Second generation knockout sickle mice: the effect of HbF.

Sickle transgenic mice expressing exclusively human globins are desirable for studying pathophysiology and testing gene therapy strategies, but they must have significant pathology and show evidence of amelioration by antisickling hemoglobins. Mice were generated that expressed exclusively human sickle hemoglobin with 3 levels of HbF using their previously described sickle constructs (cointegrated human miniLCRalpha2 and miniLCRbeta(S) [PNAS 89:12150, 1992]), mouse alpha- and beta-globin-knockouts, and 3 different human gamma-transgenes. It was found that, at all 3 levels of HbF expression, these mice have balanced chain synthesis, nearly normal mean corpuscular hemoglobin, and, in some cases, F cells. Mice with the least adult HbF expression were the most severe. Progressive increase in HbF from less than 3% to 20% to 40% correlated with progressive increase in hematocrit (22% to 34% to 40%) and progressive decrease in reticulocyte count (from 60% to 30% to 13%). Urine concentrating ability was normalized at high HbF, and tissue damage detected by histopathology and organ weight were ameliorated by increased HbF. The gamma-transgene that produces intermediate levels of HbF was introduced into knockout sickle mice described by Pàszty and coworkers that express the miniLCRalpha1(G)gamma(A)gammadeltabeta(S) transgene and have fetal but not adult expression of HbF. It was found that the level of HbF required to ameliorate low hematocrit and normalize urine concentrating defect was different for the miniLCRalpha2beta(S) and miniLCRalpha1(G)gamma(A)gammadeltabeta(S) mice. We conclude that knockout mice with the miniLCRalpha2beta(S) transgene and postnatal expression of HbF have sufficiently faithful sickle pathology to serve as a platform for testing antisickling interventions.

Hemoglobin C in transgenic mice: effect of HbC expression from founders to full mouse globin knockouts.

When present in the homozygous form, hemoglobin C (HbC, CC disease) increases red cell density, a feature that is the major factor underlying the pathology in patients with SC disease (Fabry et al., JCI 70, 1315, 1982). The basis for the increased red cell density has not yet been fully defined. We have generated a HbC mouse in which the most successful founder expresses 56% human alpha and 34% human beta(C). We introduced knockouts (KO) of mouse alpha- and beta-globins in various combinations. In contrast to many KO mice, all partial KOs have normal MCH. Full KOs that express exclusively HbC and no mouse globins have minimally reduced MCH (13. 7 +/- 0.3 pg/cell vs 14.5 +/- 1.0 for C57BL/6) and a ratio of beta- to alpha-globin chains of 0.88 determined by chain synthesis; hence, these mice are not thalassemic. Mice with beta(C) > 30% have increased MCHC, dense reticulocytes, and increased K:Cl cotransport. Red cell morphology studied by SEM is strikingly similar to that of human CC cells with bizarre folded cells. We conclude that red cells of these mice have many properties that closely parallel the pathology of human disease in which HbC is the major determinant of pathogenesis. These studies also establish the existence of the interactions with other gene products that are necessary for pleiotropic effects (red cell dehydration, elevated K:Cl cotransport, morphological changes) that are also present in these transgenic mice, validating their usefulness in the analysis of pathophysiological events induced by HbC in red cells.

Anti-beta s-ribozyme reduces beta s mRNA levels in transgenic mice: potential application to the gene therapy of sickle cell anemia.

