Free-breathing T2* mapping for MR myocardial iron assessment at 3 T.

BACKGROUND: Timely diagnosis of cardiac iron overload is important for children with transfusion-dependent anaemias and requires modern measure methods. Nowadays, myocardial iron quantification is performed by magnetic resonance (MR) breath-hold techniques, sensitive to respiratory motion and unfeasible in patients who are unable to hold their breath. Free-breathing T2* mapping sequences would allow to scan children who cannot hold their breath for a specified duration. Our aim was to test a free-breathing T2* mapping sequence, based on motion correction by multiple signal accumulation technique. METHODS: We used an electrocardiographically gated T2* mapping sequence based on multiple gradient echo at 3-T in 37 paediatric patients with haematologic disorders aged from 2 to 16. We compared T2* values of myocardium and signal-to-noise ratio of this new sequence with standard breath-holding T2* mapping sequence. T2* values were measured in the interventricular septum for both methods in studies with adequate image quality. RESULTS: All children were scanned without complications. Five patients were excluded from analysis because of the presence of respiratory artefacts on the T2* images with breath-holding technique due to patient's inability to hold their breath. Breath-holding T2* was 19.5 ± 7.7 ms (mean ± standard deviation), free-breathing T2* was 19.4 ± 7.6 ms, with positive correlation (r = 0.99, R2 = 0.98; p < 0.001). The free-breathing sequence had a higher signal-to-noise ratio (median 212.8, interquartile range 148.5-566.5) than the breath-holding sequence (112.6, 71.1-334.1) (p = 0.03). CONCLUSION: A free-breathing sequence provided accurate measurement of myocardial T2* values in children.

The hemostasis system in children with hereditary spherocytosis.

INTRODUCTION: Patients with hereditary spherocytosis (HS) are characterized by having an increased risk for thrombosis. An early manifestation of thrombotic complications can occur even in childhood, especially after surgery. Hypercoagulability can be associated with hemolytic crises. AIM: The aim of this study was to investigate the hemostatic state in children with HS using global hemostasis assays. METHODS: The hemostatic status of 62 children (38 boys and 24 girls; age range: 0.5 to 17 years) with HS during and without hemolytic crisis was assessed using clotting times (APTT, TT, and PR), fibrinogen and D-dimer levels, and global hemostasis, thromboelastography (TEG) and thrombodynamics (TD) assays. One hundred and two healthy children undergoing annual medical examination were enrolled as a control group. RESULTS: TEG and TD parameters were increased in the children with HS compared to the control group (60 ±â€¯5 mm vs. 53 ±â€¯4 mm, p < 0.05 for TEG maximum amplitude; 28 ±â€¯3 µm/min vs. 24 ±â€¯2 µm/min, p < 0.05 for TD clot growth rate), while APTT, TT and PR were not significantly different between the two groups. Patients with HS were divided into 2 groups: those during hemolytic crisis (28 patients) and those without hemolytic crisis (34 patients). TEG and TD parameters were increased in those during hemolytic crisis compared to the steady state HS group (62 ±â€¯5 mm vs. 57 ±â€¯4 mm, p < 0.05 for TEG maximum amplitude; 31 ±â€¯4 µm/min vs. 26 ±â€¯3 µm/min, p < 0.05 for TD clot growth rate). The D-dimer levels were increased in 4 HS patients, for whom the activation of blood clotting was noted. Fibrinogen levels were decreased in patients with HS compared to the control group (2.1 ±â€¯0.4 mg/ml vs. 2.6 ±â€¯0.4 mg/ml, p < 0.05). Other tests were within the reference ranges for both groups. CONCLUSIONS: The global hemostasis tests TEG and TD revealed hypercoagulability in patients with HS. More dramatic changes were observed in patients experiencing a hemolytic crisis.

[THE CYTOMETRIC TECHNIQUE OF BINDING OF EOSIN-5-MALEIMIDE IN DIAGNOSTIC OF INHERENT SPHEROCYTOSIS].

