Grey-lethal mutation induces severe malignant autosomal recessive osteopetrosis in mouse and human

The spontaneous mouse grey-lethal (gl) mutation is responsible for a coat color defect and for the development of the most severe autosomal recessive form of osteopetrosis. Using a positional cloning approach, we have mapped and isolated the gl locus from a 1.5 cM genetic interval. The gl locus was identified in a bacterial artificial chromosome (BAC) contig by functional genetic complementation in transgenic mice. Genomic sequence analysis revealed that the gl mutation is a deletion resulting in complete loss of function. The unique 3 kb wild-type transcript is expressed primarily in osteoclasts and melanocytes as well as in brain, kidney, thymus and spleen. The gl gene is predicted to encode a 338−amino acid type I transmembrane protein that localizes to the intracellular compartment. Mutation in the human GL gene leads to severe recessive osteopetrosis. Our studies show that mouse Gl protein function is absolutely required for osteoclast and melanocyte maturation and function.

Because osteopetrosis is relatively rare in humans, our understanding of osteopetrosis and osteoclast defects is based mainly on animal models, particularly the mouse. Several spontaneous, induced or targeted mouse mutants have provided most of our present knowledge of the differentiation and maturation of osteoclasts. As the osteoclast is of hematopoietic origin, genes implicated early in multiple lineage commitment, such as Sfpi1 and Nfkb, have an indirect but major role in osteoclastogenesis^{11, 12}. In the granulocyte-macrophage lineage, the absence of colony-stimulating factor-1 (resulting from the op mutation), its receptor c-fms, c-fos or RANK (receptor-activator of NF-B) affects differentiation, proliferation and survival of osteoclast progenitors^{13, 14, 15, 16, 17}. Sporadic amelioration of osteopetrosis with age has been observed for the mouse op mutant¹⁸.

We previously used positional cloning to isolate the gl gene. We first defined nine co-segregating polymorphic microsatellite markers that were associated with transmission of the gl mutation, and established the genetic interval for the gl locus as 1.5 cM²⁸. Subsequently, we generated a yeast artificial chromosome (YAC) map that covered our candidate region²⁹. In this study, we used functional rescue by BAC transgenesis to identify the gl gene and characterize its mutation, tissue expression and cellular localization in vivo and ex vivo. Our studies provide evidence that Gl is a cytoplasmic protein, with a putative single transmembrane domain, that has a role in osteoclast functional activity and melanocyte pheomelanin synthesis and localization. On the basis of the highly conserved homologous human GL gene that we isolated, we hypothesized that mutations affecting the GL gene may be responsible for the development of recessive osteopetrosis. Here, we characterize the first mutation in the human GL gene that leads to severe recessive osteopetrosis.

On the basis of our previous genetic localization of the gl locus in a 1.5 cM interval on mouse chromosome 10 (ref. 28), we generated a YAC contig covering 8.5 Mb, in which 15 mouse genes were precisely mapped²⁹. To further refine the gl region, we established a BAC contig using as entry points the markers D10Mit184 and Cd24a, previously identified from our YAC physical map (Fig. 1a). After several rounds of chromosome walking, 18 overlapping genomic clones were assembled. New polymorphic sequence-tagged sites were characterized concomitantly, reducing our non-recombinant interval to 500 kb. The boundaries of this interval, corresponding to the gl candidate region, were thus defined as the T7 end insert sequence of BAC 371K6 and D10Mit184 (Fig. 1a).

Figure 1. Isolation of the mouse gl gene. a, Physical map of the gl candidate region. A 500 kb BAC contig spanning the gl non-recombinant interval included 18 individual clones with the boundary markers D10Mit184 and BAC 371K6 T7 end. Database and sequence analysis identified 4 known genes (in bold): Lace1, Snx3, Nr2e1 and Sec63. Arrows indicate the transcriptional orientation of each. , T7 BAC ends; , Sp6 BAC ends; |, NotI restriction sites; *, 3 BACs chosen for transgenesis. b−g, Functional rescue of osteopetrosis in BAC-transgenic gl mice. Bone sections from 3-week-old mice were stained with H&E. Wild-type (b and c) and gl/gl mice transgenic for BAC 373N3 (f and g) display identical histologically normal phenotypes. Non-transgenic gl/gl mice are characterized by osteopetrosis (d and e). BM, bone marrow; CB, cortical bone; GP, growth plate. Scale bar, 250 m. Magnification 5. Full Figure and legend (81K)

