Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on August 27, 2004
Human Molecular Genetics 2004 13(20):2351-2359; doi:10.1093/hmg/ddh277

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 13/20/2351    most recent ddh277v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (43) Request Permissions Google Scholar Articles by Albrecht, A. N. Articles by Mundlos, S. Search for Related Content PubMed PubMed Citation Articles by Albrecht, A. N. Articles by Mundlos, S. Social Bookmarking          What's this?

Human Molecular Genetics, Vol. 13, No. 20 © Oxford University Press 2004; all rights reserved

A molecular pathogenesis for transcription factor associated poly-alanine tract expansions

Andrea N. Albrecht¹, Uwe Kornak¹, Annett Böddrich², Kathrin Süring¹, Peter N. Robinson¹, Asita C. Stiege¹, Rudi Lurz¹, Sigmar Stricker¹, Erich E. Wanker² and Stefan Mundlos^1,*

¹Max-Planck Institute for Molecular Genetics and Institute for Medical Genetics, Charité, Berlin, Germany and ²Department of Neuroproteomics, Max-Delbrück Center, Berlin, Germany

Received July 3, 2004; Accepted August 20, 2004

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Poly-alanine (Ala) tract expansions in transcription factors have been shown to be associated with human birth defects such as malformations of the brain, the digits, and other structures. Expansions of a poly-Ala tract from 15 to 22 (+7)–29 (+14) Ala in Hoxd13, for example, result in the limb malformation synpolydactyly in humans and in mice [synpolydactyly homolog (spdh)]. Here, we show that an increase of the Ala repeat above a certain length (22 Ala) is associated with a shift in the localization of Hoxd13 from nuclear to cytoplasmic, where it forms large amorphous aggregates. We observed similar aggregates for expansion mutations in SOX3, RUNX2 and HOXA13, pointing to a common mechanism. Cytoplasmic aggregation of mutant Hoxd13 protein is influenced by the length of the repeat, the level of expression and the efficacy of degradation by the proteasome. Heat shock proteins Hsp70 and Hsp40 co-localize with the aggregates and activation of the chaperone system by geldanamycin leads to a reduction of aggregate formation. Furthermore, recombinant mutant Hoxd13 protein forms aggregates in vitro demonstrating spontaneous misfolding of the protein. We analyzed the mouse mutant spdh, which harbors a +7 Ala expansion in Hoxd13 similar to the human synpolydactyly mutations, as an in vivo model and were able to show a reduction of mutant Hoxd13 and, in contrast to wt Hoxd13, a primarily cytoplasmic localization of the protein. Our results provide evidence that poly-Ala repeat expansions in transcription factors result in misfolding, degradation and cytoplasmic aggregation of the mutant proteins.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Trinucleotide repeat expansions are a common cause of disease. In principle, they can be categorized into two subclasses: those that have their repeats in non-coding sequences and those characterized by exonic repeats. Expansions of (CAG)n repeats that code for polyglutamine (poly-Q) tracts have been shown to be the basis of several human diseases, characterized by progressive neuronal dysfunction that typically begins in mid-life and results in severe neurodegeneration (1). Consequences of hyperexpansion of DNA-triplet repeats might include altered rates of transcription or translation, mRNA instability and aberrant hairpin structures (2). The expanded protein repeat is thought to destabilize the native configuration of the mutant protein resulting in the formation of aggregates within the cell. Although other amino acid repeats besides poly-Q are common in the human genome (3), few have been associated with disease. Recently, poly-alanine (Ala) tract expansions have been described as the causative defect in several conditions (4). To date, nine human conditions, including synpolydactyly (SPD) type II (HOXD13) (OMIM 186000) (5), cleidocranial dysplasia (RUNX2) (OMIM 119600) (6), holoprosencephaly type 5 (ZIC2) (OMIM 603073.0003) (7), hand–foot–genital syndrome (HOXA13) (OMIM 140000) (8), blepharophimosis epicanthus inversus syndrome (FOXL2) (OMIM 110100) (9), mental retardation with growth hormone deficiency (SOX3) (OMIM 300123) (10), Partington syndrome (ARX) (OMIM 309510) (11), congenital central hypoventilation syndrome (PHOX2B) (OMIM 209880) (12) and oculopharyngeal muscular dystrophy (OPMD) (PABPN1) (OMIM 164300) (13), have been described. With the exception of PABPN1, a poly-A binding protein, all mutated genes code for transcription factors with important roles during development and differentiation. The length of the normal Ala tract is similar in all transcription factors (14–20 Ala residues), and so is the expanded tract that causes disease (18–29 Ala residues), suggesting a common underlying mechanism. Poly-Ala repeats are meiotically stable and thus differ sharply from the dynamic expansions underlying conditions such as Huntington's disease or myotonic dystrophy. Furthermore, in contrast to other trinucleotide expansion diseases that cause late onset neurodegenerative conditions, poly-Ala expansions in transcription factors result in congenital defects. Thus, the underlying mutational mechanisms as well as the pathogenesis of poly-Ala repeat associated conditions appears to be distinct from poly-Q tract expansions. However, the pathogenic mechanism of Ala repeat expansions remains unknown.

Ala expansions (+7 to +14 Ala in addition to the 15 Ala in wt) in HOXD13 were the first mutations of this kind described (5). They result in a distinct hand malformation phenotype, SPD, a dominant condition characterized by syndactyly of fingers 3 and 4, with an additional finger in the polydactylous web. In a previous study, we could show that the length of the Ala-expansions correlates with severity and penetrance of the phenotype (14). Individuals with short repeats, i.e. +7 Ala, show considerable intra-familial and also intra-individual variability, whereas those with long expansions, i.e. +10 Ala and more, have a more severe and completely penetrant phenotype. The mechanisms by which these repeats cause pathology are unknown but studies in mice with similar expansions in Hoxd13 suggest a dominant negative effect (15). Likewise, inactivation of Hoxd13 in the mouse results in a rather mild phenotype not comparable with SPD, whereas inactivation of Hoxd11, Hoxd12 and Hoxd13 together results in a SPD-like malformation (16).

