increased fat mass in YAC128 mice (10). Plasma leptin levels were significantly lower in YAC128 and WT mice treated with 17β-ED compared with their respective placebo- and 17α-ED treated controls (Fig. 2C). Finally, to examine whether 17β-ED treatment leads to decreased htt levels in YAC128 mice, plasma htt levels in 17α-ED- and 17β-ED-treated YAC128 animals were measured using a time-resolved FRET assay and normalized to htt levels in placebo-treated YAC128 animals. Htt levels in 17β-ED-treated YAC128 animals were not different compared with 17α-ED-treated YAC128 animals (Fig. 2D), suggesting that treatment with 17β-ED does not lead to decreased htt levels. These findings suggest that the increased body weight observed in YAC128 animals is mediated by IGF-1 and largely reflects increased fat mass.
Given the widespread expression of IGF-1 in CNS, we also examine whether FL htt levels modulate IGF-1 gene expression in striatal, cortical, cerebellar and hypothalamic tissues using quantitative RT-PCR. IGF-1 mRNA expression in striatal tissue of YAC18 mice was significantly higher than that of WT control mice (Fig. 3A). In contrast, IGF-1 mRNA expression in cortical, cerebellar and hypothalamic tissues of YAC18 mice were similar to that of WT mice (Fig. 3A). A similar pattern of IGF-1 mRNA expression was observed in YAC128 mice, with significantly increased IGF-1 mRNA expression in striatal tissue but similar levels in cortical, cerebellar and hypothalamic tissue compared with WT animals (Fig. 3B).
Consistent with the increased levels of IGF-1 mRNA in striatal tissue of YAC128 animals, cultured primary striatal neurons derived from YAC128 pups secreted significantly more IGF-1 compared with the cultures of primary striatal neurons from WT pups (Fig. 3C). In contrast, the amount of IGF-1 secreted by primary cortical and cerebellar neurons derived from YAC128 pups was similar to that of WT pup-derived cultures (Fig. 3C). Furthermore, IGF-1 secreted by primary striatal neurons steadily accumulated over time, and was significantly higher in YAC128 cultures at day 6 and day 9 compared with WT cultures (Fig. 3D).
We next examined the effect of reduced FL htt levels on IGF-1 expression. Significantly lower levels of FL htt are detected in the clonal striatal knock-in STHdhQ111 cells (Q111) compared with STHdhQ7 cells (Q7) as assessed by immunoblotting (Fig. 4A). It is important to interpret this result carefully as this decrease in FL htt levels in Q111 cells compared with Q7 cells may reflect reduced antibody affinity or access to binding epitope in mutant htt (Q111) compared with normal (Q7) htt, and may not reflect true differences in protein levels. However, this is unlikely as this decrease in relative FL htt in Q111 levels is seen with two different antibodies to htt with distinct binding epitopes (MAB2166 recognizes an epitope between amino acids 414 and 503 in htt, whereas BKP1 recognizes an epitope between amino acids 1 and 17 in the N-terminus of htt). Moreover, our observations of decreased FL htt levels in Q111 cells is consistent with previous reports employing different anti-htt antibodies with distinct recognition epitopes, showing lower htt band intensity in Q111 lysates compared with Q7 lysates (22,23), and decrease mutant htt expression in lymphoblastoid cells and brain tissues from HD patients compared with controls (24–28). This indeed may reflect increased sequestration of htt into aggregates or increase proteolytic activity in cells expressing mutant htt. Assessment of Q111 cells revealed significantly lower IGF-1 expression levels compared with Q7 cells (Fig. 4B). Furthermore, Q111 cells secreted significantly less IGF-1 compared with Q7 cells (Fig. 4C). These data are consistent with our findings in the htt transgenic YAC and BAC mice and support FL htt dose-dependent modulation of IGF-1 expression.
To examine the impact of reduced htt levels in vivo, we assessed R6/2 HD mice, which show progressive depletion of endogenous FL htt levels, at 7, 9 and 11 weeks of age (29). We observed significantly reduced plasma IGF-1 levels in R6/2 mice at 11 weeks of age (Fig. 4D). Consistent with published observations, these mice began to lose body weight starting at 9 weeks of age, reaching statistical significance at 11 weeks of age (Fig. 4E).
To examine whether IGF-1 expression is altered in HD patients, caudate tissue and skin-derived fibroblasts from HD patients and age-matched controls were assessed for IGF-1 expression using quantitative RT-PCR. Reduced expression of IGF-1 was observed in the caudate (Table 2)2) and in fibroblasts (Table 3) of patients with HD compared with age-matched controls.