To identify potential candidate genes, we used an in vivo test of biological activity involving BAC transgenesis to functionally rescue the gl/gl phenotypes. We first ensured that C57BL/6J and C3H/HeJ genetic backgrounds were both appropriate to distinguish the microinjected BAC genotypically from the endogenous genomic region, using internal polymorphic markers. Three overlapping BACs (498E23, 373N3 and 343H5) covering 75% of our candidate region (Fig. 1a) were injected. Founder animals were obtained with BAC clones 373N3 and 343H5. Molecular analysis of the transgenic mice showed the expected vector-insert junction, the internal polymorphic marker and normal hybridization pattern with an internal probe, indicating integrity of the transgene. The transgenic founders did not develop any obvious abnormal phenotype and transgene transmission followed mendelian distribution. The transgenic BAC mice were then successively mated to heterozygous gl/+ mice to verify rescue of the gl/gl phenotypes. Offspring from these matings were analyzed using polymorphic markers not included in the injected BAC to determine the gl genotype and to test the BAC vector-insert junction for the presence of the transgene. Whereas non-transgenic homozygous gl/gl littermates exhibited the gl/gl phenotypes, all 17 transgenic gl/gl animals carrying the BAC 373N3 transgene (6 copies) displayed normal growth, an agouti coat color, tooth eruption and appropriate bone marrow development (Fig. 1b−g). In addition, these transgenic gl/gl mice had a lifespan greater than two years and had no apparent age-related abnormalities. In contrast, no functional rescue was obtained with the three 343H5 BAC transgenic lines. Together, these results indicate that the gl mutation is linked to a gene contained in the BAC 373N3.

Figure 2. Molecular characterization of the gl gene and mutation. a, Genomic structure of the wild-type (wt) mouse gl gene. , coding exons; , non-coding sequences; , 460 bp L1 sequence. The restriction sites used for Southern blot analysis are indicated as H (HindIII) and E (EcoRI). Exon and intron sizes are shown, as well as the structure of the rearranged transgene. b, Southern blot analysis of wild-type and gl genomic DNA. The fragment used as probe (thick horizontal line) is indicated in (a). The wild-type hybridization pattern consists of a 5.9 kb HindIII and a 14 kb EcoRI fragment. The deleted gl allele shows shorter fragments of 2.4 kb (HindIII) and 6.7 kb (EcoRI). Because the gl background derives from 129Sv and C57BL/6, genomic DNA from both were used as controls. c, PCR assay design to distinguish mutated and wt alleles. Specific primer pairs F1/R1 and F2/R2 (indicated as half-arrows in a) were used for PCR amplification. The PCR products (wild-type, 236 bp; gl, 330 bp) were resolved on a 1% agarose gel and revealed by ethidium bromide staining. Full Figure and legend (74K)

To verify that the candidate gene did, in fact, correspond to the gl mutation, we characterized the genomic structure of the locus in homozygous gl/gl mice. In contrast to its wild-type counterpart, genomic DNA from gl/gl mice had a 7.5 kb deletion as assessed by PCR and Southern analysis (Fig. 2a and b). This deletion included the promoter, extending 4.5 kb upstream from the initiation codon, the entire first exon and a large portion (3 kb) of the first intron (Fig. 2a and b). Detailed characterization of the deletion breakpoints showed that in the gl mice only, the gene had a 460 bp insertion. This short sequence was homologous to the 3' untranslated region (UTR) of a LINE1 (long interspersed nuclear element-1) retrotransposon (Fig. 2a). Based on this genomic structure, we designed a specific PCR assay to distinguish wild-type and gl allelic forms at this locus (Fig. 2c).

