Haematologica 2002; 87:(04)ECR12
[Medline] [prev] [index] [next]Triosephosphate isomerase deficiency. genetic, enzymatic and metabolic characterization of a new case from Spain
Ada Repiso,1 Joan Boren,2 Fernando Ortega,3 Assumpció Pujades, 4 Josep Centelles,2 Joan Lluis Vives-Corrons,4 Fernando Climent,1 Marta Cascante,2 José Carreras1
1Unitat de Bioquímica, Departament de Ciències Fisiològiques I, Facultat de Medicina, Institut d'Investigacións Biomèdiques August Pi i Sunyer (IDIBAPS), Universitat de Barcelona, Casanovas 143, 08036 Barcelona; 2Departament de Bioquímica i Biologia Molecular, Facultat de Química, Universitat de Barcelona and Centre de Química Teòrica (CeRQT-PCB), Barcelona; 3Centre de Química Teòrica (CeRQT-PCB) and Departament de Química Física, Facultat de Química, Universitat de Barcelona, Martí i Franquès, 1 08028 Barcelona, 4Unitat d'Eritropatologia, IDIBAPS, Hospital Clínic i Provincial, Universitat de Barcelona, Villarroel 170, 08036 Barcelona, Spain.
Correspondence: Prof. José Carreras, Unitat de Bioquímica, Facultat de Medicina, Universitat de Barcelona, Casanovas 143, 08036 Barcelona (Spain). Phone:34.93.402.4541. Fax: 34.93.403.5882. e-mail: jcarrera@medicina.ub.es
We describe the first triosephosphate isomerase (TPI, E.C. 5.3.1.1) Spanish deficiency fully characterized at genetic, metabolic and enzymatic levels. The propositus presented non-spherocytic hemolytic anemia and neuromuscular dysfunction, and exhibited 8% of normal TPI activity in her erythrocytes. The parents, with no clinical symptoms, each had about 50% of normal enzyme activity. Enzyme kinetic studies revealed no changes in either the Km values for DHAP and GAP or in the effects produced by several competitive inhibitors. Conversely, thermostability studies showed a significant decrease in the deficient enzyme heat stability. Accumulation of DHAP and fructose 1,6-bisphosphate (FBP) were the only metabolic abnormalities observed in the TPI-deficient red blood cells. Measurements of the glycolytic flux showed no changes induced by the TPI deficiency. A single homozygous mutation replacing Glu-104 by Asp was found in exon 3.Triosephosphate isomerase (TPI, EC 5.3.1.1) catalyzes the interconversion of dihydroxyacetone phosphate (DHAP) and glyceraldehyde-3-phosphate (GAP) and plays an important role in several crucial metabolic pathways. TPI deficiency is a rare autosomal recessive multisystem disease that has been known since 1965.1 Whereas heterozygotes are clinically normal, homozygotes and compound heterozygotes are characterized by a markedly decreased enzyme activity and clinical manifestations dominated by a lifelong hemolytic anemia and severe progressive neuromuscular degeneration leading to death of the patient in early childhood.2 Up to now, fourteen different mutations in the human TPI gene have been identified.3 Among these, the most frequent is the mutation at nucleotide 1592 (GC) leading to a change at amino acid 104 (E105D) . This mutation, first described in 1986,4 has been found in more than ten apparently unrelated families throughout the world, in contrast to the other known TPI mutations all of which have been reported in individual families.3 We report here a new case of TPI deficiency found in Spain, fully characterized from the enzymatic, metabolic and genetic points of view. In the first case reported in 19755,6 metabolic and genetic studies could not be performed due to the early death of the patient.
Case report
The propositus, a 5-year-old Spanish girl, had been hospitalized several times since the age of 8 months due to urinary tract infections and pneumonia associated with fever and hemolytic anemia. When she was 15 months old, the diagnosis of TPI deficiency was made. The RBC TPI activity was less than 10% of normal. The patient's hemoglobin was maintained at about 108 g/L by repeated blood transfusion, but progressive neuromuscular impairment appeared, leading to a respiratory insufficiency that required mechanical ventilation. The parents were not consanguineous. They did not present clinical symptoms, but each had 50-60% of normal RBC TPI activity.