The laboratory diagnostic of inherent spherocytosis is based on detection of spherocytes in peripheral blood, decreasing of index of sphericity, decreasing of osmotic resistance of erythrocytes. The new test of diagnostic of hereditary spherocytosis build on molecular defect was developed on the basis of binding extracellular fragments of protein of band 3 with eosin-5-maleimide (EMA-test). The study was carried out to implement comparative analysis of sensitivity and specificity of techniques applied to diagnose inherent spherocytosis. The sampling of 94 patients with various forms of anemias was analyzed All patients were applied complex clinical laboratory examination including analysis of osmotic resistance of erythrocytes, erythrocytometry and EMA-test as specific techniques of diagnostic of inherent spherocytosis. In 51 out of 94 patients (54%) decreasing of values of EMA-test was detected and in 47 patients diagnosis of inherent spherocytosis was confirmed. The standard values of EMA-test were established in 43 patients (46%) and 12 patients out of them with established diagnosis of inherent spherocytosis. Therefore, sensitivity of EMA-test made up to 79% and specificity - 80%. The most sensitive techniques of diagnostic remain osmotic resistance of erythrocytes (91%) and index of sphericity (up to 96%). But the highest specificity in this respect has EMA-test (80%). Nowadays, none of implemented techniques of diagnostic of inherent spherocytosis can be applied as a universal one. The implementation of complex examination is needed for proper diagnostic of disease.

[Iron deficiency anemia and anemia in chronic celiac disease in children].

UNLABELLED: Anemia in celiac disease (CD) may be associated with deficiency of iron, vitamins, macro- and micronutrients. It is also possible the development of chronic disease anemia (CDA), associated with activation of proinflammatory cytokines. The aim of this work is to optimize the diagnosis and treatment of celiac disease in children on the basis of study of iron metabolism disorders heterogeneity, including the role of CDA. PATIENTS AND METHODS: We observed 34 children with CD aged 1.5 to 17 years, 27 of them children were observed both in the active stage of the disease and in remission. The control group included 25 children aged 1.2 to 17 years who were previously excluded for any chronic (including autoimmune) disease: these children were observed with chronic functional gastrointestinal motility disorders. Special methods of examination were study of iron metabolism, including the determination of the serum hepcidin level, as well as the determination of the serum proinflammatory cytokines (tumor necrosis factor [TNF] -α, interleukin [IL] -2, IL-6, IL-10). RESULTS AND DISCUSSION: In active CD in 14.71% of children anemia of varying severity (mild or moderate) were diagnosed. Among these children we observed mild decrease of serum iron in the range 2.2-8.0 g/ml in 15 of 34 children (44%) and a marked reduction in serum ferritin level in 59% of children. In the active celiac disease in the majority of children there is a decrease in the serum hepcidin, but approximately in 20% of children serum hepcidin level was increased. This indicates the development of CDA in these children. During the active stage the average values of IL-2 was significantly increased (p < 0.05). Thus, the iron metabolism disorders in celiac disease is a result of immunopathological process which results in a reduction in iron absorption in the gut due to the intestinal mucosa villous atrophy and to improve the hepcidin production by liver cells and iron depot blocking with the CDA development in 20% of children.

[Automatic erythrocytometry in a robotized microscope MEKOS-Ts1]. / Avtomaticheskaia éritrotsitometriia v robotizirovannom mikroskope MEKOS-Ts1.

Measurements of the form and size of erythrocytes are needed in the diagnosis of a number of diseases. However, such measurements, if made manually, are a labor-consuming and often inaccurate method, which ensures the determination of a very limited number of parameters. Hardware/software unit MEKOS-C1 enables an automated examination of blood smear, thus speeding up significantly the analysis and ensuring a more complete and accurate information. The possibilities of unit MEKOS-C1 were evaluated for the diagnosis of ovalocytosis. Blood smears of 19 patients from the Russian Pediatric Clinical Hospital, including 8 patients with inherited ovalocytosis, 1 patient with spherocytic ovalocytosis and 10 patients without the disease, were made use of Erythrocytes were isolated in the images of preparations and the contour of each cell was approximated by ellipse. The ratio between ellipse semi-axes served as a measure of erythrocytes' ovality. The mean ratio of semi-axes (RS) and the index of ovalocytosis (IO), i.e. a ratio of the mean maximum diameter to the mean minimal diameter, were calculated for each smear. Manual IO measurements were made in all preparations as a control. Since an additional error can enter the result because of the irregular smear nature and impossibility to standardize completely the technology of smear preparation, the data, obtained from two different smear parts and from parallel preparations, were compared. The reliability, stability and good reproducibility of the automated measurement results were demonstrated. The mean erythrocytes' RS correlated well with IO, obtained manually, and did not virtually differ from RS, measured in the automated manner. The mean RS value of erythrocytes, obtained from patients with inherited ovalocytosis, significantly differed from the control values, which is indicative of a high information density of the discussed parameter. Therefore, RS, when measured automatically, is a reliable and convenient characteristic of erythrocytes' ovality under the conditions of using the ordinary technique of smear preparation.