Northern blot analysis of the expression of the gl candidate gene detected a single 3 kb mRNA transcript, suggesting no alternative splicing or multiple transcripts. Expression of gl was most prevalent in brain, kidney, spleen and, to a lesser extent, thymus, testis, heart, liver and primary osteoblasts (Fig. 3a). Osteoclasts and melanocytes, the two cell lineages known to be affected in the gl mouse mutant, were examined for gl expression by northern blot and RT-PCR. High expression of the gl gene was detected in wild-type osteoclast-like cells (OCLs) differentiated in co-culture, whereas no expression was observed in OCLs derived from gl/gl mice (Fig. 3a). In addition, the wild-type immortalized melanocyte (melan-a) cell line expressed gl, consistent with the gl/gl phenotype (Fig. 3a). To confirm osteoclast gl expression in vivo, we used in situ hybridization to show that gl was expressed in wild-type resident multinucleated osteoclasts (Fig. 3b), whereas no expression was detected in any of the homozygous gl/gl tissues examined. Together, these expression studies in osteoclasts and melanocytes strongly support the validity of this candidate as the gl gene.

Figure 3. gl expression pattern. a, Northern analysis of total RNA using the gl cDNA probe detected variable expression of a unique 3 kb transcript in multiple mouse adult tissues. T, testis; Th, thymus; H, heart; K, kidney; S, spleen; B, brain; L, liver; OBs, primary osteoblasts. The transcript was undetectable in gl/gl samples. Expression of gl was detected by northern analysis in OCLs and primary osteoblasts and by RT-PCR in the melan-a melanocyte cell line. b, In situ hybridization analysis of adult wild-type bone sections with an antisense (AS, left) probe showed gl expression in multinucleated osteoclasts. No signal was detected with the sense (S, right) probe. c, gl expression during development detected by in situ hybridization. Inverted images showing real-size sagittal sections from embryonic stage E12.5 to postnatal stage P10, after film exposure. Specific expression appears bright on a dark field. Tissues displaying high and specific expression are spinal cord (sc), fetal liver (fl), brain (b), placenta (p), vertebra (v), gut (g), thymus (th), submandibular gland (su), mandible (mb), hippocampus (hi), cerebellum (cb), kidney (k), spleen (sp), skin (sk), and dermis (d). d, Northern analysis detected high expression of the 3 kb gl transcript in gl/gl BAC 373N3−transgenic mice, compared with non-transgenic homozygous gl/gl mice. In situ hybridization with an AS probe (left) detected gl expression in multinucleated osteoclasts from transgenic mice. No signal was detected with the S probe (right). TG, transgenic. b and d, Scale bars, 25 m; magnification 100. Full Figure and legend (107K)

Figure 4. Expression and immunolocalization of the Gl protein. a, Kyte-Doolittle hydropathy plot of Gl protein sequence. Gl1 and Gl2 denote position of the peptides used to generate polyclonal antibodies against the Gl protein. b, Western blot analysis using Gl1 antibody on OCL protein extracts from wild-type and gl/gl cultures. Specific 34- and 38-kD proteins are present in wild-type cell extracts but are undetectable in gl/gl cell extracts. Polyclonal antibody against the 31-kD subunit of the V-ATPase was used as a positive control. c−f, Osteoclasts from wild-type animals of a gl/+ intercross were processed for immunofluorescence microscopy using Gl1 rabbit antibodies. c, overall osteoclast cell shape by phase contrast microscopy; d, multinucleated osteoclast nuclei with Hoechst stain; e, cytosolic localization of the Gl protein using Gl1 polyclonal antibody; f, merge of d and e. Approximately 10 osteoclasts were assessed, all showing identical protein sublocalization. Magnification 30. Full Figure and legend (47K)

To study the Gl protein, we raised two Gl-specific polyclonal antibodies (called Gl1 and Gl2) that correspond to two different epitopes on either side of the putative transmembrane domain of the Gl protein (Fig. 4a). Western blotting with Gl1 and Gl2 (data not shown) detected two specific bands in wild-type osteoclasts, corresponding to proteins of 34 kD and 38 kD. In contrast, no Gl protein was detected in extracts from gl/gl cells (Fig. 4b). As an internal control, a polyclonal antibody against the 31-kD subunit of the V-ATPase proton pump³¹ was used to confirm the integrity of the extracts and the pre-immune serum for antisera specificity (data not shown).

