Significantly differential diffusion of neuropathological aggregates in ...

Brain Struct Funct (2013) 218:283–294 DOI 10.1007/s00429-012-0401-x

ORIGINAL ARTICLE

Significantly differential diffusion of neuropathological aggregates in the brain of transgenic mice carrying N-terminal mutant huntingtin fused with green fluorescent protein Pei-Hsun Cheng • Chia-Ling Li • Lu-Shiun Her Yu-Fan Chang • Anthony W. S. Chan • Chuan-Mu Chen • Shang-Hsun Yang

•

Received: 22 October 2011 / Accepted: 21 February 2012 / Published online: 16 March 2012 Ó Springer-Verlag 2012

Abstract Huntington’s disease (HD) is a genetically neurodegenerative disease, affecting the central nervous system and leading to mental and motor dysfunctions. To date, there is no cure for HD; as a result, HD patients gradually suffer devastating symptoms, such as chorea, weight loss, depression and mood swings, until death. According to previous studies, the exon 1 region of the huntingtin (HTT) gene with expanded CAG trinucleotide repeats plays a critical role in causing HD. In vitro studies using exon 1 of HTT fused with green fluorescent protein (GFP) gene have facilitated discovering several P.-H. Cheng and C.-L. Li contributed equally to this work.

Electronic supplementary material The online version of this article (doi:10.1007/s00429-012-0401-x) contains supplementary material, which is available to authorized users. P.-H. Cheng Y.-F. Chang S.-H. Yang (&) Department of Physiology, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan e-mail: [email protected] C.-L. Li S.-H. Yang Institute of Basic Medical Sciences, National Cheng Kung University, Tainan 70101, Taiwan L.-S. Her Department of Life Sciences, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 70101, Taiwan A. W. S. Chan Department of Human Genetics, Emory University School of Medicine, Atlanta, GA 30329, USA C.-M. Chen Department of Life Sciences, Agricultural Biotechnology Center, National Chung Hsing University, Taichung 40227, Taiwan

mechanisms of HD. However, whether this chimera construct exerts similar functions in vivo is still not clear. Here, we report the generation of transgenic mice carrying GFP fused with mutant HTT exon 1 containing 84 CAG trinucleotide repeats, and the evaluation of phenotypes via molecular, neuropathological and behavioral analyses. Results show that these transgenic mice not only displayed neuropathological characteristics, observed either by green fluorescent signals or by immunohistochemical staining, but also progressively developed pathological and behavioral symptoms of HD. Most interestingly, these transgenic mice showed significantly differential expression levels of nuclear aggregates between cortex and striatum regions, highly mimicking selective expression of mutant HTT in HD patients. To the best of our knowledge, this is the first report showing different nuclear diffusion profiling in mouse models with transgenic mice carrying the exon 1 region of mutant HTT. Our model will be beneficial for tracing the expression of mutant HTT and accelerating the understanding of selective pathological progression in HD. Keywords Huntington’s disease (HD) Differential diffusion of neuropathological aggregates Green fluorescent protein Transgenic mice Animal model

Introduction Huntington’s disease (HD) is a genetic autosomal dominant disease which is caused by the mutation of Huntingtin protein (HTT) translated from huntingtin gene (HTT; as known as IT15 gene) (The Huntington’s Disease Collaborative Research Group 1993). HD is a neurodegeneration disease that affects the central nervous system in humans. Typical neuropathological features of HD are

123

284

nuclear aggregates, intranuclear aggregates (inclusion bodies) and neuropil aggregates, caused by misfolded HTT. Symptoms of HD include mental deterioration, motor dysfunction, emotional disturbance and cognitive deficits, leading to HD patients suffering serious pain until death, since there is no cure for HD (Li and Li 2006; Walker 2007). One of the important characteristics of HTT is the CAG trinucleotides in the exon 1 region, and an abnormal expansion of CAG trinucleotides ([37 repeats) usually leads to the onset of HD. Another important feature of HTT resulting in HD is N-terminal fragments, where smaller fragments carrying longer expanded CAG trinucleotides will cause severer phenotypes of HD. In order to study the disease progression of HD, several animal models have been generated (Yang and Chan 2011), such as flies (Williams et al. 2008), rats (von Horsten et al. 2003), mice (Schilling et al. 1999) and non-human primates (Yang et al. 2008; Chan and Yang 2009), because clinical specimens of HD are in limited supply. The mouse model, especially, is commonly used for HD studies due to its easy handling, relatively shorter life span and highly genomic and physiological similarity to humans. Several different HD transgenic mice that display molecular, pathological and behavioral abnormalities to mimic HD patients have been generated. R6/2 expressing exon 1 of HTT with 115–156 CAG repeats under the control of human HTT promoter was the first transgenic mouse model, and has been widely used in HD research (Davies et al. 1997; Mangiarini et al. 1996). This model shows motor deficits, metabolic dysfunction, abnormal cognitive ability and progressive neuronal pathology. These are typical phenotypes highly mimicking features of HD patients, suggesting that this model has high feasibility to serve as a HD animal model. However, the shorter life span of this model has led to a higher challenge of maintenance. Besides the generation of R6/2, several other HD transgenic mouse models also have been established. Commonly used examples of HD transgenic mice include N171-82Q mice expressing the first 171 amino acids of HTT with 82 CAG repeats under the control of prion promoter (Schilling et al. 1999), HTT yeast artificial chromosome (YAC) mice carrying full-length HTT with 46 or 72 CAG repeats under control of human HTT promoter (Hodgson et al. 1999), as well as HTT knock-in mice expressing mouse HTT with expanded CAG repeats under control of mouse endogenous regulatory elements (Lin et al. 2001). These transgenic mice display different onset of HD, and show pathological and behavioral characteristics at different stages. Therefore, these HD mice provide useful models for studying HD phenotypes. Nevertheless, in an attempt to investigate the underlying cellular and molecular mechanisms for HD, none of these