In this study, we aimed at identifying the molecular mechanisms behind the poly-Ala expansions in Hoxd13. Furthermore, we wanted to know whether the poly-Ala expansions in other transcription factors follow similar mechanisms resulting in a common pathogenetic cause. Our results show that an increase in the length of the poly-Ala repeat beyond a certain threshold results in cytoplasmic aggregation and degradation of the mutant protein, thus preventing nuclear localization and normal function of the transcription factor.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Mutant Hoxd13 accumulates in the cytoplasm
To investigate the mechnisms of Ala repeat expansion mutations, we made a series of constructs containing wt Hoxd13 (15 Ala), Hoxd13 with a very short Ala repeat (2 Ala) and Hoxd13 with expanded Ala repeats of varying length, similar to those observed in SPD patients. The size of the Ala repeat is indicated as the number of Ala residues in addition (+) to the 15 alanines in wt. These constructs were transiently expressed in Cos-1 cells and analyzed 40 h later by fluorescence microscopy. We observed the formation of large cytosolic aggregates, a finding that was most pronounced in cells transfected with the long expansions (Fig. 1A). In contrast, cells expressing the wt or the mutant containing only 2-Ala showed an exclusively nuclear localization. We analyzed the aggregates using electron microscopy and immunogold labeling (Fig. 1B). Wt Hoxd13 was found in the nucleus, mainly in the periphery in areas of chromatin condensations. In contrast, in cells expressing mutant Hoxd13 (+14 Ala) loose aggregates were observed in the cytoplasm that showed labeling for Hoxd13. These aggregates consisted of dense and more electron lucent areas that did not appear to have a higher structural order and were thus distinct from those observed in poly-Q disease associated aggregates such as Huntington's disease. Labeling for the mutant Hoxd13 protein was observed only in the electron dense areas.

View larger version (101K): [in this window] [in a new window]   Figure 1. Expansion of poly-Ala tracts leads to cytoplasmic inclusions. (A) Expression of wt Hoxd13, +7 Ala Hoxd13, +10 Ala Hoxd13 and +21 Hoxd13 in Cos-1 cells. Nuclear localization of the wt Hoxd13 protein. Expression of Hoxd13 with Ala expansions results in cytoplasmic aggregates. Nuclear and cytoplasmic localization in Hoxd13 with short, i.e. +7 and +10 Ala, expansions, but larger expansions result in predominantly cytoplasmic staining. (B) Electron microscopy of Cos-1 cells expressing wt (top) and +14 Ala (bottom) Hoxd13. Right side shows enlargement of boxed area. Labeled Hoxd13 protein is detected by gold particles. Wt Hoxd13 is predominantly found in chromatin condensations of the nucleus, whereas mutant Hoxd13 is found in cytoplasmic aggregates. The aggregates do not show a structural order. Hoxd13 protein is associated with the electron dense areas. Nucleus is marked with N and cytoplasm with C. (C) Distribution of cellular localization of wt and Hoxd13 with different Ala repeat length. With increasing length of the repeat more cells show a cytoplasmic localization of Hoxd13 (red bars) and less cells show a nuclear localization (blue bars). The results are given in percentage, the SD represents six experiments from three transfections. (D) Co-expression of wt and mutant (+14 Ala) Hoxd13 results in retention of the wt in the aggregates. (E) Co-expression of mutant (+14 Ala) Hoxd13 with wt Hoxd11 and Hoxd12. In spite of aggregate formation, both the transcription factors readily translocate into the nucleus. (F) Expression of Hoxa13, Runx2 and Sox3 with expanded Ala-repeats results in cytoplasmic inclusions similar to those observed for Hoxd13. Wt is shown on left. Bar, 10 µm.

 
We compared the localization of the protein between the different mutants by cell counting, performed by two independent investigators. In the majority of cells expressing the short expansions, i.e. +7 Ala and +8 Ala, Hoxd13 localized to the nucleus. The +9 variant showed nuclear localization in only 50%, and in cells expressing the longest expansions, i.e. +14 Ala and +21 Ala, the mutant protein was almost exclusively found in the cytoplasm (Fig. 1C). Thus, expansions of the Ala repeat in the N-terminal domain of Hoxd13 result in aggregation of the protein in cytoplasmic inclusions, a process that appears to prevent Hoxd13 from entering the nucleus. Cytoplasmic aggregation occurs beyond a threshold of repeat length [22 (+7) Ala] and increases with the repeat length.

Wt Hoxd13 co-localizes with mutant protein in aggregates
To study the interaction of mutant on wt Hoxd13, we co-expressed both wt and +14 Ala in Cos-1 cells. The results show co-localization of mutant and wt protein in the aggregates indicating that the mutation prevents the wt protein from entering the nucleus (Fig. 1D). This finding may partly explain the dominant character of the Ala expansion mutations. However, the observed interaction appears to be limited to wt Hoxd13 as other Hox genes that are normally co-expressed with Hoxd13 in the limbs, i.e. Hoxd11, Hoxd12 or Hoxa13 are not influenced and readily translocate into the nucleus (Fig. 1E). Other transcription factors such as Runx2 were also not affected. The Ala repeat appears not to be the critical region of interaction as the 2 Ala mutant is also retained. Thus, in vitro the result of the repeat expansion is compatible with a dominant negative effect but there is no influence on the cellular localization of other Hox-genes.

Aggregate formation in other poly-Ala expansion mutations
Poly-Ala expansion diseases are a growing group of disorders that are characterized by abnormal development. Given the similarity between the HOXD13 and the other poly-Ala repeat diseases, we asked the question whether these conditions may follow a similar pathology. We cloned wt and mutant of RUNX2 [+10 Ala mutant (6)], HOXA13 (+11 Ala, unpublished case with hand–foot–genital syndrome) and SOX3 [+11 Ala (10)] and expressed them in Cos-1 cells. We obtained the same results as for the Hoxd13 mutants. The Ala expansion mutant was completely retained in the cytoplasm, whereas the wt was located in the nucleus (Fig. 1F). Similar to Hoxd13, co-expression of the mutant together with the wt protein resulted in retention of the wt in the aggregate. This indicates that the above-described mechanisms may be a general one explaining not only Hoxd13 mutations but also other Ala expansion mutations.