Figure 5. Alignment of the predicted amino acid sequences of mouse and human Gl proteins (using ClustalW). Shaded residues are conserved between the 2 proteins. Arrowhead indicates a potential cleavage site for the signal peptide. The C-terminal transmembrane domain is underlined. Full Figure and legend (87K)

Genome sequence analysis of 19 autosomal recessive osteopetrotic patients was carried out on the entire coding exons (including flanking intronic regions), the 5' UTR and the polyadenylation site of the GL gene. One mutation, found in a patient of Italian origin, consisted of a GA transition at position +5 of the donor splice site of intron V (IVS5+5 GA; Fig. 6a). The patient was homozygous for this mutation, which correlates with the recessive nature of the disease. This mutation was not found in 100 control chromosomes that were tested using the ApoI (R/AATTY) restriction site generated by this mutation (wild-type allele: gTAAGTTt; mutated allele: gTAAATTt). Conversely, sequence analysis of DNA from the asymptomatic parents showed heterozygosity for this mutation (Fig. 6a).

Figure 6. DNA and RNA analysis of a patient with infantile malignant osteopetrosis. a, Pedigree of the affected patient harboring a mutation in the GL gene. Chromatograms show a homozygous GA mutation at position +5 of the donor splice site in intron V of the GL gene for a patient with recessive infantile malignant osteopetrosis () as compared with an unrelated wild-type control (WT). The mutation co-segregates with the disease in the family, as both parents are heterozygous (diagonally shaded square, father; diagonally shaded circle, mother). Ex, exon. b, RT-PCR amplification across exon 5 of the GL gene. Identity and sequence of PCR products are indicated. Underlined sequences correspond to the primers used in RT-PCR. M, markers (1-kb ladder); C, unrelated wild-type control; OP1, osteopetrotic patient 1. Sizes and corresponding structure of the PCR products are marked and . Boxes delineate exon junctions. Exonic sequences are in upper-case letters; intronic sequences are in lower-case letters. c, Alignment of the wild-type Gl amino acid sequence with the 2 aberrant hypothetical proteins. Frameshifts and nonsense sequences with the new stop codons are shown in red. Both sequences lack the region corresponding to the putative transmembrane domain (underlined residues). *, shared amino acid residues. Full Figure and legend (58K)

To study the effect of the mutation on transcription of the GL gene, an RT-PCR assay was designed to span exon 5. RT-PCR on total RNA isolated from cultured fibroblasts that usually express GL did not detect the 469-bp product in the osteopetrotic patient compared with wild-type control. In fact, three new products of smaller size were present, which may most likely correspond to aberrant splicing transcripts (Fig. 6b). Direct sequencing revealed that all three forms lacked exon 5. The sequence of the major 355-bp form indicated the presence of an additional 52-bp fragment from intron IV that resulted from use of cryptic acceptor (cag) and donor (gtatgt) splice sites (localized -66 bp and -113 bp upstream of exon 5, respectively). The minor 303-bp form corresponds to a fusion between exon 4 and exon 6 and the skipping of exon 5. Finally, the third amplification product represented a heteroduplex between the 355-bp and 303-bp forms (Fig. 6b). Thus, the IVS5+5 GA mutation leads to exon skipping and produces aberrant transcripts and hypothetical aberrant proteins (Fig. 6c).

For our positional cloning of the gl locus, a BAC physical map of the gl candidate region was established and the gl genomic interval was reduced to 500 kb. Functional identification of the gl gene was then undertaken by using BAC transgenesis for phenotypic rescue. Because the gl grey coat color is visible only on an agouti background (Aa or AA), the transgenic mice were crossed successively to heterozygous gl/+ AA animals to ensure phenotypic coat color scoring. All F₂ genotypically homozygous gl/gl transgenic animals produced from a 180 kb BAC-transgenic line had a completely wild-type phenotype, as evidenced by agouti coat color and normal tooth eruption, bone marrow development and life span. This transgenic approach provided a positive biological assay and reduced our candidate region to a 180 kb physical interval.

Additional support for the identification of the gl gene was obtained from genomic structural characterization of the gl mutation. First, comparison of the structures of the candidate locus in wild-type and gl mice detected a deletion in the gl mice. This deletion spanned 7.5 kb, including the promoter, the first exon and a large part of the first intron. Second, the gl genomic sequence had an additional 460 nucleotides, identical to the 3' UTR of a LINE1 element, at the deletion breakpoints. As a result of this genomic structure, the wild-type and mutated alleles can be distinguished at any age and before phenotypic scoring. Most importantly, the deletion was compatible with the lack of detectable mRNA in homozygous gl/gl mice, strongly indicating that this deletion is the gl mutation. The 3' UTR of the LINE1 portion of the gl mutated allele is virtually identical to that of the beige allele³². This result indicates that we have identified the gl locus. Definite evidence that we have cloned the gl gene was obtained recently by coat color rescue in gl/gl transgenic mice specifically expressing gl cDNA in melanocytes (M.P. and J.V., unpublished data).

