Polycythemia is associated with bone loss and reduced osteoblast activity in mice

Increased fragility has been described in humans with polycythemia vera (PV). Herein, we describe an osteoporotic phenotype associated with decreased osteoblast activity in a mouse model of PV and another mouse of polycythemia and elevated circulating erythropoietin (EPO). Our results are important for patients with PV or those treated with recombinant EPO (rEPO).

INTRODUCTION

PV and other myeloproliferative syndromes have been recently associated with an increased risk for fractures. However, the presence of osteoporosis in these patients has not been well documented. EPO, a hormone primarily known to stimulate erythropoiesis, has been shown recently to regulate bone homeostasis in mice. The aim of this study was to examine the bone phenotype of a mouse model of PV and compare it to that of animals with polycythemia caused by elevated circulating EPO.

METHODS

Bone mass and remodeling were evaluated by micro-computed tomography and histomorphometry. The JAK2(V617F) knock-in mouse, a model of human PV, manifests polycythemia and low circulating EPO levels. Results from this mouse were compared to wild type (wt) controls and the tg6 transgenic mouse that shows polycythemia caused by increased constitutive expression of EPO.

RESULTS

Compared to wt, both JAK2(V617F) and tg6 mice had a decrease in trabecular bone mass. Tg6 mice showed an additional modest decrease in cortical thickness and cortical bone volume per tissue volume (P < 0.01) suggesting a more severe bone phenotype than JAK2(V617F). Decreased osteoblast numbers and bone formation along with normal osteoclast numbers and activity were found in both mice.

CONCLUSIONS

This study indicates that PV is associated with low bone mass and decreased osteoblast activity in mice. Our results support future studies of osteoporosis in affected humans. Polycythemia caused by chronically elevated circulating EPO also results in bone loss, and implications on patients treated with rEPO should be evaluated.

Therapeutic haemoglobin synthesis in beta-thalassaemic mice expressing lentivirus-encoded human beta-globin

The stable introduction of a functional beta-globin gene in haematopoietic stem cells could be a powerful approach to treat beta-thalassaemia and sickle-cell disease. Genetic approaches aiming to increase normal beta-globin expression in the progeny of autologous haematopoietic stem cells might circumvent the limitations and risks of allogeneic cell transplants. However, low-level expression, position effects and transcriptional silencing hampered the effectiveness of viral transduction of the human beta-globin gene when it was linked to minimal regulatory sequences. Here we show that the use of recombinant lentiviruses enables efficient transfer and faithful integration of the human beta-globin gene together with large segments of its locus control region. In long-term recipients of unselected transduced bone marrow cells, tetramers of two murine alpha-globin and two human betaA-globin molecules account for up to 13% of total haemoglobin in mature red cells of normal mice. In beta-thalassaemic heterozygous mice higher percentages are obtained (17% to 24%), which are sufficient to ameliorate anaemia and red cell morphology. Such levels should be of therapeutic benefit in patients with severe defects in haemoglobin production.

The cHS4 insulator increases the probability of retroviral expression at random chromosomal integration sites

Retroviruses are highly susceptible to transcriptional silencing and position effects imparted by chromosomal sequences at their integration site. These phenomena hamper the use of recombinant retroviruses as stable gene delivery vectors. As insulators are able to block promoter-enhancer interactions and reduce position effects in some transgenic animals, we examined the effect of an insulator on the expression and structure of randomly integrated recombinant retroviruses. We used the cHS4 element, an insulator from the chicken beta-like globin gene cluster, which has been shown to reduce position effects in transgenic Drosophila. A large panel of mouse erythroleukemia cells that bear a single copy of integrated recombinant retroviruses was generated without using drug selection. We show that the cHS4 increases the probability that integrated proviruses will express and dramatically decreases the level of de novo methylation of the 5' long terminal repeat. These findings support a primary role of methylation in the silencing of retroviruses and suggest that cHS4 could be useful in gene therapy applications to overcome silencing of retroviral vectors.

Selection and mapping of replication origins from a 500-kb region of the human X chromosome and their relationship to gene expression

In higher eukaryotes the mechanism controlling initiation of DNA replication remains largely unknown. New technologies are needed to shed light on how DNA replication initiates along the genome in specific regions. To identify the human DNA sequence requirements for initiation of replication, we developed a new method that allows selection of replication origins starting from large genomic regions of human DNA. We repeatedly isolated 15 new putative replication origins (PROs) from a human DNA region of 500 kb in which 17 genes have previously been characterized. Fine-mapping of these PROs showed that DNA replication can initiate at many specific points along actively transcribed DNA in the cell lines used for our selection. In conclusion, in this paper we describe a new method to identify PROs that suggests that the availability of initiation sites is dependent on the transcriptional state of the DNA.

