Skip Navigation


Human Molecular Genetics Advance Access originally published online on April 23, 2007
Human Molecular Genetics 2007 16(11):1279-1292; doi:10.1093/hmg/ddm076
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/11/1279    most recent
ddm076v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Bradley, S. V.
Right arrow Articles by Ross, T. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bradley, S. V.
Right arrow Articles by Ross, T. S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Degenerative phenotypes caused by the combined deficiency of murine HIP1 and HIP1r are rescued by human HIP1

Sarah V. Bradley1,{dagger}, Teresa S. Hyun1,{dagger}, Katherine I. Oravecz-Wilson1, Lina Li1, Erik I. Waldorff2, Alexander N. Ermilov3, Steven A. Goldstein2, Claire X. Zhang4, David G. Drubin5, Kate Varela1, Al Parlow6, Andrzej A. Dlugosz3 and Theodora S. Ross1,*

1 Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, MI 48109-0942, USA, 2 Department of Orthopaedic Surgery, 3 Department of Dermatology and Comprehensive Cancer Center, University of Michigan, Ann Arbor, MI 48109, USA, 4 Institute of Molecular Medicine, Peking University, Beijing 100871, China, 5 Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720-3202, USA and 6 National Hormone and Peptide Program, Harbor-UCLA Medical Center, Torrance, CA 90509, USA

* To whom correspondence should be addressed at: 1500 E. Medical Center Dr., Ann Arbor, MI 48109, USA. Tel: +1 7346155509; Fax: +1 7346153947; Email: tsross{at}umich.edu

Received February 3, 2007; Accepted March 20, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
The members of the huntingtin-interacting protein-1 (HIP1) family, HIP1 and HIP1-related (HIP1r), are multi-domain proteins that interact with inositol lipids, clathrin and actin. HIP1 is over-expressed in a variety of cancers and both HIP1 and HIP1r prolong the half-life of multiple growth factor receptors. To better understand the physiological importance of the HIP1 family in vivo, we have analyzed a large cohort of double Hip1/Hip1r knockout (DKO) mice. All DKO mice were dwarfed, afflicted with severe vertebral defects and died in early adulthood. These phenotypes were not observed during early adulthood in the single Hip1 or Hip1r knockouts, indicating that HIP1 and HIP1r compensate for one another. Despite the ability of HIP1 and HIP1r to modulate growth factor receptor levels when over-expressed, studies herein using DKO fibroblasts indicate that the HIP1 family is not necessary for endocytosis but is necessary for the maintenance of diverse adult tissues in vivo. To test if human HIP1 can function similar to mouse HIP1, transgenic mice with ‘ubiquitous’ expression of the human HIP1 cDNA were generated and crossed with DKO mice. Strikingly, the compound human HIP1 transgenic DKO mice were completely free from dwarfism and spinal defects. This successful rescue demonstrates that the human HIP1 protein shares some interchangeable functions with both HIP1 and HIP1r in vivo. In addition, we conclude that the degenerative phenotypes seen in the DKO mice are due mainly to HIP1 and HIP1r protein deficiency rather than altered expression of neighboring genes or disrupted intronic elements.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Huntingtin-interacting protein 1 (HIP1) and its only known mammalian relative, HIP1-related (HIP1r), are phosphoinositide (1), clathrin (2–6) and actin-binding proteins (7,8) found mainly in the cytoplasm of the cell. HIP1 was identified originally as a protein that bound huntingtin, the protein whose gene is mutated in the germline of patients with Huntington's disease (9,10). Subsequently, HIP1r was identified due to its homology with HIP1 (11). The sequence homology, similarity to molecular-interacting domains (clathrin, actin and phosphoinositides), together with overlapping expression patterns, suggested that HIP1 and HIP1r have similar functions. However, there is evidence that the two members of the HIP1 family also have non-overlapping functions. For example, HIP1r tissue distribution is more widespread than HIP1 (12). Furthermore, HIP1r and HIP1 have overlapping and distinct localization within the cell (1). In addition, HIP1 is frequently elevated in multiple tumor types (13,14), whereas HIP1r is not (Hyun T.S. and Ross T.S., unpublished data).

Differences in the phenotypes of single Hip1 and Hip1r knockout mice also suggest that the proteins encoded by these genes have at least partially non-overlapping functions. The only phenotype so far observed in HIP1r-deficient mice is a progressive depletion of acid-secreting gastric parietal cells that leads to asymptomatic gastric hyperplasia and intestinal metaplasia (Samuelson et al., manuscript in preparation). In contrast, HIP1-deficient mice do not exhibit this phenotype. HIP1 deficiency results in male infertility associated with testicular degeneration due to apoptosis of post-meiotic spermatids (5), increased cataracts due to cell death in the lens (15) and kypholordosis (a spinal defect) by 1 year of age (15,16). The mechanisms of the spinal defect are not well understood but are reflected histologically in disordered vertebrae with a normal spinal cord and paraspinal musculature.

Embryonic fibroblasts derived from either Hip1 (15) or Hip1r (12) single knockout mice did not have consistent endocytic or signaling abnormalities. Overall growth and survival of single knockout mouse embryonic fibroblasts (MEFs) did not consistently deviate from those of wild-type MEFs. In contrast, when either HIP1 or HIP1r expression have been knocked down by RNAi in human cell lines, defects have been observed in endocytosis, actin dynamics (2,17) and nuclear hormone receptor signaling (18). The contrasting effects of RNAi on cell lines when compared with gene targeting on primary mouse cells raise a number of possibilities. One possibility is that the function of HIP1 family members in human cells or in cell lines differs from their function in mouse cells or in primary cells. Another possibility is that the failure to observe these phenotypes in gene-targeted mice is caused by compensatory changes that occur over time after elimination of a single HIP1 family member. Since there are only two members of the HIP1 family, the most likely potential source of compensation would be increased function of the second family member after the other family member is knocked out.

To test whether HIP1 and HIP1r have redundant or compensatory functions, we have generated double knockout (DKO) mice and compared their phenotype with that of single knockout mice or DKO mice that are transgenic for the human HIP1 transgene. We found that DKO mice had a more severe kypholordosis than single Hip1 knockout mice (which develop modest kypholordosis) or Hip1r knockout mice. DKO mice also exhibited progressive weight loss associated with metabolic failure and premature death. In addition, proliferation of early passage embryonic fibroblasts in culture was diminished. These phenotypes were either not observed in Hip1 or Hip1r single knockout mice or were greatly reduced in their severity. Furthermore, expression of the human HIP1 transgene rescued many aspects of these DKO phenotypes. Despite the extensive defects in DKO mice, we could find no evidence for a disruption of nuclear hormone function, PtdIns 3-kinase signaling, receptor stability or actin dynamics in embryonic fibroblasts derived from these mice. These data indicate that HIP1 and HIP1r are necessary for adult tissue maintenance, despite not being widely required for trafficking, PtdIns 3-kinase signaling or actin dynamics. The lack of clear defects in clathrin trafficking, signaling or actin paths in MEFs raises the possibility that these proteins may play novel roles in the regulation of cell survival and proliferation. In addition, these data indicate that the mouse HIP1 and HIP1r proteins are interchangeable with the human HIP1 protein, providing impetus for future studies of disease physiology using this mouse model.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Premature death and weight loss in DKO mice
Since the Hip1 and Hip1r alleles are linked on chromosome 5 in mice, DKO mice were derived by a complex series of matings of Hip1r–/– and Hip1null/null mice as described (12). Previously, we reported that a small group of Hip1/Hip1r DKO mice had accelerated development of the kypholordosis phenotype and had a reduced body mass when compared with single Hip1 mutant mice (12). We have now generated significant numbers of DKO mice and littermates by intercrossing Hip1r–/–; Hip1null/+ mice with each other. These matings generated 25% Hip1r–/–;Hip1+/+, 61% Hip1r–/–;Hip1null/+ mice and only 14.6% DKO mice (n = 342 total mice). The number of DKO adult mice (more than 3 weeks of age) was significantly reduced compared with the expected Mendelian ratio (P < 0.005, expected 25%), indicating that the HIP1 family is necessary in a subset of offspring to survive early development. This partial lethality was observed previously for the single Hip1 knockout mice [27, 55 and 17% yield, respectively (15)] but not in the single Hip1r knockout mice [normal Mendelian ratios born (12)]. Overall, these data indicate that DKO mice survive into adulthood at a similar frequency as seen previously in Hip1null/null mice (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Summary of observed pathologies in mice with HIP1 family alterations

