Huntington disease (HD) is caused by polyglutamine [poly(Q)] expansion in the protein huntingtin (htt). Although the exact mechanism of disease progression remains to be elucidated, altered interactions of mutant htt with its protein partners could contribute to the disease. Using the yeast two-hybrid system, we have isolated a novel htt interacting protein, HIP14. HIP14's interaction with htt is inversely correlated to the poly(Q) length in htt. mRNAs of 9 and 6 bp are transcribed from the HIP14 gene, with the 6 kb transcript being predominantly expressed in the brain. HIP14 protein is enriched in the brain, shows partial co-localization with htt in the striatum, and is found in medium spiny projection neurons, the subset of neurons affected in HD. HIP14 localizes to the Golgi, and to vesicles in the cytoplasm. The HIP14 protein has sequence similarity to Akr1p, a protein essential for endocytosis in Saccharomyces cerevisiae. Expression of human HIP14 results in rescue of the temperature-sensitive lethality in akr1Δ yeast cells and, furthermore, restores their defect in endocytosis, demonstrating a role for HIP14 in intracellular trafficking. Our findings suggest that decreased interaction between htt and HIP14 could contribute to the neuronal dysfunction in HD by perturbing normal intracellular transport pathways in neurons.
Although polyglutamine [poly(Q)] expansion, the causative mutation in Huntington disease (HD) was discovered several years ago, there is as yet no proven mechanism for the selective vulnerability of neurons in the striatum and cortex observed in the disease. The protein huntingtin (htt) is ubiquitously expressed in all tissues. In neurons, htt is associated with synaptic vesicles and microtubules, and is enriched in dendrites and nerve terminals (1,2).
In order to understand the normal function of htt and mechanisms of pathogenesis in HD, and to expand the pathways of interaction with htt, we performed yeast two-hybrid screens to identify htt-binding proteins and determined whether poly(Q) length modulated these interactions. In the event that the interaction is modulated by CAG expansion, this would suggest that the interaction is obviously disturbed in the presence of the mutation and that this could therefore contribute to the pathogenesis of HD. In this regard, it is notable that certain classes of proteins cluster in their pattern of interaction with htt. For example, functions of normal htt can be gleaned from analysis of its protein partners. If these partners do not show altered interactions in the presence of the mutation, then these interactions are unlikely to underlie HD. Mutant htt does not show altered interaction with proteins such as HIP2 and p53, suggesting that the role of htt in cell cycle control and cell survival is part of its normal function, and does not contribute directly to the pathogenesis.
Seven proteins involved in transcriptional regulation interact with htt. These are the nuclear receptor co-repressor (3), the transcriptional co-repressor C-terminal binding protein (4), HYP-A (FBP-11), HYP-B(p231HBP) (5), cAMP response element binding protein (CREB)-binding protein (CBP) (6), mSin3a (7) and p300/CBP-associated factor(P/CAF) (7). Interestingly, almost all of htt's interacting protein partners implicated in transcriptional regulation show altered interaction with htt in the presence of the poly(Q) expansion, suggesting that disturbance of this interaction could contribute to the observed pathogenesis.
Furthermore, mutant htt shows altered interaction with every one of its protein partners implicated as having functions in neuronal intracellular transport, providing strong evidence that altered transport could contribute to the pathogenesis observed in HD. Evidence in support of this is that several interacting proteins, including htt-interacting protein 1 (HIP1) (8), htt-associated protein 1 (HAP1) (9–11), calmodulin (12), SH3GL3 (13) and cystathione β-synthase (14), have been identified with functions in transport and cytoskeletal organization. We describe here the identification and functional determination of the htt-interacting protein, HIP14, which also shows altered interaction with mutant htt and has a function in endocytosis. HIP14 is predominantly expressed in neurons in the brain but not in glia, and localizes to the Golgi complex and to cytoplasmic vesicles. HIP14 shares a high degree of amino acid similarity to proteins in a variety of organisms, including Akr1p, a protein essential for endocytosis of pheromone receptors in the yeast Saccharomyces cerevisiae (15). Yeast cells lacking functional AKR1 (akr1Δ) show decreased viability at an elevated temperature and display defects in endocytosis. We determined that HIP14 can rescue the temperature-sensitive viability of akr1Δ yeast cells and restore their block in endocytosis, suggesting that HIP14 may play a role in intracellular transport processes, and providing further proof for the role of htt in neuronal transport. This data leads to the hypothesis that a disturbed interaction between htt and HIP14 could lead to alterations in neuronal transport contributing to the pathogenesis in HD.
Isolation of HIP14 cDNA clones
We performed yeast two-hybrid library screens to identify htt's protein partners. The yeast two-hybrid screen identified 14 positive clones (16). Of these, a single HIP14 clone (pGADHIP14-500) was found to specifically interact with the pGBT9–1955-44 fusion protein, as yeast co-transformants gave a His+, β-galactosidase+ phenotype, and the transformation of pGADHIP14-500 with pGBT9 vector did not induce β-galactosidase activity.