Mutant Hoxd13 does not locate to the aggresome
Misfolded proteins have been shown to accumulate at a specific site in the cells around the microtubule-organizing center (MTOC), where they get degraded by the proteasome. One consequence of aggresome formation is the rearrangement of components of the intermediate filament cytoskeleton such as vimentin (17,18). Vimentin is a common component of the filament, normally forming a delicate array of peripheral fibers. Overexpression of GFP250, for example, results in rapid aggregate formation within the MTOC and in drastic collapse of vimentin intermediate filaments in a ‘cage-like’ structure around the aggregate (17,18). Formation of the Hoxd13 aggregate, however, does not result in the displacement of vimentin in contrast to the dense aggregates at the MTOC associated with overexpression of, for example, GFP-250 (Fig. 2A). Hoxd13 aggregates are loosely arranged protein clumps surrounding the nucleus (Fig. 2B). Co-staining with -tubulin showed that the bulk of Hoxd13 aggregates localized in the area of the MTOC, but we did not observe co-localization with -tubulin and vimentin, as observed with GFP-250. Furthermore, the Golgi apparatus which is in close proximity to the MTOC as well as the 20S proteasome subunit which is recruited to the MTOC did not co-localize with the Hoxd13 aggregates (Fig. 2C and D). These results indicate that Hoxd13 associated aggregates are different from those classically associated with aggresome formation.

View larger version (78K): [in this window] [in a new window]   Figure 2. Localization of aggregates, involvement of chaperones and degradation. (A) Expression of GFP-250 in Cos-1 cells via transient plasmid transfection results in aggresome formation and in collapse of vimentin filaments around the misfolded protein. In contrast, (B) mutant Hoxd13 (+14 Ala) expressing cells show normal distribution of vimentin filaments. (C) Co-staining of cells containing cytoplasmic aggregates with -tubulin shows no co-localization, but the major part of the Hoxd13 aggregate is located in the region of the MTOC. (D) Staining of the Golgi-apparatus and the 11S and 20S proteasome subunits show lack of co-localization. (E) Hoxd13 aggregates co-localize with Hsp40 and Hsp70. (F) Reduction of aggregate formation and re-localization of Hoxd13 in the nucleus after treatment with geldanamycin. (G) Retroviral infection of cells with virus (RCAS) expressing wt Hoxd13 and Hoxd13 with a +14 Ala expansion and detection of Hoxd13 protein by fluorescently labeled anti-HA antibody (top). Bottom panel shows merge with nuclear (DAPI) staining. Note nuclear localization of wt protein but only very small amounts of Hoxd13 protein with expanded Ala repeat. Appearance of cytoplasmic inclusions after treatment with the proteasome inhibitor ALLN. Bar, 10 µm.

 
Involvement of chaperones in aggregate formation
The folding of newly synthesized proteins to their proper conformations involves chaperones such as heat shock proteins (Hsp) which also play a role in the recognition of misfolded proteins. Moreover, Hsp70 and Hsp40 can catalyze the refolding of denatured or partially denatured molecules into active forms and are able to disassemble intracellular protein aggregates into soluble native species (19). We were able to demonstrate co-localization of Hsp40 and Hsp70 with the aggregated protein indicating that both proteins are recruited to the perinuclear aggregates (Fig. 2E). Treatment of cells with geldanamycin, a naturally occurring tumor drug that interacts with Hsp90, results in an up-regulation of Hsp40 and Hsp70 (20). We found that treatment of Hoxd13 (+14 Ala) transfected cells with geldanamycin results in a drastic reduction of aggregate formation and a consecutive increase in nuclear localization (Fig. 2F), indicating that Hsps, when activated early in the aggregation process, can slow down the perinuclear accumulation of mutant Hoxd13. This finding opens the possibility for therapeutic intervention through the manipulation of chaperone activity.

Mutant Hoxd13 is degraded via the proteasome
Aggregate formation is due to overproduction and/or inefficient degradation of the mutant protein. To test the effect of low levels of Hoxd13 protein expression, we used a retroviral infection system (RCAS) permitting incorporation of a single gene copy per cell and thus resulting in a more physiological dose of expression level (21). We infected chick fibroblasts with RCAS virus containing wt or mutant Hoxd13 (+14 Ala). Although the wt protein showed nuclear staining in the vast majority of cells, only very few cells stained for the Hoxd13 with expanded repeats suggesting efficient degradation of the mutant protein. Treatment of virus infected cells with the proteasome inhibitor ALLN (calpain inhibitor I) resulted in the appearance of mutant protein in cytoplasmic aggregates similar to those observed in Cos-1 cells (Fig. 2G). Similar results were obtained for two other more specific proteasome inhibitors, MG132 and lactacystin. Thus, at low production rates, the mutant protein can be efficiently degraded via the proteasome.

Mutant Hoxd13 aggregates in vitro
To investigate the effect of the poly-Ala expansion on the native Hoxd13 protein, we produced recombinant wt Hoxd13 and mutant protein with a +21 Ala expansion. Using a filter precipitation assay, we can show that the mutant but not the wt protein spontaneously forms aggregates (Fig. 3). This indicates that the poly-Ala expansion is necessary and sufficient for aggregation of the protein.

View larger version (85K): [in this window] [in a new window]   Figure 3. In vitro aggregation of mutant Hoxd13. Aggregation of recombinant affinity purified Hoxd13 protein detected by filter retardation assay and anti-Hoxd13 antibody. (A) Time scale of captured aggregates and detection on filter (B). Mutant Hoxd13 (+21 Ala) but not wt Hoxd13 forms aggregates beginning 48 h after incubation.