[Soluble transferrin receptors: significance and diagnostic value in anemia]. / Rastvorimye retseptory transferrina: znachenie v diagnostike anemii.

The study was undertaken for the purpose of demonstrating a diagnostic importance of determining the level of soluble transferrin receptors not only in the diagnosis of iron-deficiency conditions (IDC) or in the differential diagnosis of anemia in a chronic disease (ACD) and of iron-deficiency anemia (IDA) but also in making a diagnosis in case of transfusion-dependent patients with Cooley's anemia. The iron metabolism (including determination of the serum iron (SI), of the total iron-binding ability (TIBA), of saturation of transferrin with iron (STFI), of serum transferrin (ST), and of soluble transferrin receptors (s-STR) was examined in 31 patients with different-genesis anemias, aged 3 to 7. Diagnoses were made for all children on the basis of the standard clinical-and-laboratory examinations and tests. A pattern of iron-metabolism parameters, typical of this pathology, was detected in all IDC patients; it is noteworthy, that the s-STR concentration amounted in this category to 9.4 +/- 1.35 mg/l, which essentially topped the normal values. It can be concluded on the basis of the obtained results that s-STR can be regarded as a key IDC marker. There was an increased SF (up to 324 +/- 64.5 mkg/L) alongside with hypoferremia in the ACD patients; the s-STR content was found to be within the reference values (3.21 +/- 0.55 mg/L), the determination of s-STR can be regarded as a test in the differential diagnosis of IDC and ACD. The results of s-STR research ranged, in patients with Cooley's anemia, drastically from 0.8 mg/L to 17 mg/L; it is noteworthy, that there was a reliable dependence on an adequacy of the conducted therapy in case of such patients. The highest sSTR values were found in children with the hereditary spherocytic hemolytic anemia (11.85 +/- 2.5 mg/L), which is, apparently, explained by a proliferative super-activity of the bone marrow observed in this pathology.

Comparison of the relative quantities of gamma-mRNAs and fetal hemoglobin in SS patients with different haplotypes.

We have studied the relative levels of gamma-mRNA [%gamma/(gamma + beta)], Ggamma- and Agamma-mRNAs [%Ggamma/(Ggamma + Agamma)], hemoglobin (Hb) F, and the Ggamma and Agamma chains in some 50 patients with sickle cell anemia (SS) and with different haplotypes. As expected, the Hb F levels varied greatly and were high in patients with the Saudi Arabian-Indian haplotype. Similarly, the Ggamma values varied greatly (from 19.5 to 76.5%) and depended on the haplotypes. A rare haplotype, named Mor, was found in 3 SS patients, 1 of whom was a homozygote Mor/Mor; this haplotype is associated with the lowest Ggamma value (19.5% in the homozygote) and with a C-->T mutation at position -202 of the Agamma promoter. The levels of gamma-mRNA roughly parallel those of Hb F, but older patients have increased levels of mRNA, which appears not to be efficiently translated into Hb F. Similar observation have been reported for other hemoglobinopathies such as deltabeta-thalassemia heterozygotes and Hb Lepore heterozygotes. The relative quantity of Ggamma-mRNA was closely related to that of the Ggamma chain in the 15 patients who were studied; the Ggamma- to Agamma-mRNA ratio did not change with age.

Analysis of mRNA from red cells of patients with thalassemia and hemoglobin variants.