123

Brain Struct Funct (2013) 218:283–294

mouse models can provide trackable and visible markers to monitor expression profiling of mutant HTT. Green fluorescent protein (GFP) is a powerful reporter gene widely used in biomedical research (Chalfie et al. 1994). Upon gene engineering to fuse GFP with target genes, expression profiles of target genes are easily observed without any staining, which accelerates a better understanding of protein characteristics at the cellular level. In HD studies, GFP has also been broadly used to trace the expression profiling of HTT in vitro (Cornett et al. 2005). Using these chimera constructs not only provides the information of expressional locations of HTT, but also shows the typical pathological features, such as intranuclear aggregates. In addition, GFP fusion protein also serves as an excellent marker to trace axonal transportation via time lapse photography (McGuire et al. 2006), which provides a unique methodology to observe the real-time progression of transportation and to elucidate variant biofunctions of HTT. However, in vivo studies using GFPHTT fusion protein to generate transgenic mice have not been reported, and so whether this fusion protein system could work or play similar roles for physiological and pathological phenotypes in vivo is not known. In this study, we generated GFP-HTT transgenic mice to mimic phenotypes of HD patients. These mice showed the expression of mutant HTT protein, and displayed abnormally pathological phenotypes, such as nuclear aggregates and intranuclear aggregates, and behavioral phenotypes of motor deficits. More importantly, the analysis of pathological brain sections showed the significantly differential expression level of nuclear aggregates between the cortex and striatum regions, highly mimicking selective expression of mutant HTT in HD patients. Thus, the achievement of this specific transgenic mouse line may accelerate the understanding of selective pathological progression in HD.

Materials and methods Preparation of transgene construct The transgene of Ubiquitin-GFP-HTT84Q (Ubi-G-HTT84Q) was used in the generation of transgenic mice. This transgene included exon 1 of the human HTT gene with the expansion of 84 CAG trinucleotide repeats, which encodes 153 amino acids, in the 30 end of GFP gene, where the stop codon was removed and mutant HTT was ligated in frame, under the control of human ubiquitin promoter. This transgene for the generation of transgenic mice was purified using QIAquick gel extraction kit (Qiagen), and then the concentration of DNA was adjusted to 2 ng/lL for transgenesis.


Functional test of transgene in vitro In order to confirm the functional expression of the Ubi-GHTT84Q transgene, in vitro transfection was performed in human embryonic kidney (293 FT; Invitrogen) and mouse neuroblastoma (N2a; ATCC) cells. 293 FT cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen) containing 10% heat-inactivated fetal bovine serum (Hyclone), 2 mM L-glutamine (Invitrogen) and 100 U/mL penicillin/100 lg/mL streptomycin (Invitrogen). N2a cells were cultured in modified Eagle’s medium (MEM; Invitrogen) with 10% heat-inactivated fetal bovine serum (Hyclone), 2 mM L-glutamine (Invitrogen), 1 mM sodium pyruvate (Sigma) and 100 U/mL penicillin/ 100 lg/mL streptomycin (Invitrogen). Transient transfection was preceded by using Lipofectamine 2000 transfection reagent (Invitrogen), and fluorescent images were captured using Leica Application Suite software (Leica) under a DM IRB fluorescent microscope (Leica) 48 h after transfection. In addition, cell samples were collected for Western blotting to determine protein expression profiling as described below. Generation of Ubi-G-HTT84Q transgenic mice All animal procedures performed in this study were approved by the Institutional Animal Care and Use Committees at National Cheng Kung University. In order to perform super-ovulation, 4-week-old FVB female mice received an intraperitoneal injection of 5 IU pregnant mare’s serum gonadotropin (PMSG; Sigma) followed by a 5 IU human chorionic gonadotropin (hCG; Sigma) injection administered 48 h following the PMSG injection. These injected mice were then mated with male FVB mice, and killed at 18 h after hCG injection for the collection of fertilized oocytes. Pronuclear DNA microinjection followed the protocol as previously described (Hogan 1994). Simply, 2 ng/lL purified Ubi-G-HTT84Q DNA was injected into each pronucleus of fertilized oocytes via a micromanipulation system (Eppendorf) under a DM IRB inverted microscope (Leica). The injected zygotes were transferred into the oviducts of pseudopregnant female mice, and then carried to term. Genotyping for transgenic mice Mouse tail snips were collected from pups at 10–14 days after birth for genomic DNA extraction, and then subjected for genotyping via following PCR detection. To detect the Ubi-G-HTT84Q gene, forward primer (50 -GGCGAC CCTGGAAAAGCTGA-30 ) and 30 reverse primer (50 -TG AGGAAGCTGAGGAGGCGG-30 ) were used for PCR

