Use of a new mouse beta-globin haplotype (Hbbs2) to study hemoglobin expression during development

Mice of the mutant haplotype (Hbbs2) produce a variant beta-s globin (beta-s2major) which can be distinguished from beta-smajor and beta-sminor by cellulose acetate electrophoresis and ion exchange chromatography. Mice homozygous for this mutation were used to study the relative quantities of the mutant beta-s2major and normal beta-sminor globins specified by the two adult beta-globin genes of the Hbbs2 haplotype during development. At 11.5 days of gestation, beta-s2major comprises under 20% and beta-sminor over 80% of the adult beta-globin. The relative level of beta-sminor decreases through fetal development; at birth beta-sminor represents 33.7% of the beta-globin. The adult value of 71.0% beta-s2major and 29.0% beta-sminor globin is expressed in mice 6 days after birth. In mildly anemia alpha-thalassemic heterozygotes (Hbab2(th)/Hbab;Hbbs2/Hbbs2) the level of beta-sminor globin increases from 29.0 to 37.9%, but beta-sminor is elevated only slightly (29.0 to 33.9%) in asymptomatic beta-thalassemic heterozygotes (Hbab/Hbab;Hbbd3(th)Hbbs2). The relative quantity of beta-sminor is increased significantly (29.0 to 41.4%) in doubly heterozygous alpha-thalassemic, beta-thalassemic mice (Hbab2(th)/Hbab;Hbbd3(th)/Hbbs2). The relative levels of expression of the beta 1s2major and beta 2sminor globin genes of Hbbs2/Hbbs2 mice correlates well with the expression of the beta 1dmajor and beta 2dminor globin genes of Hbbd/Hbbd mice during development and in response to hematological stress caused by thalassemia. Expression of the beta 1sminor globin gene should not have been affected by the ENU-induced base substitution in the beta 1smajor gene. Therefore, we propose that the beta 1sminor gene is also expressed in mice of the Hbbs haplotype. The results also indicated that the two adult beta-globin genes of the Hbbs2 and, presumably, of the Hbbs haplotypes are regulated independently as are the beta 1dmajor and beta 2dminor genes of the Hbbd and Hbbp haplotypes.     

A new haplotype of the beta-globin gene complex, Hbbw1, in Chinese wild mouse

A new pattern was observed in the electrophoretic survey of the hemoglobin beta chain (Hbb) in Chinese wild mice, Mus musculus. The electrophoretic mobility of the major component of the new Hbb was identical to that of Hbbs on cellulose acetate plate, although it was almost identical to that of Hbbd or Hbbp on acrylamide gel. This suggests that the major component of Chinese Hbb has a unique primary structure. A minor component of the new Hbb was completely different from that of the other three Hbb haplotypes well known. These results indicate that the Hbb-b1 and Hbb-b2 of the new Hbb haplotype, assigned Hbbw1, are unique genes in their molecular structure. So far, Hbbw1 has been observed in northwestern China.     

The sequence of a mouse embryonic beta-globin gene. Evolution of the gene and its signal region

We have determined the nucleotide sequence of an embryonic beta-globin gene of the Balb/c mouse. It possesses structural features typical of known expressed beta-globins, including 5'-untranslated region, potential capping site, initiation and terminator codons, poly(A) addition signal, splicing signals, and intervening sequences. There is a striking 15-base homology between the putative RNA polymerase binding site of this gene and the corresponding region of a human embryonic beta-gene which may be important in the control of expression during embryonic development. The sequence of the gene predicts the sequence of the entire y2 protein, which has not previously been available. As expected from evolutionary considerations, it is more homologous to human embryonic beta-globin than to mouse adult beta-globin.     

Two mouse early embryonic beta-globin gene sequences. Evolution of the nonadult beta-globins

We have determined the complete nucleotide sequence of two early embryonic beta-globin genes of the BALB/c mouse: beta h0 and beta h1 X beta h1 codes for the embryonic z protein, while the beta h0 gene may be a minor early embryonic beta-globin gene. The general sequence organization of both genes is entirely analogous to other functional globin genes. There is, however, a 220-base pair insertion of unique sequence within the first intron of beta h0 X beta h0 and beta h1 are 96% homologous for 260 base pairs 5' to the AUG initiation codon, and 93% homologous throughout their coding regions. Analysis of the 5'-flanking sequence demonstrates that these genes are more nonadult-like than adult-like. The sequences show evidence for gene conversions among the mouse nonadult beta-globin genes that were limited to individual exons, presumably by the presence of non-homologous introns. We propose that this arrangement has the beneficial evolutionary effect of allowing gene conversion to act independently on regions of the protein with different structural or functional responsibilities. beta h0 and beta h1 are evolutionary homologs to the human fetal and rabbit beta 3 genes, while their manner of expression is similar to rabbit beta 3 and dissimilar to human fetal expression. The evolutionary history of the human beta-globin genes, therefore, includes the recruitment of an embryonic gene to fetal developmental control.     

History and origin of beta-thalassemia in Turkey: sequence haplotype diversity of beta-globin genes.