Genetic treatment of severe hemoglobinopathies: the combat against transgene variegation and transgene silencing

Gene addition strategies are rational approaches to the treatment of sickle cell anemia and thalassemia. The goal of such genetic treatments is to introduce a functional globin transcription unit in hematopoietic stem cells and express the transgene in a manner that is erythroid-specific, elevated, relatively constant from one cell to another, and sustained over time. Gene transfer is mediated by an expanding array of viral and nonviral vectors. High-titer retroviral vectors harboring the human beta-globin gene and the core sequences of the human beta-globin locus control region yield erythroid-specific gene expression in erythroid cell lines and in short-term murine bone marrow chimeras. However, we show that expression remains subject to position effect variegation and often decreases over time in vivo. Rather than a progressive transcriptional silencing in all cells, we ascribe the waning expression to the gradual emergence in blood of erythroid progeny derived from more and more primitive precursor cells in the months after transplantation. In our model, transgene expression is therefore determined by the integration site and the differentiation stage of the transduced cell at the time of integration. Globin expression is thus different in the progeny of a transduced erythroid progenitor cell and in the erythroid progeny of a transduced hematopoietic stem cell, reflecting the effect of flanking chromatin in differentiated cells and of chromatin remodeling at the site of integration in the progeny of multipotential cells. This model predicts that insulators and matrix attachment regions could be highly valuable to gene therapy in combination with potent transcriptional activators. When efficient gene transfer in hematopoietic stem cells is achieved at last, the challenge will be to regulate gene expression in vivo and overcome transgene variegation and transgene silencing.

Selection and fine mapping of chromosome-specific cDNAs: application to human chromosome 1

We have developed a methodology for identification and fine mapping of chromosome-specific transcripts. Combining digestion of DNA with different restriction enzymes, ligation to "bubble" linkers, and PCR amplification from Alu and "bubble" primers, we have synthesized human chromosome 1-specific sequences from DNA of a somatic cell hybrid, A9Neol. After hybridization to human fetal brain cDNA, we could efficiently capture chromosome 1-specific cDNAs. The cDNAs were sequenced and used as probes in hybridizations to high-density filters containing the arrayed CEPH Mega-YAC library and to the arrayed cDNA library from infant brain made by B. Soares, which has been extensively sequenced. By this approach we have been able to select chromosome 1-specific cDNAs, to map them to chromosome 1 YAC contigs, and to identify and map corresponding longer cDNAs and ESTs.

A comparative transcriptional map of a region of 250 kb on the human and mouse X chromosome between the G6PD and the FLN1 genes

The transcriptional organization of the region of the mouse X chromosome between the G6pd and the Fln1 genes was studied in detail, and it was compared with the syntenic region of the human chromosome. A cosmid contig of 250 kb was constructed by screening mouse cosmid libraries with probes for human genes and with whole cosmids. Overlapping cosmids were aligned by comparing EcoRI and rare-cutter restriction enzyme digestions. The gene order and the orientation of transcription were determined by hybridization with fragments from the 5' and 3' moieties of each cDNA. Our work demonstrates that all of the new genes identified in human are present in the mouse. The size of the region, 250 kb, is also very similar, as are gene order and gene organization: the transcriptional organization in "domains" described in human is found to be identical in the mouse. The major difference detected is the much lower content in rare-cutter restriction sites, which is related to the lower G+C and CpG content of mouse DNA. The very high conservation that we have described suggests that a potent selective pressure has contributed to such conservation of gene organization.

Identification of a novel X-linked gene responsible for Emery-Dreifuss muscular dystrophy

Emery-Dreifuss muscular dystrophy (EDMD) is an X-linked recessive disorder characterized by slowly progressing contractures, wasting of skeletal muscle and cardiomyopathy. Heart block is a frequent cause of death. The disease gene has been mapped to distal Xq28. Among many genes in this region, we selected eight transcripts expressed at high levels in skeletal muscle, heart and/or brain as the best candidates for the disease. We now report, in all five patients studied, unique mutations in one of the genes, STA: these mutations result in the loss of all or part of the protein. The EDMD gene encodes a novel serine-rich protein termed emerin, which contains a 20 amino acid hydrophobic domain at the C terminus, similar to that described for many membrane proteins of the secretory pathway involved in vesicular transport.