 
Long-term observation of adult DKO mice demonstrated that they also have a shortened lifespan. A group of 10 DKO mice and 15 littermates were observed for the first 8 months of life. During this period, all of the DKO mice died between 2 and 6 months of age, whereas there were no deaths among littermate controls (Fig. 1A). During the course of these studies, we have generated 95 DKO mice for observation and experimentation. Almost all of these DKO mice died spontaneously or became moribund and were euthanized between the ages of 8 and 23 weeks [n = 83 (88%)]. Despite significant efforts to catalogue behavioral or health status changes prior to death, only increased respiration and pulse associated with severe mechanical limitations secondary to their spinal deformities were observed. It is therefore probable that the mice suffered progressive pulmonary restrictive disease leading to difficulties with clearing secretions and with inflation of all lung segments. These mechanical problems could lead not only to chronic hypoxia and pneumonia but predict that the mice ultimately will die of pulmonary failure. Histological analysis of lungs from DKO and control mice did not show significant changes indicating that restrictive disease rather than widespread pneumonia was likely. To substantiate this hypothesis, we have observed a cohort of the DKO mice and their littermates in metabolic cages and confirmed that they do have diminished metabolism reflected by diminished oxygen intake, CO2 generation and body temperature (data not shown).


Figure 1
View larger version (44K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Hip1null/null, Hip1r–/– (DKO) mouse phenotype. (A) Kaplan–Meier survival curve of DKO mice (n = 10; red line), their littermate controls (n = 15; solid black line) and human HIP1 transgene expressing DKO (hHIP1;DKO) mice (n = 3; dotted black line). (B) Photograph of the deformed spinal column and associated rib cage of a 6-month-old preterminal DKO mouse. The kypholordosis is usually accompanied by alopecia (secondary to mechanical stretch) at the apex of the thoraco-lumbar curvature. Note as well the clouded pupil of the eye reflective of a clinically defined lens cataract. (C) Severe kypholordosis is observed in DKO mice as early as 21 days. Nine littermates from a Hip1r–/–;Hip1 null/+ intercross were sacrificed and skin removed for visualization of spine curvature at 21 days. Kypholordosis was seen in all of the DKO mice (n = 3), but not in the littermate controls (n = 6). Shown here are a representative DKO mouse and its littermate Hip1r–/–;Hip1null/+ control. (D) Growth of female and male adult DKO mice, their littermate controls and hHIP1;DKO mice. Female DKO mice (n = 14) weigh significantly less than female control littermates (n = 18) at every time point (P < 0.01; Student's t-test). The average weight of the DKO females (n = 14) by 8 weeks was 15 gm, the control female (n = 18) was 20 gm, and the hHIP1;DKO female (n = 2) was 21 gm. Male DKO mice (n = 9) weigh significantly less than male control littermates (n = 22) at every time point (P < 0.001; Student's t-test). The average weight of the DKO males (n = 9) by 8 weeks was 15 gm and the control male (n = 22) average weight was 25 gm. (E) Total body weights and organ weights of DKO mice (n = 4) relative to littermate control total body and organ weights (n = 4). Both HIP1 and HIP1r are normally expressed in the liver, spleen, kidney, testes, heart, lung and brain of wild-type mice. The asterisk denotes a significant difference between DKO and control (P < 0.05; Student's t-test).

 
Although 100% of the male and female DKO mice developed severe kypholordosis (Fig. 1B) as early as 3 weeks of age (Fig. 1C), difficulties with mobility or obtaining food and water were rarely observed. To control for any difficulty obtaining food due to dwarfism, moistened food pellets were placed on the floor of the cage every day. In addition, thorough necropsies of healthy and moribund DKO mice were remarkable for their lack of obvious gross or histological defects in skin, muscle, cardiac, pulmonary, genito-urinary, brain, blood or abdominal organs. Normal complete blood counts, serum electrolytes, glucose, liver enzymes, amylase, lipase, calcium, alkaline phosphatase, creatinine, cholesterol and total protein indicated that the mice were not dying from severe metabolic defects, organ failure or malnourishment (data not shown). Potential cellular and molecular mechanisms underlying the weight loss and spinal defects were evaluated further.

Both male and female DKO mice failed to gain as much weight after weaning as their littermate controls (Fig. 1D) and when observed carefully, demonstrated a gradual body mass decline prior to early death (Supplementary Material, Fig. S1A). Because endocrine abnormalities are a significant cause of body mass derangements, we considered the possibility that the cause of growth retardation could be alteration of the growth hormone (GH)/insulin-like growth factor-1 (IGF-1) axis (19), hypothyroidism or diabetes. We therefore tested serum IGF-1, GH, thyroid-stimulating hormone (TSH) and fasting glucose levels in DKO mice (Table 2). Surprisingly, there were no abnormalities in DKO mice for any of these tests.


View this table:
[in this window]
[in a new window]

  		Table 2. Endocrine function is intact in DKO mice

 
To further explore the nature of the growth defects, we harvested, weighed and generated protein extracts from the organs of both newborn and adult DKO and littermate control mice. Newborn DKO mice were found to be identical in size to their littermates (data not shown). In young adult DKO mice, however, the mass of the liver, kidneys, testes, heart and lungs were consistently reduced in proportion to total body weight. These organs, as well as the spleen and brain, express at least low levels of both HIP1 and HIP1r in wild-type mice. Interestingly, in the young DKO adult mice, the spleen mass, although also usually proportionately reduced in size, was in some mice larger than that expected for the diminished size of the mouse. In contrast, the brain mass in the young adult DKO mice was consistently the same as the littermate controls (Fig. 1E). The increase in relative size might be expected for cerebral tissue, which reaches its adult size by 3 weeks of age and is known to be the last organ affected by nutritional deficiencies (20). Histological analysis of these organs displayed no gross abnormalities aside from the cataracts of the eye and testicular degeneration previously reported in Hip1null/null mice (Supplementary Material, Fig. S2).

Since HIP1 or HIP1r over-expression stabilizes and thereby up-regulates growth factor receptors (1), HIP1 and HIP1r bind to the lipid substrates of PTEN, and HIP1 over-expression is associated with decreased levels of AP2 (13), we wanted to test if loss of function had the opposite effect. To test whether physiological levels of HIP1 and/or HIP1r are required for the normal expression of these endocytic and signaling proteins in vivo, we examined each organ to determine if endocytic proteins (clathrin, AP2), PTEN and receptor (AR, PDGFßR and EGFR) levels were altered. To our surprise, none of these proteins were altered with respect to their absolute levels in any organ (data not shown). We have also tested embryonic brain extracts for EGFR, PDGFßR, actin and PTEN levels and again found no differences between DKO and control brains (Supplementary Material, Fig. S1B). Thus, we observed no evidence for global changes in endocytosis or signaling in tissues of DKO mice as were previously observed as a result of over-expression of HIP1 or HIP1r in cultured cells (1,21).