285

amplification and the expected PCR amplicon was 345 bp. The PCR program was performed at 94°C for 5 min followed by 35 cycles of 94°C for 30 s, 62°C for 30 s and 72°C for 20 s, and finally at 72°C for 5 min. Genomic DNA from wild-type mice was used as a control. To determine the number of CAG repeats via DNA sequencing, a forward primer (50 -GGCGACCCTGGAAAAGC TGA-30 ) was used for sequencing. RNA extraction, reverse transcription (RT), RT real-time quantitative PCR (RT-Q-PCR) analysis Tissue samples were subjected to RNA extraction by using the Trizol reagent (Invitrogen) followed by DNase treatment (New England Biolabs). RT was performed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems), and cDNA was used for Q-PCR. To quantify the relative HTT RNA expression levels, the following specific primer sets were designed for detecting exon 1 (HD HD-Exon 1-F and HD-Exon 1-R) and endogenous HTT (HD-Exon 10-F and HD-Exon 12-R): HD-Exon 1-F: 50 -ATGGCGACCCTGGAAAAGCT-30 , HD-Exon 1-R: 50 -TGCTGCTGGAAGGACTTGAG-30 , HD-Exon 10-F: 50 -TCAAGAAAACAAAAAGGCAAAGTG-30 and HD-Exon 12-R: 50 -GTGGAAACCCCTGAAGAAGCA-30 . 29 Power SYBRÒ Green PCR Master Mix (Applied Biosystems) was mixed with the primers and cDNA, and subjected to the StepOnePlus Real Time PCR system (Applied Biosystems) at 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 60 s. 18S primers (18S-F: 50 -CGGCTACCA CATCCAAGGAA-30 and 18S R: 50 -CCTGTATTGTT ATTTTTCGTCACTACCT-30 ) were used as an internal control to normalize expression levels. Western blotting analysis Cell and tissue samples were lysed for crude protein extraction by using a sonicator (Qsonica), and protein concentration was measured using the Bradford assay (Pierce). 9% sodium dodecyl sulfate polyacrylamide gel electrophoresis (Bio-Rad) was performed to separate crude proteins, and then proteins were transferred onto a PVDF membrane (Bio-Rad) using protein mini trans-blot cells (Bio-Rad). For the Western blotting, PVDF membranes were blocked in 5% skimmed milk, and then incubated with the primary antibodies, including mouse monoclonal mEM48 (gift from Dr. Xiao-Jiang Li; 1:50 dilution), GFP (Roche; 1:2,000 dilution) and c-tubulin (Sigma; 1:10,000 dilution) antibodies. Secondary peroxidase-conjugated antibodies (Jackson ImmunoResearch laboratories) were then applied, and protein expression levels were measured by an Amersham ECL kit (PerkinElmer).

123

286

Immunohistochemical (IHC) staining for mouse brain Mice were anesthetized and perfused using 4% paraformaldehyde (Sigma). Postmortem brain tissues were then removed and subjected to frozen sectioning, with sections being cut at 20 lm. These brain sections were examined using 3,30 -diaminobezidine (DAB) and fluorescent immunohistochemistry staining. For DAB immunohistochemistry, sections were incubated with 0.3% hydrogen peroxide for 15 min, blocked for 1 h at room temperature, and incubated with mEM48 (1:50 dilution) at 4°C overnight. After washing with DPBS, the brain sections were processed with Vectastain Elite ABC kit (Vector Laboratories), and immediately stained with DAB (Vector Laboratories) as required. Brain sections were mounted on the slides with mounting media (Ted Pella), and images were examined by a DM2500 microscope (Leica) and captured by Leica Application Suite software (Leica). Typical intranuclear aggregates, nuclear aggregates and neuropil aggregates were described in previous studies (Schilling et al. 1999; Yang et al. 2008; Maat-Schieman et al. 2007), and indicated in Fig. 5 and online resource 4 as well. Quantitative analyses for different aggregates were performed by using Metamorph software (Molecular Devices). For the fluorescent immunohistochemistry staining, sections were processed with similar steps, including blocking for 1 h, treating with mEM48 (1:50 dilution), washing using DPBS, incubating with goat antimouse IgG Alexa 594 secondary antibody (Invitrogen; 1:2,000 dilution), staining nucleus with Hoechst 33342 (Sigma) and the images were then examined under a DM2500 fluorescent microscope (Leica). Behavioral tests The genotypes of mice used for behavioral testing were confirmed by using PCR analysis, and the mice were housed together with mixed genotypes under constant temperature and a 12 h:12 h light:dark cycle. All of the following behavioral testing was performed blindly. Life span Observation was recorded everyday. Clasping Mice were held upside down by the tail for 1 min, while the clasping phenotype was observed. Clasping is defined as the mice tightly holding their four limbs together when suspended by the tail. Scoring was evaluated based on the severity of clasping. A score of 2 represents the more extreme condition of all four limbs clasping, while a score