We identified five cDNA clones encoding HIP14 by screening human brain cDNA libraries with pGADHIP14-500. Database searches identified several expressed sequence tag (EST) clones encoding regions of HIP14. DNA sequences of these cDNA clones were determined and aligned. The composite HIP14 cDNA sequence encompasses 5272 nucleotides containing a single open reading frame (ORF) that is 1902 nucleotides long, which is predicted to encode a HIP14 protein consisting of 633 amino acids. A htt-interacting protein HYPH, which is identical to the 5′ portion of HIP14, has been reported (5), providing independent confirmation of the interaction of HIP14 and htt. The nucleotide sequence for HIP14 has been deposited in the DDBJ/EMBL/GenBank database under accession no. AB024494.
Identification of HIP14-related protein (HIP14L)
Database searches have identified human EST clones encoding a gene homologous to HIP14. Of these, one clone contained a single ORF that is 1869 nucleotides long, encoding a predicted protein product of 622 amino acids, with 48% identity and 57% similarity to HIP14 (Fig. 1 and Table 1). We designated this EST HIP14-related protein (HIP14L). The nucleotide sequence for HIP14L has been deposited in the DDBJ/EMBL/GenBank database under accession no. AB024495. PCR-based somatic cell hybrid mapping revealed that HIP14L mapped to human chromosome 11 (data not shown).
HIP14 maps to human chromosome 12q14–q15
In order to determine the chromosomal localization of HIP14, FISH analysis was performed, revealing that HIP14 mapped to a single genomic locus at 12q14–q15. The GDB and OMIM database search revealed that no genetic disease had been mapped to this region (data not shown).
HIP14 and HIP14L show homology to the ankyrin repeat containing the S. cerevisiae protein Akr1p
HIP14 and HIP14L amino acid sequences were subjected to a homology search, which identified significant homology between HIP14, HIP14L, two S. cerevisiae ankyrin repeat-containing proteins (Akr1p, a protein essential for endocytosis of pheromone receptors and Akr2p), and other predicted transcripts of unknown function in Caenorhabditis elegans (accession no. AAD12801), Drosophila melanogaster (accession no. AAF49554), Mortierella alpina (accession no. CAB56510) and Schizosaccharomyces pombe (accession no. CAA90497). Multiple sequence alignments between HIP14, HIP14L, Akr1p, Akr2p, C. elegans H32C10.a, D. melanogaster, M. alpina and S. pombe have confirmed significant sequence similarity to each other (Table 1). Structural and motif analyses of these proteins revealed that each has a series of 33-residue ankyrin repeats and four to six putative transmembrane helices (Fig. 1). The relative positions of the ankyrin repeats and the putative transmembrane helices in these proteins were found to be well conserved, suggesting that these proteins may have functional or structural similarities.
Decreased interaction of HIP14 with mutant htt
To confirm the interaction of htt with HIP14, and to assess the influence of poly(Q) length on this interaction, we performed β-galactosidase liquid assays (Fig. 2A). Five isolated colonies from each of three separate transformations of htt constructs pGBT9–1955-15 or pGBT9–1955-128, or pGBT9 as control, were individually assayed for their strength of interaction with pGAD–HIP14 (Fig. 2A; pGBT9; 1.91±0.05 SEM; n=15, pGBT9–1955-15; 29.00±0.48 SEM; n=15, pGBT9–1955-128; 17.93±0.82 SEM; n=15). Poly(Q) expansion in htt resulted in a significant decrease in its interaction with HIP14 (P<0.001). Western blot analysis of yeast cells transformed with pGAD–HIP14-500, and either pGBT9–1955-15 or pGBT9–1955-128 showed equivalent protein expression levels (data not shown), indicating that the decreased interaction between HIP14 and expanded htt in yeast was not attributable to differences in translation or to altered stability of htt with different poly(Q) lengths.
To confirm the dependence of the HIP14–htt interaction on poly(Q) length in htt, in vitro binding assays were performed using immobilized HIP14–GST fusion protein and extracts from HEK293T cells overexpressing the N-terminal region of htt containing either 15 or 128 poly(Q) residues. GST–HIP14 bound to glutathione sepharose beads consistently precipitated more wild-type than mutant htt (n=2). Neither wild-type nor mutant huntingtin was precipitated with GST alone (Fig. 2B). This finding confirms that the physical interaction between HIP14 and htt decreases as poly(Q) length increases.