Our current strategy for gene therapy of sickle cell anemia involves retroviral vectors capable of transducing "designer" globin genes that code for novel anti-sickling globins (while resisting digestion by a ribozyme), coupled with the expression of a hammerhead ribozyme that can selectively cleave the human beta s mRNA. In this report, we have tested in vivo an anti-beta s hammerhead ribozyme embedded within a cDNA coding for the luciferase reporter gene driven by the human beta-globin promoter and hyper-sensitive sites 3 and 4 of the locus control region. We have created mice transgenic for this luciferase-ribozyme construct and bred the ribozyme transgene into mice that were already transgenic for the human beta s gene. We then measured expression of the beta s transgene at the protein and RNA levels by HPLC and primer extension. The presence of the ribozyme was associated with a statistically significant reduction in the level of beta s mRNA in spleen stress reticulocytes (from 60.5 +/- 4.1% to 52.9 +/- 4.2%) and in the percentage of beta s globin chains in very young mice (from 44.5 +/- 0.6% to 40.8 +/- 0.7%). These results demonstrate that it is possible to decrease the concentration of beta s chains and mRNA with the help of a hammerhead ribozyme. While the enormous amount of globin mRNA in reticulocytes is a challenge for ribozyme technology, the exquisite dependence of the delay time for formation of Hb S nuclei on the concentration of Hb S in red blood cells suggests that even a modest reduction in Hb S concentration would have therapeutic value.

Fetal hemoglobin expression in compound heterozygotes for -117 (G-->A)A gamma HPFH and beta zero 39 nonsense thalassemia.

The -117(G-->A)A gamma hereditary persistence of fetal hemoglobin (Greek HPFH) and beta zero 39-thal mutations are rather frequent in Sardinia so that their interaction is to be expected. Characterization of eight compound heterozygotes for these defects indicated that HPFH was linked to haplotype VII and beta zero 39-thal to haplotype II. Haplotype II beta zero 39-thal chromosome carries the A gamma T gene which is a useful marker of gamma-gene expression. Since the Hb F level in these compound heterozygotes was significantly higher than in 46 -117 HPFH carriers, the A gamma I, A gamma T, and G gamma globin level was determined. A gamma T was underexpressed while G gamma was significantly increased, which suggest that in -117 A gamma HPFH/beta zero 39-thal healthy subjects the increase in Hb F production is determined only by the -117 mutated A gamma gene and the adjacent G gamma gene.

Molecular cloning of mouse erythrocyte protein 4.2: a membrane protein with strong homology with the transglutaminase supergene family.

We report the molecular cloning and characterization of mouse erythrocyte protein 4.2 (P4.2). Mouse erythrocyte P4.2 is a 691-amino-acid protein with a predicted MW of 77 kDa. Northern blot analysis detected a 2.2-kb transcript in mouse reticulocytes, compared with a 2.4- to 2.5-kb transcript in human reticulocytes, which is consistent with the absence of the 30-amino-acid splicing insert in mouse erythrocyte P4.2 that is found in the human protein (isoform I). Like the human erythrocyte P4.2, mouse erythrocyte P4.2 contains regions strikingly homologous with the transglutaminase (TGase) proteins although it too most likely lacks TGase crosslinking activity. Mouse P4.2 is on average 73% identical with human erythrocyte P4.2, although regional variations exist, with greatest conservation in the regions of the molecule that contain the TGase active site, the TGase calcium-binding site, and a band 3 binding site. Hydropathy analysis reveals a protein containing a series of hydrophobic domains, similar to the situation for human P4.2 and consistent with its tight binding to the membrane, although the mouse P4.2 is missing both the strongly hydrophilic region and adjacent highly charged region that are present in the human protein, suggesting that the two proteins could differ in their physical characteristics, binding associations, or functional properties. The availability of the complete mouse erythrocyte P4.2 cDNA should help in the design of P4.2-deficient animal models (for example, ribozyme or homologous recombinant "knockout" models) that should accelerate the understanding of P4.2 function in both erythroid and non-erythroid cells.

Mild beta+(-87)-thalassemia CACCC box mutation is associated with elevated fetal hemoglobin expression in cis.

