The change from glutamic acid to valine in sickle hemoglobin was reported by Ingram (1959). Ingram (1956) had reported that the difference between hemoglobin A and hemoglobin S lies in a single tryptic peptide. His analysis of this peptide, peptide 4, was possible by the methods developed by Sanger for determining the structure of insulin and Edman's stepwise degradation of peptides.
Kan and Dozy (1978) used the HpaI restriction endonuclease polymorphism (actually the linkage principle) to make the prenatal diagnosis of sickle cell anemia (603903). As described in 143020, when 'normal' DNA is digested with HpaI, the beta-globin gene is contained in a fragment 7.6 kilobases long. In persons of African extraction 2 variants were detected, 7.0 kb and 13.0 kb long. These variants resulted from alteration in the normal HpaI recognition site 5000 nucleotides to the 3-prime side of the beta-globin gene. The 7.6 and 7.0 kb fragments were present in persons with Hb A, while 87% of persons with Hb S had the 13.0 kb variant. The method is sufficiently sensitive that the cells in 15 ml of uncultured amniotic fluid sufficed. Restriction enzyme studies indicate that whereas Hb S and Hb C originated against the same genetic background (as independent mutations) and the Hb S in the Mediterranean littoral probably is the same mutation as the West African Hb S, Hb S in Asia is apparently a separate mutation. It does not show association with the noncoding polymorphism (Kan and Dozy, 1979).
Mears et al. (1981) used the linkage of the sickle gene with restriction polymorphisms to trace the origin of the sickle gene in Africa. They found evidence that 2 different chromosomes bearing sickle genes were subjected to selection and expansion in 2 physically close but ethnically separate regions of West Africa, with subsequent diffusion to other areas of Africa. The restriction enzyme MnlI recognizes the sequence G-A-G-G, which also is eliminated by the sickle mutation. The MstII enzyme recognizes the sequence C-C-T-N-A-G-G. Predictably, the resulting fragments are larger than those produced by some other enzymes, and MstII is, therefore, particularly useful in prenatal diagnosis (Wilson et al., 1982). The sickle cell mutation can be identified directly in DNA by use of either of 2 restriction endonucleases, DdeI or MstII (Geever et al., 1981; Kazazian, 1982). The nucleotide substitution alters a specific cleavage site recognized by each of these 2 enzymes. The fifth, sixth, and seventh codons of Hb A are CCT-GAG-GAG; in Hb S, they are CCT-GTG-GAG. The recognition site for DdeI is C-T-N-A-G, in which N = any nucleoside. Chang and Kan (1982) and Orkin et al. (1982) found that the assay using the restriction enzyme MstII is sufficiently sensitive that it can be applied to uncultured amniotic fluid cells. The enzyme DdeI requires that the amniotic cells be cultured to obtain enough DNA for the assay.
Antonarakis et al. (1984) applied the Kazazian haplotype method to the study of the origin of the sickle mutation in Africans. Among 170 beta-S bearing chromosomes, 16 different haplotypes of polymorphic sites were found. The 3 most common beta-S haplotypes, accounting for 151 of the 170, were only rarely seen in chromosomes bearing the beta-A gene in these populations (6 out of 47). They suggested the occurrence of up to 4 independent mutations and/or interallelic gene conversions. By haplotype analysis of the beta-globin gene cluster in cases of Hb S in different parts of Africa, Pagnier et al. (1984) concluded that the sickle mutation arose at least 3 times on separate preexisting chromosomal haplotypes. The Hb S gene is closely linked to 3 different haplotypes of polymorphic endonuclease restriction sites in the beta-like gene cluster: one prevalent in Atlantic West Africa, another in central West Africa, and the last in Bantu-speaking Africa (equatorial, East, and southern Africa). Nagel et al. (1985) found hematologic differences between the first 2 types explicable probably by differences in fetal hemoglobin production. Ramsay and Jenkins (1987) found that 20 of 23 sickle-associated haplotypes in southern-African Bantu-speaking black subjects were the same as those found commonly in the Central African Republic, a finding providing the first convincing biologic evidence for the common ancestry of geographically widely separated speakers of languages belonging to the Bantu family. The 3 haplotypes seen with the beta-S gene in Africa are referred to as Senegal, Benin, and Bantu. The 'Bantu line' extends across the waist of Africa; south of the line, Bantu languages are spoken. Based on their study, Ramsay and Jenkins (1987) suggested that the sickle cell mutation arose only once in the Bantu speakers, presumably in their nuclear area of origin, before the Bantu expansion occurred about 2,000 years ago. In Yaounde, the capital city of Cameroon, Lapoumeroulie et al. (1992) observed a novel RFLP pattern in the study of beta-S chromosomes. This chromosome contained an A-gamma-T gene and the RFLP haplotype was different from all the other beta(S) chromosomes in both the 5-prime and 3-prime regions. All the carriers of this specific chromosome belonged to the Eton ethnic group and originated from the Sanaga river valley.