In the present study we report the sequence haplotypes associated with 22 beta-globin gene mutations present in Turkey. Nine nucleotide polymorphisms and an (AT)xTy motif located at the 5' end of the beta-globin gene form the sequence haplotypes that were investigated in 204 unrelated beta-thalassemia and wild-type chromosomes from Turkey. Twelve sequence haplotypes were observed in the chromosomes analyzed and haplotypic heterogeneity was found in the wild-type beta-globin genes. Samples from the Black Sea region demonstrated a remarkable level of haplotypic heterogeneity in contrast to the homogeneity present in Central Anatolian samples. Of the 22 beta-globin mutations analyzed, 18 were related with single sequence haplotypes. This simple association led to the attempt to determine the origin of these mutations by comparing their frequencies in Turkey with those in other countries and/or the world distribution of the haplotypes carrying them. However, the presence of several exceptions for the "one haplotype/one mutation" rule showed that the beta-globin gene cluster is far from static. Each of the IVS-I-110 (G-->A), Cd 39 (C-->T), IVS-I-6 (T-->C), and -30 (T-->A) beta-globin mutations was associated with a minimum of two sequence haplotypes. This fact is best explained by the likelihood of strong recombination mechanisms taking place, rather than by assuming multiple origins for each of these alleles. According to our results, malarial selection for the oldest beta-thalassemia allele in Anatolia (i.e., IVS-I-110 G-->A) may have occurred between 6500 and 2000 B.C. From that date on, most of the common beta-thalassemia mutations in Turkey were established, and by the 13th century A.D. most of them were brought to frequencies close to those observed at present.     

Evolution of tuf genes: ancient duplication, differential loss and gene conversion

The tuf gene of eubacteria, encoding the EF-tu elongation factor, was duplicated early in the evolution of the taxon. Phylogenetic and genomic location analysis of 20 complete eubacterial genomes suggests that this ancient duplication has been differentially lost and maintained in eubacteria.     

Rat kappa-chain J-segment genes: two recent gene duplications separate rat and mouse

We have cloned DNA segments containing the Jk genes from LOUVAIN rat liver, and have determined their nucleotide sequence. Seven readily identifiable Jk-coding regions (six expressible) are evident in the rat, compared with five in the mouse (four expressible). The two additional J segments in the rat appear to be the result of two sequential gene duplications occurring since the divergence of rats and mice. The first involved a homologous but unequal crossing-over in a 14 bp region spanning the 3' end of the coding region of J1 and J2. The second involved a crossing-over following unequal pairing of the two newly duplicated regions. We propose that the probability of a second duplication was greatly increased following the first as a result of the increased target for unequal pairing (370 bp of good homology versus 27 bp in the original pairing). Comparisons of rat and mouse J genes show a surprisingly high degree of sequence conservation, both inside and outside the coding regions, similar to the pattern we reported previously for the kappa constant-region gene. This provides additional evidence that constraints exist on the nucleotide sequences of these genes independent of the function of the encoded proteins.     

Isolation and characterization of the complete chicken beta-globin gene region: frequent deletion of the adult beta-globin genes in lambda.

A library of bacteriophage lambda clones containing chicken chromosomal DNA was screened, using the adult beta-globin cDNA plasmid pHb 1001 as a probe. Sixteen overlapping clones were isolated containing 35 kilobase pairs (kbp) of chicken DNA. Characterization of these clones revealed four beta-like globin genes, some genomically repeated sequences, but no pseudo-genes. The four beta-like genes have an average intergenic distance of less than half of that found for the mammalian beta-like globin gene clusters so far characterized. The overall features of the map were confirmed by genomic Southern analysis. Frequent deletions were shown to occur between the various beta-like globin genes during phage propagation. The presumptive hatching gene in particular was always associated with abnormal lambda clones although we were able to find one such clone that did contain a normal copy of the hatching gene itself. Probably such deletions explain the failure to recover this gene in previous attempts.     

Evolution of the mouse polyubiquitin-C gene

The polymeric ubiquitin (poly-u) genes are composed of tandem 228-bp repeats with no spacer sequences between individual monomer units. Ubiquitin is one of the most conserved proteins known to date, and the individual units within a number of poly-u genes are significantly more similar to each other than would be expected if each unit evolved independently. It has been proposed that the rather striking similarity among poly-u monomers in some lineages is caused by a series of homogenization events. Here we report the sequences of the polyubiquitin-C (Ubc) genes in two mouse strains. Analysis of these sequences, as well as those of the previously reported Chinese hamster and rat poly-u genes, supports the assertion that the homogenization of the ubiquitin-C gene in rodents is due to unequal crossing-over events. The sequence divergence of noncoding DNA was used to estimate the frequency of unequal crossing-over events (6.3 × 10⁻⁵ events per generation) in the Ubc gene, as well as to provide evidence of apparent selection in the poly-u gene.     