During the past decade new procedures have been developed for the isolation of RNA from a few mL of freshly collected blood. This material is reverse transcribed and the resulting cDNA can be used for the determination of the ratios between different types of globin mRNA, namely alpha 2/alpha 1, alpha/zeta, alpha/beta, gamma/beta, beta A/beta X, delta beta Lep/beta, and G gamma/A gamma. Details about these polymerase chain reaction-based methods are reviewed, and information about their usefulness in studying alpha-thalassemia, beta-thalassemia, sickle cell anemia and other beta-globin gene abnormalities, Hb Lepore heterozygosity, and heterozygosity for alpha 2- or alpha 1-globin gene mutations will be provided. The methods are also most useful in characterizing the mRNA types in single, in vitro cultured, BFU-E colonies; in colonies derived from cells of a Hb S heterozygote; for instance, the beta A- and beta(S)-mRNAs were present in all colonies and in about equal quantities, while many of those cells from a subject with a somatic cell mutant (Hb Costa Rica) contained beta A-mRNA and no beta-Costa Rica mRNA, and only a few had both types. The techniques described have considerable diagnostic value and offer a rather simple approach to the study of some of the listed diseases.

Hb Lepore-Baltimore (delta 68Leu-beta 84Thr) and Hb Lepore-Washington-Boston (delta 87Gln-beta IVS-II-8) in central Portugal and Spanish Alta Extremadura.

Hb Lepore is one of the most common abnormal haemoglobins in Caucasians in Central Portugal and in the Spanish Alta Extremadura (0.28% in a survey of school children). A group of 19 Portuguese and 14 Spanish Hb Lepore carriers (all unrelated) was characterised at the molecular level by the polymerase chain reaction, sequencing and restriction enzyme analysis. The Portuguese and one Spanish carrier were heterozygous for Hb Lepore-Baltimore, whereas all other Spanish subjects were Hb Lepore-Washington-Boston carriers. Sequencing of the Hb Lepore-Baltimore gene further established the crossover at delta 68-beta 84, a region two codons (CDs) shorter than that previously described and easily confirmed by digestion with MaeI and BanI. Data from haplotype analysis suggest that this crossover occurred as an independent event on the Iberian Peninsula. The haematological data were similar in both groups except for the levels of Hb F and the G gamma chain, which were significantly higher in the Hb Lepore-Baltimore heterozygotes. Quantification of the globin chains and the mRNA transcripts showed that the delta beta gene is transcribed at a higher level than the delta gene with levels of translation giving rise to 10%-15% of Hb Lepore. The different levels of Hb F observed in the two groups are the results of the higher transcription rate of the gamma genes in Hb Lepore-Baltimore heterozygotes and an apparently less efficient translation of G gamma genes in Hb Lepore-Washington-Boston heterozygotes.

Relative levels of alpha-, beta-, and gamma-mRNA from patients with severe and intermediate beta-thalassemia major.

We have determined the relative quantities of gamma- and beta-mRNAs and the alpha/beta-mRNA ratios in 37 patients with beta-thalassemia major with specific genotypes, namely 8 with a homozygosity for codon (CD) 39 (C-->T), 7 with a homozygosity for IVS-I-110 (G-->A), 5 with a homozygosity for IVS-I-6 (T-->C), for 15 patients with compound heterozygosities for 2 of these 3 mutations, and for 2 patients with the IVS-I-110 (G-->A)/-87 (C-->G) mutations. None had an alpha-thalassemia. Twelve patients had thalassemia intermedia and the remainder, transfusion-dependent severe conditions. Differences in phenotype were observed for compound heterozygotes involving the IVS-I-6 (T-->C) mutation in combination with either the IVS-I-110 (G-->A) or the CD 39 (C-->T) mutations: patients with thalassemia intermedia had a lower alpha/beta-mRNA ratio, about half of that of the patients with severe beta-thalassemia major. This might suggest a higher beta-mRNA synthesis in some patients than in others with the same genotype; mutations in promoter, enhancer, and/or locus control region sequences may be responsible for these differences. In vitro chain synthesis data were too incomplete to be helpful in this study. The RT-PCR procedure allowed the separation of abnormal (extended) mRNA from normal beta-RNA in subjects carrying the IVS-I-110 (G-->T) mutation. The relative quantities of this beta Th-mRNA (% of beta A + beta Th) were determined by scanning of the appropriate autoradiograms; they averaged 25% for homozygotes and about 4% for heterozygotes, indicating a considerable instability of the message.