In this study, we demonstrate that FL htt modulates IGF-1 expression in a dose-dependent fashion. We demonstrate that increased FL htt levels in htt transgenic YAC and BAC mice are associated with increased expression of IGF-1. This novel function of htt is independent of polyglutamine expansion as it is seen in both htt transgenic animals expressing normal (YAC18 mice) and expanded htt (YAC128 and BACHD mice), and requires FL htt as it is not observed in shortstop mice which only express an N-terminal fragment of htt.
Our findings showing modulation of IGF-1 expression by levels of not only WT but also mutant htt suggest that this novel function of htt is CAG-independent but rather dose-dependent. However, proteolytic cleavage and aggregation of FL htt over time are seen to a greater extent in the presence of CAG expansion. These events are expected to lead to a greater and more rapid decrease in FL htt levels in HD patients compared with controls, leading as a consequence to a greater decrease in IGF-1 levels in HD patients compared with controls.
Weight loss is a feature of HD (30), and is seen despite increased appetite and caloric intake in HD patients (31–33). The cause of the weight loss remains largely unknown. Our findings suggest that loss of htt function, specifically htt-mediated IGF-1 expression, as a result of decreased FL htt levels with disease progression, may contribute to the weight loss observed, and raise the possibility that IGF-1 replacement may be a viable approach to mitigate weight loss in HD. Consistent with this notion, the levels of FL htt are reduced in HD patients compared with controls (34). Furthermore, it has previously been shown that plasma IGF-1 levels are reduced in HD patients compared with controls (33), although another study has failed to confirm this (35). This may be due to differences in the methodology employed in collecting plasma samples and measuring IGF-1 levels in the two studies, and additional studies are necessary to resolve this discrepancy. Indeed, further indirect evidence in support of this is provided by early studies examining growth hormone (GH) responses in HD, where more prompt and increased GH release was observed in HD patients following stimulation compared with control (36–43). As IGF-1 release is elicited by GH and acts as part of a negative feedback loop to inhibit GH release (44), this enhanced GH release may reflect delayed feedback inhibition in HD patients due to reduced levels of plasma IGF-1.
Weight loss is observed in R6/2 mice, which express a small N-terminal fragment of mutant htt, and in a Hdh(CAG)150 knock-in mice with 150 CAG repeats (45–48). In R6/2 mice, the progressive weight loss is associated with progressive depletion of endogenous FL htt (29). An association between FL htt levels and body weight has been observed in other mouse models as well. For example, conditional htt knockout mice with reduced htt levels in adulthood show a decrease in body weight with proportional decreases in organ weight (49). Chimeric mice, in which 20–75% of cells do not express FL WT htt, show decreases in body weight as great as 40% (50). Mice expressing only 10–20% levels of endogenous FL htt have also been shown to have significantly lower body weight compared with WT (51).
In contrast, weight loss is not seen in transgenic YAC and BAC mouse models of HD, which express two copies of FL WT htt as well as one or more copies of FL mutant htt (10,15). Furthermore, body weight increases in these animals correlate with the levels of FL htt (10,15).
The role of IGF-1 in influencing body weight is well established (11). The findings from this study suggest that the body weight changes observed in the different HD mouse models likely reflect the impact of FL htt levels on IGF-1. Indeed, we observe decreases in IGF-1 levels in a mouse model that exhibits weight loss (R6/2 mice).
Significant alterations in the IGF-1 family of genes in HD have been observed in a number of gene expression and proteomic studies in HD patients and animal models. For example, decreased expression of IGF-1 and the IGF-1-binding proteins 2, 5 and 7 is observed in gene expression profiling studies performed on striatal and muscle tissues from HD mice and primary striatal neuron models of HD (52–55). Reduced induction of IGF-1 expression in clonal striatal cells is observed in differential gene expression profiling studies performed on stimulated cells expressing N548-Q105 or Q118 compared with WT (56). Furthermore, proteomic profiling of plasma from HD patients demonstrates that the levels of the IGFALS, a component of the IGF-1 ternary binding complex, is reduced in the plasma of HD patients (57). These findings from gene expression and proteomic studies support IGF-1-related alterations in HD, and are consistent with our findings of reduced IGF-1 expression in the caudate and skin-derived fibroblasts from HD patients compared with controls.
How might FL htt modulate IGF-1 levels? Htt may modulate the levels of IGF-1 by influencing its transcription. Htt has been shown to interact with a range of transcriptional factors and its role in transcriptional regulation is well established (4). This possibility is supported by our observation that htt-mediated changes in IGF-1 levels are seen not only on the protein but also on the mRNA level. Htt may modulate IGF-1 transcription by interacting with and influencing localization, level or the activity of transcription factor(s) mediating IGF-1 expression.