The beta zero-thalassemia codon 39 nonsense mutation predominant in Sardinia is severe, and homozygotes are transfusion dependent. Two-thirds of beta zero 39 alleles are linked to A gamma T (haplotype II). One-fourth are linked to A gamma I (haplotypes I and IX), as is the mild beta +-thalassemia -87 C-->G mutation (haplotype VIII). beta +/beta zero-Thalassemia VIII/II compound heterozygotes have significantly higher A gamma I:A gamma T (23:7) than beta zero-thalassemia I/II (24:20) or IX/II (16:17) cases. This suggests that the beta + -87 mutation is associated with elevated gamma expression in cis, which may contribute to the lack of transfusion-dependence in beta +/beta zero cases.

Direct demonstration that the A gamma T globin gene is linked to the 4 bp promoter deletion in the beta A chromosome of sickle cell traits.

In beta zero-thalassemia and sickle cell patients, a 4 bp deletion at -222 to -225 of the A gamma globin promoter was associated with low expression of the A gamma T variant (threonine at codon 75 of A gamma), whereas A gamma I (isoleucine at 75) had the normal A gamma promoter and higher expression. However, it has been reported that the beta A chromosomes of sickle cell trait cases have the 4 bp deletion as a common polymorphism unlinked to the A gamma T allele. We now present data demonstrating the association of the A gamma T allele with the 4 bp deletion in beta A chromosomes of sickle cell traits.

The 32.6 kb Indian delta beta-thalassaemia deletion ends in a 3.4 kb L1 element downstream of the beta-globin gene.

The Indian delta beta-thalassaemia, with elevated fetal gamma globin gene expression, was previously found to have a large deletion beginning 1 kb 3' of the (A) gamma globin gene at GenBank HUMHBB coordinate 42151, and extending into a new L1 sequence. We have now determined the 3' breakpoint of this deletion, and in doing so we have extended the known beta-globin gene cluster DNA sequence from its end at 73326 to projected GenBank coordinate 79016. These data show that the deletion is 32.6 kb long, terminating 11 kb 3' of the beta-globin gene. This 3' breakpoint is at 74772, within a 3.4 kb partial L1 repeat at 74263-77665; the Black ((A) gamma delta beta)(0)-thalassaemia also terminates in this L1, at 76508. In addition, two Alu sequences were found, at 73692-73816 and 78171-78441. Among the protein-binding DNA sequence motifs 3' to the Indian delta beta-thalassaemia breakpoint, at 76581/76607 there is a TGATAA/ACACCC pair that binds the erythroid-specific GATA-1 and ubiquitous CACCC-box binding proteins. We hypothesize that elevated fetal haemoglobin may be due to an enhancer or enhancers 3' to the deletion breakpoints and may involve the TGATAA/ACACCC pair.

G gamma and A gamma globin genes are identical from -471 of the promoter midway through gamma IVSII in a Benin beta s haplotype associated with elevated fetal hemoglobin.

In seven kindreds in which sickle cell (SS) patients had elevated (greater than 12%) fetal hemoglobin (Hb F), Milner and colleagues reported that a determinant for elevated Hb F and elevated F cells was linked to the beta s gene. Independently, the Senegal (SEN) beta s haplotype has been found in association with elevated Hb F in SS and beta-thalassemia patients. We have used the kindreds of Milner and colleagues to characterize further the association of haplotype and gamma gene DNA sequence variation with Hb F expression. For the largest kindred, Wi, all four SS had high (greater than 14%) Hb F and both SEN and Benin (BEN) haplotypes. Two AS cases carrying SEN had low Hb F and low F cells, while three AS and one CS carrying BEN had elevated Hb F and elevated F cells; only one AS carrying BEN had low Hb F and low F cells. In order to look for genetic alterations that could account for the elevated Hb F of kindred Wi, we sequenced both the G gamma and A gamma genes of the Wi BEN haplotype. The data showed largely identical G gamma and A gamma genes which may have been generated by two gene conversions: the A gamma promoter was like that of G gamma 3' to -471, while the G gamma IVSII was like that of A gamma in its 5' half. In addition, three new mutations were found in gamma IVSII.(ABSTRACT TRUNCATED AT 250 WORDS)

Diminished A gamma T fetal globin levels in Sardinian haplotype II beta 0-thalassaemia patients are associated with a four base pair deletion in the A gamma T promoter.

