A. J. Bellingham

Evidence for founder effect of the glu104asp substitution and identification of new mutations in triosephosphate isomerase deficiency

Triosephosphate isomerase (TPI) deficiency is an autosomal recessive disorder of glycolysis characterized by multisystem disease and lethality in early childhood. Among seven unrelated Northern European kindreds with clinical TPI deficiency studied, a single missense mutation at codon 104 (GAG;Glu→GAC;Asp) predominated, accounting for 11/14 (79%) mutant alleles. In three families molecular analysis revealed compound heterozygosity for Glu104Asp and novel missense mutations. In two cases the second mutation was a Cys to Tyr substitution at codon 41(TGT→TAT) and in one an Ile to Val substitution at codon 170(ATT→GTT). The origin of the Glu104Asp mutation was defined by haplotype analysis using a novel G/A polymorphism at nucleotide 2898 of the TPI gene. Cosegregation of the low frequency 2898A allele with the G→C base change at nucleotide 315 supports a single origin for the Glu104Asp mutation in a common ancestor. Hum Mutat 10:290–294, 1997.

Tucaresol increases oxygen affinity and reduces haemolysis in subjects with sickle cell anaemia.

The primary pathophysiological event in sickling is the intracellular polymerization of deoxygenated haemoglobin S. Tucaresol (589C80;4[2‐formyl‐3‐hydroxy‐phenoxymethyl] benzoic acid), a substituted benzaldehyde, was designed to interact with haemoglobin to increase oxygen affinity and has been shown to inhibit sickling in vitro. We administered tucaresol to sickle cell patients in the steady state to examine the anti‐sickling effect in vivo. Oral doses of tucaresol or placebo were given to nine stable sickle cell patients (aged 17–39 years; tucaresol, six; placebo, three) for 10 d. The first two patients on tucaresol were scheduled to receive a loading dose of 800 mg or 1200 mg (depending on bodyweight) for the first 4 d, followed by maintenance doses of 200 or 300 mg for the next 6 d. Due to concerns over the sharp rise in haematocrit in one patient, subsequent cohorts received 300 mg tucaresol daily throughout the dosing period. The oxygen affinity of haemoglobin S was increased in all patients receiving tucaresol, with between 10% and 24% of the haemoglobin modified, dependent on dose. In all patients on tucaresol, haemolysis was reduced with rises in haemoglobin of 0.9–3.7 g/dl (mean 2.2 g/dl), falls in lactate dehydrogenase of 16–52%, and a halving of the irreversibly sickled cell counts. These effects were apparent within a few days and persisted for 1–2 weeks following discontinuation of the drug. Three of the six patients on tucaresol developed fever and cervical lymphadenopathy, with onset between days 7 and 11 from start of drug. Further evaluation of the tolerability and efficacy of tucaresol in sickle cell patients is necessary.

Red cell 3-phosphoglycerate level as a diagnostic aid in pyruvate kinase deficiency.

Summary. A case of pyruvate kinase (PK) deficiency is described in which the diagnosis was aided by measurement of the 3‐phosphoglycerate (3PG) concentration. Review of the literature on the levels of red cell metabolites in 52 families with PK deficiency confirmed that a rise in 3PG is a valuable indicator of a functional deficiency of PK. Estimation of 3PG is relatively easy (and accurate). Furthermore, reticulocytosis, which sometimes makes the diagnosis of PK deficiency more difficult, has minimal effect on the level of 3PG in comparison with all other glycolytic intermediates or PK activity.

Normal levels of ATP, total nucleotides and activities of three enzymes related to nucleotide metabolism in fetal erythrocytes

Pure fetal blood was obtained by direct‐vision fetoscopy from 30 fetuses at 17–23 weeks gestation. The erythrocyte concentrations of ATP and total nucleotides and the activities of the enzymes pyrimidine‐5′‐nucleotide nucleosidase (Fyr5N), phosphoribosylpyrophosphate (PRPP) synthetase and adenylate kinase (AK) were analysed by established techniques to find the normal ranges for this gestational age. The ranges were relatively narrow and could serve as reference values for the prenatal diagnosis of defects in nucleotide metabolism.

Hydroxyurea, sickle cell disease and renal transplantation.