Characterization of HIP1/HIP1r DKO MEFs
To study the effects of the complete loss of the HIP1 pathway on cellular phenotypes, a series of DKO (n = 8) and control embryos (n = 31) were isolated and the fibroblasts were studied. The parents of these embryos were both Hip1r–/–; Hip1null/+ so that all offspring were Hip1r–/–. All eight of the DKO embryos and their 31 control littermates were without gross abnormalities and were of normal size (data not shown). Despite the normal appearance of the embryos, many of the cultured DKO MEFs were larger, appeared flattened and grew to a lower density compared with their littermate MEFs (Supplementary Material, Fig. S3A, phase contrast). These early passage MEFs were analyzed for chromosomal number, proliferation (using MTT assay, Fig. 2B) and cellular survival (Fig. 2A, DAPI stain). Chromosomal spreads showed no increase in aneuploidy or other chromosomal aberrations (data not shown). As would be expected based on the normal appearance of the DAPI-stained nuclei (Fig. 2A), there was no increase in apoptosis by TUNEL analysis in the early passage DKO MEFs. Although DKO MEFs differed in their appearance relative to control MEFs and did not achieve as high a cell density at confluence as observed in littermate control cell cultures, their overall cell cycle distribution was similar based on propidium iodide staining (data not shown). This suggested that progression through all phases of the cell cycle was delayed in DKO MEFs.


Figure 2
View larger version (62K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Analysis of DKO MEFs. (A) Phalloidin (red) and DAPI (blue) stain of early passage (P4) MEFs from control and DKO embryos. (B) Early passage (P4) 6-day-old embryonic fibroblasts were plated at 2000 cells per 96 wells in 10% FCS and samples were assayed in triplicate for viability using the MTT assay. Dotted lines are DKO cell and solid lines are littermate controls. (C) Comparison of PDGFßR levels after stimulation with PDGFßß. MEF cell lines were harvested 0, 1 and 2 h after starvation and stimulation with PDGFßß and subject to western blot analysis for HIP1, PDGFßR, Akt, Phospho-Akt, PTEN and phospho-PTEN. Actin and EGFR were used as loading controls. All MEFs were null for HIP1r and either wild-type for Hip1 (+/+) or null for Hip1 (–/–). (D) Comparison of EGFR levels after stimulation with EGF. MEF cell lines were harvested 0, 1, 2 and 4 h after starvation and stimulation with EGF and subject to western blot analysis for HIP1, EGFR, Akt, Phospho-Akt, PTEN and phospho-PTEN. Actin and PDGFßR were used as loading controls.

 
In order to test the effects of HIP1 and HIP1r deficiency on ligand-induced endocytosis of PDGFßR and EGFR, the same DKO MEF lines that displayed growth defects and control MEFs were starved for 24 h in serum-free medium and stimulated with PDGFß or EGF. Whole cell lysates were collected at various time points after stimulation and subject to western blot analysis to determine PDGFßR and EGFR levels. Ligand-mediated degradation or downstream Akt signaling of PDGFßR or EGFR was not altered in the DKO MEFs compared with control MEFs. Total or phosphorylated PTEN levels also were not altered in the PDGFß and EGF-stimulated DKO MEFs compared with control MEFs (Fig. 2C and D). In sum, although HIP1/HIP1r over-expression resulted in increased receptor levels and signaling (1), HIP1 family deficiency did not display diminished receptor levels or signaling.

Next, we utilized fluorescently labeled transferrin to test the effects of HIP1 and HIP1r deficiency on constitutive endocytosis. Wild-type and DKO 3T3 immortalized MEFs were starved for 3 h and then incubated with Alexa Fluor 633-conjugated transferrin. Samples were collected at 0, 10, 30, 60 and 90 min after stimulation. Analysis by flow cytometry showed no significant difference in accumulation of transferrin in DKO MEFs compared with wild-type MEF lines (Supplementary Material, Fig. S3B). These data from the transferrin and PDGFßR/EGFR experiments indicated that HIP1 and HIP1r are not required for either constitutive or stimulated endocytosis in embryonic fibroblasts. In addition, the steady-state levels of the androgen receptor, IGF-1R and insulin R were, like the PDGFßR and EGFR, not altered in the early or late passage MEFs (data not shown). We confirmed HIP1/HIP1r genotypes both by western blot (Fig. 2C and D) and Southern blot analyses.

As noted above, there was no alteration in the steady-state levels of several growth factor receptors in the various tissues of the DKO mice. Nonetheless, there might be functional alteration in endocytosis in primary cells that have not been cultured for significant periods of time. To begin to test if there was altered endocytosis of transferrin in primary cells, we have harvested spleens from DKO and littermate mice and tested for transferrin uptake. To do this, we cultured dissociated splenocytes overnight. They were then incubated in serum-free medium for 2 h and Alexa Fluor 633-conjugated transferrin was then added to the medium. Samples were collected at 0, 30, 60 and 90 min after stimulation. Analysis by flow cytometry showed no significant difference in accumulation of transferrin between control versus DKO splenocytes (Supplementary Material, Fig. S3C). These data indicated that HIP1 and HIP1r are not absolutely required for constitutive endocytosis in the DKO spleen.

Since the data ruled out a complete disruption of endocytosis in the HIP1/HIP1r-deficient fibroblasts and splenocytes, we investigated other potential functions of the HIP1 family. For example, HIP1 and HIP1r have TALIN homology domains that have the capacity to interact with actin (7,8). To evaluate whether actin structures were altered with the deficiency of HIP1 and HIP1r, we stained DKO MEFs with the fluorescently labeled actin-binding compound phalloidin (Fig. 2A). Previous data indicated that the HIP1 family is involved in promoting a productive actin-mediated endocytic control mechanism (22). Although early passage DKO MEFs were larger, flatter and less dense, the actin filaments were intact and displayed no obvious differences between the DKO and control fibroblasts (Fig. 2A). In addition, functional experiments that involved testing for the abnormalities in actin dynamics as they related to endocytosis were carried out and no differences were found between the DKO and control MEFs (data not shown). For example, it was previously shown that in HeLa cells, HIP1r was required for maintenance of normal endocytic structures associated with actin cytoskeleton proteins and that HIP1r depletion resulted in the dispersal of the trans-golgi network (23). Furthermore, both the clathrin-mediated endocytosis and golgi-to-endosome trafficking that were inhibited by RNAi-mediated depletion of HIP1r were rescued by ectopic expression of wild-type HIP1r in HeLa cells (22). Since actin, Golgi and overall endocytic abnormalities were not observed in the DKO MEFs, despite the observation of altered growth characteristics, we conclude that HIP1 and HIP1r are not required for these pathways in all cell types but are required to maintain several adult tissues in vivo.