In Sardinia, the beta-39 nonsense mutation is the primary cause of beta 0-thalassaemia. This mutation is found mainly on beta-globin gene cluster haplotypes I and II, which differ in their A gamma globin types (A gamma I and A gamma T, respectively). This report presents data on G gamma, A gamma I and A gamma T levels, and the presence or absence of a 4 base pair (bp) deletion at -225 to -222 of the A gamma globin promoter, in 55 poly-transfused beta 0-thalassaemia major patients. Six patients were homozygotes for the normal (N) A gamma promoter lacking the 4 bp deletion, had no A gamma T globin, and their mean G gamma:A gamma I: A gamma T ratio was 52.9:47.1:0. Twenty-five patients were homozygotes for the mutant (M) A gamma promoter with the 4 bp deletion, had no A gamma I globin, and the mean G gamma:A gamma I: A gamma T ratio was 62.1:0:37.9. For M/M compared to N/N, the lower A gamma T than A gamma I was significant by the t-test (P less than 0.001). Twenty-four N/M cases had mean G gamma:A gamma I:A gamma T of 56:24.4:19.6, and the lower A gamma T than A gamma I was also significant (P less than 0.001). Partial haplotype analysis on these and 17 other beta 0-thalassaemia patients suggested that the 4 bp deletion was strongly associated with haplotype II. Of 33 M/M, 32 were haplotype II/II and one was II/5a; of 31 N/M, 29 were I/II and two were II/IX; of eight N/N, seven were haplotype I/I and one was I/IX. These data show a strong association of the 4 bp promoter deletion with decreased expression of the A gamma T globin gene on haplotype II.

Polymerase chain reaction amplification applied to the direct detection of a 4 bp deletion in the promoter region of the A gamma gene.

We report here the identification of a 4 bp deletion in the A gamma T globin gene promoter by means of Fnu4HI digestion of DNA amplified by polymerase chain reaction (PCR). This deletion has been previously associated with haplotype II beta-thalassemia in Sardinia. This simple, non-radioactive procedure should facilitate the screening of various populations of normal and beta-thalassemic subjects for this specific genetic alteration.

The DNA deletion in an Indian delta beta-thalassaemia begins one kilobase from the A gamma globin gene and ends in an L1 repetitive sequence.

High fetal haemoglobin levels of 5-15% are present in adult heterozygotes for delta beta-thalassaemia as the result of large deletions of DNA. We have cloned DNA spanning the deletion breakpoint for a new Indian delta beta-thalassaemia associated with mild anaemia. The 5' breakpoint is at 42151 of GenBank file HUMHBB, which is about 1 kb 3' of the A gamma globin gene poly A site at 41003. On the 3' side of the breakpoint, the sequence is homologous to L1 (KpnI) repetitive DNA located 3.6-10 kb 3' of the beta-globin gene: Indian delta beta-thalassaemia DNA is 74% homologous to the inverted complement of HUMHBB from 69849 to 70020, followed by a region 78% homologous to the direct sequence of HUMHBB from 70534 to 71010. The precise location of the 3' endpoint of this deletion has not been determined, but it is within L1 sequences located more than 10 kb 3' of the beta-globin gene.

Distal CCAAT box deletion in the A gamma globin gene of two black adolescents with elevated fetal A gamma globin.

Point mutations in G gamma and A gamma globin gene promoters are associated with increased production of G gamma and A gamma globin, respectively. To determine whether an upstream promoter mutation could account for elevated A gamma in a Black adolescent with A gamma-beta+-HPFH and sickle cell trait, we cloned the 13 kb BglII fragment containing both gamma genes into phage lambda vector EMBL3. For one clone, the A gamma upstream promoter showed no hybridization to a 19 bp oligonucleotide whose sequence centered at -117. A gamma promoter sequence data for this mutant clone revealed a 13 bp deletion which eliminated the A gamma distal CCAAT box. Amplified A gamma genomic DNA of this and a similar case showed hybridization to both deletion-mutant and normal oligonucleotide probes. We propose that this 13 bp deletion removes part of the binding site for a repressor protein which is abundant in adult erythroid cells.