Dr. Andrew Allen, Renal Unit, Department of Medicine, Hammersmith Hospital, Du Cane Road, London W12 0NN (UK) Dear Sir, The use of hydroxyurea to increase HbF production in patients with homozygous sickle cell disease is increasing, particularly after recent publication of data demonstrating a significant reduction in painful crises [1]. We here describe the successful use of hydroxyurea in a sickle cell patient with severe painful crises occurring after renal transplantation and an associated rise in haematocrit. A 26-year-old Jamaican with homozygous sickle cell disease had a stable haemoglobin of 8-9 g/dl and painful crises once every 3 years. He developed proteinuria and renal impairment, and renal biopsy demonstrated mesangiocapillary glomerulonephri-tis. Renal function slowly declined with a parallel fall in haematocrit and he commenced haemodialysis in early 1994. At this point his haemoglobin was 4 g/dl but he failed to respond to subcutaneous erythro-poietin despite doses of up to 600 U/kg/week – higher doses were not tolerated because of pain on injection and the usual causes of erythropoietin resistance were excluded. In December 1994 he received a cadaveric renal allograft after a 4-unit exchange transfusion, with standard triple immuno-suppression (cyclosporin 8 mg/kg b.d. initially, prednisolone 20 mg and azathioprine 1 mg/kg). Three further units of packed red blood cells were given perioperatively. Cre-atinine fell to 159μmol/l (1.8mg/dl) and haemoglobin rose to 11 g/dl with HbS > 80% once the transfused blood had gone. The ele120-1 3⁄4 100 300 400 Time (days) Fig. 1. Haemoglobin concentration and globin fractions in relation to renal transplantation and hydroxyurea therapy. Renal transplantation in homozygous sickle cell patients with end-stage renal disease is well described with good graft and patient survival rates [2, 3]. The usual rise in haematocrit, seen when the graft functions well, is sometimes associated with the development of painful crises, ascribed to the concurrent rise in plasma viscosity [4, 5]. Previously these crises have been managed with analgesia and venesection or exchange trans-

Red cell glycolytic intermediates in normal, anaemic and transfused human fetuses.

Normal fetal ranges for red cell glycolytic intermediates at 18–24 weeks gestation, which are useful as reference values for the prenatal diagnosis of erythroenzymopathies, were established for the first time. Characteristic increases in glucose‐6‐phosphate (G6P), fructose‐6‐phosphate (F6P) and particularly fructose‐ 1,6‐diphosphate (FDP) suggest that there is no metabolic block at the phospho‐fructokinase (PFK) step of glycolysis as previously suggested by others for premature infants on the first day of life.

2,3-Diphosphoglycerate and 3-phosphoglycerate in red cell pyruvate kinase deficiency.

600-800 in two, higher than 800 in four (in the whole group respectively 59%, 29%, 7% of the cases): the values do not however allow any significant statistical analysis. Unfortunately, the PVSG data do not take into account the degree of elevation of the platelet count, and do not analyse the platelet function. From the published series of primary thrombocytosis (including that of PSVG), it seems clear that the thrombotic risk increases. even in young patients, when the platelet count exceeds 800 x 109/1, and possibly a lower limit in presence of another risk factor (for instance high haematocrit). In this way, I think, as apparently the authors of the above letter do, that the maintenance of the platelet count to a level lower than 600 x 109/1 is probably useful, even if not statistically demonstrated. Note that two female patients have had one or several abortions, with a platelet count higher than 600 (is it a vascular event I ) . I cannot answer the second question. No valuable statistical multiparametric analysis is possible when only 10 cases of vascular events are retrospectively analysed. However, our data are in good accord with previously published data, which ‘reinforce what many. . . . . think are the logical risk factors for thrombosis’. It remains a very encouraging fact: when early complications are avoided, the long-term survival seems to be very long. I have re-analysed the charts for writing this reply and no further deaths have occurred even in the patients with myeloid metaplasia or spent phase. patients with myeloproliferative disease we found the same risk factors as the PVSG in the older patients (except age and vascular antecedents, obviously): (i) The frequency of thrombotic events in the first year (or first 3 years for PVSG) after diagnosis; in our study 10 of the 15 accidents were observed at the very beginning or during the first year after diagnosis. (ii) The quality of the maintenance: contrary to the third sentence of the second paragraph of the above letter, the PVSG data obviously show that the maintenance of a normal haematocrit is a chief parameter for the vascular risk (physiological studies, some of them done in England, demonstrate a decrease of the cerebral blood flow as soon as the haematocrit increases beyond 50% or even 48%); we have observed exactly the same phenomenon, even if a delay between the last blood examination and the accident does not allow a precise correlation (the same is true in the PVSG data). (iii) The choice of therapy is. in the PVSG documents, another parameter concerning the risk of thrombotic event, at least during the first 3 years. We also found that more accidents occurred in the phlebotomized than in the myelosuppressed cases (in fact no case in the 32P-treated patients). So our data are in accord with those of the PVSG. In our group the high number of early thrombotic events may be due to the severity of the disease at the time of diagnosis, higher than in older patients, a fact which could be related to a better clinical tolerance of the high blood viscosity in the young patients. A possible difference between our cohort and that of the PVSG is the platelet count. In our 10 cases with thrombotic events the platelet count was 400-600 x 109/1 in four cases, Department of Nuclear Medicine and Haematology, Saint-Louis Hospital, 7501 0 Paris, France YVES NAJEAN