The role of htt in vesicular transport is well established (3) and studies have shown that increasing the levels of htt increases the rate of vesicular transport, whereas decreasing the levels of htt decreases the rate of vesicular transport (58). Indeed, htt has been shown to influence microtubule-dependent transport of BDNF-containing vesicles (58). Therefore, htt may also modulate IGF-1 levels by influencing the rate of transport of IGF-1-containing vesicles to the plasma membrane for fusion and release, although this possibility is less likely given that the changes in IGF-1 levels are detected at the mRNA level.
It is noteworthy that of the CNS tissues examined, modulation of IGF-1 expression was observed in the region of most prominent atrophy, namely the striatum. This raises the possibility that decreased IGF-1 expression with disease progression in these tissues specifically may at least partly be responsible for the selective atrophy observed in HD.
Studies in animal models of HD have demonstrated that the severity of the disease is dependent on the level of expression of mutant htt (59). Motivated by the observations that the removal of mutant htt ameliorates the behavioral and neuropathological HD-like phenotypes in mouse models of HD (60,61), several therapeutic initiatives employing such techniques as RNA interference or antisense oligonucleotide-based technology aimed at silencing the HD gene in the CNS are currently underway (62–67). For these studies, it is most important to have a measure of htt knockdown preferably without having to sacrifice the animal. The development of a biomarker that would preclude the need for invasive approaches would be of great importance for improving ease and efficiency of clinical trials and subsequent application of gene silencing strategies in patients.
Our data demonstrating modulation of IGF-1 expression by both FL WT and mutant htt levels raise the intriguing possibility that IGF-1 levels may serve as a biomarker of FL htt levels. The findings also provide the impetus for further studies to examine the effects of CNS htt knockdown on IGF-1 levels in the cerebrospinal fluid and in plasma to assess its utility as a biomarker of htt expression levels.
Furthermore, studies in human populations and mice reveal that IGF-1 levels are genetically regulated (68). There is a significant variance in IGF-1 levels in different human races and between different mouse strains (14,68). Furthermore, in mice, age and gender also influence IGF-1 levels. Of note, in this and other studies of the FVB/N strain, minimal differences in the levels of IGF-1 between the different genders are present compared with other strains such as 129S1 mice where gender can confound the results (14). Also, in some strains such as C57Bl/6, IGF-1 levels are low in both genders compared with the FVB/N strain. The low baseline levels of IGF-1 in this strain may provide too small a window for detecting weight changes in the presence of FL htt in this strain (14). Clearly, these studies suggest that the assessment of IGF-1 levels in any trials involving knockdown of htt will be useful in the consideration of IGF-1 as a biomarker for htt levels.
Experiments were carried out on YAC18 (line 212), YAC128 (line 53) and BACHD mice generated to express FL htt with 18, 128 and 97 glutamines, respectively (15,16,69), and shortstop mice expressing an N-terminal fragment of htt with 128 glutamines (17). YAC18, YAC128, BACHD and shortstop mice were maintained on the FVB/N strain background (Charles River, Wilmington, MA, USA). All animals were bred at the animal facility of the Centre for Molecular Medicine and Therapeutics at the University of British Columbia and were group-housed in numerical birth order with littermates of mixed genotype, with the exception of the R6/2. Female R6/2 mice, which were originally obtained from Jackson ImmunoResearch Laboratories, were bred at the animal facility of Brigham and Women's Hospital, Harvard Medical School, and have a shorter CAG repeat than other R6/2 mice (110–115 versus 150) (70). Mice were kept on a normal light/dark cycle where lights were turned off at 8:00 p.m. and on at 6:00 a.m. Experimenters were blind to the genotype of the mice. Body weights were taken at 9:00 a.m. at the specified time points.
Immunoblots were performed on tissue samples frozen immediately following euthanasia of mice by terminal intraperitoneal injection of 2.5% avertin. Protein lysates were separated on a NuPAGE 4–12% Bis-Tris gel (Invitrogen). Following transfer to a membrane, htt protein was detected using the htt-specific MAB2166 (1:2500; Chemicon, USA) or monoclonal BKP1 htt antibody (1:2500). β-Tubulin protein was detected with an anti-β-tubulin mouse monoclonal antibody (1:2000; Sigma). An IR dye 800CW goat anti-mouse secondary antibody (1:10 000; Rockland) and the LiCor Odyssey Infrared Imaging system were used for detection and quantification.