 
Mutant Hoxd13 is retained in the cytoplasm and degraded in vivo
To test whether the effects observed in vitro are also of relevance in vivo, we investigated the localization of wt and mutant Hoxd13 protein in wt and synpolydactyly homolog (spdh) mutant mice. Spdh is a recessive mouse mutant with a +7 Ala expansion in Hoxd13 identical to the expansions observed in patients with SPD (22). The mutant phenotype is similar to the human phenotype consisting of polydactyly, syndactyly and brachydactyly. The mice show a reduction in proliferation of cells in the autopod, a finding that may in part explain the shortening of the digits (23). We analyzed the distribution of Hoxd13 protein using a Hoxd13 antibody (24) in E13.5 wt limbs and found a distribution of Hoxd13 similar to that of Hoxd13 mRNA. Positive cells showed a predominantly nuclear staining. In contrast, the overall signal was much weaker in spdh/spdh mice, suggesting a significant reduction in the amount of Hoxd13 protein (Fig. 4B). Furthermore, distribution of the protein was abnormal with stronger staining in the cytoplasm than in the nucleus. In areas of high Hoxd13 expression, i.e. around the cartilaginous anlagen, some cells showed accumulated cytoplasmic staining (Fig. 4D). Western blot (WB) analysis of Hoxd13 protein in wt, spdh/+ and spdh/spdh limbs showed a reduction by 50% in the heterozygous and by 90% in the homozygous animals (Fig. 4E), thus confirming our immunohistochemical findings.

View larger version (11K): [in this window] [in a new window]   Figure 4. Distribution of wt and mutant Hoxd13 in vivo. Sections through wt (A, C) and spdh/spdh (B, D) limb buds at stage E13.5. Staining of Hoxd13 protein using a primary anti-Hoxd13 antibody and a secondary FITC labeled antibody is shown in at top panel. Nuclear staining in red (propidium iodine), merge is shown in bottom panel. (A, B) Distal limb tip with ectoderm (magnification 200x). (C, D) Mesenchymal cells surrounding the cartilage anlagen (magnification 630x). Bar, 10 µm. Wt sections show a predominantly nuclear staining with the strongest expression around the condensations of the digits. In contrast, spdh/spdh mutant mice show less staining and the majority of staining is present in the cytoplasm. Some cells show cytoplasmic accumulation of Hoxd13 protein. (E) Detection of Hoxd13 in E13.5 limb buds by WB shows moderate reduction of Hoxd13 protein in spdh/+ limbs and strong reduction in homozygous spdh/spdh limbs.

 
Poly-Ala tracts are associated with nuclear proteins and have a maximal length of 20
Poly-Ala tracts are not uncommon in the human genome. To find a possible association between Ala tracts, gene function and mechanisms of disease, we performed a genome-wide search based on Swiss-Prot and TrEMBL for poly-Ala tracts longer than 8 Ala. We identified 137 proteins of which seven could not be assigned to any functional category. Of the remaining 130 proteins, 102 (77%) were described to be located in the nucleus and 32 (23%) were cytoplasmic, membrane bound, or extracellular proteins. Of the nuclear proteins, 66 (50% of total) were transcription factors. The relative percentage of transcription factors increased with the length of the repeat. The longest repeat consisting of 20 Ala was found in PHOX2B, the transcription factor mutated in congenital central hypoventilation syndrome. The results indicate that Ala repeats are likely to have a function related to the nucleus and thus in the regulation of transcription.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Our results indicate a common pathogenetic mechanism for poly-Ala repeat expansions in transcription factors. Expansions of the repeat beyond a certain threshold appear to result in degradation of the protein and/or, dependent on the efficacy of degradation, the level of expression and the length of the Ala-repeat, cytoplasmic aggregation. First, the mutant protein is translocated into the nucleus, but prolonged exposure of the cell results in the formation of small aggregates that subsequently form large inclusions around the nucleus. These inclusions appear to capture mutant and wt protein, thus preventing it from reaching its destination, the nucleus. Similar to our in vitro results, we could show in a previous clinical study that the length of the Ala expansion correlates with the severity and penetrance of the SPD phenotype (14). Individuals with short repeats, i.e. +7 Ala, show considerable intra-familial and also intra-individual variability and low penetrance, whereas those with long expansions, i.e. +10 Ala and more, have a more severe phenotype with complete penetrance. This scenario is likely to apply also to other poly-Ala repeat diseases.

The strong association of long (>10) Ala repeats with nuclear proteins suggests a function related to the nucleus and thus regulation of transcription. In previous studies, Ala tracts have been shown to be necessary for transcription factor activity (25). The fact that no repeats longer than 20 Ala were found in our Swiss-Prot and Trembl search and that the length of the repeat is highly constant indicates a evolutionary constraint on the repeat length. This is in agreement with the clinical observation that the shortest repeat to give rise to SPD is 22 (+7) Ala long and that other transcription factor associated poly-Ala expansions are of similar size. In addition, our in vitro results demonstrate a similar threshold. It thus appears likely that an increase of Ala repeat length above 18–22 Ala results in misfolding and/or aggregation of the protein due to biophysical limitations. This hypothesis is supported by our in vitro finding that recombinant Hoxd13 with a +21 Ala expansion but not wt Hoxd13 spontaneously forms aggregates as detected by a filter retardation assay. It is thus conceivable that mutant proteins with expanded Ala tracts form aggregates in a dose and environment dependent manner. Once present, such aggregates would grow because more and more protein (wt as well as mutant) is drawn into the aggregate. Furthermore, the formation of protein clumps hinders degradation because all proteins have to be unfolded before they can be degraded.

The discrimination between native and non-native proteins involves a quality control that primarily depends on chaperones (19). This quality control can take place in the cytosol or in the endoplasmic reticulum (ER). Terminally misfolded proteins are recognized within the ER and ‘retrotranslocated’ across the ER membrane and ubiquitinated to be degraded by cytosolic proteasomes. If production exceeds degradation, distinct perinuclear inclusion bodies will form that contain the mutant protein. These inclusions are located at a specific site of the cell around the centrosome/the MTOC (17,18) as shown by co-localization of the aggresome with -tubulin (an established centrosome marker) and vimentin. At high levels of misfolded proteins, these structures expand and recruit the cytosolic pools of ubiquitin, the HSP70 and the 20S proteasome (26). Together these findings suggest that mammalian cells contain a specific organelle, which subsequently has been named ‘aggresome’, that specializes in the degradation of misfolded proteins (17). Our results indicate that Hoxd13 aggregates are different from those classically associated with aggresome formation as we do not find involvement of the vimentin intermediate filaments, no localization of the aggregates at the MTOC and no co-localization with the 20S proteasome subunit. A possible explanation is that Hoxd13, as other transcription factors, is translated in the cytoplasm by free ribosomes, and thus does not involve endoplasmic reticulum associated degradation (ERAD). However, similar structures have been observed in cells overexpressing GFP–GCP170, a GFP-tagged Golgi protein (18), indicating that this phenomenon is not restricted to proteins with poly-Ala expansions.

