Comparison of factor IX methylation on human active and inactive X chromosomes: implications for X inactivation and transcription of tissue-specific genes.

Maintenance of dosage compensation for housekeeping genes on the human X chromosome is mediated through differential methylation of clustered CpG nucleotides associated with these genes. To determine if methylation has a role in maintaining inactivity of X-linked genes which show tissue-specific expression, we examined the locus for blood clotting Factor IX. The analysis encompassed 91% of the HpaII and HhaI sites in the 41-kb region that includes the presumed promoter region, 5 kb of 5'- and 4 kb of 3'-flanking sequences. Although there are sex differences in methylation of the locus in leukocytes, the methylation pattern in liver, where the gene is expressed, is essentially the same for loci on the active and inactive X chromosome. The lack of differences in methylation of active and inactive genes makes it unlikely that methylation within the locus has a role in expression of the Factor IX gene. These findings, along with the absence of clustered CpG dinucleotides within the Factor IX locus, suggest that functional differences in DNA methylation related to X chromosome dosage compensation may be limited to CpG clusters. In any event, dosage compensation seems to be maintained regionally, rather than locus by locus.     

Selecting representative model micro-organisms

Background  

Micro-biological research relies on the use of model organisms that act as representatives of their species or subspecies, these are frequently well-characterized laboratory strains. However, it has often become apparent that the model strain initially chosen does not represent important features of the species. For micro-organisms, the diversity of their genomes is such that even the best possible choice of initial strain for sequencing may not assure that the genome obtained adequately represents the species. To acquire information about a species' genome as efficiently as possible, we require a method to choose strains for analysis on the basis of how well they represent the species.     

Molecular studies of marsupial X chromosomes reveal limited sequence homology of mammalian X-linked genes

To explore the extent to which the X chromosome has been conserved during mammalian evolution, we compared six loci that are X-linked in the human genome with the corresponding genes of the North American marsupial, the Virginia opossum (Didelphis virginiana). Our analysis shows that in the opossum genome there are sequences highly homologous to those of human cDNAs for housekeeping genes, glucose-6-phosphoribosyltransferase (HPRT), phosphoglycerate kinase A (PGK1), and alpha-galactosidase A (GLA). However, ornithine transcarbamylase and blood clotting Factor IX--tissue-specific genes that are X-linked in eutherians mammals--have no highly conserved homologs in the marsupial genome. By cloning opossum G6PD and HPRT, we found that these genes are X-linked in the opossum and that homologous sequences are limited to coding regions. As all genomic fragments hybridizing with the human GLA probe show dosage effects, it is likely that the opossum counterpart is X-linked. Finally, the pattern of hybridization suggests that the autosomal pseudogenes of HPRT and PGK1 in the opossum have remained highly homologous to the human X-linked genes.     

Prevalence of polymorphic 21-hydroxylase gene (CA21HB) mutations in salt-losing congenital adrenal hyperplasia

Using genomic restriction analysis of 14 unrelated patients with salt-losing congenital adrenal hyperplasia, we identified three different CA21HB mutation patterns: no detectable restriction fragment abnormalities (16/28 haplotypes), deletion of the active CA21HB gene (9/28), and apparent conversion of the active CA21HB gene to the pseudogene CA21HA (3/28). CA21HB gene deletion was associated with HLA-Bw47 in 6 haplotypes and with absent C4B expression in 7. A variety of HLA and C4 types was associated with the other mutations. Apparent conversion of CA21HB to CA21HA was identified by the disparity between the intensity ratios for the major TaqI and BglII hybridization fragments.     

Frequent derepression of G6PD and HPRT on the marsupial inactive X chromosome associated with cell proliferation in vitro

X chromosome dosage compensation in Marsupials is like that in eutherian mammals except that the paternal X chromosome is always inactive, and silence of this chromosome is not well maintained. We previously showed that the unstable inactivation of the paternal G6PD allele is associated with the lack of DNA methylation in the 5' CpG cluster. Even though this CpG island is unmethylated, the paternal allele (marked by an enzyme variant) is at least partially and often severely repressed in most tissues of the opossum, so that factors other than methylation must inactivate the locus. Here we report that when cell cultures are established from these tissues, the silent G6PD locus is depressed. Although often complete, the extent of derepression differs among tissues and within different cell types in the same tissue, and is not accompanied by obvious changes in the pattern of chromosome replication. Studies of the HPRT locus in these cells show that the paternal HPRT allele also derepresses in cultured cells. These observations suggest that without DNA methylation to maintain the silence of the locus, tissue or cell-specific factors act to repress the silent locus, but are unable to maintain inactivity through cell division, or are lost as cells proliferate in culture.     

Expression of the G6PD locus on the human X chromosome is associated with demethylation of three CpG islands within 100 kb of DNA.