In order to assess whether htt and HIP14 interact in vivo, we performed co-immunoprecipitation experiments from cells overexpressing both htt and HIP14. We found that lysates from cells expressing HIP14 and wild-type htt (pCI1955-15) immunoprecipitated more HIP14 (n=2) than lysates from cells containing HIP14 and mutant htt (pCI1955-128), proving that HIP14 and htt interact in vivo, and that this interaction is modulated by the poly(Q) tract length in htt.
HIP14 is not toxic in 293T cells
The interaction between HIP14 and mutant htt has been shown to be reduced, which could lead to excess unbound HIP14 in patient neurons generating toxicity. A similar finding has been shown for HIP1, with toxicity being mediated through its death effector domain (DED) (17,18). We studied the effect of overexpression of HIP14 in transiently transfected 293T cells. Measurement of toxicity in these cells using the modified MTT assay showed that, when overexpressed, HIP14 was not significantly more toxic (P>0.05) than a control LacZ-expressing vector (data not shown).
HIP14 generates two transcripts in various tissues
To determine expression of the HIP14 transcripts, northern blot analysis was performed (Fig. 3), revealing the presence of two transcripts of ∼9 and 6 kb in various tissues. Both transcripts show a similar pattern of expression, although the 6 kb transcript showed much higher expression when compared with the 9 kb transcript. Both transcripts were expressed in all tissues examined, with the highest expression of the 6 kb transcript occurring in the brain.
HIP14 is expressed at the highest level in the brain
Western blot analysis on human tissue detected a single immunoreactive band at 70 kDa in all brain regions. Preincubation of the anti-HIP14 antibody with the antigenic peptide completely eliminated the band (data not shown). HIP14 expression is highest in the cortex, cerebellum, occipital lobe and caudate, and lowest in the spinal cord (Fig. 4A). In human peripheral tissues, HIP14 protein was detected in testis, pancreas, heart and kidney, with no expression detected in the liver and lung (Fig. 4B). In mouse tissues, HIP14 was detected in heart, pancreas, kidney, testis, liver and brain, with the highest expression in the brain, and none in the lung (Fig. 4C). Therefore the pattern of HIP14 expression is similar in mouse and human tissues.
Since HD is characterized by cell death in the striatum and cortex, we further characterized the localization of HIP14 in mouse brain slices. Immunohistochemical analysis of HIP14 combined with neuronal and glial markers revealed that HIP14 localized to the perinuclear region of neurons from the cortex (Fig. 5J), striatum (Fig. 5K) and hippocampus (Fig. 5L), but not glial cells (data not shown). No HIP14 immunoreaction was detected in the nuclei. HIP14 was also found in a variety of other brain regions including cerebellum, brainstem and thalamus. No staining was observed in control sections when primary antibody was omitted or in the presence of blocking antigen (data not shown).
htt and HIP14 co-localize in medium spiny neurons of the striatum
To determine the localization of HIP14 and its co-localization with htt in cells of the striatum, we performed co-staining of htt and HIP14 in mouse brain sections. As shown in Figure 6, htt (Fig. 6A, green) and HIP14 (Fig. 6B, red) showed partial overlapping staining (Fig. 6C, yellow) in cytoplasmic and perinuclear regions of the striatal cells. These abundant and medium sized cells in the striatum had the morphology of medium sized spiny neurons. A low-powered image of HIP14 and htt staining in the striatum is shown in Figure 6D, also displaying partial overlap in the staining of HIP14 and htt (yellow).
HIP14 is found in projection neurons affected in HD
The specific subset of neurons that are selectively vulnerable to neurodegeneration in HD are the medium spiny neurons in the striatum that project to the medial and lateral globus pallidus (19). To determine whether HIP14 is present in the striatopallidal neurons that took up the marker in the globus pallidus and transported it back to the striatum, we injected fluorogold tracer into the globus pallidus of mice. As seen in Figure 6E, HIP14 (red) is present in the specific subset of striatal neurons that project to the globus pallidus and uptake the fluorogold tracer (blue), conclusively indicating that HIP14 is found in the specific subset of neurons most affected in HD. Fluorogold injections were confined to the globus pallidus, as confirmed by anatomical assessment of the injection site in the immunohistochemistry slices. The fluorogold injection sites did not overlap with the striatal marker DARPP-32 (Chemicon) (data not shown).
HIP14 localizes to the Golgi and to cytoplasmic vesicles in neuronal precursor NT2 cells and ES cell-derived neurons
In an effort to elucidate the subcellular localization of HIP14, we assessed its immunolocalization in neuronal NT2 cells and ES cell-derived neurons. HIP14 immunoreactivity was localized to a perinuclear Golgi-like structure and to cytoplasmic vesicles, as analyzed by conventional and laser confocal microscopy. No overlap was found between HIP14 and the ER markers calreticulin and Grp78. By confocal microscopy, HIP14 showed partial co-localization with the cis-Golgi marker GM130 (Fig. 5A–C), which is involved in structural maintenance of the Golgi stack and interacts with the vesicle docking protein p115 (20). In addition, HIP14 was found to partially co-localize with the vesicle markers γ-adaptin, a component of the AP-1 complex (Fig. 5D–F), and clathrin (Fig. 5G–I) in the Golgi region. HIP14 also showed overlapping staining with the Golgi marker β-COP (data not shown). Biochemical fractionation of the human cortex, cerebellum and caudate using differential centrifugation revealed that HIP14 was highly enriched in the membrane (P3) fraction (data not shown) similar to the distribution of htt and HIP1 (21). These immunolocalization studies together with structural analysis suggested that HIP14 was associated with Golgi and cytoplasmic vesicle membranes.