Isolation of new genes in distal Xq28: transcriptional map and identification of a human homologue of the ARD1 N-acetyl transferase of Saccharomyces cerevisiae

In this paper, we describe the physical and transcriptional organization of a region of 140 kb in Xq28, 5' to the L1CAM gene. By isolation and mapping of CpG islands to the physical map of the region, isolation of cDNAs, determination of partial nucleotide sequences and study of the pattern of expression and of the orientation of the transcripts identified we have established a transcriptional map of this region. In this map, previously identified genes (L1CAM, V2R, HCF1 and RnBP) have been positioned as well as 3 new genes. All genes in the region are rather small, ranging in size from 2 to 30 kb, and very close to one another. With the exception of the V2R gene, they are housekeeping, have a CpG island at their 5' end and the same orientation of transcription. This kind of organization is consistent with the one previously described for the more distal portion of Xq28, between the Color Vision (CV) and the G6PD genes and indicates that genes with housekeeping and tissue specific pattern of expression are interspersed in the genome but they are probably found in different 'transcriptional domains'. Among the new genes, TE2 demonstrated 40% identity with the protein N-acetyl transferase ARD1 of S. cerevisiae: TE2 may be the human homologue of the S. cerevisiae gene.

Comparative mapping of the actin-binding protein 280 genes in human and mouse

Two genes encode actin-binding protein 280 isoforms. ABP-280 or filamin (FLN1) is present in the cytoskeleton of many cell types, whereas expression of FLN2 is limited to skeletal muscle and heart. FLN1 maps to human chromosome Xq28, and, by physical mapping in YAC clones, we have mapped the homologous murine locus (Fln1) to mouse chromosome X, in a region of syntenic homology with human chromosome X. We mapped FLN2 to human chromosome 7q32-q35 by analysis of somatic cell hybrids containing portions of chromosome 7, and, by using a mapping panel from an interspecific murine cross, we mapped the corresponding murine locus (Fln2) to murine chromosome 6 in a region homologous to human chromosome 7.

Transcriptional organization of a 450-kb region of the human X chromosome in Xq28

In this paper, we report the transcriptional organization of a 450-kb gene cluster in Xq28, flanked by the glucose-6-phosphate dehydrogenase and the color vision genes. CpG islands previously identified and mapped to distal Xq28 have helped in construction of a continuous contig of cosmids and in identification of cDNAs corresponding to eight transcripts. Thirteen to 16 small genes with CpG islands are clustered in a region of 250-300 kb. Many are highly expressed in muscle or brain and may be the genes responsible for muscle or neurological disorders mapped to distal Xq28. Our analysis indicates that, in this region of the genome, genes not related in sequence are organized in transcriptional domains of 100 kb and that this organization may be important for establishing and regulating gene expression in relation to tissue distribution and X chromosome inactivation.

Mapping of two genes encoding isoforms of the actin binding protein ABP-280, a dystrophin like protein, to Xq28 and to chromosome 7

ABP-280 is a ubiquitous actin binding protein present in the cytoskeleton of many different cell types. ABP-280 was mapped to distal Xq28, 50-60 kb downstream of the Green Colour Pigment (GCP) genes. To establish if ABP-280 may be a candidate for one of the muscle disease localized by linkage analysis to distal Xq28 we looked for alternative forms of ABP-280 mRNA. Several different ABP-280 mRNAs were indeed identified: two are X-linked and are produced by alternative splicing of a small exon of 24 nucleotides. At least one additional gene encoding a RNA more than 70% identical to ABP-280 in the 1700 bp sequenced has also been found. It was mapped to chromosome 7. While both forms of the X-linked ABP-280 are ubiquitous, the gene on chromosome 7 is highly expressed only in skeletal muscle and heart. The two genes were therefore excellent candidates for the X-linked and for the autosomal dominant form of the Emery-Dreifuss Muscular Dystrophy (EDMD) both of which have been described. So far, however we were unable to demonstrate mutations in the coding region or affecting the alternative splicing of the X-linked form of ABP-280, in several patients studied, and we think that it is quite unlikely that this is the gene responsible for EDMD.

Methylation and sequence analysis around EagI sites: identification of 28 new CpG islands in XQ24-XQ28

Thirty-two probes for CpG islands of the distal long arm of the human X chromosome have been identified. From a genomic library of DNA of the hamster-human cell hybrid X3000.1 digested with the rare cutter restriction enzyme EagI, 53 different human clones have been isolated and characterized by methylation and sequence analysis. The characteristic pattern of DNA methylation of CpG islands at the 5' end of genes of the X chromosome has been used to distinguish between EagI sites in CpG islands versus isolated EagI sites. The sequence analysis has confirmed and completed the characterization showing that sequences at the 5' end of known genes were among the clones defined CpG islands and that the non-CpG islands clones were mostly repetitive sequences with a non-methylated or variably methylated EagI site. Thus, since clones corresponding to repetitive sequences can be easily identified by sequencing, such libraries are a very good source of CpG islands. The methylation analysis of 28 different new probes allows to state that demethylation of CpG islands of the active X and methylation of those on the inactive X chromosome are the general rule. Moreover, the finding, in all instances, of methylation differences between male and female DNA is in very strong support of the notion that most genes of the distal long arm of the X chromosome are subject to X inactivation.