Human HD and control brain postmortem tissues were obtained from the Canadian Brain Tissue Bank and used in this study following review and approval by the University of British Columbia Children and Women's Research Review Committee (CW06-0171/H05-70532). Skin-derived fibroblasts from HD patients and age-matched controls were obtained from the University of Copenhagen and the work performed on these cells complies with current Danish laws. Total RNA was isolated using RNeasy Micro Kit (Qiagen) according to the manufacturers’ instructions. Five hundred nanograms of DNAse I-treated RNA were reverse-transcribed with SuperScript II (Invitrogen) to generate cDNA for real-time PCR using SYBR Green PCR Master Mix (Applied Biosystems) in an ABI PRISM 7500 Sequence Detection system. Primers were designed using Primer Express software (Applied Biosystems) or retrieved from Primer Bank (71). The primer sequences used are tabulated in Table 1.
Anterior striata from embryonic day 15.5 or postnatal day 0 mice were dissected, and neurons were dissociated and cultured as described previously (72). Culture media were collected at days 10 and 11 for cultures prepared from E15.5 mice, and on days 4, 6 and 9 for cultures prepared from P0 mice. Primary neuronal cultures were prepared from dissociated cerebellum of postnatal day 7 mice or cortex of day 15.5 embryos. An equal number of neurons were plated per well per genotype for each of the cultures. Culture media were collected and frozen immediately at −80°C for IGF-1 measurements.
17β-estradiol, 17α-estradiol and placebo pellets were obtained from Innovative Research of America. Body weights were taken at 9:00 a.m. every 2 weeks from 2 to 12 months of age. At 7.5 months of age, placebo, 17β-estradiol (1.7 mg; ~0.55 mg/kg/day) or 17α-estradiol (1.7 mg; ~0.55 mg/kg/day) pellets were implanted in WT and YAC128 mice by making a skin incision in the nape of the neck of each animal and placing the pellet subcutaneously. At 12 months of age, blood was obtained by cardiac puncture with heparinized syringes, collected into EDTA-coated tubes and spun immediately to obtain plasma. The plasma was frozen immediately at −80°C for IGF-1 and leptin measurements.
Detection of mutant htt in plasma samples was performed as described previously (73). To each 10-µl sample, 2 µl of antibody master mix was added, resulting in a final dilution of 1 ng/well 2B7 terbium-labeled antibody and 10 ng/well MW1-D2-labeled antibody in 50 mm NaH2PO4, 0.1% bovine serum albumin (BSA), and 0.05% Tween 20. Plates were measured with a xenon lamp Victor Plate Reader (Perkin Elmer) after excitation at 340 nm (time delay 50 µs, window 200 µs). The signal measured at 615 nm resulted from the emission of the terbium-labeled antibody and was used for normalization of potential signal artifacts. The htt-specific signal at 665 nm resulted from emission of the D2-labeled antibody after time-delayed excitation by the terbium. The relative htt concentration is represented by the 665/615-nm ratio. Signal from plasma samples obtained from age-matched control WT animals was used for background correction.
For plasma, animals were bled from the sephanous vein with EDTA-coated capillary tubes (Sarstedt, Germany). Blood was centrifuged for 10 min at 3500 rpm and plasma was transferred to an eppendorf tube and stored at −80°C until analysis. IGF-1 and leptin levels in culture media and in plasma were measured using Quantikine Mouse IGF-1 and leptin ELISA assays (R&D Systems) as per the manufacturer's instructions.
Data are expressed as means ± SEM. Where appropriate, results were interpreted using one-way ANOVA with a Tukey post-hoc test. Pairwise comparisons between genotypes were assessed with a Student's t-test. Linear regression analyses for r2 and P-values were calculated with GraphPad Prism version 5.0a. Differences were considered statistically significant when P < 0.05.
X.W.Y. is supported by NINDS/National Institutes of Health (NS049501 and NS049501-05S1), Hereditary Disease Foundation and CHDI Foundation. M.R.H. is supported by grants from the Canadian Institutes of Health Research, the Huntington Society of Canada, the Huntington's Disease Society of America and the CHDI Foundation.
We thank Lily Liu, Li-Ping Cao, Crystal Doty and Ge Lu for technical assistance. We also thank Dr. Marcy MacDonald (Massachusetts General Hospital, Boston, MA, USA) for the generous gift of the STHdh striatal cell lines.
Conflict of Interest statement. M.A.P. is the recipient of doctoral awards from the Canadian Institute of Health Research and the Michael Smith Foundation for Health Research (MSFHR). M.R.H. is a Killam University Professor and holds a Canada Research Chair in Human Genetics.