To conclusively determine the localization of HIP14 within the Golgi apparatus, we performed pre-embedding immunogold labeling on mouse brain tissue sections using the anti-HIP14 antibody. At the electron-microscopic level, most of the immunogold particles were associated with the Golgi apparatus (Fig. 7A). Gold particles were also found in coated vesicles in the cytoplasm. Occasional staining was associated with the endoplasmic reticulum. Within the Golgi apparatus, particles were associated with the medial cisternae and the cis-Golgi cisternae (Fig. 7B). Particles were also associated with other Golgi compartments including trans cisternae, TGN and the pre-Golgi network.
HIP14 rescues the temperature-sensitive phenotype and the defects in endocytosis of akr1Δ cells
In S. cerevisiae, it has been demonstrated that Akr1p is necessary for cellular viability at elevated temperatures. Considering that there is a 41% amino acid similarity between HIP14 and Akr1p, we explored whether HIP14 could rescue the temperature-dependant lethality of akr1Δcells. AKR1 and akr1Δ yeast cells were transformed with the pAD5 vector alone, pAD5–HIP14, or the AKR1-containing positive control plasmid pPB575 (2µ, AKR1, LEU2) (22), streaked onto SD–leu plates and incubated at 25°C (permissive temperature for akr1Δ cells), and 37°C (restrictive temperature for akr1Δ cells). As can be seen in Figure 8A, akr1Δ cells carrying pPB575 grew well at both temperatures, whereas cells carrying pAD5 were viable only at 25°C. akr1Δ cells carrying pAD5–HIP14 grew well at 25°C, and were able to grow more vigorously at 37°C than cells carrying pAD5.
We determined whether HIP14 could specifically rescue the block in constitutive endocytosis of the yeast pheromone receptor Ste3p seen in akr1Δ cells. Akr1p is required for the constitutive endocytosis of Ste3p (15). Ste3p is normally targeted to the plasma membrane, then endocytosed and transported through vesicular intermediates to the vacuole, where it is degraded. In wild-type cells, Ste3p has a half-life of ∼15 minutes, whereas in akr1Δ cells this half-life is increased >5-fold (15). The longer half-life of Ste3p in akr1Δ cells was shown to result from a block in constitutive endocytosis (15). We transformed akr1Δ cells with pAD5, pAD5–HIP14 or pPB575, subjected these strains to a cycloheximide-induced protein synthesis block, and removed samples at various time-points. Using a Ste3p antibody, we determined that Ste3p degradation is restored upon the expression of HIP14 in yeast, as can be seen in Figure 8B, demonstrating that HIP14 can functionally rescue the endocytosis defect in akr1Δ cells.
In this study, we have identified and characterized a novel htt-interacting protein, HIP14, that shows decreased interaction with mutant htt. A 542 bp fragment of HIP14 had been previously described (HYPH) (5) as an htt interacting protein, providing independent confirmation of this interaction. htt is widely expressed throughout the central nervous system (23,24). Unlike htt, which has ubiquitous expression, HIP14 is localized predominantly in the brain of both humans and mice. In the brain, it is localized, among other regions, in the medium spiny neurons of the striatum, the site of early and prominent pathology in HD.
HIP14 and htt show partial co-localization in the medium spiny neurons of the striatum, and, in addition, they are found in medium spiny neurons that project into the globus pallidus, which is the subset of neurons selectively vulnerable to neurodegeneration in HD. Furthermore, HIP14 is localized in the Golgi apparatus and in discrete cytoplasmic vesicles. Immunofluorescence microscopy has revealed that htt is also enriched on membranes of the Golgi, and associates with clathrin-coated and non-coated vesicles in the cytoplasm and along plasma membranes (25). The specific cellular and subcellular colocalization of HIP14 and htt provides further support for the functional interaction between these two proteins in living cells.