123


of 1 represents the clasping of either the forelimbs or the hindlimbs, and a score of 0 represents the absence of clasping. Rotarod Mice were trained at 12 months of age every 3 days for three times under the condition of a constant 6 revolutions per minute (rpm) speed for 5 min in 2 trials, and then mice were tested for three consecutive days. During the testing phase, each mouse was placed on an accelerating rotarod (Singa) at an increasing speed, ranging from 0 to 40 rpm for up to 5 min per trial. Three trials were performed with 1 min intervals and the time when mice fell off was recorded. Footprinting Mice were trained to walk inside and along a closed tunnel with 10-cm wide and -high walls three times before the tests. For the footprinting tests, white paper was placed on the floor of tunnel, while the forefeet of mice were painted with blue ink and the hindfeet were painted with red ink to aid in observation. The gait features for each mouse were observed and recorded as each mouse was put into one end of the tunnel for the trial. Grasping strength Mice were trained twice at 3 months of age, and then grasping strength was longitudinally recorded twice a month until 13 months of age. To measure the grasping strength, the mouse was held by the tail and the four limbs were placed on the apparatus (Bioseb). The mouse was gently pulled back until it released its grasp from the apparatus, and then the grasping strength was recorded by the apparatus. Each mouse was tested for 3 trials with 1 min intervals. Statistical analysis Data were expressed as mean ± standard deviation. Differences between groups were analyzed using one-way analysis of variance with post hoc tests in commercial statistical software (GraphPad Prism 4.02; GraphPad Software). In some situations, Student’s t test was used to compare differences between particular groups. Statistical significance was set at p \ 0.05.

Results Since HTT is expressed ubiquitously in all tissues (Li and Li 2006; DiFiglia et al. 1995), a stronger promoter, human ubiquitin promoter, was used in this study, as described


287

before (Yang et al. 2008). To accelerate the onset of HD in transgenic mice, mutant HTT containing only the exon 1 region with 84 CAG trinucleotide repeats was used to construct the functional gene. In order to generate the GFPHTT fusion protein, the stop codon of the GFP gene was removed and the mutant HTT gene was ligated into the 30 end of GFP gene in frame. This construct was driven by an ubiquitin promoter and was named Ubi-G-HTT84Q. We first tested whether this transgene could properly express mutant HTT. The Ubi-G-HTT84Q transgene was transiently expressed in cell models, including human embryonic kidney (293 FT) and mouse neuroblastoma (N2a) cells, via transfection. As shown in Fig. 1, the expression of GFP-HTT signal was observed under fluorescent microscope in both 293 FT (Fig. 1a–d) and N2a cells (Fig. 1e–h). Not only nuclear aggregates were observed in both transfected cell lines, but also intranuclear aggregates (inclusion bodies) appeared inside the cells. In

addition, Western blotting using the HTT-specific antibody (mEM48) and GFP antibody showed the expression of HTT oligomers (Fig. 1i, j), suggesting that expressed HTTs were misfolded and moved to a position of higher molecular weight. These results indicate that the Ubi-G-HTT84Q transgene can functionally express mutant HTT protein and lead to pathological phenotypes of HD in embryonic kidney cells or neuroblastoma cells. Next, we used this Ubi-G-HTT84Q construct to generate transgenic mice using pronuclear microinjection as described previously (Hogan 1994). After transferring injected embryos into pseudopregnant female mice and carrying to term, 6 out of 25 transgenic mouse founders were obtained and confirmed by PCR and Southern blotting. Copy numbers of transgene in these transgenic mice were estimated by using Southern blotting, and different transgenic founders displayed various copy numbers ranging from 3 to 12 (Online recourse 1). Transgenic

Fig. 1 Expression of Ubi-G-HTT84Q in vitro. Ubi-G-HTT84Q transgene was transiently transfected into 293 FT (a–d) and N2a cells (e–h), and expression of GFP-HTT signals was observed via a fluorescent microscope. a, e show bright-field images; b, f show nucleus staining using Hoechst 33342; c, g show GFP fluorescent

images; d, h show merged images from nucleus staining and GFP signals. Arrow heads indicate nuclear aggregates, and arrows indicate intranuclear aggregates. Western blotting was also performed in transfected 293 FT (i) and N2a (j) cells using the mEM48, GFP and T-tubulin antibodies. Scale bar 50 lM