We have previously reported that expression of the G6PD locus is correlated with the methylation status of two islands of CpG dinucleotides which are 3' to the locus and in the 5' region of two adjacent genes of unknown function, P3 and GdX. We have now examined the methylation of a third CpG island in the promoter region of the G6PD gene itself in DNA from males, females and reactivants that express G6PD on the inactive X chromosome. Our results show that expression of the G6PD gene is associated with concordant demethylation of all three CpG islands in this 100-kb region of DNA.     

Restriction fragment analysis of duplication of the fourth component of complement (C4A)

The two genes encoding the fourth component of complement (C4A and C4B) reside between HLA-B and HLA-DR on human chromosome 6. Two kilobases downstream from each C4 gene lies a 21-hydroxylase gene (CA21HA and CA21HB, respectively). Utilizing the method of Southern blotting and a 5'-end 2.4-kb BamHI/KpnI fragment of the C4 cDNA, we have analyzed TaqI-digested DNA from four pedigrees with one or more extended haplotypes containing a C4A duplication, as demonstrated by protein electrophoresis and segregation analysis. Two C4A protein duplications (C4A*2,A*3,C4B*QO and C4A*3,A*5,C4B*QO) segregated with two large TaqI DNA restriction fragments (7.0 and 6.0). In pedigree Fi, one individual homozygous for HLA-A3,B35,C4,DR1,DQ1,BFF,C2C,-C4A2,3,C4BQO had TaqI 7.0- and 6.0-kb restriction fragments with equal hybridization intensities as measured by two-dimensional densitometry (7.0/6.0 kb = 0.83, SD = 0.12, N = 7). A hybridization probe for the 21-hydroxylase gene also demonstrated equal gene dosage (CA21HA/CA21HB = 1.01). DNA from another individual (Ma I-2) with a different C4A gene duplication (C4A*3,A*5,C4B*QO) also had equal densitometry measurements (7.0/6.0 kb = 1.07). We conclude that two extended haplotypes from unrelated pedigrees have two C4 genes and both C4 genes encode separate C4A alleles. These findings are compatible with a gene conversion event of C4B to C4A.     

Three related centromere proteins are absent from the inactive centromere of a stable isodicentric chromosome

We developed an aqueous spreading procedure that permits simultaneous analysis of human chromosomes by Q-banding and indirect immunofluorescence. Using this methodology and anticentromere antibodies from an autoimmune patient we compared the active and inactive centromeres of an isodicentric X chromosome. We show that a family of structurally related human centromere proteins (CENP-A, CENP-B, and CENP-C) is detectable only at the active centromere. These antigens therefore may be regarded both as morphological and functional markers for active centromeres.     

Influence of cell culture conditions on aromatase activity in human genital skin fibroblasts

Summary Because the measurement of aromatase activity in cultured human genital skin fibroblasts has been proposed as a means of studying estrogen production in men, we investigated the influence of culture conditions on aromatase activity. Genital skin fibroblasts were seeded onto culture plates at a density of 1×10⁶ cells/plate and aromatase activity was determined over a 1-mo. period. Enzyme activity rose slowly over the first 14 d but then rose rapidly to a 10-fold higher plateau by Day 28. The rise in aromatase activity was similar whether activity was normalized for protein or for DNA content. When cells were seeded at the usual density of 1×10⁶ or at 0.25×10⁶ cells/plate, aromatase activity was consistently lower during the first 2 wk in cells plated at lower density, but thereafter the levels of enzyme activity in the two groups converged. In cells plated at the lower density, the lower activity observed in the first 2 wk was associated with a lower V _max . Preincubation of cells plated at one density with conditoned medium from cells plated at the other density did not change the relatve levels of activity in the two groups. By contrast, dihydrotestosterone (DHT) receptor binding and 5α-reductase activity were similar at all time points, despite differences in plating density. In additional experiments, the culture medium was replaced daily rather than every 3rd d, and aromatase activity was assayed on Day 7. In cells fed daily, DNA and protein content were twice that of cells fed every 3rd d. By contrast, aromatase activity declined to 30% of the in the latter group. DHT and dexamethasone receptor binding and 5α-reductase activity were similar in the two groups. In summary, factors such as plating density, culture density, and frequency of media replacement dramatically influence aromatase activity in cultured human genital skin fibroblasts. Therefore, the interpretation of aromatase activity data obtained from cultured cells in relation to physiologic or pathologic states should be viewed with appropriate caution. The work was supported in part by grants R01 DK 35339 and R01 DK 00180 from the National Institutes of Health, Bethesda, MD, and by RR 00035 from CLINFO Systems at the Johns Hopkins University School of Medicine, Baltimore, MD.