Alpha-, beta-, and gamma-mRNA levels in beta-thalassemia; transcriptional and translational differences in heterozygotes, homozygotes, and compound heterozygotes.

We have determined the relative levels of alpha-, beta-, and gamma- (G gamma- and A gamma-) mRNAs in the reticulocytes of patients with mild beta-thalassemia intermedia due to combinations of promoter mutations and a classical type of beta-thalassemia, as well as in their relatives. The expected differences in the alpha/beta-mRNA ratio confirmed the mild suppression of beta-mRNA synthesis, particularly in heterozygotes for the -101 (C-->T) promoter mutation and the large increase in the relative gamma-mRNA level in compound heterozygotes. A significant discrepancy between Hb F and gamma-mRNA levels, observed in previously published studies, was confirmed indicating a less efficient gamma-mRNA translation. When the two different gamma-mRNA (G gamma- and A gamma-) levels were determined it was observed that in beta-thalassemia heterozygotes the extra gamma-mRNA was primarily of the G gamma type suggesting a more efficient translation of the A gamma-mRNA. This difference disappeared in homozygotes and compound heterozygotes: both mRNAs (G gamma- and A gamma-) translate with an equal efficiency.

Detection of the alpha-thalassemia-2 (3.7 kb) deletion in DNA extracted from 20-year-old blood smears.

Using a polymerase chain reaction procedure we have identified the 3.7 kb alpha-thalassemia-2 deletion in the DNA that was isolated from slides of blood smears stained with Wright's stain some 20 years ago. Details about the procedure are presented. The success of this approach depends entirely on the amount of DNA that could be isolated; thick smears always gave good data provided they were not covered with immersion oil. The success rate in this study was 45%.

Variability in the fetal hemoglobin level of the normal adult.

We analyzed blood samples from more than 200 normal adults, and quantified their Hb F by cation-exchange high-performance liquid chromatography. In several subjects with slightly elevated Hb F (0.4-4.3%), we determined the Ggamma levels in the Hb F and DNA sequence variations in the locus control region II and in the Ggamma and Agamma promoters. About 25% of the approximately 200 normal teenaged high school students had elevated Hb F; detailed analyses of some 20 students, selected at random, identified most as females with a homozygosity for the C-->T variation at position -158 (Ggamma). One 11-year-old boy was heterozygous for the A-->G change at position -161 (Ggamma); he and two of his relatives had approximately 4% Hb F, high Ggamma values, and a high level of (mainly) Ggamma-mRNA. Nearly 40 normal adults from Macedonia and from Georgia (mostly Caucasians) were tentatively identified as Swiss HPFH heterozygotes because slightly elevated Hb F levels were observed at least once. Many of these persons were heterozygous or homozygous for the C-->T mutation at -158 (Ggamma), and a few carried a gamma-globin gene triplication. The C-->T change appears to be an important factor predisposing the adult to increased Hb F production. Evidence suggests a gene dose effect in (mildly) anemic adults; however, other factors besides the C-->T change at -158 (Ggamma), including factors not linked to the beta-globin region, may cause an increase in gamma-chain synthesis.

The relative levels of alpha 2-, alpha 1-, and zeta-mRNA in HB H patients with different deletional and nondeletional alpha-thalassemia determinants.