Kulozik et al. (1986) found that the sickle gene in Saudi Arabia and on the west and east coasts of India exists in a haplotype not found in Africa. They concluded that the data are most consistent with an independent Asian origin of the sickle cell mutation. The distribution of the Asian beta-S-haplotype corresponded to the reported geographic distribution of a mild clinical phenotype of homozygous SS disease. Ragusa et al. (1988) found that the beta-S gene in Sicily is in linkage disequilibrium with the Benin haplotype, the same haplotype observed among sickle cell anemia patients from Central West Africa. In addition, this haplotype is either nonexistent or very rare among nonsickling Sicilian persons. They concluded that the beta-S gene was introduced into Sicily from North Africa and that the gene flow originated in Central West Africa, traveling north through historically well-defined trans-Saharan commercial routes.
Zeng et al. (1994) indicated that 5 different haplotypes associated with Hb S had been described, 4 in Africa (Bantu, Benin, Senegal, and Cameroon) and 1 found in both India and Saudi Arabia (Chebloune et al., 1988). There is a correlation between disease severity and haplotype for at least the 2 extremes of severity: patients with the Indian/Arabian haplotype have the mildest course of disease, while those with the Bantu haplotype exhibit the most severe course. Nucleotide -530 is a binding site for a protein called BP1 (601911), which may be a repressor of the HBB gene. BP1 binds with the highest affinity to the Indian haplotype sequence and with the weakest affinity to the Bantu sequence, which might explain the differences in clinical course in these different population groups. Zeng et al. (1994) demonstrated the same sequence at -530 bp in patients with the Arabian haplotype as in Indian sickle cell anemia patients. This supports the idea of a common origin of the sickle cell mutation in individuals in India and Saudi Arabia.
Sammarco et al. (1988) presented further strong evidence that the Hb S gene in Sicily was brought by North African populations, probably during the Muslim invasions.
Currat et al. (2002) studied the genetic diversity of the beta-globin gene cluster in an ethnically well-defined population, the Mandenka from eastern Senegal. The absence of recent admixture and amalgamation in this population permitted application of population genetics methods to investigate the origin of the sickle cell mutation (Flint et al., 1993) and to estimate its age. The frequency of the sickle cell mutation in the Mandenka was estimated as 11.7%. The mutation was found strictly associated with the single Senegal haplotype. Approximately 600 bp of the upstream region of the beta-globin gene were sequenced for 94 chromosomes, showing the presence of 4 transversions, 5 transitions, and a composite microsatellite polymorphism. The sequence of 22 chromosomes carrying the sickle mutation was also identical to the previously defined Senegal haplotype, suggesting that the mutation is very recent. Maximum likelihood estimates of the age of the mutation using Monte Carlo simulations were 45 to 70 generations (1,350-2,100 years) for different demographic scenarios.