Rescue of the DKO mice by transgenic expression of human HIP1
To further study why the DKO mice developed such a severe degenerative phenotype, we attempted to rescue the phenotypes caused by deficiency of HIP1 and HIP1r by expressing a single human HIP1 transgene in DKO mouse tissues. We chose to express the human HIP1 cDNA rather than the mouse HIP1 cDNA to determine if the mouse and human genes were functionally interchangeable and to allow for convenient differentiation of the transgenically expressed protein from endogenous mouse HIP1. In these mice, the hHIP1 cDNA expression was driven via the CAG promoter, which is a chimera of the CMV-IE enhancer and a modified chicken ß-actin promoter (Fig. 3A) (24). The transgenic construct also contained a floxed EGFP cDNA upstream of the human HIP1 sequence. Ubiquitous or tissue-specific expression of Cre recombinase would then result in the excision of the EGFP and expression of human HIP1 protein. For this study, the EGFP expressing transgenic founder mice (pCle.hHIP1) were mated with EIIa-cre mice. EIIa-cre mice express the Cre transgene under the control of the adenovirus EIIa promoter, resulting in ubiquitous expression of the Cre recombinase in the early mouse embryo and frequently in the germline (25). We selected for founder mice that produced EGFP-negative, hHIP1-positive progeny reflective of germline recombination. Two original pCle.hHIP1 transgenic mouse lines were used to mate with the EIIa-cre mice, and the derivative EGFP-negative lines expressing human HIP1 were designated: hHIP1hi and hHIP1lo. The hHIP1hi transgenic mice displayed ubiquitous tissue expression of human HIP1 protein in every tissue examined (Fig. 3B), whereas hHIP1lo transgenic mice displayed restricted expression of human HIP1 protein that was limited to skeletal and cardiac muscle (Fig. 3B). To test for the degree of over-expression, the ratio of amount of human HIP1 protein relative to endogenous mouse HIP1 protein in the hHIP1hi line of mice was examined in a variety of tissues by western blot using an HIP1 polyclonal antibody (UM354). This antibody recognizes both mouse and human HIP1. The full-length human HIP1 protein is larger than mouse HIP1 and as a result can be distinguished from the mouse protein by its slightly slower migration on SDS–PAGE (Fig. 3B). The highest ratios of human HIP1 expression in the HIP1hi mouse line, compared with mouse HIP1 expression, were detected in the liver (39x), eye (34x), and skeletal muscle (40x) (Supplementary Material, Fig. S4A). To begin to test for growth factor receptor pathway alterations, western blot analysis for total levels of the EGFR in the livers from transgenic and control mice was performed and demonstrated only slight variability in the total and phosphorylated levels of the EGFR (Supplementary Material, Fig. S4B). Immunohistochemical analysis, using the HIP1 polyclonal antibody UM323 (Fig. 4), demonstrated over-expression of the HIP1 protein in the endocrine (Islets of Langerhans) and exocrine pancreas (Fig. 4A), skeletal muscle fibers (Fig. 4B), the anterior epithelial layer and the youngest peripheral fibers in the lens of the eye (Fig. 4C), the transitional epithelium of the bladder (Fig. 4D) as well as many other tissues including cells of the vertebral column (Fig. 4E). Not all cell types expressed HIP1 within a tissue. For example, in the bladder the protective ‘umbrella’ cells that cover the transitional epithelium (Fig. 4D, arrow) were completely negative for hHIP1 expression and the vast array of bone marrow cells also displayed variable expression of the hHIP1 transgene (Fig. 4E).


Figure 3
View larger version (37K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Generation of human HIP1 transgenic mice. (A) Conditional human HIP1 transgenic mice were generated using the pCle.hHIP1 vector that includes a CMV enhancer, ß-actin promoter and loxP sites surrounding the EGFP cDNA upstream of the human HIP1 cDNA. Germline Cre expression resulted in excision of the EGFP sequence and expression of human HIP1 protein in two different mouse lines designated hHIP1hi and hHIP1lo. (B) Extracts from a variety of tissues from wild-type, hHIP1hi and hHIP1lo transgenic mice were analyzed for levels of HIP1 using the 4B10 antibody that detects human HIP1 (lower blots) and the UM354 antibody that detects both mouse and human HIP1 (upper blots). Human HIP1 migrated slightly slower than mouse HIP1. The most significant relative increases in HIP1 were observed in the liver (39x), eye (34x) and skeletal muscle (40x). The HIP1lo transgenic mouse had high levels of hHIP1 expression only in skeletal and cardiac muscles.

 


Figure 4
View larger version (136K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Immunohistochemical analysis of human HIP1 in tissues derived from hHIP1hi transgenic mice. (A–E) Immunohistochemical staining with the HIP1 polyclonal antibody UM323 of a variety of tissues from HIP1hi transgenic mice and control mice. Endogenous mouse HIP1 was not detectable in any tissue with this antibody. (A) Human HIP1 expression in the endocrine and exocrine pancreas, with a hematoxylin counter-stain (10x). (B) Human HIP1 expression in paraspinal muscle (10x). (C) Human HIP1 expression in the anterior epithelial layer and the more recently formed peripheral fibers in the lens of the eye (10x and 40x). (D) Human HIP1 expression in the transitional epithelium of the bladder, but not the umbrella cell layer (arrows) (10x). (E) Human HIP1 expression in the bone marrow and trabeculae of a lumbar vertebra of a hHIP1hi;DKO mouse and a wild-type littermate (100x). 1, osteocytes; 2, bone; 3, hyaline cartilage; 4, ostecoclast; 5, osteoblast; 6, bone marrow cells; 7, blood vessel.

 
The hHIP1hi line was chosen to attempt to ‘rescue’ the DKO mice. This was decided due to the broad spectrum of tissues where human HIP1 was expressed in this line (Figs 3 and 4). The rationale was that this ‘ubiquitous’ expression would more likely result in a successful DKO rescue than the hHIP1lo line where hHIP1 expression was limited (Fig. 3). Transgenic hHIP1hi mice were crossed with Hip1r–/–; Hip1+/null mice, hHIP1hi;Hip1r+/-;Hip1+/null progeny were then crossed with Hip1r–/–;Hip1+/null mice to generate hHIP1hi;Hip1r–/–;Hip1null/null (hHIP1hi;DKO) mice (genotypes displayed in Supplementary Material, Fig. S5). So far, this cross has resulted in 25% hHIP1hi;DKO mice (n = 3; total n = 12), which, although the numbers of mice were small, was an increased frequency compared with the expected Mendelian frequency of 6.25%. This suggests that the perinatal lethality associated with HIP1 deficiency (15) was rescued in these mice (Table 1). The three hHIP1hi;DKO mice that have been generated to date were free from spinal defects (Fig. 5), maintained their weight (Fig. 1D) and survived to at least 8 months of age (Fig. 1A).


Figure 5
View larger version (60K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Spinal defects and diminished weight of DKO mice are rescued by transgenic expression of the human HIP1 protein. HIP1/HIP1r DKO, littermate control and human HIP1 transgene expressing HIP1/HIP1r DKO (hHIP1;DKO) mice were compared for spinal defects and diminished weight. (A) Photograph of 5-month-old DKO, littermate control and hHIP1;DKO mice. (B) In vivo micro-CT images of the spine of 5-month-old DKO, littermate control and hHIP1;DKO mice. (C) Aggregate spine angle [angle between the T8–T13 and L1–L6 vertebrae sections of the spine, which is indicated by the angle between the black lines in (B)]. There was a significant difference (P < 0.003) between DKO and the corresponding littermate control mice.

 
To further evaluate the spines of these hHIP1hi;DKO mice, 5-month-old hHIP1hi;DKO (n = 3), DKO (n = 4) and their control littermates (n = 7) were anesthetized and scanned with an in vivo micro-computed tomography (micro-CT) system. Representative images from these scans are shown in Figure 5B. The angle of the spine at the T8–T13 and L1–L6 vertebrae sections (lines in Fig. 5B) were computed. The DKO mice displayed a significantly smaller thoraco-lumbar spine angle (P < 0.003) than the wild-type littermates (Fig. 5C). The spinal angles for the hHIP1hi;DKO mice were not different from the wild-type littermates, again demonstrating the complete rescue of the spinal defects with human HIP1 expression (Fig. 5C).