Structure analysis showed that the Gl protein contains a putative signal peptide at the N-terminal end and one membrane-spanning domain in the C-terminal region. Western blot analysis of protein extracts from OCLs, using Gl-specific antibodies raised against distinct regions of the protein, detected two bands of 34 kD and 38 kD. The latter probably represents post-translational modifications of the protein; neither protein was present in gl/gl extracts. Cellular localization of the Gl protein seems to be cytosolic and excluded from the nucleus. This agrees with our previous results showing that the gl/gl phenotype in osteoclasts is associated with a defect in cytoskeletal rearrangement at the late cell maturation stage²⁷. At this stage, several ongoing processes could be altered, leading to dysfunction. Possible roles for Gl include cellular polarization, transcytosis pathways (exocytosis and endocytosis) and extracellular acidification^{33, 34, 35, 36}. Similar to the osteoclast impairment, the coat color defect associated with the gl mutation does not affect differentiation, but rather melanocyte maturation. In gl melanocytes, migration of only the yellow pigment is hindered, causing the coat to appear grey²⁵. The clumping of pheomelanin granules may result from a defect in localization, trafficking or exocytosis. Together, the coat color and osteopetrotic defects indicate that the Gl protein has a role in intracellular protein trafficking or localization or both.

In this study, we showed for the first time that a mutation in the human GL gene can underlie the development of severe autosomal recessive infantile malignant osteopetrosis. Nineteen autosomal recessive osteopetrotic patients with no mutations in the TCIRG1 and CLCN7 genes were further analyzed for the presence of mutations in the human GL gene. One mutation was identified that corresponds to a GA substitution at position +5 of the donor splice site of intron V of the GL gene. The patient was homozygous for this mutation, whereas both asymptomatic parents were heterozygous for the same mutation, consistent with the genotype-phenotype correlation. The patient died at one month of age, indicating that loss of GL expression results in a severely abnormal phenotype and an early death, similar to the mouse gl phenotype. The severity of the osteopetrotic disease may explain why only one mutation was found among the 19 patients tested, as mutations in the GL gene may cause prenatal or perinatal death of affected individuals.

	Top

Circular BAC DNA (1 ng/l) was injected into fertilized mouse oocytes isolated from F₁ (C3H C57BL/6) C57BL/6 crosses. Transgenic founders were identified by PCR using specific BAC end-junction sequences and internal polymorphic markers. Each founder was successively crossed with agouti AA heterozygous gl/+ mice to obtain gl/gl BAC transgenic mice. The F₂ gl/gl transgenic mice were then identified by homozygosity at the polymorphic D10Mit55 and D10Mit255 loci²⁸. Histology was performed on bone samples as described²⁷. To characterize BAC gene content, a shotgun M13 phage library was generated and sequenced. BLAST analysis was used in parallel to define transcription units.

Protein extracts (25 g) were resolved on 12% SDS-PAGE gels, transferred onto nitrocellulose membranes and probed²⁷ with polyclonal Gl1 (1:100) or V-ATPase−specific antibody (1:500).