123

288


Fig. 2 Characterization of UbiG-HTT84Q transgenic mice. a PCR results show the germ line transmission of the No. 22 transgenic founder. 22-1*2219 were selected positive F1 offspring from No. 22 founder. b Real-time RT-PCR results using brain samples show that transgenic mice (GHD-1*4) overexpressed mutant HTT exon 1 without disrupting endogenous HTT compared to those of non-transgenic mice (WT-1*4). c Western blotting using brain samples shows that transgenic mice (GHD-1*4) expressed mutant HTT, whereas non-transgenic mice did not (WT-1*4). Upper panel is the result using a mEM48 antibody, and bottom panel is the result using a T-tubulin antibody as internal control. GHD cells were the transfected cells serving as a positive control

founders were selected for breeding, and germline transmission was confirmed upon genotyping of F1 offspring (Fig. 2a). Additionally, the length of CAG trinucleotide repeats was confirmed by PCR (Fig. 2a) and sequencing (data not shown), showing that 84 complete CAG trinucleotide repeats were integrated into the mouse genome without CAG instability (Yang et al. 2008). To confirm the expression of mutant HTT mRNA in transgenic mice, brain tissues from F1 offspring were subjected to RNA extraction, and expression profiles were quantified using realtime RT-PCR. In Ubi-G-HTT84Q transgenic mice, the mRNA expression level of exon 1 region was higher than that of non-transgenic littermates, where the mRNA expression of endogenous HTT showed similar expression profiles (Fig. 2b). Since the Ubi-G-HTT84Q transgene only contains the exon 1 region of HTT, this data suggests that these transgenic mice overexpressed mutant Ubi-G-HTT84Q transgene without disrupting endogenous HTT. Expression of mutant HTT protein leads to pathological and behavioral phenotypes in transgenic mice (Schilling et al. 1999; Davies et al. 1997), so we first examined whether Ubi-G-HTT84Q transgenic mice expressed mutant HTT. Similar to the in vitro cell models, a specific antibody, mEM48, against mutant HTT protein was used for Western blotting. Since the brain is the most susceptible tissue in HD (Li and Li 2006; Walker 2007), the expression of mutant HTT protein in brain tissues was first assessed. Brain samples from transgenic mice and non-transgenic littermates at 8 weeks of age were examined and mutant

123

HTT proteins were detected in Ubi-G-HTT84Q transgenic mice, but not in non-transgenic littermates (Fig. 2c). Since mutant HTT is expressed ubiquitously in HD patients, and the misfolded proteins increasingly accumulate to manifest pathological characteristics (Li and Li 2006; Walker 2007), we next examined protein levels in different tissues of transgenic mice at 2, 6 and 12 months of age. The expression of mutant HTT was detected in the heart, liver, lung, spleen, kidney, muscle, and brain at different ages, with the highest levels being found in the brain (Fig. 3). In addition, the expression of a higher molecular weight HTT, which represents the presence of aggregates of mutant HTT, was observed to steadily increase along with age in different tissues, especially in brain tissues. These results suggest the progression of HD in these transgenic mice at the protein level, in a manner highly similar to that found in HD patients. Misfolded HTT with a higher molecular weight leads typically to neuropathological features, such as nuclear aggregates, intranuclear aggregates (inclusion bodies) and neuropil aggregates, both in human patients and animal models (Schilling et al. 1999; Yang et al. 2008; MaatSchieman et al. 2007). Since mutant HTT was aggregated in Western blotting results, neuropathological characteristics were examined in three transgenic mouse lines at 2, 6 and 12 months of age. At 2 months of age, we did not observe any significant pathological characteristics, such as nuclear aggregates or neuropil aggregates; however, weaker signals of neuropil aggregates were observed before


Fig. 3 Expression of mutant HTT in different tissues at different stages. Different tissues were collected at 2 (a), 6 (b) and 12 (c) months of age, and then subjected to Western blotting using the mEM48 antibody. GHD cells were the 293 FT transfected cells serving as a positive control, and 293 FT cells served as a negative control (WT cells)

the onset of motor dysfunction, which will be described in the following behavioral tests (Online resource 2). Based on the GFP fluorescent signal, more significant nuclear aggregates and inclusion bodies were observed in the cortex and striatum regions, and these signals were colocalized with those from mEM48 immunofluorescent staining at 12 months of age (Fig. 4). The pathological characteristics were easily visualized when 3, 30 -diaminobezidine (DAB) staining was used to stain postmortem brain sections. Nuclear aggregates, intranuclear aggregates and neuropil aggregates were more obviously stained in the cortex and striatum of transgenic mice at 12 months of age (Fig. 5), whereas mutant HTTs were not detected in non-transgenic littermates (Online resource 3). Further