We have found that osteoclast differentiation in vivo and ex vivo in DKO mice is intact as evidenced by normal levels of serum tartrate-resistant acid phosphatase (TRAP) and normal differentiation of DKO bone marrow cells to osteoclasts (Supplementary Material, Fig. S6A). Quantitation of osteoblast and osteoclast frequency in bone sections from the femurs and vertebra of DKO mice was also not significantly abnormal (data not shown). Further analysis of the bone remodeling process in fractured or PTHrp challenged DKO mice will be of interest. Because the hHIP1hi;DKO mice express HIP1 in osteoclasts, osteoblasts and osteocytes (Fig. 4E), as well as in the muscles and spinal cord neural components, we are not able to predict from the current rescue mice which, if any, of these cell types are required for the maintenance of the spine integrity.

Severe microopthalmia was observed in all DKO mice (arrow in the middle panel of Fig. 5A). This phenotype was completely rescued by human HIP1 expression. However, it is likely that the hHIP1;DKO mice are still afflicted with cataracts to some degree, albeit less than the DKO mice, as all six pupils of the hHIP1;DKO eyes were visibly opaque. This pupil opacity has been observed in Hip1null/+ mice (15), suggesting that human HIP1 levels from the transgene may not have been high enough to completely rescue this ‘dose-dependent’ phenotype (Supplementary Material, Fig. S6B).

The Hip1null/null male infertility was not rescued in the one male hHIP1;DKO mouse. This male mouse was sacrificed at 8 months of age and displayed mild testicular degeneration, with a lack of sperm in the testes and epididymis, compared with its Hip1r–/–;Hip1+/null littermate (Supplementary Material, Fig. S6C). However, since the levels of human HIP1 expression in the seminiferous tubules of the testes and the epididymis were low, a very possible explanation of this failed rescue is that the human HIP1 was not expressed at high enough levels in the right cells of the testis to compensate for the deficiency of mouse HIP1.

The rescue of the DKO adult degenerative phenotypes (spinal defects, weight loss, premature adult death) by the expression of human HIP1 protein demonstrates for the first time the functional similarity of the human and mouse HIP1 protein family. The result also demonstrates that the severe kypholordosis is due to the loss of mouse HIP1 and HIP1r and not due to the effects of altered expression of neighboring genetic elements (genes, microRNAs, etc.) on chromosome 5 due to the Hip1/Hip1r ‘knockout’ mutations.

In light of the successful hHIP1hi transgenic rescue experiment, future attempts to rescue the phenotypes of the DKO mice or their derivative cells with tissue-specific transgene expression, such as in muscle, cartilage, osteoclasts, osteoblasts or endothelial cells, will be made. Also, HIP1 transgenes that are lipid, clathrin and actin binding-deficient could be used to determine whether these domains are necessary for the in vivo maintenance function of HIP1. Results from these experiments will provide us with a deeper understanding of the in vivo molecular pathways this protein family acts through to maintain normal adult tissues or, when over-expressed, to promote tumor growth.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
The HIP1 family has been implicated in a variety of fundamental cellular pathways including endocytosis (2–6,17), actin dynamics (17,26,27), cellular survival (5,14,28,29) and nuclear hormone receptor signal transduction (18). These proteins also have been implicated in complex in vivo physiological disease processes such as aging (30), neurodegeneration (9,10,16) and tumorigenesis (13,14,31–33). For example, alteration of HIP1 in cancer is widespread. HIP1 was first implicated in cancer as the chromosome 7 partner in the leukemogenic t(5;7) PDGFßR translocation (32). Subsequently, HIP1 itself was found to be up-regulated in multiple epithelial tissues including colon, prostate and breast cancers (14) and to possess transforming activity in vitro (13). This transformation was associated with increased EGFR signaling (1,13), and recently, we found that HIP1 deficiency inhibits prostate tumorigenesis in vivo (31). As a result, we proposed that altered endocytosis promotes tumorigenesis. As a corollary, we proposed that targeted therapeutics might manipulate endocytosis of growth factor receptors to induce senescence or cell death to avoid or reverse transformation.

To gain a better understanding of the physiological function of the HIP1 family in the context of a whole mammalian organism, we have studied a large cohort of mice deficient for both HIP1 and HIP1r. These mice were viable and appeared normal as newborns but subsequently all of them developed multiple degenerative abnormalities including early death and cachexia, which could be linked to the severe spinal defects that they exhibit. These early adulthood phenotypes were not observed in the single Hip1 and Hip1r knockout mice (Table 1). Despite these severe phenotypes, there was no evidence for widespread or severe disruptions in endocytosis, PtdIns 3-kinase signaling or actin dynamics. In addition, DKO mice displayed relatively normal cellular differentiation as reflected by normal embryonic and perinatal development as well as normal differentiation of fibroblasts to fat cells (data not shown) and bone marrow progenitors to macrophages and osteoclasts.

We have not discovered a single cause of early death after an extensive and detailed analysis of both ‘healthy’ and moribund DKO mice. Despite the wasting phenotype in early adulthood, moribund DKO mice showed no evidence of cancer or infection at necropsy. Histology of all major organs other than the disordered thoraco-lumbar spinal column, eye and testes was normal. The pituitary axis, pancreatic function and the steady-state hematopoietic system were normal. For example, we did not find alteration in the steady-state levels of nuclear hormone receptors, endocytic proteins, glucose, GH, TSH or IGF-1. There also were no changes in fibroblast or macrophage actin patterns, transferrin uptake by MEFs or splenocytes or the stability of the EGFR or PDGFßR in response to EGF or PDGFßß stimulation. Since the DKO mice display the most severe kyphosis ever reported among surviving adult mutant mice with this type of defect (34–41), we surmise that pulmonary and cardiac complications due to restricted skeletal cage expansion likely contribute to the early death of DKO mice.

To further understand the mechanism of how the DKO phenotype develops, we generated human HIP1 transgenic mice to determine whether ‘ubiquitous’ expression of human HIP1 protein would rescue the DKO phenotype. Three hHIP1hi;DKO mice were successfully generated and displayed no evidence for kypholordosis, weight loss or early death at 8 months of age, whereas 100% of DKO mice were dwarfed and hunched and 88% of them were dead by this age. Furthermore, using in vivo micro-CT analysis of the spines, we have demonstrated that the reduced spinal angle of DKO mice is reverted to a normal spinal angle by human HIP1 expression.

Interestingly, the three hHIP1;DKO mice did have mild clinical cataracts, and the one male hHIP1;DKO mouse that was sacrificed displayed a mild pathological cataract (Supplementary Material, Fig. S5B). Although there was expression of human HIP1 in the hHIP1;DKO lens (Supplementary Material, Fig. S5B), it did not completely rescue this characteristic of the Hip1null/null and DKO phenotype (Table 1). Recently, we have found that the two female hHIP1hi;DKO mice are fertile, whereas the male hHIP1hi;DKO mouse was infertile with low sperm counts and testicular degeneration similar to the Hip1null/null phenotype (Table 1). Since the cataracts and the male infertility were not completely rescued by the human transgene, it is possible that the expression of the transgene is not at the correct levels in the correct cell types in those tissues. In fact, we did find that the human HIP1 protein was only expressed in the capsule and the interstitial cells of the hHIP1;DKO mouse testis, and not the sertoli cells and spermatids of the seminiferous tubules (Supplementary Material, Fig. S5C). HIP1 expression has been shown previously to be highly expressed in the postmeiotic spermatids of human testis (5) and in the elongating spermatids and sertoli cells of mouse testis (42). The lack of any human HIP1 expression in the seminiferous tubules of the hHIP1;DKO testis could explain the inability to rescue the HIP1 null phenotype of testicular degeneration. Less likely are the possibilities that the human HIP1 does not provide all of the functions that the mouse HIP1 provides or that neighboring genes or intragenic micro-RNA elements contribute to the testis phenotype.