Nothing is known about the function of HIP14. However, clues to its normal function can be derived from the insights gained from the study of the gene in other organisms. The HIP14 protein contains several transmembrane and ankyrin repeat domains, and shares significant homology with the yeast protein Akr1, which, when deleted, causes defects in endocytosis in yeast (15). Interestingly, expression of HIP14 protein in akr1Δ yeast cells restores their endocytic function. This functional complementation by HIP14, along with its strong sequence similarity to Akr1p, provides evidence that Akr1p is the yeast homologue for HIP14 and gives clues to the role of HIP14 in intracellular protein trafficking. Akr1p has been recently shown to play a role in the targeting of type I casein kinases to the plasma membrane, where they are involved in the ubiquitin-mediated endocytosis of yeast pheromone receptors (26). Ubiquitin plays a central role in early endocytic events in yeast by providing a sorting determinant that initiates uptake. In addition, it has been shown to play a role in the internalization of several mammalian proteins as well (26,27). Ubiquitination, and the subsequent internalization and degradation, are thought to be the critical mechanism in the down-regulation of several receptors at the plasma membranes of mammalian cells (28). Since HIP14 can functionally compensate for the loss of activity of Akr1p, it is likely to have a similar role as Akr1p in the sorting or targeting of critical proteins involved in the initiating events of endocytosis at the plasma membrane.
Interestingly, HIP14 is not the only protein with which htt interacts that has been implicated in intracellular transport. Other htt-interacting proteins with roles in intracellular transport include α-adaptin, a component of the AP-2 complex that binds to clathrin and is involved in endocytosis at the plasma membrane (25), and HAP1 (9), which in turn binds to the p150Glued subunit of dynactin (29), a protein involved in retrograde transport. HAP1 is also directly involved in the regulation of vesicular trafficking from early endosomes to the late endocytic compartment (11). htt has also been shown to bind HIP1 (21), whose yeast homologue (Sla2p) mutants show defects in endocytosis and cytoskeletal organization (30,31). HIP1 has been recently shown to bind to F-actin, clathrin and adaptor protein 2 (32–35), all of which are proteins essential for vesicle assembly and cytoskeletal organization in mammalian cells, thus providing further evidence for similar functions between mammalian and yeast homologues in intracellular transport pathways. In addition, htt interacts with SH3GL3, a protein belonging to the SH3 domain-containing protein family that plays a role in synaptic vesicle transport (13). Interestingly, HAP1, HIP1, SH3GL3 and HIP14 all show altered interactions with mutant htt, supporting the possibility that the pathogenesis in HD involves dysfunction in intracellular transport in the affected neurons. Our study provides additional evidence for the role of htt in intraneuronal transport processes.
Several putative functions have been suggested for htt. These include roles in transcriptional regulation, intracellular transport and cell cycle regulation. Interestingly, almost all of htt's interaction partners implicated in the functions of intracellular transport and transcriptional regulation show altered interaction with htt in the presence of the poly(Q) expansion, thus implicating disturbance of these two proposed functions of htt as contributing to the pathogenesis in HD. In contrast, htt does not show altered interactions with proteins such as HIP2 and p53, suggesting that the functions of htt in cell cycle control and cell survival is part of its normal function, and do not contribute directly to the pathogenesis. The isolation and characterization of HIP14 as an htt-interacting protein with altered interaction in the presence of the mutation, together with the fact that htt and HIP14 co-localize, and that HIP14 is able to rescue the defect in endocytosis in Akr1Δ cells, suggests another link in the function of htt in intracellular transport and endocytosis. Although important, this study does not provide a clear mechanism as to how an HIP14-expanded htt interaction could contribute to the selective vulnerability of cortical and striatal neurons in HD. However, altered interactions with HIP14 caused by the mutation in htt further implicate altered pathways of intraneuronal transport and endocytosis in the pathology of HD, a hypothesis that now deserves formal assessment.
MATERIALS AND METHODS
Yeast two-hybrid screening
The N-terminal htt plasmid construct pGBT9–1955-44 (16) was transformed into the yeast strain Y190 (MATa, ura3-52, his3-200, ade2-101, lys2-801, trp1-901, leu2-3, 112, gal4Δ, gal80Δ, cyhr2, LYS2::GAL1UAS- HIS3TATA-HIS3, URA3::GAL1UAS-GAL1TATA-lacZ ). A human adult brain Matchmaker cDNA library (Clontech) was then transformed into Y190 cells already harboring the pGBT9–1955-44 construct (36). The transformants were plated onto synthetic complete (SC) medium (Trp−/Leu−/His−) with 25 mm of 3-aminotriazole (3-AT) to suppress the growth of false His+ colonies. β-Galactosidase filter assays were performed, and the plasmids containing activating domain cDNAs were isolated by electroporating the bacterial strain KC8 with the yeast lysate (37). The resulting colonies were transformed into DH5-α for further manipulation.