289

examination revealed that more nuclear aggregates and inclusion bodies were observed in the striatum region, whereas neuropil aggregates were predominantly present in the cortex region (Fig. 5a–c). In order to show the more detailed locations of these aggregates, nucleus staining using hematoxylin were performed (Online resource 4), further supporting the neuropathological phenotypes in these transgenic mice. Moreover, quantitative analyses were also performed to consolidate that significantly differential diffusion of neuropathological aggregates in the different brain regions of Ubi-G-HTT84Q transgenic mice (Fig. 5d). Behavioral phenotypes are important evidence to confirm neuropathological symptoms in HD. In order to demonstrate whether the observed pathological characteristics lead to behavioral symptoms, and particularly since motor deficit is an important symptom in HD, tests of clasping, grasping strength, footprinting and rotarod were performed to examine motor functions in No. 22 transgenic mouse line. Mice with mixed gender were used in these behavioral studies. Owing to severe symptoms of HD in these transgenic mice, behavioral tests were terminated at 13 months of age, and mice were killed for sample collections. First, basic information of life span was recorded, based on daily observation. Results showed that some transgenic mice started to die at 5 months of age, and the survival rate at 13 months of ages was approximately 33%, whereas the survival rate of wild-type mice was 100% (Fig. 6a). For the longitudinal behavioral tests, clasping and grasping strength were recorded from 4 months of age to 13 months of age. In Ubi-G-HTT84Q transgenic mice, significant clasping behavior was continuously observed after 8.5 month of age (p \ 0.05; Fig. 6b). At the end stage, most of the transgenic mice showed the worst clasping phenotypes by holding four limbs together. Similar to the results of clasping phenotypes, transgenic mice showed significantly weaker grasping strength after 8.5 month of age (p \ 0.05; Fig. 6c), and this deficit was more significant with increasing age, suggesting that motor deficit in these transgenic mice became progressively worse. We also performed footprinting and rotarod testing at the 4-, 6-month-old and end stage to further confirm these abnormal motor functions. At 4 and 6 month of age, we did not observe significant difference between Ubi-GHTT84Q transgenic mice and wild-type mice in footprinting and rotarod tests (Online resource 5). At end stage, in the footprinting tests, disordered and shorter stride length, dragging and uneven step pattern and unbalanced gaits were observed in transgenic mice when walking through the tunnel, whereas a regular, straight and consistent stride length was observed in non-transgenic mice (Fig. 6d). In addition to footprinting tests, transgenic mice also appeared significantly unbalanced on the rotarod while running, and

123

290 Fig. 4 Expression of GFP-HTT in the brain of transgenic mice. Brain sections from one representative mouse at 12 months of age were subjected to immunofluorescent staining by using the mEM48 antibody. a–e are images from cortex (CTX) regions, and f–j are images from striatum (STR) regions. a, f are brightfield images. b, g are nucleus staining using Hoechst 33342. c, h are GFP fluorescent signals. d, i are immunofluorescent signals using the mEM48 antibody. e, j are merged images from Hoechst 33342, GFP and mEM48 signals in the CTX and STR

123



291

Fig. 5 Neuropathological phenotypes of HD in the brain of Ubi-GHTT84Q transgenic mice. Brain sections were sampled from one representative transgenic mouse at 12 months of age, and stained with the mEM48 antibody via DAB immunohistochemistry staining. a Lower magnification shows nuclear aggregates (small spots) in striatum (STR) region than in cortex (CTX) or white matter region (WM). b A higher power image of STR displays more nuclear aggregates, intranuclear aggregates and neuropil aggregates. c A

higher power image shows neuropil aggregates are dominant in CTX. Arrow heads indicate nuclear aggregates, arrows indicate intranuclear aggregates and stars indicate neuropil aggregates. d Quantitative analyses shows different neuropil (NPA), nuclear (NA) and intranuclear (INA) aggregates in cortex (CTX) and striatum (STR) regions. Aggregate numbers are accumulated from 5 randomly different 120 9 120 lM2 regions and counted by Metamorph software

exhibited a significantly higher tendency to fall off during the three consecutive days compared to wild-type mice (p \ 0.05; Fig. 6e). Taken together, the observed neuropathological characteristics confirmed the dysfunctions of behavioral phenotypes, especially in motor deficit, and these phenotypes were highly comparable to previous HD transgenic mice (Schilling et al. 1999; Davies et al. 1997; Carter et al. 1999).