Aggregation of misfolded proteins has been described in conditions associated with repeat expansions of poly-Q encoding tracts (1). In contrast to the congenital conditions discussed here, they lead to progressive neuronal dysfunction and severe neurodegeneration typically beginning in mid-life. Likewise, expansions of a short Ala repeat from 8 Ala to 11–17 Ala in PABNP1, a gene coding for a poly-A binding molecule, result in the neurological phenotype OPMD, a condition characterized by late onset progressive muscular weakness. Similar to the poly-Q associated conditions, such as Huntington's disease, OPMD patients show characteristic nuclear inclusions composed of tubular filaments that are often arranged in palisades or tangles (13,27). In contrast, electron microscopy (Fig. 1B) of the inclusions associated with Hoxd13 overexpression show amorphous material consisting of dense and more electron lucent areas that does not appear to follow a higher structural order. The differences in onset, pathology and clinical presentation argue that the pathogenesis of PABNP1 mutations is similar to other neurological repeat expansion diseases and thus different from poly-Ala expansions in transcription factors that are predominantly associated with developmental defects.

In several poly-Ala associated conditions, other mutations such as deletions, non-sense mutations or frameshifts have been described suggesting that the phenotype can be caused by a loss of protein function. In these cases, Ala repeat expansions are likely to result in efficient degradation of the misfolded protein by the proteasome and thus a non-functional protein. In other conditions such as SPD (5), cleidocranial dysplasia (6) and hand–foot–genital syndrome (8), however, there appears to be a dominant negative effect of the mutations resulting in additional or more pronounced phenotypes. For example, inactivation of Hoxd13 in the mouse causes a rather mild phenotype which is distinct from SPD, whereas inactivation of Hoxd11, Hoxd12 and Hoxd13 together result in a SPD-like malformation (16). This finding and genetic complementation experiments in the mouse mutant spdh suggested a negative effect of mutant Hoxd13 over other Hox genes (15). Our in vitro results show that the repeat expansion is unlikely to have a specific effect on other Hox genes. Instead, the presence of small amounts of mutant Hoxd13 in the cytoplasm of spdh/spdh mice suggests inefficient degradation of the mutant protein and an overload of the cell's degradative capacity. This effect can be expected to be more pronounced in repeats that are longer than the +7 expansion in spdh. The pathology associated with excessive protein degradation is still controversial, but it has become clear that other components besides the mutant protein get sequestered and thus at least temporarily inactivated during the degradation process, resulting in compromised cellular function and/or cell death (28). This may ultimately compromise the folding of other proteins and possibly influence more general cellular functions, giving rise to phenotypes that are different and/or more severe than loss of function mutations. On the basis of model presented here, pathogenicity of the mutant protein is influenced by the length of the repeat expansion the level of expression, and the efficacy of folding, a process controlled by chaperones. The level of expression of chaperones can be influenced in the living organism by environmental factors such as stress induction or elevation of temperature. Such epigenetic effects are likely to have an effect on the strong variability in expressivity and in penetrance observed in families with SPD.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Plasmid construction
The mouse Hoxd13, Hoxd12 and Hoxd11 genes were amplified by RT–PCR and cloned into pTL1-HA and pTL10-Flag (21). The +7 Ala construct was amplified by PCR using DNA from the spdh mouse mutant as template. All other Ala constructs (+8, +9, +10, +14 and +21) were cloned by inserting an oligo consisting of the additional alanines into the NotI site inside the Ala-coding repeat of the mouse Hoxd13 gene. For the Hoxa13 constructs, exon 1 was PCR amplified from DNA of a patient with a heterozygous +11 Ala mutation, ligated to the RT–PCR amplified mouse exon 2 and cloned into pTL1-HA. The Ala repeat of human RUNX2 was PCR amplified from a patient harboring a +10 Ala expansion and inserted into the coding sequence of mouse Runx2 in pTL-HA. The human Sox3 gene was PCR amplified from a patient with a heterozygous +11 Ala mutation and cloned into pTL1-HA and pTL10-Flag. The mouse Hoxd13 wt and +14 Ala constructs were cloned into the RCAS-BPA-vector. Virus production was done as described previously (21).

Cell lines and cell transfection
Cos-1 cells were grown in Dulbecco's modified Eagle's medium high glucose supplemented with 5% fetal calf serum and containing penicillin (100 U/ml) and streptomycin (100 µg/ml). Transfection was performed using PolyFect (Qiagen) according to manufacturer's instruction. After 16 h, the medium was removed and fresh medium was added. The cells were washed with PBS and fixed with 4% PFA 40 h after transfection. Geldanamycin (InvivoGen) was used as described previously (20).

DF-1 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 2% fetal chicken serum and containing penicillin (100 U/ml) and streptomycin (100 µg/ml). Cells were infected with RCAS virus containing Hoxd13 wt and Hoxd13 +14 Ala. For inhibition of the proteasome activity after infection, cell cultures were adjusted to 25 µM Calpain Inhibitor I (ALLN, Sigma) and incubated for an additional 24 h. Control cells were incubated with an equivalent amount of DMSO.

Immunofluorescence and microscopy
Cos-1 cells grown on coverslips were fixed 40 h after transfection with 4% paraformaldehyde for 10 min at room temperature. Cells were permeabilized with 0.2% Triton X-100 in PBS and incubated with primary antibody diluted in PBS, 10% FCS and 0.05% NaN₃ for 1 h at room temperature. Secondary antibodies were diluted in PBS, 10% FCS and 0.05% NaN₃ and incubated on coverslips for 1 h at room temperature. The cells were washed and nuclei were counterstained with DAPI. The coverslips were mounted on slides in Fluoromount-G (Electron Microscopy Sciences). The samples were examined with a fluorescence microscope Axiovert-200 (Zeiss). For the -tubulin-antibody, the cells were fixed and permeabilized with ice-cold methanol for 10 min at –20°C.