Embury et al. (1987) described a new method for rapid prenatal diagnosis of sickle cell anemia by DNA analysis. The first step involved a 200,000-fold enzymatic amplification of the specific beta-globin DNA sequences suspected of carrying the sickle mutation. Next, a short radiolabelled synthetic DNA sequence homologous to normal beta-A-globin gene sequences is hybridized to the amplified target sequence. The hybrid duplexes are then digested sequentially with 2 restriction endonucleases. The presence of the beta-A or beta-S gene sequence in the amplified target DNA from the patient determines whether the beta-A hybridization probe anneals perfectly or with a single nucleotide mismatch. This difference affects the restriction enzyme digestion of the DNA and the size of the resulting radiolabelled digestion products which can be distinguished by electrophoresis followed by autoradiography. The method was sufficiently sensitive and rapid that same-day prenatal diagnosis using fetal DNA was possible. The same test could be applied to the diagnosis of hemoglobin C disease. Hemoglobin C (Georgetown) also sickles. See Herrick (1910), Sherman (1940), Neel (1949), Pauling et al. (1949), Allison (1954), Ingram (1956, 1957, 1959), Chang and Kan (1981), and Shalev et al. (1988).
Barany (1991) described a new assay designed to detect single base substitutions using a thermostable enzyme similar to the DNA polymerase used in PCR. This enzyme, DNA ligase, specifically links adjacent oligonucleotides only when the nucleotides are perfectly base-paired at the junction. In the presence of a second set of adjacent oligonucleotides, complementary to the first set and the target, the oligonucleotide products may be exponentially amplified by thermal cycling of the ligation reaction. Because a single base mismatch precludes ligation and amplification, it will be easily distinguished. Barany (1991) demonstrated the utility of the method in discriminating between normal and sickle globin genotypes from 10 microliter blood samples.
Prezant and Fischel-Ghodsian (1992) described a trapped-oligonucleotide nucleotide incorporation (TONI) assay for the screening of a mitochondrial polymorphism and also showed that it could distinguish the genotypes of hemoglobins A/C, A/A, A/S, and S/S. The method was considered particularly useful for diagnosing mutations that do not produce alterations detectable by restriction enzyme analysis. It also requires only a single oligonucleotide and no electrophoretic separation of the allele-specific products. It represents an improved and simplified modification of the allele-specific primer extension methods. (TONI, the acronym for the method, is also the given name of the first author.)
Grosveld et al. (1987) identified dominant control region (DCR) sequences that flank the human beta-globin locus and direct high-level, copy-number-dependent expression of the human beta-globin gene in erythroid cells in transgenic mice. By inserting a construct that included 2 human alpha genes and the defective human beta-sickle gene, all driven by the DCR sequences, Greaves et al. (1990) produced 2 mice with relatively high levels of human Hb S in their red cells. Use of this as an animal model for the study of this disease was suggested.
Turhan et al. (2002) presented evidence suggesting that a pathogenetic mechanism in sickle cell vasoocclusion may reside in adherent leukocytes. Using intravital microscopy in mice expressing human sickle hemoglobin, they demonstrated that SS red blood cells bind to adherent leukocytes in inflamed venules, producing vasoocclusion of cremasteric venules. SS mice deficient in P- and E-selectins, which display defective leukocyte recruitment to the vessel wall, were protected from vasoocclusion. Thus, drugs targeting SS RBC-leukocyte or leukocyte-endothelial interactions might prevent or treat the vascular complications of this disease.
Nitric oxide (NO), essential for maintaining vascular tone, is produced from arginine by NO synthase. Plasma arginine levels are low in sickle cell anemia, and Romero et al. (2002) reported that the sickle transgenic mouse model has low plasma arginine. They supplemented these mice with a 4-fold increase in arginine over a period of several months. Mean corpuscular hemoglobin concentration decreased and the percent high-density red cells was reduced. Romero et al. (2002) concluded that the major mechanism by which arginine supplementation reduces red cell density in these mice is by inhibiting the Ca(++)-activated K(+) channel.