demonstrated that the expression of the Ubi-G-HTT84Q transgene caused the misfolded aggregates of mutant HTT that led to typical neuropathological phenotypes (Figs. 4, 5). In addition, neuropathological and behavioral phenotypes showed the increasing deficit of motor functions with age, suggesting the progression of HD in this transgenic mouse model (Fig. 6). The molecular, pathological and behavioral phenotypes appeared to be comparable to those characteristics of other HD transgenic mice (Schilling et al. 1999; Davies et al. 1997), indicating that this novel model can be used for further HD studies. Furthermore, we have recently also generated Ubi-G-HTT19Q transgenic mice which carry exon 1 of HTT with 19 CAG repeats under the control of ubiquitin promoter (Online resource 6). This model is expected to serve as an excellent control for Ubi-G-HTT84Q transgenic mice. Since few HD transgenic mice have a parallel control carrying the same constructs except for the shorter CAG repeats (usually less than 23

Discussion The overall objective of this study is to generate novel transgenic mice carrying the GFP-HTT fusion protein. This transgenic mouse model provides a fluorescent signal to trace the expression of the HTT protein and manifests subsequent pathological characteristics resembling the neuropathological hallmarks of HD patients. Our results

123

292


Fig. 6 Behavioral phenotypes of Ubi-G-HTT84Q transgenic mice. a Life span. b Clasping tests. c Grasping strength tests. d Footprinting tests. e Rotarod tests. Asterisk represents statistically significant

difference (p \ 0.05) between Ubi-G-HTT84Q transgenic mice (GHD) and non-transgenic mice (WT) at the same stages. Data represented mean ± SD

CAG repeats), our Ubi-G-HTT84Q transgenic mice have an advantage for application to further studies. HTT protein is expressed ubiquitously inside the human body; however, only a few studies have focused on peripheral tissues (Chiang et al. 2011; Ciammola et al. 2011; Moffitt et al. 2009), because HD is defined as a neurodegeneration disease. To address the effects on peripheral tissues in vivo, developing a higher expression of mutant HTT in these tissues is an important and also straightforward strategy. In this study, ubiquitin promoter was used to drive mutant G-HTT84Q to generate HD transgenic mice, and this model showed the selective and progressive expression of mutant HTT in vivo (Fig. 3). According to a previous study in the non-human primate model (Yang et al. 2008), expression profiles of mutant

HTT under control of the ubiquitin promoter also showed high similarity to HD patients, suggesting that this alternative promoter can be used to study HD. In particular, this unique model would be beneficial to serve as an in vivo model focusing on the effects on peripheral tissues in HD because ubiquitin promoter has stronger ability to drive gene expression in peripheral tissues than HTT promoter. In addition, the higher expression level of mutant HTT in the liver and muscle of Ubi-G-HTT84Q transgenic mice also raises the important question concerning the pathological roles of HTT in HD patients (Fig. 3), echoing recent studies showing abnormality of liver and muscle cells in HD (Chiang et al. 2011; Ciammola et al. 2011). Another important issue is the onset of HD in our transgenic mice. Since mutant HTT containing only the

123


exon 1 region with expanded CAG repeats is considered to be a more toxic element leading to HD (Wang et al. 2008), earlier onset of HD and shorter life span were expected in our transgenic mice. In R6/2 (Davies et al. 1997) and N171-82Q (Schilling et al. 1999) HD transgenic mice, which are carrying exon 1 region of HTT with expanded CAG repeats, both mouse models showed the onset of HD at approximately 3 months of age and died at 4–7 months of age. However, our Ubi-G-HTT84Q transgenic mice displayed a relatively longer life span and a later onset of HD at approximately 8 months of age (Fig. 6), and therefore confer the advantage of greater ease of maintaining the transgenic colony, compared to the shorter-lived R6/2 and N171-82Q HD transgenic mice. The most interesting findings of this study concern the neuropathological phenotypes. In these transgenic mice, while the striatum region showed more nuclear aggregates and inclusion bodies, the neuropil aggregates, by contrast, were predominantly present in the cortex region (Fig. 5). This is the first report showing different nuclear diffusion profiling in mouse models of transgenic mice carrying exon 1 region of mutant HTT with expanded CAG repeats. Because the selective expression of mutant HTT is observed in HD patients, and because the striatum is one of the most susceptible regions in the brain, our model not only provides a visual marker to trace mutant HTT but also may provide this unique characteristic for HD studies. However, how this chimera construct causes the different nuclear diffusion profiling in different regions still needs to be clarified. Our previous transgenic mouse study using a similar construct without the GFP gene did not give rise to different nuclear diffusion profiling in different regions (Online resource 7). In addition, the study of transgenic monkeys carrying exon 1 region of mutant HTT with expanded CAG repeats also revealed widespread aggregates without significant differences among different brain regions (Yang et al. 2008). These results suggest that the GFP protein may play an unknown role leading to this unique pathological phenotype. Therefore, investigating the influence of GFP on different nuclear diffusion profiling may give an insight into the mechanism of selective patterning with regard to mutant HTT or pathology in HD. In summary, we have generated Ubi-G-HTT84Q transgenic mice, and these mice displayed molecular, pathological and behavioral phenotypes similar to previous HD transgenic mice. In addition to the conventional phenotypes, this model also possesses the convenient advantage of being able to trace the mutant HTT via GFP fluorescent signals. Most interestingly, the striatum of these transgenic mice was the most susceptible region for mutant HTT, achieving selective expression of HTT profiling. Taking these results together, this unique model is thus a promising one for HD studies, and will be beneficial for some specific