For electron microscopy, COS-1 cells were fixed and embedded in LR White (London Resin Company) 40 h after transfection. Post-embedded immunogold labeling was performed as described using the rabbit-anti-HA antibody (1:80) followed by secondary antibody conjugated with 10 nm gold (1 : 100; British Bio Cell). The samples were viewed in a Philips CM 100 electron microscope.

For immunofluorescence (IF) on mouse tissues, E13.5 limb buds were paraformaldehyde fixed, paraffin embedded and serial sections of 6 µm were cut. Two different primary anti-Hoxd13 antibodies (24) were applied to the dewaxed sections, washed and the signal detected by AlexaFluor 488 goat-anti-rabbit IgG.

Antibodies
The following antibodies were used for IF and/or WB analyses: mouse anti-HA (Sigma), rabbit anti-HA (Sigma), mouse anti-Flag (Sigma) and rabbit anti-Flag (Sigma) were diluted 1 : 250 (IF), mouse anti-vimentin (Sigma) was diluted 1 : 400 (IF), mouse anti- Tubulin (Sigma) was diluted 1 : 100 (IF), goat-anti-Hsp40 and goat-anti-Hsp70 (Santa Cruz Biotechnology) were diluted 1 : 50 (IF), anti-Golgi (anti--adaptin) was diluted 1 : 300, anti-11S (Affiniti) was diluted 1 : 250, anti-proteasome 20S -Type 1 subunit (Calbiochem) was diluted 1 : 60 (IF), rabbit anti-Hoxd13 (a generous gift from Atsushi Kuroiwa) was diluted 1:8000 (WB), rabbit anti-actin (Sigma) was diluted 1:8000 (WB), peroxidase goat-anti-rabbit IgG (Oncogene) was diluted 1 : 2000 (WB), AlexaFluor 546 goat-anti-rabbit IgG, AlexaFluor 546 goat-anti-mouse IgG, AlexaFluor 488 goat-anti-rabbit IgG, AlexaFluor 488 goat-anti-mouse IgG, AlexaFluor 488 donkey-anti-goat IgG and AlexaFluor 546 donkey-anti-rabbit IgG (Molecular Probes) were diluted 1 : 500 (IF).

Western blot
The Hoxd13-expressing limb-tissue of three embryos of embryonic stage E13.5 were pooled in 4 mM HEPES–NaOH pH 7.4 and 320 mM sucrose. Proteinase-inhibitor (Complete Mini, Roche) and ß-mercaptoethanol (10 mM) were added and the limbs lysed on ice in 1.5 ml tubes using a rotary homogenizer. The protein concentration was estimated using the BioRad microassay procedure (Bradford method) with BSA as standard. Ten micrograms of total protein per lane was loaded on a 12.5% SDS–PAGE and run for 2–3 h at 40 mA. SDS–PAGE-separated samples were transferred to Immobilon PVDF membrane and analyzed by immunoblotting with an anti-Hoxd13 antibody. To remove primary and secondary antibodies, the membranes were washed twice with PBS, incubated for 2x10 min in 150 mM glycine pH 2.5, 0.4% SDS, neutralized with 1 M Tris–HCl pH 6.8 and washed three times with PBS and reprobed with an anti-actin-antibody. Chemiluminescent reaction was performed to detect the signals.

Filter retardation assay
Wt Hoxd13 and mutated Hoxd13 (+21 Ala) were expressed as maltose binding protein (MBP) fusion proteins in E. coli and affinity purified on amylose resin according to manufacturer's instructions (NEB, Frankfurt am Main, Germany). For in vitro aggregation experiments 7.2 µM of MBP-Hoxwt and MBP-Hox +21 Ala were incubated with FaktorXa at 37°C. At indicated times, reactions were stopped by the addition of an equal volume of 4% SDS/100 mM DTT and boiling for 5 min. Aliquots corresponding to 11 µg of MBP fusion protein were diluted into 0.1 ml 0.1% SDS and analyzed by filter retardation assay as described previously (29). Captured aggregates were detected by incubation with anti-Hoxd13 serum (1 : 1500), followed by incubation with alkaline phosphatase conjugated anti-rabbit secondary antibody and the fluorescent substrate attophos.

Mice
Spdh mice were obtained from the Jackson Lab, Bar Harbor, ME, USA. Spdh is a spontaneous mutation that arose in a B6C3 mouse and was subsequently maintained on this background (22). Analysis of the phenotype was performed on the same background. Mutant offspring was generated by breeding spdh/+ with spdh/+ mice. For analysis of mutant embryos, timed matings were produced and noon of the day the vaginal plug was observed was counted as day 0.5 of gestation. DNA from liver or tail tips served for genotyping. Genotyping of mice was performed as described (22).

	ACKNOWLEDGEMENTS

 
This work was supported by a grant from the Deutsche Forschungsgemeinschaft to S.M. We thank A. Kuroiwa for generously providing us with Hoxd13 antibodies and E. Sztul for providing us with a GFP-250 clone for cell expression.

	FOOTNOTES

 
^* To whom correspondence should be addressed at: Max-Planck-Institute for Molecular Genetics, FG Development and Disease, Ihnestrasse 73, 14195 Berlin, Germany. Tel: +49 84131263; Fax: +49 84131385; Email: mundlos{at}molgen.mpg.de

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Cummings, C.J. and Zoghbi, H.Y. (2000) Fourteen and counting: unraveling trinucleotide repeat diseases. Hum. Mol. Genet., 6, 909–916.

2. Devys, D., Yvert, G., Lunkes, A., Trottier, Y. and Mandel, J.L. (2001) Pathological mechanisms in polyglutamine expansion diseases. Adv. Exp. Med. Biol., 487, 199–210.[ISI][Medline]

3. Karlin, S., Chen, C., Gentles, A.J. and Cleary, M. (2000) Associations between human disease genes and overlapping gene groups and multiple amino acid runs. Proc. Natl Acad. Sci. USA, 26, 17008–17013.