In a Jamaican study, Serjeant et al. (1968) described 60 patients with homozygous sickle cell disease who were 30 years of age or older, and Platt et al. (1994) estimated a median survival of 42 to 48 years. Serjeant et al. (2007) stated that the sickle cell clinic at the University of West Indies had treated 102 patients (64.7% women) who survived beyond their 60th birthday. None of the patients received hydroxyurea, and only 2 patients with renal impairment received regular transfusions. The ages of the patients ranged from 60.2 to 85.6 years. Measurement of fetal hemoglobin levels suggested that higher fetal hemoglobin levels probably conferred protection in childhood. The major clinical problems emerging with age were renal impairment and decreased levels of hemoglobin.
Kwiatkowski (2005) noted that HbS homozygotes have sickle-cell disease, whereas heterozygosity confers a 10-fold increase in protection from life-threatening malaria (611162) and lesser protection against mild malaria.
Cholera et al. (2008) found that P. falciparum (Pf)-infected HbA/HbS erythrocytes did not bind to microvascular endothelial cells as well as Pf-infected HbA/HbA erythrocytes. Reduced binding correlated with altered display of the major Pf cytoadherence ligand on erythrocyte membranes. Cholera et al. (2008) noted that this protective mechanism had features in common with that of HbC (141900.0038), and they suggested that weakening of cytoadherence interactions may influence the degree of malaria protection in HbA/HbS children.
Modiano et al. (2008) adopted 2 partially independent haplotypic approaches to study the Mossi population in Burkina Faso, where both the HbS and HbC alleles are common. They showed that both alleles are monophyletic, but that the HbC allele has acquired higher recombinatorial and DNA slippage haplotypic variability or linkage disequilibrium decay and is likely older than HbS. Modiano et al. (2008) inferred that the HbC allele has accumulated mainly through recessive rather than a semidominant mechanism of selection.
Gouagna et al. (2010) used cross-sectional surveys of 3,739 human subjects and transmission experiments involving 60 children and over 6,000 mosquitoes in Burkina Faso, West Africa, to test whether the HBB variants HbC and HbS, which are protective against malaria, are associated with transmission of the parasite from the human host to the Anopheles mosquito vector. They found that HbC and HbS were associated with significant 2-fold in vivo (P = 1.0 x 10(-6)) and 4-fold ex vivo (P = 7.0 x 10(-5)) increases of parasite transmission from host to vector. In addition, mean oocyte densities were particularly high in mosquitoes fed from HbS carriers.
Ferreira et al. (2011) demonstrated that wildtype mice or mice expressing normal human Hb, but not mice expressing Hbs, developed experimental cerebral malaria (ECM) 6 to 12 days after infection with the murine malaria parasite, Plasmodium berghei. The Hbs mice eventually succumbed to the unrelated condition of hyperparasitemia-induced anemia. Tolerance to Plasmodium infection was associated with high levels of Hmox1 (141250) expression in hematopoietic cells, and mice expressing Hbs became susceptible to ECM when Hmox1 expression was inhibited. Hbs induced expression of Hmox1 in an Nrf2 (NFE2L2; 600492)-dependent manner, which inhibited the production of chemokines and Cd8-positive T cells associated with ECM pathogenesis. Ferreira et al. (2011) concluded that sickle hemoglobin suppresses the onset of ECM via induction of HMOX1 and the production of carbon monoxide, which inhibits the accumulation of free heme, affording tolerance to Plasmodium infection.
Cyrklaff et al. (2011) found that HbS and HbC affect the trafficking system that directs parasite-encoded proteins to the surface of infected erythrocytes. Cryoelectron tomography revealed that P. falciparum generates a host-derived actin cytoskeleton within the cytoplasm of wildtype red blood cells that connects the Maurer clefts with the host cell membrane and to which transport vesicles are attached. The actin cytoskeleton and the Maurer clefts were aberrant in erythrocytes containing HbS or HbC. Hemoglobin oxidation products, enriched in HbS and HbC erythrocytes, inhibited actin polymerization in vitro and may account for the protective role in malaria.