293

investigations, such as HTT effects on peripheral tissues or selective expression in HD brain. Acknowledgments We thank Jonathan Courtenay for critical reading of the manuscript, Dr. Xiao-Jiang Li for providing mEM48 antibodies, Dr. Chauying Jen and Pi-Hsueh Shirley Li for providing equipment and Dr. Shaw-Jeng Tsai and Dr. H. Sunny Sun for support and suggestions. This work was supported by National Science Council grants (NSC 99-2320-B-006-026-MY3 and NSC 100-2627B-006-023) and in part by grant of the Ministry of Education, Taiwan, Republic of China, under the ATU plan. Conflict of interest

None.

References Carter RJ, Lione LA, Humby T et al (1999) Characterization of progressive motor deficits in mice transgenic for the human Huntington’s disease mutation. J Neurosci 19:3248–3257 Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994) Green fluorescent protein as a marker for gene expression. Science 263:802–805 Chan AW, Yang SH (2009) Generation of transgenic monkeys with human inherited genetic disease. Methods 49:78–84 Chiang MC, Chern Y, Juo CG (2011) The dysfunction of hepatic transcriptional factors in mice with Huntington’s Disease. Biochim Biophys Acta 1812:1111–1120 Ciammola A, Sassone J, Sciacco M et al (2011) Low anaerobic threshold and increased skeletal muscle lactate production in subjects with Huntington’s disease. Mov Disord 26:130–137 Cornett J, Cao F, Wang CE et al (2005) Polyglutamine expansion of huntingtin impairs its nuclear export. Nat Genet 37:198–204 Davies SW, Turmaine M, Cozens BA et al (1997) Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell 90:537–548 DiFiglia M, Sapp E, Chase K et al (1995) Huntingtin is a cytoplasmic protein associated with vesicles in human and rat brain neurons. Neuron 14:1075–1081 Hodgson JG, Agopyan N, Gutekunst CA et al (1999) A YAC mouse model for Huntington’s disease with full-length mutant huntingtin, cytoplasmic toxicity, and selective striatal neurodegeneration. Neuron 23:181–192 Hogan B (1994) Manipulating the mouse embryo: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor Li S, Li XJ (2006) Multiple pathways contribute to the pathogenesis of Huntington disease. Mol Neurodegener 1:19 Lin CH, Tallaksen-Greene S, Chien WM et al (2001) Neurological abnormalities in a knock-in mouse model of Huntington’s disease. Hum Mol Genet 10:137–144 Maat-Schieman M, Roos R, Losekoot M et al (2007) Neuronal intranuclear and neuropil inclusions for pathological assessment of Huntington’s disease. Brain Pathol 17:31–37 Mangiarini L, Sathasivam K, Seller M et al (1996) Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87:493–506 McGuire JR, Rong J, Li SH, Li XJ (2006) Interaction of Huntingtinassociated protein-1 with kinesin light chain: implications in intracellular trafficking in neurons. J Biol Chem 281:3552–3559 Moffitt H, McPhail GD, Woodman B, Hobbs C, Bates GP (2009) Formation of polyglutamine inclusions in a wide range of non-

123

294 CNS tissues in the HdhQ150 knock-in mouse model of Huntington’s disease. PLoS ONE 4:e8025 Schilling G, Becher MW, Sharp AH et al (1999) Intranuclear inclusions and neuritic aggregates in transgenic mice expressing a mutant N-terminal fragment of huntingtin. Hum Mol Genet 8:397–407 The Huntington’s Disease Collaborative Research Group (1993) A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell 72:971–983 von Horsten S, Schmitt I, Nguyen HP et al (2003) Transgenic rat model of Huntington’s disease. Hum Mol Genet 12:617–624 Walker FO (2007) Huntington’s disease. Lancet 369:218–228

123

Brain Struct Funct (2013) 218:283–294 Wang CE, Tydlacka S, Orr AL et al (2008) Accumulation of N-terminal mutant huntingtin in mouse and monkey models implicated as a pathogenic mechanism in Huntington’s disease. Hum Mol Genet 17:2738–2751 Williams A, Sarkar S, Cuddon P et al (2008) Novel targets for Huntington’s disease in an mTOR-independent autophagy pathway. Nat Chem Biol 4:295–305 Yang SH, Chan AW (2011) Transgenic animal models of Huntington’s Disease. Curr Top Behav Neurosci 7:61–85 Yang SH, Cheng PH, Banta H et al (2008) Towards a transgenic model of Huntington’s disease in a non-human primate. Nature 453:921–924