4. Brown, L.Y. and Brown, S.A. (2004) Alanine tracts: the expanding story of human illness and trinucleotide repeats. Trends Genet., 1, 51–58.

5. Muragaki, Y., Mundlos, S., Upton, J. and Olsen, B.R. (1996) Altered growth and branching patterns in synpolydactyly caused by mutations in HOXD13. Science, 272, 548–551.[Abstract]

6. Mundlos, S., Otto, F., Mundlos, C., Mulliken, J.B., Aylsworth, A.S., Albright, S., Lindhout, D., Cole, W.G., Henn, W., Knoll, J.H. et al. (1997) Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell, 89, 773–779.[CrossRef][ISI][Medline]

7. Brown, L.Y, Odent, S., David, V., Blayau, M., Dubourg, C., Apacik, C., Delgado, M.A., Hall, B.D., Reynolds, J.F., Sommer, A. et al. (2001) Holoprosencephaly due to mutations in ZIC2: alanine tract expansion mutations may be caused by parental somatic recombination. Hum. Mol. Genet., 8, 791–796.

8. Utsch, B., Becker, K., Brock, D., Lentze, M.J., Bidlingmaier, F. and Ludwig, M. (2002) A novel stable polyalanine [poly(A)] expansion in the HOXA13 gene associated with hand–foot–genital syndrome: proper function of poly(A)-harbouring transcription factors depends on a critical repeat length? Hum. Genet., 110, 488–494.[CrossRef][ISI][Medline]

9. De Baere, E., Beysen, D., Oley, C., Lorenz, B., Cocquet, J., De Sutter, P., Devriendt, K., Dixon, M., Fellous, M., Fryns, J.P. et al. (2003) FOXL2 and BPES: mutational hotspots, phenotypic variability, and revision of the genotype–phenotype correlation. Am. J. Hum. Genet., 2, 478–487.

10. Laumonnier, F., Ronce, N., Hamel, B.C., Thomas, P., Lespinasse, J., Raynaud, M., Paringaux, C., Van Bokhoven, H., Kalscheuer, V., Fryns, J.P. et al. (2002) Transcription factor SOX3 is involved in X-linked mental retardation with growth hormone deficiency. Am. J. Hum. Genet., 71, 1450–455.[CrossRef][ISI][Medline]

11. Stromme, P., Mangelsdorf, M.E., Shaw, M.A., Lower, K.M., Lewis, S.M., Bruyere, H., Lutcherath, V., Gedeon, A.K., Wallace, R.H., Scheffer, I.E. et al. (2002) Mutations in the human ortholog of Aristaless cause X-linked mental retardation and epilepsy. Nat. Genet., 4, 441–445.

12. Amiel, J., Laudier, B., Attie-Bitach, T., Trang, H., de Pontual, L., Gener, B., Trochet, D., Etchevers, H., Ray, P., Simonneau, M. et al. (2003) Polyalanine expansion and frameshift mutations of the paired-like homeobox gene PHOX2B in congenital central hypoventilation syndrome. Nat. Genet., 4, 459–461.

13. Brais, B., Bouchard, J.P, Xie, Y.G., Rochefort, D.L., Chretien, N., Tome, F.M., Lafreniere, R.G, Rommens, J.M., Uyama, E. and Nohira, O. (1998) Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy. Nat. Genet., 2, 164–167.

14. Goodman, F.R., Mundlos, S., Muragaki, Y., Donnai, D., Giovannucci-Uzielli, M.L., Lapi, E., Majewski, F., McGaughran, J., McKeown, C., Reardon, W.P. et al. (1997) Synpolydactyly phenotypes correlate with size of expansions in HOXD13 polyalanine tract. Proc. Natl Acad. Sci. USA, 94, 7458–7463.[Abstract/Free Full Text]

15. Bruneau, S., Johnson, K.R., Yamamoto, M., Kuroiwa, A. and Duboule, D. (2001) The mouse Hoxd13(spdh) mutation, a polyalanine expansion similar to human type II synpolydactyly (SPD), disrupts the function but not the expression of other Hoxd genes. Dev. Biol., 237, 345–353.[CrossRef][ISI][Medline]

16. Zakany, J. and Duboule, D. (1996) Synpolydactyly in mice with a targeted deficiency in the HoxD complex. Nature, 384, 69–71.[CrossRef][Medline]

17. Johnston, J.A., Ward, C.L. and Kopito, R.R. (1998) Aggresomes: a cellular response to misfolded proteins. J. Cell Biol., 143, 1883–1898.[Abstract/Free Full Text]

18. Garcia-Mata, R., Bebok, Z., Sorscher, E.J. and Sztul, E.S. (1999) Characterization and dynamics of aggresome formation by a cytosolic GFP-chimera. J. Cell Biol., 146, 1239–1254.[Abstract/Free Full Text]

19. Hartl, F.U. and Hayer-Hartl, M. (2002) Molecular chaperones in the cytosol: from nascent chain to folded protein. Science, 295, 1852–1858.[Abstract/Free Full Text]

20. Sittler, A., Lurz, R., Lueder, G., Priller, J., Lehrach, H., Hayer-Hartl, M.K., Hartl, F.U. and Wanker, E.E. (2001) Geldanamycin activates a heat shock response and inhibits huntingtin aggregation in a cell culture model of Huntington's disease. Hum. Mol. Genet., 10, 1307–1315.[Abstract/Free Full Text]

21. Morgan, B.A. and Fekete, D.M. (1996) Manipulating gene expression with replication-competent retroviruses. Methods Cell Biol., 51, 185–218.[ISI][Medline]

22. Johnson, K.R., Sweet, H.O., Donahue, L.R., Ward-Bailey, P., Bronson, R.T. and Davisson, M.T. (1998) A new spontaneous mouse mutation of Hoxd13 with a polyalanine expansion and phenotype similar to human synpolydactyly. Hum. Mol. Genet., 7, 1033–1038.[Abstract/Free Full Text]