We have analyzed the alpha 2/alpha 1-, alpha/beta-, zeta/(alpha + zeta)-mRNA ratios in the retic-ulocytes of 40 patients with Hb H disease. 21 patients had deletional Hb H disease (- -/- alpha), namely combinations of one of four types of alpha-thal-1 (MED-I, MED-II, -(alpha)20.5, SEA) and one of two types of alpha-thal-2 (-3.7 or -4.2 kb); 13 had Hb H disease because of combinations of one of these alpha-thal-1 deletions with either a 5 nt deletion at the 5' splicing site of IVS-I, or a terminating codon mutation (Hb CS), or a poly(A) mutation, and six were homozygous for either a poly(A) mutation or the 5 nt deletion. Significant differences were observed between the deletional types (- -/- alpha; alpha 2/alpha 1 ratio of zero; alpha/beta ratio of approximately 1) and non-deletional types (- -/alpha T alpha; alpha 2/alpha 1 ratio of 0.05-0.3 for those with T = the 5 nt deletion or the terminating codon mutant, and approximately 1.0 for those with T = a poly(A) mutation; alpha/beta ratio in all types of approximately 0.7). Comparable data were found for the nondeletional alpha-thal-2 homozygotes. The noted differences were highly significant and the determination of the two ratios may be diagnostically of considerable value. The low alpha 2/alpha 1-mRNA ratio in the two patients with - -/alpha-5nt alpha and the one patient with alpha-5nt alpha/alpha-5nt alpha indicates the presence of minute amounts of alpha 2-mRNA; apparently splicing at the donor site is greatly impaired by this deletion but not eliminated. The high alpha 2/alpha 1-mRNA ratio in the four patients with - -/alpha PA-2 alpha and the five patients with alpha PA-1 alpha/ alpha PA-1 alpha (PA-1 and PA-2 are poly(A) mutations) is due to the presence of an elongated alpha 2-mRNA which uses an alternate location as polyadenylation site. The relative levels zeta-mRNA varied considerably; the highest levels were found in patients with the -(alpha)20.5/-alpha or - -SEA/-alpha deletional types but not in those with the -(alpha)20.5/alphaPA-2 alpha, -(alpha)20.5/alpha-5nt alpha, or - -SEA/alphaCS alpha nondeletional types. No definitive explanation can be given for these differences; perhaps certain sequences that are part of some of the alpha-thal-1 deletions are important for the suppression of the zeta-globin gene.

MRNA analysis in reticulocytes of subjects with Hb D, Hb Porto Alegre, Hb E, and different types of unstable hemoglobin variants.

Using a reverse transcription-polymerase chain reaction (RT-PCR) technique we determined the alpha 2/alpha 1, alpha/beta, and gamma/beta mRNA ratios in reticulocytes of 11 patients with seven different unstable beta chain variants, of 4 patients with two unstable alpha chain variants, in hemoglobin (Hb) D, Hb Porto Alegre, and Hb E heterozygotes, and in 8 patients with Hb X-beta 0-thalassemia (thal) (three D-beta 0-thal, one Porto Alegre = beta 0-thal, one Lulu Island-beta 0-thal, and three E-beta 0-thal). In addition, we determined the beta X/beta A mRNA ratios (X = unstable) in some Hb D heterozygotes and in 6 subjects with an unstable beta chain variant. Normal alpha/beta and beta X/beta A mRNA ratios were found in all heterozygotes tested, indicating that the respective mutations did not alter the stability of the mRNAs. The alpha/beta mRNA ratio in four Hb E heterozygotes averaged 4.21 (normal, 4.47), and that in 2 patients with Hb E-beta 0-thal and four alpha-globin genes (alpha alpha/alpha alpha) averaged a high 22.4. The gamma mRNA level in the Hb E heterozygotes was < 1% but varied greatly in patients with Hb E-beta 0-thal; the alpha/(gamma + beta) mRNA ratios in the 2 patients were 15.5 and 16.7, respectively. The large differences in alpha/beta and alpha/(gamma + beta) mRNA ratios in reticulocytes of subjects with AE and with E-beta 0-thal may be due to differences in the levels of normally-spliced beta E and abnormally-spliced beta E mRNAs. Only the latter is unstable and is preferentially produced in bone marrow and reticulocytes of Hb E-beta 0-thal patients, where it is rapidly degraded.

The relative levels of different types of beta-mRNA and beta-globin in BFU-E derived colonies from patients with beta chain variants; further evidence for somatic mosaicism in the Hb Costa Rica carrier [beta 77(EF1)His-->Arg].