Yeast two-hybrid interactions
The yeast strain Y190 was co-transformed with pGADHIP14-500, the clone isolated from the yeast two hybrid library screen, and pGBT9–1955-15, pGBT9–1955-128 (38) or pGBT9 using a modified lithium acetate transformation protocol as described previously (36). β-Galactosidase chromogenic filter assays (16,37) and semiquantitative liquid assays using o-nitrophenyl-β-d-galactopyranoside (ONPG) as a substrate were performed as recommended (Clontech). Statistical analyses were performed using the Newman–Keuls multiple-comparison test.
In vitro binding assays
The insert DNA of pGADHIP14-500 was re-cloned into pGEX4T2 (Pharmacia), grown and induced with 1 mm IPTG. Cell pellets were resuspended in 5 ml of lysis buffer [150 mm NaCl, 50 mm Tris–HCl (pH 7.5), 0.5% NP40, 2.5 mm MgCl2, 2.5 mm KCl, 1.5 mm CaCl2, and protease inhibitor cocktail (Boehringer Mannheim)] containing 0.5 mg/ml lysozyme, sonicated and centrifuged. The resulting supernatant was bound to GST–Sepharose 4B beads (Pharmacia). To prepare htt, human embryonic kidney (HEK) 293T cells were transfected with pCI1955-15 or pCI1955-128 (38). Cell pellets were resuspended in 250 µl of HEPES buffer [10 mm HEPES (pH 8.3), 1.5 mm MgCl2 and protease inhibitor cocktail], homogenized and centrifuged. 500 µg of protein was bound to beads and washed extensively. GST–HIP14 or GST-only beads were boiled in SDS sample buffer, resolved on 7.5% polyacrylamide gels and transferred to polyvinylidene difluoride (PVDF) membranes (39). The anti-htt antibody BKP1 (1:100) (16) was used for immunodetection.
Co-immunoprecipitation of HIP14 with htt
Twenty-five microliters per sample of protein G sepharose beads (Pharmacia) were washed four times with lysis buffer (5 mm HEPES and 1% Triton X-100) and bound to the anti-htt antibody 2166 (Chemicon), along with an antibody-omitted control. The beads were then washed and 100 µl of slurry was used for each reaction. To express htt and HIP14, HeLa cells were co-transfected with HIP14 and htt pCI1955-15 or pCI1955-128, using FuGene 6 transfection reagent (Roche) as indicated by the manufacturer. Twenty-four hours later, cells were lysed and 400 µl of lysate (∼2 mg protein) were added to the beads and incubated end-over-end at 4°C overnight. Samples were washed five times with RIPA buffer [50 mm Tris (pH 8), 150 mm NaCl, 0.1% SDS, 1% NP-40 and 0.5% deoxycholic acid], boiled in SDS sample buffer, separated on 7.5% acrylamide gels and transferred to PVDF membrane (Bio-Rad). Western blots were performed using anti-HIP14 polyclonal antibody and developed using ECL (Amersham) chemiluminescence reagent.
cDNA cloning and the identification of EST clones
Approximately 1×106 pfu from two human brain cDNA libraries (gifts from Drs Montal and Johnston at John Hopkins University) were screened using standard procedures (40). HIP14 cDNA from pGADHIP14-500 was used as a probe. After secondary screening, all positive clones were subcloned into pBluescript SK(−) using the in vivo excision procedure (41).
To identify additional cDNA clones, we searched the EST database (dbEST) using a BLAST program (42) at NCBI (http://www.ncbi.nlm.nih.gov). The EST clones 41870 and 838424 were obtained from Research Genetics Inc. The entire coding region was assembled in pBluescriptII SK(−).
DNA sequencing and DNA/amino acid sequence analysis
DNA was sequenced using the ABI 373A DNA sequencer. The DNA sequence and the translated amino acid sequences were compared with sequence databases using BLASTN and BLASTP, respectively (42). The colinear alignment of the predicted amino acid sequences was performed using Pileup [Wisconsin Package Version 10.0, Genetics Computer Group (GCG), Wisconsin] and displayed using BOXSHADE (http://www.isrec.isb-sib.ch/software/BOX_form.html). The transmembrane helices were predicted by using PSORTII (http://www.genome.ad.jp), and TMpred (http://www.hgsc.bcm.tmc.edu:8088/search-launcher/launcher.html).
Chromosomal mapping of the HIP14 gene
Chromosomal mapping of the HIP14 gene was performed by fluorescence in situ hybridization (FISH) to normal human lymphocyte chromosomes counterstained with propidium iodide and 4′,6-diamidin-2-phenylindole dihydrochloride (DAPI) (43). Two HIP14 genomic bacterial artificial chromosome (BAC) clones (44), 363B17 and 463M12, were identified. Biotinylated genomic clones were detected with avidin–fluorescein isothiocyanate (FITC). Images were captured by a thermoelectrically cooled charge-coupled device (CCD) camera (Photometrics, Tucson, AZ). Separate images of DAPI-banded chromosomes and FITC-targeted chromosomes were obtained. Pseudocolored images of the DAPI and FITC were overlaid and merged electronically (45).