23. Albrecht, A.N., Schwabe, G.C., Stricker, S., Boddrich, A., Wanker, E.E. and Mundlos, S. (2002) The synpolydactyly homolog (spdh) mutation in the mouse—a defect in patterning and growth of limb cartilage elements. Mech. Dev., 112, 53–67.[CrossRef][ISI][Medline]

24. Suzuki, M. and Kuroiwa, A. (2002) Transition of Hox expression during limb cartilage development. Mech. Dev., 118, 241–245.[CrossRef][ISI][Medline]

25. Han, K. and Manley, J.L. (1993) Functional domains of the Drosophila engrailed protein. EMBO J., 12, 2723–2733.[ISI][Medline]

26. Wigley, W.C, Fabunmi, R.P, Lee, M.G., Marino, C.R., Muallem, S., DeMartino, G.N. and Thomas, P.J. (1999) Dynamic association of proteasomal machinery with the centrosome. J. Cell Biol., 3, 481–490.

27. Calado, A., Tome, F.M., Brais, B., Rouleau, G.A., Kuhn, U., Wahle, E. and Carmo-Fonseca, M. (2000) Nuclear inclusions in oculopharyngeal muscular dystrophy consist of poly(A) binding protein 2 aggregates which sequester poly(A) RNA. Hum. Mol. Genet., 15, 2321–2328.

28. Selkoe, D.J. (2003) Folding proteins in fatal ways. Nature, 426, 900–904.[CrossRef][Medline]

29. Wanker, E.E., Scherzinger, E., Heiser, V., Sittler, A., Eickhoff, H. and Lehrach, H. (1999) Membrane filter assay for detection of amyloid-like polyglutamine-containing protein aggregates. Methods Enzymol., 309, 375–386.[ISI][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				L. Moumne, A. Dipietromaria, F. Batista, A. Kocer, M. Fellous, E. Pailhoux, and R. A. Veitia Differential aggregation and functional impairment induced by polyalanine expansions in FOXL2, a transcription factor involved in cranio-facial and ovarian development Hum. Mol. Genet., April 1, 2008; 17(7): 1010 - 1019. [Abstract] [Full Text] [PDF]

				Y. Oma, Y. Kino, K. Toriumi, N. Sasagawa, and S. Ishiura Interactions between homopolymeric amino acids (HPAAs) Protein Sci., October 1, 2007; 16(10): 2195 - 2204. [Abstract] [Full Text] [PDF]

				D. Kelberman and M. T. Dattani Hypothalamic and pituitary development: novel insights into the aetiology Eur. J. Endocrinol., August 1, 2007; 157(suppl_1): S3 - S14. [Abstract] [Full Text] [PDF]

				K. Anan, N. Yoshida, Y. Kataoka, M. Sato, H. Ichise, M. Nasu, and S. Ueda Morphological Change Caused by Loss of the Taxon-Specific Polyalanine Tract in Hoxd-13 Mol. Biol. Evol., January 1, 2007; 24(1): 281 - 287. [Abstract] [Full Text] [PDF]

				E. M. Berry-Kravis, L. Zhou, C. M. Rand, and D. E. Weese-Mayer Congenital Central Hypoventilation Syndrome: PHOX2B Mutations and Phenotype Am. J. Respir. Crit. Care Med., November 15, 2006; 174(10): 1139 - 1144. [Abstract] [Full Text] [PDF]

				Z. Berger, J. E. Davies, S. Luo, M. Y. Pasco, I. Majoul, C. J. O'Kane, and D. C. Rubinsztein Deleterious and protective properties of an aggregate-prone protein with a polyalanine expansion Hum. Mol. Genet., February 1, 2006; 15(3): 453 - 465. [Abstract] [Full Text] [PDF]

				J. E. Davies, S. Sarkar, and D. C. Rubinsztein Trehalose reduces aggregate formation and delays pathology in a transgenic mouse model of oculopharyngeal muscular dystrophy Hum. Mol. Genet., January 1, 2006; 15(1): 23 - 31. [Abstract] [Full Text] [PDF]

				M. Garcia-Barcelo, M.-t. So, D. K.-c. Lau, T. Y.-y. Leon, Z.-w. Yuan, W.-s. Cai, V. C.-h. Lui, M. Fu, J.-A. Herbrick, E. Gutter, et al. Population Differences in the Polyalanine Domain and 6 New Mutations in HLXB9 in Patients with Currarino Syndrome Clin. Chem., January 1, 2006; 52(1): 46 - 52. [Abstract] [Full Text] [PDF]

				L. Moumne, M. Fellous, and R. A. Veitia Deletions in the polyAlanine-containing transcription factor FOXL2 lead to intranuclear aggregation Hum. Mol. Genet., December 1, 2005; 14(23): 3557 - 3564. [Abstract] [Full Text] [PDF]

				D. Trochet, S. J. Hong, J. K. Lim, J.-F. Brunet, A. Munnich, K.-S. Kim, S. Lyonnet, C. Goridis, and J. Amiel Molecular consequences of PHOX2B missense, frameshift and alanine expansion mutations leading to autonomic dysfunction Hum. Mol. Genet., December 1, 2005; 14(23): 3697 - 3708. [Abstract] [Full Text] [PDF]

				T. Bachetti, I. Matera, S. Borghini, M. D. Duca, R. Ravazzolo, and I. Ceccherini Distinct pathogenetic mechanisms for PHOX2B associated polyalanine expansions and frameshift mutations in congenital central hypoventilation syndrome Hum. Mol. Genet., July 1, 2005; 14(13): 1815 - 1824. [Abstract] [Full Text] [PDF]

				J. W. Innis, D. Mortlock, Z. Chen, M. Ludwig, M. E. Williams, T. M. Williams, C. D. Doyle, Z. Shao, M. Glynn, D. Mikulic, et al. Polyalanine expansion in HOXA13: three new affected families and the molecular consequences in a mouse model Hum. Mol. Genet., November 15, 2004; 13(22): 2841 - 2851. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 13/20/2351    most recent ddh277v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (43) Request Permissions Google Scholar Articles by Albrecht, A. N.