Somewhat to our surprise, DKO fibroblasts did not display defects in stability or endocytosis of PDGFß, EGF and transferrin receptors despite the fact that these receptors have been shown to be affected by over-expression and/or RNAi-mediated knockdown of HIP1 and HIP1r. This suggests that although over-expression or acute knockdown of the HIP1 family perturbs endocytosis, the HIP1 family is not physiologically necessary for endocytosis. One possible explanation for the failure to observe endocytic defects in DKO mice, despite the report that HIP1 is required for endocytosis in cell lines based on RNAi knockdown experiments (22), is that HIP1 is required for endocytosis in certain human cell lines, but is not physiologically required for endocytosis in most mouse cells. Levels of a large array of endocytic proteins, all HIP1/HIP1r-interacting proteins (huntingtin, AP2, clathrin, actin) and cell surface receptors were also unchanged. Although we have not yet found endocytic defects in the single HIP1 (15), HIP1r (12) or DKO mice and their derivative MEFs, it remains possible that endocytic or 3-kinase pathway defects in the HIP-deficient state may be present in response to certain types of signals or in cell types that we did not study. Finally, it is possible that compensatory mechanisms that do not involve HIP1 family members are induced after Hip1/Hip1r knockout in mice. Further experiments will be required to distinguish between these possibilities.

The adult-onset defects observed in DKO mice are in stark contrast to the embryonic lethality observed in huntingtin (43–45) and AP2 (46)-deficient mice. This lack of complete embryonic lethality in the DKO mice indicates that the loss of Htt or AP2 interaction with the HIP1 family does not lead to a complete loss of AP2 or huntingtin functions. In addition, knockin of mutant CAG-expanded huntingtin resulted in adult brain phenotypes (47) that were not modified by HIP1 deficiency, and the HIP1 deficiency phenotypes (listed in Table 1) were not modified by the huntingtin mutant allele (Oravecz-Wilson et al., manuscript in preparation). These data indicate that there is no genetic interaction between HIP1 and huntingtin mutations. Taken together, HIP1 and HIP1r may function as modulators of huntingtin, clathrin, AP2 and actin functions but their activities are not necessary for these interacting proteins to perform their baseline ‘housekeeping’ functions.

We propose that the HIP1 family provides a necessary function involved in general maintenance of adult tissues and speculate that HIP1 may have important functions in aging tissues. A previous observation that HIP1 was up-regulated in fibroblasts cultured from progeria patients and elderly individuals (30), together with the degenerative phenotypes of the DKO mice, predict that HIP1 and HIP1r may participate in pathways of aging. Further analyses of pathways that contribute to aging or cellular senescence such as the INK4A/ARF pathway are underway and will be of interest. In addition, HIP1 over-expression transforms fibroblasts (13), HIP1 is over-expressed in a variety of cancers (14) and HIP1 deficiency inhibits prostate tumorigenesis (31). In contrast to these observations, deletion of tumor suppressor genes tends to increase cancer incidence in mice and may even delay the onset of aging phenotypes (48).

Finally, the phenotypes in the DKO mice, together with the location of the human HIP1 gene to 7q11, suggest that there may be autosomal recessive mutations of HIP1 in human patients with genetic syndromes that include dwarfism and well as eye, testicular and skeletal abnormalities. The chromosomal locus of 7q11 and its syntenic 5q region in mice is a gene-rich region and is somatically deleted in leukemias and pre-leukemia syndromes. Whether germline mutations in the HIP1 gene (and the HIP1r gene at 12q in humans and 5q in mice) occur in genetic syndromes that include cataracts, spinal defects and dwarfism will be important to determine.

In sum, although single Hip1 and Hip1r knockout mice are relatively normal in young adulthood, double-deficient mice, as might be expected for a tumor-promoting protein family, were dwarfed, hunched and died young. The profound effects of HIP1/HIP1r deficiency indicate that the HIP1 family is important for several aspects of organism homeostasis and provide avenues for further study. Also, the rescue of the DKO adult phenotype by the expression of human HIP1 transgene demonstrates for the first time that the human HIP1 protein has a similar function to the mouse HIP1 family. Finally, our study provides the first evidence that the HIP1 family is necessary not only for tumorigenesis (31), but also for a normal adulthood and lifespan in mammals without being widely required for endocytosis or PtdIns 3-kinase signaling.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 REFERENCES
 
Animals
Hip1r knockout (12), Hip1 knockout (15) and DKO mice (12) were generated, monitored and maintained as described previously. Human HIP1 transgenic mice were generated by standard procedures that included microinjection of C57BL/6xSJL fertilized eggs with vector DNA (pCle.hHIP1) and founder mouse generation. The transgenic vector, pCLE.hHIP1, contains the CMV-IE enhancer and ß-actin promoter (CAG promoter) and loxP sites surrounding an EGFP transgene upstream of the human HIP1 sequences (Fig. 3A). Because expression of the human HIP1 cDNA was expected only if EGFP was excised, fluorescent green founder mice (lines 395 and 396) were initially mated with EIIa-cre transgenic mice to recombine the loxP sites and remove the EGFP sequence in the germline DNA. We reasoned that in the event that the HIP1 over-expression was not lethal to the development of the mouse, expression of the human HIP1 transgene from the ubiquitous CAG promoter would increase the chances of a successful DKO rescue experiment. Two different germline transgenic lines were developed from two different EGFP founders: HIP1hi and HIP1lo. HIP1hi transgenic mice displayed expression of HIP1 in every tissue tested, whereas HIP1lo transgenic mice displayed expression of HIP1 only in skeletal and cardiac muscles. This difference is presumably due to different integration sites of the transgenes where the HIP1hi line (derived from the original line 395) has the transgene integrated in a more widely transcribed region of the genome.

Mice were genotyped either by Southern blot for mouse Hip1 (15) and Hip1r (12) targeted mutations or in the case of the human HIP1 transgene by PCR for either the unrecombined EGFP-containing transgene or a more generic set of primers that amplify the human HIP1 sequences. The PCR primers for the unrecombined EGFP/HIP1 construct included a forward primer in the EGFP cDNA and a reverse primer in the hHIP1 cDNA. The respective sequences were as follows: EGFP: forward (5'-CGG CAA CAT CCT GGG GCA CAA GCT GGA GTA CAA CTA CAA CAG-3') and huHIP1: reverse (5'-GCA CCT TGG GCA GTG GGT TGG GCA CCT GCT TCA TGG AGC-3'). The PCR primers that amplify positions between nucleotides 1262 and 1391 of the human cDNA, and detect both the recombined transgene and the unrecombined, EGFP containing, transgene were as follows: 5'EcoR1: forward (5'-CTG AGA GCC AGC GGG TTG TGC TGC AGC TGA-3') and 3'EcoR1: reverse (5'-CTC CTT TAG CTT GCT ATA TCG CTG TTC ATT GGC-3').

Serum assays
Blood was collected from the saphenous vein of mice that were maintained and generated as described previously (12). Serum IGF-1 was measured in duplicate using a mouse/rat IGF-1 enzyme immunoassay (EIA) kit according to manufacturer's directions (Diagnostic Systems Laboratories, Inc.). mGH RIA [anti-serum NIDDK-Anti-Rat GH-RIA-5 (AFP)] and mTSH RIA [anti-serum Anti-mTSH AFP98991 (GP)] were used to test for GH and TSH levels. Active tartrate-resistant acid phosphatase form 5b (TRAP 5b) levels in serum were assessed in duplicate with the MouseTRAP Assay from SBA Sciences according to manufacturer's directions. Serum osteocalcin levels were measured in duplicate with a mouse osteocalcin EIA kit according to manufacturer's recommendations (Biomedical Technologies, Inc.). Absorbance for each test was read on a VersaMax microplate reader.