Four categories of gamma-globin gene triplications: DNA sequence comparison of low G gamma and high G gamma triplications.

The human fetal gamma chains are produced by closely linked G gamma and A gamma genes, and unequal crossing over between them leads to gamma gene deletions and triplications. Nine gamma gene triplications from seven ethnic groups were analyzed for G gamma and hemoglobin F (Hb F) values of heterozygotes and for the presence of polymorphic XmnI restriction sites 5' to the gamma genes. Four categories of triplication were found: I had low G gamma and low Hb F values and lacked XmnI sites 5' to the three gamma genes [---]. II had high G gamma and slightly elevated Hb F values but was also [---]. III was similar to II, except that XmnI was [+--]. IV had very high G gamma and slightly elevated Hb F values, and XmnI was [++-]. One case each of triplications I and IV were cloned into Charon 35. For both, the two 5' gamma gene code for G gamma chain, while the 3' gamma gene codes for A gamma chain. DNA sequencing showed that the unequal crossover occurred between 472 and 398 base pairs (bp) 5' to the gamma gene Cap sites (-472 and -398) for the type IV triplication and between -271 and codon 136 for the type I triplication. In addition, type I had a 4-bp deletion of AGCA from -225 to -222. The high G gamma values of the type IV triplication are explained by its -G gamma-G gamma-A gamma-gene arrangement and the XmnI sites 5' to the G gamma genes. We hypothesize that the low G gamma value of the type I triplication, which is also -G gamma-G gamma-A gamma-, is due to inactivation of the middle G gamma gene by the AGCA deletion at -225 to -222.

Upstream promoter mutation associated with a modest elevation of fetal hemoglobin expression in human adults.

In hereditary persistence of fetal hemoglobin, Hb F (alpha 2 gamma 2) is elevated after birth. Screening of sickle cell patients has revealed a family with elevated Hb F and high A gamma values. The propositus was a sickle cell patient with approximately 25% Hb F and 68.4% A gamma. He was heterozygous for the Benin (#19) and Mor beta S haplotypes. Five AS relatives with the Mor haplotype had 2.5% +/- 0.9% fetal hemoglobin and 92.8% +/- 2.8% A gamma, whereas two with the Benin haplotype had normal fetal hemoglobin (0.5%). The Mor haplotype is thus associated with the elevated Hb F in this family. The 13-kilobase (kb) Bg/II fragment containing the G gamma and A gamma genes of the Mor haplotype was cloned, and the G gamma and A gamma promoters sequenced from -383 to beyond the Cap sites. The Mor G gamma gene was normal, but the A gamma gene had a unique C----T mutation at -202. A different mutation at -202 of G gamma (C----G) was previously detected by other researchers in association with considerably higher Hb F in AS cases (15% to 25%). These data suggest either that -202 mutations affect the G gamma and A gamma promoters differently or that different nucleotide substitutions at -202 have divergent effects. Alternatively, additional unknown mutations could cause the differences in gene expression.

Four base-pair DNA deletion in human A gamma globin-gene promoter associated with low A gamma expression in adults.

Fetal haemoglobin (alpha 2 gamma 2) is predominant in red cells of the fetus and newborn baby, and is largely replaced after birth by adult haemoglobin (alpha 2 beta 2). The two types of gamma chains (A gamma and G gamma) are generally less than 1% of total beta-like chain in adults, and the G gamma: A gamma ratio is typically 40:60. Higher G gamma values (greater than 50% of gamma chain) are frequently associated with a T for C nucleotide substitution 158 base pairs 5' of the G gamma Cap site (-158). The first exception to this rule was a beta o-thalassaemia in a Black family that was associated with about 60% G gamma in heterozygotes. A DNA fragment containing the G gamma and A gamma genes of the high G gamma haplotype of this case has now been cloned. DNA sequencing from -383 to the Cap site showed no differences from normal for the G gamma gene, except for C at -158. For the A gamma gene, however, a deletion of four base pairs (AGCA) at -222 to -225 was found. It is hypothesized that this deletion causes reduced A gamma globin gene expression in adults, which suggests that promoter elements important for the regulation of fetal haemoglobin expression in adults extend upstream at least to -225.