	Top

1. Karsenty, G. The genetic transformation of bone biology. Genes Dev. 13, 3037–3051 (1999). | Article | PubMed  | ISI | ChemPort |
2. Teitelbaum, S.L. Bone resorption by osteoclasts. Science 289, 1504–1508 (2000). | Article | PubMed  | ISI | ChemPort |
3. Lazner, F., Gowen, M., Pavasovic, D. & Kola, I. Osteopetrosis and osteoporosis: two sides of the same coin. Hum. Mol. Genet. 8, 1839–1846 (1999). | Article | PubMed  | ISI | ChemPort |
4. Reeves, J.D., August, C.S., Humbert, J.R. & Weston, W.L. Host defense in infantile osteopetrosis. Pediatrics 64, 202–206 (1979). | PubMed  | ISI | ChemPort |
5. Wilson, C.J. & Vellodi, A. Autosomal recessive osteopetrosis: diagnosis, management, and outcome. Arch. Dis. Child 83, 449–452 (2000). | Article | PubMed  | ISI | ChemPort |
6. Gerritsen, E.J. et al. Bone marrow transplantation for autosomal recessive osteopetrosis. A report from the Working Party on Inborn Errors of the European Bone Marrow Transplantation Group. J. Pediatr. 125, 896–902 (1994). | PubMed  | ISI | ChemPort |
7. Eapen, M., Davies, S.M., Ramsay, N.K. & Orchard, P.J. Hematopoietic stem cell transplantation for infantile osteopetrosis. Bone Marrow Transplant. 22, 941–946 (1998). | Article | PubMed  | ChemPort |
8. Frattini, A. et al. Defects in TCIRG1 subunit of the vacuolar proton pump are responsible for a subset of human autosomal recessive osteopetrosis. Nat. Genet. 25, 343–346 (2000). | Article | PubMed  | ISI | ChemPort |
9. Sobacchi, C. et al. The mutational spectrum of human malignant autosomal recessive osteopetrosis. Hum. Mol. Genet. 10, 1767–1773 (2001). | Article | PubMed  | ISI | ChemPort |
10. Kornak, U. et al. Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell 104, 205–215 (2001). | Article | PubMed  | ISI | ChemPort |
11. Tondravi, M.M. et al. Osteopetrosis in mice lacking haematopoietic transcription factor PU.1. Nature 386, 81–84 (1997). | Article | PubMed  | ISI | ChemPort |
12. Iotsova, V. et al. Osteopetrosis in mice lacking NF-B1 and NF-B2. Nat. Med. 3, 1285–1289 (1997). | Article | PubMed  | ISI | ChemPort |
13. Yoshida, H. et al. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442–444 (1990). | Article | PubMed  | ISI | ChemPort |
14. Dai, X.M. et al. Targeted disruption of the mouse colony-stimulating factor 1 receptor gene results in osteopetrosis, mononuclear phagocyte deficiency, increased primitive progenitor cell frequencies, and reproductive defects. Blood 99, 111–120 (2002). | Article | PubMed  | ISI | ChemPort |
15. Johnson, R.S., Spiegelman, B.M. & Papaioannou, V. Pleiotropic effects of a null mutation in the c-fos proto-oncogene. Cell 71, 577–586 (1992). | Article | PubMed  | ISI | ChemPort |
16. Wang, Z.Q. et al. Bone and haematopoietic defects in mice lacking c-fos. Nature 360, 741–745 (1992). | Article | PubMed  | ISI | ChemPort |
17. Kong, Y.Y. et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397, 315–323 (1999). | Article | PubMed  | ISI | ChemPort |
18. Begg, S.K. et al. Delayed hematopoietic development in osteopetrotic (op/op) mice. J. Exp. Med. 177, 237–242 (1993). | Article | PubMed  | ISI | ChemPort |
19. Thesingh, C.W. & Scherft, J.P. Fusion disability of embryonic osteoclast precursor cells and macrophages in the microphthalmic osteopetrotic mouse. Bone 6, 43–52 (1985). | Article | PubMed  | ISI | ChemPort |
20. Hodgkinson, C.A. et al. Mutations at the mouse microphthalmia locus are associated with defects in a gene encoding a novel basic-helix-loop-helix-zipper protein. Cell 74, 395–404 (1993). | Article | PubMed  | ISI | ChemPort |
21. Soriano, P., Montgomery, C., Geske, R. & Bradley, A. Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693–702 (1991). | Article | PubMed  | ISI | ChemPort |
22. Lomaga, M.A. et al. TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes Dev. 13, 1015–1024 (1999). | PubMed  | ISI | ChemPort |