Toxicity of HIP14
HEK293T cells were cultured and transfected with pCIneoHIP14. Toxicity was measured 48 h after CaPO4 transfection using a modified MTT assay as previously described (46). htt with 128 repeats (1955-128) and a LacZ-expressing vector were used as positive and negative controls, respectively.
Northern blot analysis
Multiple human adult tissue northern blots (Clontech) were hybridized with [α-32P] dCTP-labeled HIP14-500 or human actin cDNA in ExpressHyb hybridization solution (Clontech) at 68°C. The membranes were washed under stringent conditions (0.1×SSC, 0.1% SDS, 55°C) and exposed to X-ray film (X-OMAT, Kodak) at −80°C with intensifying screens (DuPont).
Generation of antibodies
The generation of the htt-specific antibody BKP1 is described elsewhere (16). To generate a polyclonal antibody for HIP14 (HIP14PEP1), a peptide corresponding to amino acids 49–60 (RKTHIDDYSTWD) that represents a unique amino acid sequence was synthesized and coupled to keyhole limpet hemocyanin (KLH) (Pierce) with succinimidyl 4-(N-maleimidomethyl)cyclohexane-1-carboxylate (Pierce). Female New Zealand White rabbits were injected with HIP14 peptide and Freund's adjuvant. Antibody was purified using standard Thiol–sepharose affinity columns (Pharmacia).
Western blot analysis
Protein from frozen human tissues and fresh mouse tissues was isolated as stated above, denatured in SDS sample buffer, loaded onto 7.5% acrylamide SDS gel and transferred onto PVDF membranes. Western blot analysis was conducted using the purified HIP14PEP1 antibody (1:50), and then detected using enhanced chemiluminescence (Amersham). Equivalent protein loading was assessed by GAPDH (Chemicon) immunoblotting of the PVDF membranes.
Immunocytochemistry in neuronal precursor NT2 cells and ES cell-derived neurons
Double immunofluorescence was performed with neuronal NT2 cells and ES-cell-derived neurons. Cells were fixed in 4% (w/v) paraformaldehyde, permeabilized in PBS containing 1% paraformaldehyde and 0.3% Triton X-100 for 5 min, washed, and incubated with anti-HIP14 antibody overnight at 4°C. Slides were incubated with monoclonal antibodies (mAbs) directed against GM130 (Transduction Laboratories), β-COP (Sigma), clathrin heavy chain (Transduction Laboratories), γ-adaptin (Transduction Laboratories), calreticulin or Grp78 (StressGen, BC) in phosphate-buffered saline (PBS) and 2% normal goat serum for 2 h. Slides were then incubated with goat anti-rabbit Alexa 488 and/or goat anti-mouse Alexa 568, respectively (Molecular Probes), counterstained with propidium iodide or DAPI and mounted in Mowiol (Aldrich). Immunofluorescence was detected using a laser confocal microscope (Biorad) or conventional immunofluorescence microscopy. Images were captured digitally with a CCD camera (Princeton Instruments Inc.).
Injections of the fluorogold retrograde tracer were performed to label the population of projection neurons in the striatum, which are the first to degenerate in HD (47), and to determine if HIP14 was localized in these neurons. Fluorogold tracer injected into the globus pallidus is retrogradely transported to the cell body of the striatopallidal neurons in the striatum. To perform the injections, mice were anaesthetized with inhaled isofluorane and placed in a stereotaxic apparatus, and a small craniotomy was performed at the injection site. Using pulled-glass pipettes and the nanojector microinjector (World Precision Instruments), 20 nl of 1% fluorogold (Fluorochrome Inc.) in dH2O was slowly injected (10 nl/min) into the globus pallidus (from Bregma: anterior −2.63 mm, lateral±1.8 mm, depth 4.33 mm at a 30° angle). Five minutes following injection, the glass pipette was slowly withdrawn and the scalp was closed with surgical sutures. The animal was given postoperative care in a heated chamber prior to returning to its home cage. After a 24 h survival time, mice were sacrificed with isofluorane and perfused intercardially with 3% paraformaldehyde. The brain was removed and allowed to fix in 3% paraformaldehyde for 24 h. Twenty-five micrometer brain sections were cut on a vibratome and then stained for HIP14 and htt (HD3 antibody) (1) using standard immunohistochemical techniques. Images were taken from either a Zeiss Axioscope or a Bio-Rad Radiance 2000 confocal imaging system.