Histology
Tissue obtained at necropsy was fixed in 10% formalin/PBS. Bone samples were decalcified using CalExII decalcifying solution (Fischer). Paraffin embedding and standard hematoxylin and eosin staining were performed by the University of Michigan Comprehensive Cancer Center Research and Histology and Immunoperoxidase Laboratory. Immunohistochemistry for human HIP1 expression was performed with the HIP1 polyclonal antibody UM323 [generated with the human 3' HIP1 antigen (49)] at a 1:500 dilution (8-7-2000 bleed).

Generation and genotyping of MEFs
MEF generation and genotyping for Hip1 and Hip1r by Southern blot have been described previously (12,15). MEFs that were less than 10 passages were defined as ‘early passage’ (50).

Phalloidin staining
Cells were plated on irradiated cover slips and cultured to 60–70% confluence. Media were removed and cells were washed twice with PBS. Cells were then fixed with 3.7% formaldehyde/PBS for 10 min at room temperature, washed twice with PBS, incubated with 200 µl of PBS/0.1% Triton X-100 for 3–5 min, and washed twice with PBS. Five microliters of phalloidin-Texas red (Molecular Probes) in 200 µl of PBS/1% BSA was then incubated for 20 min at room temperature in the dark with the fixed cells. Cells were then washed twice with PBS, allowed to air dry, permanently mounted and then visualized with fluorescence microscopy.

Transferrin accumulation in splenocytes and MEFs
Spleens were harvested, minced and dispersed using sterile frosted microscope slides with sterile PBS. The cells were transferred into 15 ml tubes and debris allowed to settle for 5 min. Cells were transferred to another tube and pelleted (8 min at 1000 r.p.m. in a IEC HN-SII centrifuge). The pellet was resuspended in RBC lysis buffer (15 mM potassium chloride, 1 mM potassium bicarbonate, 0.1 mM EDTA) for 1 min and diluted with 2 ml of media. Cells were pelleted, resuspended in 15 ml of RPMI-1640 media (10% FCS) and cultured overnight. Spleen cells were then starved in serum-free media for 3 h, washed twice in cold PBS/1% BSA and resuspended in serum-free media. MEFs were starved for 3 h in serum-free medium, trypsinized, washed in cold 1% BSA/PBS and resuspended in cold DMEM. Cells were incubated with Alexa Fluor 633-conjugated transferrin (Molecular Probes) for 1 h at 4°C for 1 h with rotation, then shifted to 37°C for 0, 30, 60 or 90 min. Cells were washed twice with cold PBS and fixed in fresh 1% paraformaldehyde/PBS. The samples were then analyzed by flow cytometry for transferrin accumulation as described previously (1).

Growth factor receptor stability assays
Experiments involving stimulation with PDGFß or EGF were modified from protocols described previously (1). Briefly, early passage MEFs were grown in 100 mM dishes DMEM media (10% FBS). Cells were starved 20–24 h in serum-free media. Cells were then treated with cycloheximide (100 µg/ml) for 30 min and stimulated with EGF (100 ng/ml) or PDGF (50 ng/ml) in the presence of cycloheximide. Samples were collected at 0, 1 and 2 h after stimulation. Ten micrograms of protein was separated on SDS–PAGE gels and transferred to nitrocellulose. Membranes were probed with anti-EGFR (Cell Signaling, 1:500), anti-PDGFßR (BD Pharmigen, 1:1000), anti-HIP1 (UM354, 1:5000), anti-AKT (Cell Signaling, 1:1000), anti-phospho-AKT (Cell Signaling, Thr-308, 1:1000) or anti-actin (Sigma, 1:1000) antibodies and signals were detected by SuperSignal West Pico Chemiluminescent Substrate (Pierce).

Micro-computed tomography
Five-month-old mice [DKO (n = 4; three of these DKO mice were part of an earlier experiment that indicated that their treatment with zoledronate did not alter the spinal defects, control littermate (n = 7) and hHIP1;DKO (n = 3)] were anesthetized and scanned with an in vivo cone beam micro-CT system (GE Healthcare BioSciences) and reconstructed at a voxel size of 45 µm. A linear interpolation in the vertical plane of the location of the T8–T13 vertebrae and the location of the L1–L6 vertebrae was done. The angle between the two lines was then determined, representing the angle between the T8–T13 and L1–L6 vertebrae sections of the spine (Fig. 5B). All results are expressed as mean ± SD. One-way ANOVA with a post hoc Bonferroni multiple comparisons was used to test for differences between DKO, littermate control and hHIP1;DKO groups. P-values less than 0.05 were considered significant.

In vitro osteoclast differentiation and TRAP stain
Mouse bone marrow cells were flushed from dissected femurs with complete DMEM medium using a 3 ml syringe and 27 G needle. Cells from each femur were plated on sterilized cover slips in six-well plates and treated with M-CSF (1 ng/ml) for 1 week to generate macrophages. Fresh medium containing M-CSF (30 ng/ml, Peprotech, Inc.) and soluble RANKL (300 ng/ml, Peprotech, Inc.) was added on day 7 and cells cultured an additional 7 days to generate osteoclasts (51). The protocol for fixing and TRAP staining of the osteoclasts was obtained from the BD BioCoat Osteologic Bone Cell Culture System (BD Biosciences).


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at HMG Online.


   ACKNOWLEDGEMENTS
 
Thanks to Dr. Jun-ichi Miyazaki for the CAG promoter. We would like to thank Steven Philips, Chiron Graves and Danielle Antonuk for critical review of the manuscript. Special thanks to Dr David Ferguson for providing analysis of fibroblast DNA content in DKO MEFs. This work was supported by grants R01 CA82363-01A1 (T.S.R.), R01 CA098730-02 (T.S.R.), a leukemia and lymphoma Society Scholar award #1035-06 (T.S.R.), T32-AG000114 (E.I.W.), CA87837 (AAD) and P30 AR46024 (S.A.G.).

Conflict of Interest statement. None declared.


   FOOTNOTES
 
{dagger} The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Hyun T.S., Rao D.S., Saint-Dic D., Michael L.E., Kumar P.D., Bradley S.V., Mizukami I.F., Oravecz-Wilson K.I., Ross T.S. HIP1 and HIP1r stabilize receptor tyrosine kinases and bind 3-phosphoinositides via epsin N-terminal homology domains. J. Biol. Chem. (2004) 279:14294–14306.[Abstract/Free Full Text]

  2. Engqvist-Goldstein A.E., Warren R.A., Kessels M.M., Keen J.H., Heuser J., Drubin D.G. The actin-binding protein Hip1R associates with clathrin during early stages of endocytosis and promotes clathrin assembly in vitro. J. Cell Biol. (2001) 154:1209–1223.[Abstract/Free Full Text]

  3. Metzler M., Legendre-Guillemin V., Gan L., Chopra V., Kwok A., McPherson P.S., Hayden M.R. HIP1 functions in clathrin-mediated endocytosis through binding to clathrin and adaptor protein 2. J. Biol. Chem. (2001) 276:39271–39276.[Abstract/Free Full Text]

  4. Mishra S.K., Agostinelli N.R., Brett T.J., Mizukami I., Ross T.S., Traub L.M. Clathrin- and AP-2-binding sites in HIP1 uncover a general assembly role for endocytic accessory proteins. J. Biol. Chem. (2001) 276:46230–46236.[Abstract/Free Full Text]

  5. Rao D.S., Chang J.C., Kumar P.D., Mizukami I., Smithson G.M., Bradley S.V., Parlow A.F., Ross T.S. Huntingtin interacting protein 1 Is a clathrin coat binding protein required for differentiation of late spermatogenic progenitors. Mol. Cell. Biol. (2001) 21:7796–7806.[Abstract/Free Full Text]