Expression of G gamma and A gamma globin genes in human adults.

Expression of the human fetal G gamma and A gamma globin genes declines shortly after birth, and adults generally have less than 1% fetal hemoglobin or Hb F (alpha 2 gamma 2). However, some adults with hereditary persistence of fetal hemoglobin (HPFH) have elevated expression of either the G gamma or A gamma gene due to a mutation in its upstream promoter. Mutations with strong effects on expression have been found at -175 and -202 of the G gamma gene and at -117, -196, -198 and -202 of the A gamma gene. Mutations at -158 and -161 of G gamma have weaker effects, which are observable primarily as increases in the G gamma:A gamma ratio. Published data are reviewed which suggest that the -158 mutation may lead to observable elevations of Hb F in SS and beta(0)-thal patients and occasionally in normal non-anemic individuals. These data also suggest that additional high Hb F determinants are linked to Benin, Bantu and Asian beta S haplotypes in some instances. A model based on data from SV40 is presented which suggests that specific DNA sequence motifs of the gamma globin gene may bind regulatory proteins. It is proposed that the -158 and -161 mutations have weak effects because they are located on the fringe of regulatory sequence motifs.

The 12.6 kilobase DNA deletion in Dutch beta zero-thalassaemia.

The Dutch beta zero-thalassaemia has few clinical symptoms in homozygotes, elevated fetal haemoglobin (4-11%) in heterozygotes, and has a DNA deletion previously estimated as 10 kb which removes the beta-globin gene (Gilman et al, 1984). A DNA fragment containing the breakpoints of the Dutch beta zero-thalassaemia deletion has now been cloned. Sequencing across the deletion junction region showed the 3' endpoint to be about 3 kb further 3' than originally thought, so that the deletion covers 12.6 kb. The 3' endpoint lies in a region of Kpn I (L1) repeated sequences, which is also the case for several other deletions. A six bp region of homology (AAATTT) between the 5' and 3' normal sequences at the breakpoint may have contributed to the non-homologous recombination event that led to the Dutch beta zero-thalassaemia deletion. The 12.6 kb Dutch beta zero-thalassaemia deletion is now seen to be a member of a 12-13 kb size category of deletions which also includes two delta beta-thalassaemias.

DNA sequence analysis of the Dutch beta zero-thalassemia deletion.

The Dutch beta zero-thalassemia has relatively few clinical symptoms in the homozygote. Heterozygotes have levels of fetal hemoglobin (4-11%) comparable to delta beta-thalassemia, and much higher than in typical beta zero-thalassemia. To analyze this deletion, we have cloned the abnormal 10 kb Bgl II fragment that contains the delta-globin gene into phage lambda vector EMBL4. A Hind III-Xba I fragment that includes the endpoints of the deletion was subcloned into plasmid pUC 19 and partially sequenced by the dideoxy method. The sequence of the Dutch beta zero-thalassemia DNA is like that of normal 5' DNA to 8920 (the numbering system is from reference 2). The sequence from 8915-8920 is AAATTT, which is also the normal 3' sequence from 21537-21542. From this point on, the Dutch beta zero-thalassemia sequence is like that of the normal 3' DNA. The deletion is therefore 12.6 kilobases, rather than 10 as originally estimated. It is similar to several other beta-globin gene deletions in terminating in a region of Kpn I repeated sequences, and having several base pairs of homology between 5' and 3' sequences at the break points.