Immunohistochemical analyses of mouse brains
To assess the expression of HIP14 in vivo, the brains of adult FVB/N mice were processed with either a neuron-specific antibody (NeuN, Chemicon) or an astroglial-specific marker (GFAP, Sigma) and HIP14PEP1. Mice were perfused with cold 3% paraformaldehyde in 0.1 m PBS, brains were post-fixed, and 30 µm sections were cut on a vibrating microtome (Vibratome). Sections were rinsed in 0.1 m PBS with 0.3% Tween-20, blocked (0.1% PBS with 0.3% Tween-20, 3% whole goat serum and 5% bovine serum albumin) and placed into primary antisera against HIP14PEP1 and NeuN or GFAP for 24 h at 4°C. The samples were then blocked and incubated in either 1:100 goat anti-rabbit CY-3 (Molecular Probes) with anti-HIP14 antibody, or 1:250 goat anti-mouse Alexa 488 with NeuN or GFAP. Sections were mounted on gelatin-coated slides, dehydrated by serial ethanol washes, mounted with Fluoromount (Gurr), and analyzed using a laser confocal microscope (Bio Rad). Digital images were captured on a CCD camera (Princeton Instruments Inc.), and processed into double immunofluorescence figures using the NIH-Image program.
Light and electron-microscopic level immunocytochemical analysis of mouse brains
Mice were perfused intracardially with 200 ml of 3% paraformaldehyde and 0.15% glutaraldehyde in 0.1 m phosphate buffer (PB) at pH 7.2 for electron-microscopic analysis. Brains were processed as described elsewhere (1,48). Controls included the omission of primary antisera. Following DAB visualization, some sections were osmicated [1% osmium tetroxide (OsO4) in 0.1 m cacodylate buffer], rinsed, and stained overnight in 2% aqueous uranyl acetate.
For ultrastructural analysis, ultra-small colloidal gold conjugated secondary antibody (Aurion, Wageningen, The Netherlands) was used. Following a post-fixation with 2.5% glutaraldehyde, gold particles were intensified using the R-gent SE-EM silver enhancement kit (Aurion). Sections were further fixed with 0.5% OsO4 in 0.1 M PB, dehydrated in ethanol and propylene oxide (1:1) and flat-embedded in Eponate 12 (Ted Pella, Redding, CA). Ultrathin sections (90 nm) were cut using a Leica Ultracut S ultramicrotome, and counterstained with 5% aqueous uranyl acetate for 5 min followed by lead citrate for 5 min. Thin sections were examined using a HITACHI H-7500 electron microscope. Following DAB staining, some sections were mounted for light microscopy.
Transformation of HIP14 into akr1Δ yeast
The entire coding region of HIP14 was isolated from pBluescriptHIP14 and ligated to pAD5 (2µ, ADH1 promoter, LEU2). AKR1 and akr1Δ yeast cells were transformed with pAD5–HIP14, pPB575 (2µ, LEU2, AKR1) (22) and pAD5. The cells were plated onto SD-leu, grown at 25°C and streak-purified twice. Random colonies were picked from each streak, and incubated at 25°C and 37°C. Colony growth and size were monitored 72 h later.
Assay of Ste3p stability
Individual colonies from the transformants above were innoculated and grown overnight, re-innoculated and grown until OD600 reached 0.8. Cycloheximide 10 µg/ml was added, and an aliquot removed at ‘time 0’ and other time-points. Protein extracts were prepared as previously described (15). Ten microliter of sample was separated on a 7.5% acrylamide gel, and transferred to PVDF membranes. The blots were immunoreacted against anti-Ste3p monoclonal antibody, and visualized using ECL (Pharmacia) as described previously (15).
We thank Michael Kalchman for originating these studies, Dr Philip Hieter for helpful discussions throughout this work, Dr Steven Sherer for the chromosomal mapping analysis and Dr Dan Gietz for providing guidance in the yeast library screens. S.G is a postdoctoral fellow in Dr Philip Hieter's laboratory. This work was funded by the Canadian Genetic Diseases Network, the Centre for Molecular Medicine and Therapeutics (CMMT), a Medical Research Council (MRC) of Canada operating grant to M.R.H., National Institutes of Health Grant NS35255 to S.M.H. and C.-A.G., and the Huntington Disease Society of America (M.M., C.-A.G., S.M.H. and M.R.H.). M.R.H is a holder of a Canada Research Chair.
To whom correspondence should be addressed at: Centre for Molecular Medicine and Therapeutics, University of British Columbia, 980 West 28th Avenue, Vancouver, Canada BC V5Z 4H4. Tel: +1 6048753535; Fax:+1 6048753819; Email: email@example.com
|hHIP14||hHIP14L||Akr1||Akr2||C. elegans||D. melanogaster||M. alpina||S. pombe|
|hHIP14||hHIP14L||Akr1||Akr2||C. elegans||D. melanogaster||M. alpina||S. pombe|
- yeast one/two-hybrid system
- huntington's disease
- clone cells
- corpus striatum
- disease progression
- dna, complementary
- golgi apparatus
- rna, messenger
- saccharomyces cerevisiae
- htt gene
- projection neuron