In vitro methylation of the 5'-flanking regions of the mouse beta-globin gene

The enzymatic methylation of the 5'-flanking region of the mouse beta-globin (major) gene containing putative regulatory regions has been investigated. In vitro methylation of this 368-base pair regulatory DNA by a DNA methyltransferase obtained from mouse erythroleukemia cells yields an asymmetric methylation pattern. Of the 10 available CG pairs, only 5-6 are modified, leading to one hemimethylated site and two apparently fully methylated sites. Only CG pairs which are localized in a 29-base pair cluster are methylated. The data suggest that a CG cluster approximately 100 base pairs upstream from the CAP site may be the in vivo site of methylation in the 5'-regulator region of the mouse beta-globin gene.     

Expression of the two adult beta-globin genes in mouse yolk sac and fetal liver erythrocytes

The two adult beta-globin genes (beta 1s2major and beta 2sminor) of the Hbbs2 haplotype are differentially expressed during development. Centrifugal elutriation was used to separate the two populations of erythrocytes present in developing fetuses. Hemoglobin analysis showed that the larger, nucleated erythrocytes (yolk sac-derived) have relatively larger amounts of beta-sminor hemoglobin than do smaller, nonnucleated cells (fetal liver-, spleen-, and bone marrow-derived) at the same stage of development.     

Joining the loops: beta-globin gene regulation

The mammalian beta-globin locus is a multigene locus containing several globin genes and a number of regulatory elements. During development, the expression of the genes changes in a process called "switching." The most important regulatory element in the locus is the locus control region (LCR) upstream of the globin genes that is essential for high-level expression of these genes. The discovery of the LCR initially raised the question how this element could exert its effect on the downstream globin genes. The question was solved by the finding that the LCR and activate globin genes are in physical contact, forming a chromatin structure named the active chromatin hub (ACH). Here we discuss the significance of ACH formation, provide an overview of the proteins implicated in chromatin looping at the beta-globin locus, and evaluate the relationship between nuclear organization and beta-globin gene expression.     

New genes originated via multiple recombinational pathways in the beta-globin gene family of rodents

Species differences in the size or membership composition of multigene families can be attributed to lineage-specific additions of new genes via duplication, losses of genes via deletion or inactivation, and the creation of chimeric genes via domain shuffling or gene fusion. In principle, it should be possible to infer the recombinational pathways responsible for each of these different types of genomic change by conducting detailed comparative analyses of genomic sequence data. Here, we report an attempt to unravel the complex evolutionary history of the beta-globin gene family in a taxonomically diverse set of rodent species. The main objectives were: 1) to characterize the genomic structure of the beta-globin gene cluster of rodents; 2) to assign orthologous and paralogous relationships among duplicate copies of beta-like globin genes; and 3) to infer the specific recombinational pathways responsible for gene duplications, gene deletions, and the creation of chimeric fusion genes. Results of our comparative genomic analyses revealed that variation in gene family size among rodent species is mainly attributable to the differential gain and loss of later expressed beta-globin genes via unequal crossing-over. However, two distinct recombinational mechanisms were implicated in the creation of chimeric fusion genes. In muroid rodents, a chimeric gamma/epsilon fusion gene was created by unequal crossing-over between the embryonic epsilon- and gamma-globin genes. Interestingly, this gamma/epsilon fusion gene was generated in the same fashion as the "anti-Lepore" 5'-delta-(beta/delta)-beta-3' duplication mutant in humans (the reciprocal exchange product of the pathological hemoglobin Lepore deletion mutant). By contrast, in the house mouse, Mus musculus, a chimeric beta/delta fusion pseudogene was created by a beta-globin --> delta-globin gene conversion event. Although the gamma/epsilon and beta/delta fusion genes share a similar chimeric gene structure, they originated via completely different recombinational pathways.     

Detecting gene conversion: primate visual pigment genes.

The effects of gene conversion can be detected in the DNA sequences of multigene families. We develop a permutation test of the significance of patterns of sequence mismatches, and apply it to the sequences of the red- and green-sensitive visual pigment genes of human and the diana monkey. Whereas conventional tests of the rate of sequence divergence are equivocal, the permutation test convincingly excludes divergence in the absence of gene conversion (p = 10(-6)).     

Evolution of RH genes in hominoids: characterization of a gorilla RHCE-like gene

The human RH locus is responsible for the expression of the Rh blood group antigens. It consists of two closely linked genes, RHD and RHCE, that exhibit 92% similarity between coding regions. These observations suggest that they are derived from a relatively recent duplication event. Previously a study of nonhuman primate RH-like genes demonstrated that ancestral RH gene duplication occurred in the common ancestor of man, chimpanzees and gorillas. By amplification of intron 3 and intron 4 of gorilla RH-like genes, we have now shown that, like man, gorillas possess two types of RH intron 3 (RHCE intron 3 being 289 bp longer than the RHD intron 3) and two types of intron 4 (RHCE intron 4 being 654 bp longer than the RHD intron 4). Here we report the characterization of a cDNA encoded by a gorilla RH-like gene which possesses introns 3 and 4 of the RHCE type. A comparison of this gorilla RHCE-like coding sequence with previously characterized human and ape cDNA sequences suggests that RH genes experienced complex recombination events after duplication in the common ancestor of humans, chimpanzees and gorillas.