23. Li, Y.P., Chen, W., Liang, Y., Li, E. & Stashenko, P. Atp6i-deficient mice exhibit severe osteopetrosis due to loss of osteoclast-mediated extracellular acidification. Nat. Genet. 23, 447–451 (1999). | Article | PubMed  | ISI | ChemPort |
24. Scimeca, J.C. et al. The gene encoding the mouse homologue of the human osteoclast-specific 116-kDa V-ATPase subunit bears a deletion in osteosclerotic (oc/oc) mutants. Bone 26, 207–213 (2000). | Article | PubMed  | ISI | ChemPort |
25. Gruneberg, H. Grey-lethal, a new mutation in the house mouse. J. Heredity 27, 105–109 (1936).
26. Walker, D.G. Bone resorption restored in osteopetrotic mice by transplants of normal bone marrow and spleen cells. Science 190, 784–785 (1975). | PubMed  | ISI | ChemPort |
27. Rajapurohitam, V. et al. The mouse osteopetrotic grey-lethal mutation induces a defect in osteoclast maturation/function. Bone 28, 513–523 (2001). | Article | PubMed  | ISI | ChemPort |
28. Vacher, J. & Bernard, H. Genetic localization and transmission of the mouse osteopetrotic grey-lethal mutation. Mamm. Genome 10, 239–243 (1999). | Article | PubMed  | ISI | ChemPort |
29. Chalhoub, N., Benachenhou, N. & Vacher, J. Physical and transcriptional map of the mouse Chromosome 10 proximal region syntenic to human 6q16-q21. Mamm. Genome 12, 887–892 (2001). | Article | PubMed  | ISI | ChemPort |
30. Zhang, Q.H. et al. Cloning and functional analysis of cDNAs with open reading frames for 300 previously undefined genes expressed in CD34⁺ hematopoietic stem/progenitor cells. Genome Res. 10, 1546–1560 (2000). | Article | PubMed  | ISI | ChemPort |
31. Brown, D., Hirsch, S. & Gluck, S. Localization of a proton-pumping ATPase in rat kidney. J. Clin. Invest. 82, 2114–2126 (1988). | PubMed  | ISI | ChemPort |
32. Perou, C.M. et al. Identification of the murine beige gene by YAC complementation and positional cloning. Nat. Genet. 13, 303–308 (1996). | Article | PubMed  | ISI | ChemPort |
33. Baron, R., Neff, L., Louvard, D. & Courtoy, P.J. Cell-mediated extracellular acidification and bone resorption: evidence for a low pH in resorbing lacunae and localization of a 100-kD lysosomal membrane protein at the osteoclast ruffled border. J. Cell Biol. 101, 2210–2222 (1985). | Article | PubMed  | ISI | ChemPort |
34. Baron, R. et al. Polarized secretion of lysosomal enzymes: co-distribution of cation-independent mannose-6-phosphate receptors and lysosomal enzymes along the osteoclast exocytic pathway. J. Cell Biol. 106, 1863–1872 (1988). | Article | PubMed  | ISI | ChemPort |
35. Blair, H.C., Teitelbaum, S.L., Ghiselli, R. & Gluck, S. Osteoclastic bone resorption by a polarized vacuolar proton pump. Science 245, 855–857 (1989). | PubMed  | ISI | ChemPort |
36. Palokangas, H., Mulari, M. & Vaananen, H.K. Endocytic pathway from the basal plasma membrane to the ruffled border membrane in bone-resorbing osteoclasts. J. Cell Sci. 110, 1767–1780 (1997). | PubMed  | ISI | ChemPort |
37. Vacher, J. & Tilghman, S.M. Dominant negative regulation of the mouse -fetoprotein gene in adult liver. Science 250, 1732–1735 (1990). | PubMed  | ISI | ChemPort |
38. Bennett, D.C., Cooper, P.J. & Hart, I.R. A line of non-tumorigenic mouse melanocytes, syngeneic with the B16 melanoma and requiring a tumour promoter for growth. Int. J. Cancer 39, 414–418 (1987). | PubMed  | ISI | ChemPort |
39. Emerson, J.A., Vacher, J., Cirillo, L.A., Tilghman, S.M. & Tyner, A.L. The zonal expression of -fetoprotein transgenes in the livers of adult mice. Dev. Dyn. 195, 55–66 (1992). | PubMed  | ISI | ChemPort |
40. Marcinkiewicz, M., Savaria, D. & Marcinkiewicz, J. The pro-protein convertase PC1 is induced in the transected sciatic nerve and is present in cultured Schwann cells: comparison with PC5, furin and PC7, implication in pro-BDNF processing. Mol. Brain Res. 59, 229–246 (1998). | Article | PubMed  | ISI | ChemPort |

	Top

Competing interests statement:  The authors declare that they have no competing financial interests.