  6. Waelter S., Scherzinger E., Hasenbank R., Nordhoff E., Lurz R., Goehler H., Gauss C., Sathasivam K., Bates G.P., Lehrach H., Wanker E.E. The huntingtin interacting protein HIP1 is a clathrin and alpha-adaptin-binding protein involved in receptor-mediated endocytosis. Hum. Mol. Genet. (2001) 10:1807–1817.[Abstract/Free Full Text]

  7. Brett T.J., Legendre-Guillemin V., McPherson P.S., Fremont D.H. Structural definition of the F-actin-binding THATCH domain from HIP1R. Nat. Struct. Mol. Biol. (2006) 13:121–130.[CrossRef][ISI][Medline]

  8. Senetar M.A., Foster S.J., McCann R.O. Intrasteric inhibition mediates the interaction of the I/LWEQ module proteins Talin1, Talin2, Hip1 and Hip12 with actin. Biochemistry (2004) 43:15418–15428.[CrossRef][Medline]

  9. Kalchman M.A., Koide H.B., McCutcheon K., Graham R.K., Nichol K., Nishiyama K., Kazemi-Esfarjani P., Lynn F.C., Wellington C., Metzler M., et al. HIP1, a human homologue of S. cerevisiae Sla2p, interacts with membrane-associated huntingtin in the brain. Nat. Genet. (1997) 16:44–53.[CrossRef][ISI][Medline]

  10. Wanker E.E., Rovira C., Scherzinger E., Hasenbank R., Walter S., Tait D., Colicelli J., Lehrach H. HIP-I: a huntingtin interacting protein isolated by the yeast two-hybrid system. Hum. Mol. Genet. (1997) 6:487–495.[Abstract/Free Full Text]

  11. Seki N., Muramatsu M., Sugano S., Suzuki Y., Nakagawara A., Ohhira M., Hayashi A., Hori T., Saito T. Cloning, expression analysis, and chromosomal localization of HIP1R, an isolog of huntingtin interacting protein (HIP1). J. Hum. Genet. (1998) 43:268–271.[CrossRef][ISI][Medline]

  12. Hyun T.S., Li L., Oravecz-Wilson K.I., Bradley S.V., Provot M.M., Munaco A.J., Mizukami I.F., Sun H., Ross T.S. Hip1-related mutant mice grow and develop normally but have accelerated spinal abnormalities and dwarfism in the absence of HIP1. Mol. Cell. Biol. (2004) 24:4329–4340.[Abstract/Free Full Text]

  13. Rao D.S., Bradley S.V., K.P.D., H.T.S., Saint-dic D., O.-W.K.I., K.C.G., R.T.S. Altered receptor trafficking in huntingtin interacting protein 1-transformed cells. Cancer Cell (2003) 3:471–482.[CrossRef][ISI][Medline]

  14. Rao D.S., Hyun T.S., Kumar P.D., Mizukami I.F., Rubin M.A., Lucas P.C., Sanda M.G., Ross T.S. Huntingtin-interacting protein 1 is overexpressed in prostate and colon cancer and is critical for cellular survival. J. Clin. Invest. (2002) 110:351–360.[CrossRef][ISI][Medline]

  15. Oravecz-Wilson K.I., Kiel M.J., Li L., Rao D.S., Saint-Dic D., Kumar P.D., Provot M.M., Hankenson K.D., Reddy V.N., Lieberman A.P., Morrison S.J., Ross T.S. Huntingtin interacting protein 1 mutations lead to abnormal hematopoiesis, spinal defects and cataracts. Hum. Mol. Genet. (2004) 13:851–867.[Abstract/Free Full Text]

  16. Metzler M., Li B., Gan L., Georgiou J., Gutekunst C.A., Wang Y., Torre E., Devon R.S., Oh R., Legendre-Guillemin V., et al. Disruption of the endocytic protein HIP1 results in neurological deficits and decreased AMPA receptor trafficking. EMBO J. (2003) 22:3254–3266.[CrossRef][ISI][Medline]

  17. Legendre-Guillemin V., Metzler M., Charbonneau M., Gan L., Chopra V., Philie J., Hayden M.R., McPherson P.S. HIP1 and HIP12 display differential binding to F-actin, AP2, and clathrin: identification of a novel interaction with clathrin-light chain. J. Biol. Chem. (2002) 277:19897–19904.[Abstract/Free Full Text]

  18. Mills I.G., Gaughan L., Robson C., Ross T., McCracken S., Kelly J., Neal D.E. Huntingtin interacting protein 1 modulates the transcriptional activity of nuclear hormone receptors. J. Cell Biol. (2005) 170:191–200.[Abstract/Free Full Text]

  19. Isaksson O. GH, IGF-I, and growth. J. Pediatr. Endocrinol. Metab. (2004) 17(Suppl. 4):1321–1326.[ISI]

  20. Rodier P.M. Chronology of neuron development: animal studies and their clinical implications. Dev. Med. Child Neurol. (1980) 22:525–545.[ISI][Medline]

  21. Rao D.S., Bradley S.V., Kumar P.D., Hyun T.S., Saint-Dic D., Oravecz-Wilson K., Kleer C.G., Ross T.S. Altered receptor trafficking in huntingtin interacting protein 1-transformed cells. Cancer Cell (2003) 3:471–482.[CrossRef][ISI][Medline]

  22. Engqvist-Goldstein A.E., Zhang C.X., Carreno S., Barroso C., Heuser J.E., Drubin D.G. RNAi-mediated Hip1R silencing results in stable association between the endocytic machinery and the actin assembly machinery. Mol. Biol. Cell (2004) 15:1666–1679.[Abstract/Free Full Text]

  23. Carreno S., Engqvist-Goldstein A.E., Zhang C.X., McDonald K.L., Drubin D.G. Actin dynamics coupled to clathrin-coated vesicle formation at the trans-Golgi network. J. Cell Biol. (2004) 165:781–788.[Abstract/Free Full Text]

  24. Sawicki J.A., Morris R.J., Monks B., Sakai K., Miyazaki J. A composite CMV-IE enhancer/beta-actin promoter is ubiquitously expressed in mouse cutaneous epithelium. Exp. Cell Res. (1998) 244:367–369.[CrossRef][ISI][Medline]

  25. Lakso M., Pichel J.G., Gorman J.R., Sauer B., Okamoto Y., Lee E., Alt F.W., Westphal H. Efficient in vivo manipulation of mouse genomic sequences at the zygote stage. Proc. Natl Acad. Sci. USA (1996) 93:5860–5865.[Abstract/Free Full Text]

  26. Engqvist-Goldstein A.E., Drubin D.G. Actin assembly and endocytosis: from yeast to mammals. Annu. Rev. Cell Dev. Biol. (2003) 19:287–332.[CrossRef][ISI][Medline]

  27. Engqvist-Goldstein A.E., Kessels M.M., Chopra V.S., Hayden M.R., Drubin D.G. An actin-binding protein of the Sla2/Huntingtin interacting protein 1 family is a novel component of clathrin-coated pits and vesicles. J. Cell Biol. (1999) 147:1503–1518.[Abstract/Free Full Text]

  28. Gervais F.G., Singaraja R., Xanthoudakis S., Gutekunst C.A., Leavitt B.R., Metzler M., Hackam A.S., Tam J., Vaillancourt J.P., Houtzager V., et al. Recruitment and activation of caspase-8 by the Huntingtin-interacting protein Hip-1 and a novel partner Hippi. Nat. Cell Biol. (2002) 4:95–105.[CrossRef][ISI][Medline]