We have identified and quantitated the different types of mRNA in single BFU-E derived colonies from Hb S and Hb Atlanta [beta 75 (E19)Leu-->Pro] heterozygotes and observed that the normal and mutated mRNAs were present in equal quantities. Similar studies for the different protein products gave less accurate data because high performance liquid chromatography methods were not sensitive enough for the analysis of a single colony, and as many as five colonies needed to be combined. The level of Hb S (approximately 40%) was the same as in red cell lysates of the Hb S heterozygote, while that of the unstable beta-Atlanta chain was lower than expected from the values observed in red cells. Similar studies for a carrier of the stable Hb Costa Rica [beta 77(EF1) His-->Arg] which was reported to be the result of a somatic cell mutation (1) gave quite different data. Dot-blot analysis with 32P-labeled probes and allele specific amplification methodology identified numerous colonies with beta A-mRNA only, while 12-15% of the colonies contained both beta A- and beta-Costa Rica-mRNA. This limited distribution of the beta-Costa Rica-mRNA was confirmed by hemoglobin analysis with anion exchange high performance liquid chromatography. These results are considered to provide additional and convincing evidence for a somatic mosaicism for the CAC-->CGC mutation at codon 77 of the beta gene which occurred during the development of the embryo and results in the presence of only some 6-8% of the abnormal Hb Costa Rica in the circulating red cells.

The dominant beta-thalassaemia in a Spanish family is due to a frameshift that introduces an extra CGG codon ( = arginine) at the 5' end of the second exon.

We have discovered a Spanish family with a dominant type of beta-thalassaemia. Carriers are characterized by mild anaemia, hypochromia, microcytosis, elevated Hb A2 and Hb F levels, reticulocytosis, and splenomegaly. The molecular basis of this condition is the introduction of a CGG triplet between codons 30 and 31 of the beta gene; this was determined by sequencing of amplified DNA and confirmed by dot-blot analysis. The abnormal mRNA (beta Th-mRNA) is stable and present in quantities similar to that of normal beta A-mRNA. cDNA fragments derived from beta Th- and beta A-mRNAs can be separated on a denaturing polyacrylamide gel electrophoresis because the beta Th fragment is three nucleotides (nts) longer than the beta A fragment. The beta Th-mRNA translates into a beta chain that is 147 amino acid residues long and carries an extra arginine residue between residues 30 and 31. This beta X chain has not been detected. It may be unstable and does not bind to the alpha chain. It probably is continuously digested by proteolytic enzymes in red cell precursors in the bone marrow. The abnormal chain probably binds haem that is excreted after proteolysis causing a darkening of the urine.

Hb Costa Rica or alpha 2 beta 2 77(EF1)His --> Arg: the first example of a somatic cell mutation in a globin gene.

We have identified a minor hemoglobin component (approximately 5%) in the blood of a healthy Costa Rican female, but not in her mother and two brothers (father not studied), that has an His --> Arg replacement at position beta 77 (Hb Costa Rica). No other amino acid replacements were observed and no beta- or gamma-chain-like peptides were present. Hb Costa Rica has abnormal stability. Sequence analyses of numerous polymerase chain reaction (PCR)-amplified segments of DNA that contain exon 2 of the beta gene failed to identify a CAC --> CGC (His --> Arg) mutation. The same was the case when cDNA was sequenced, indicating that a beta-Costa Rica-mRNA could not be detected with this procedure. Gene mapping of genomic DNA with Bg/II, BamHI, and HindIII gave normal fragments only and with the same intensity as observed for the fragments of a normal control. The quantities of the beta chain variants Hb J-Iran and Hb Fukuyama with related mutations at beta 77 vary between 30% and 45% in heterozygotes, whereas that of Hb F-Kennestone with the same His --> Arg mutation but in the G gamma-globin gene, is a high 40%-45% (as percentage of total G gamma) in a heterozygous newborn. These different observations exclude a heterozygosity of the A --> G mutation at codon beta 77, as well as a deletion comparable to that of Hbs Lepore or Kenya, or a beta-globin gene duplication, and point to a nontraditional inheritance of Hb Costa Rica. Allele-specific amplification of cDNA with appropriate primers identified the presence of a low level of mutated mRNA in the reticulocytes of the patient, which was confirmed by dotblot analysis of the same material with 32P-labeled probes. Comparable amplification products were not observed in genomic DNA. The A --> G mutation apparently occurred in a somatic cell at a relatively early stage in the development of the hematopoietic cell system, and Hb Costa Rica accumulated through rapid cell divisions in patchy areas in the bone marrow (somatic mosaicism). An unequal distribution of Hb Costa Rica over the red cells supports this possibility.