Identification of novel RFLPs in the vicinity of CpG islands in Xq28: application to the analysis of the pattern of X chromosome inactivation

Probes for CpG islands were cloned from the distal long arm of the human X chromosome; three of them were found to be polymorphic. A HindIII RFLP was identified by the probe 2-25 (DXS606), and it was mapped to the Xq27-Xq28 boundary. Probes 2-19 (DXS605) and 2-55 (DXS707), which identify EcoRI and MspI polymorphisms, respectively, have been mapped to the distal part of Xq28, in the G6PD-RCP/GCP gene region. Probe 2-19 has been further localized about 16 kb from the 3' end of the G6PD gene. The new RFLPs may be useful for the precise mapping of the many disease genes localized in this part of the human X chromosome. Probe 2-19 is highly informative, and it has been studied in greater detail. Using the methylation-sensitive rare-cutter enzyme EagI in conjunction with the polymorphic EcoRI site, we were able to demonstrate that the RFLP may be used both to study randomness of X chromosome inactivation and for carrier detection in X-linked syndromes where nonrandom X inactivation occurs. It is conceivable that the combined use of 2-19 and of the probes described so far (pSPT-PGK and M27 beta) will make analysis of X inactivation feasible in virtually every female.

The nucleotide sequence of a CpG island demonstrates the presence of the first exon of the gene encoding the human lysosomal membrane protein lamp2 and assigns the gene to Xq24

An EagI-EcoRI clone of human genomic DNA, p2-7, mapped to Xq24 has been sequenced. This analysis has confirmed the presence of a CpG island and has identified the first exon of the human LAMP2 gene, encoding a glycoprotein of the lysosomal membrane. Since the p2-7 clone corresponds to single-copy DNA, we can assign the human LAMP2 gene to Xq24.

Four chromosomal breakpoints and four new probes mark out a 10-cM region encompassing the fragile-X locus (FRAXA)

We report the validation and use of a cell hybrid panel which allowed us a rapid physical localization of new DNA probes in the vicinity of the fragile-X locus (FRAXA). Seven regions are defined by this panel, two of which lie between DXS369 and DXS296, until now the closest genetic markers that flank FRAXA. Of those two interesting regions, one is just distal to DXS369 and defined by probe 2-71 (DXS476), which is not polymorphic. The next one contains probes St677 (DXS463) and 2-34 (DXS477), which are within 130 kb and both detect TaqI RFLPs. The combined informativeness of these two probes is 30%. We cloned from an irradiation-reduced hybrid line another new polymorphic probe, Do33 (DXS465; 42% heterozygosity). This probe maps to the DXS296 region, proximal to a chromosomal breakpoint that corresponds to the Hunter syndrome locus (IDS). The physical order is thus Cen-DXS369-DXS476-(DXS463,DXS477)-(DXS296, DXS465)-IDS-DXS304-tel. We performed a linkage analysis for five of these markers in both the Centre d'Etude du Polymorphisme Humain families and in a large set of fragile-X families. This establishes that DXS296 is distal to FRAXA. The relative position of DXS463 and DXS477 with respect to FRAXA remains uncertain, but our results place them genetically halfway between DXS369 and DXS304. Thus the DXS463-DXS477 cluster defines presently either the closest proximal or the closest distal polymorphic marker with respect to FRAXA. The three new polymorphic probes described here have a combined heterozygosity of 60% and represent a major improvement for genetic analysis of fragile-X families, in particular for diagnostic applications.

Probes for CpG islands on the distal long arm of the human X chromosome are clustered in Xq24 and Xq28

We have isolated and characterized 55 EagI-containing genomic DNA clones from the distal long arm of the human X chromosome. The presence of additional sites for rare-cutter restriction enzymes and the demethylation of the corresponding genomic DNA demonstrate that at least 30 clones correspond to CpG islands of the Xq24-Xqter region. All clones were regionally mapped with a hybrid panel. The majority are in Xq28 and Xq24 (18 and 14 clones, respectively), 15 are in the Xq26-Xq27 interval, and none is in Xq25. This analysis demonstrates a nonuniform distribution of CpG islands that may reflect the distribution of coding regions in this part of the genome.