MATERIALS AND METHODS
Measurements of enzyemes and glycolytic intermediates
RBC glycolytic and several non-glycolytic enzyme activities were measured in hemolysates7 accordingly to the methods recommended by the International Committee for Standarization in Hematology.8 The substrates FBP, DHAP and GAP were quantified in deproteinized extracts according to Minamaki et al.9 All the substrates, cofactors and auxiliary enzymes were from Boehringer (Mannheim, Germany) and Sigma (St. Louis, MO, U.S.A)
Kinetic studies
The kinetic constants were determined using whole hemolysates. The Km values for GAP were obtained by determining the enzyme activity at 37¼C in 50 mM TEA-HCl buffer, pH 7.2, containing 1 mM EDTA, 0.3 mM NADH, 2 IU/mL glycerophosphate dehydrogenase and different D-GAP concentrations (from 50 mµM to 2 mM). The Km values for DHAP were obtained by determining TPI activity at 37¼C in 50 mM TEA-HCl buffer, pH 7.6, containing 1 mM EDTA, 0.3 mM NAD, 13 mM arsenate, 1 IU/mL glyceraldehyde 3-phosphate dehydrogenase and different DHAP concentrations (from 250 µM to 9 mM). The apparent Km values were calculated by the unweighted non-linear regression method using the PC program GraphPad In PlotTM™ (Copyright ©1992 Graph Pad Software Inc. Version 4.0). The effect of the competitive TPI inhibitors D-a-glycerophosphate, glycerate-3-phosphate, glycolate-2-phosphate, FBP, ATP and inorganic phosphate was determined by using GAP as substrate and standard conditions of assay. The concentrations of the inhibitors were selected in order to inhibit approximately 50% of the activity of the normal enzyme according to the reported data.10,11
Heat stability
Hemolysates were diluted 1/40 in 100 mM Tris-HCl buffer, pH 8, containing 500 µM EDTA and 1% bovine serum albumin, and were incubated at 37¼C and at 57¼C. Enzyme activity was measured using D-GAP as substrate under standard conditions7 at intervals of 10 min during a 60 min period.
Glycolytic flux measurement
The flux through the glycolytic segment including the phosphofructokinase (PFK), aldolase and TPI catalyzed reactions was measured by two different approaches: a) in a mixture containing the substrates fructose-6-phosphate (2 mM), Mg-ATP (2 mM) and NADH (0.16 mM), and glycerophosphate dehydrogenase (1 IU/mL) as auxiliary enzyme, and b) in a mixture of the substrates fructose-6-phosphate (2 mM), Mg-ATP (2 mM), NAD (6 mM), arsenate (4 mM), and glyceraldehyde-3-phosphate dehydrogenase (1 IU/ml) as auxiliary enzyme.
Computer modelling
The kinetic model of the segment of glycolysis from hexokinase to aldolase was constructed using an IBM-PC version of the simulation program Gepasi®. The rate equations and the kinetic parameters of the enzymes used for the kinetic model12,13 are listed in Table 1.
DNA extraction and sequencing
DNA was isolated from leukocytes according to Sambrook et al.,14 and amplified by PCR using the primers and the PCR program described by others.15,16 PCR products were purified with QIAGEN QIAquick™ Kit and sequenced using the same set of oligonucleotides as for PCR amplification.
Results
Red blood cell enzyme activities and glycolytic intermediates
As shown in Table 2, TPI was reduced to about 8% of normal in the propositus and to about 50% in her parents. Aldolase and glucose-6-phosphate dehydrogenase were found to be increased, in accordance with the highly increased proportion of reticulocytes (about 20%) in the patient's peripheral blood. The determination of glycolytic intermediates showed a marked increase of DHAP (23 fold) and a lower increase of FBP (10 fold) and GAP (8 fold) in the propositus. Both parents presented a much lower increase (2-3 fold) of these metabolites.
TPI kinetic properties
As summarized in Table 3, no significant differences were observed between the apparent Km values of the normal enzyme and the deficient TPI. The reported values are for the total triose phosphates in solution and do not take into account the relative distribution of hydrated gem diols or enediols which are in equilibrium with the free aldehyde and ketone.17,18 Furthermore, inhibition by arsenate when DHAP was the substrate was not considered and no correction of the apparent Km values was made. As shown, the normal and the mutant TPI were similarly affected by the competitive inhibitors studied.
TPI heat stability
When compared to the controls, heat stability of deficient TPI was significantly decreased. In the patient's hemolysate TPI activity was reduced to 30% after 12 min of incubation at 37¼C, and to 20% after 10 min of incubation at 57¼C. In a contrast, control's hemolysate showed 90% TPI activity after 60 min of incubation at 37¼C and 60% activity after 60 min of incubation at 57¼C.
Glycolytic flux measurements and computer modelling
As shown in Table 2, when DHAP was considered the final product, all flux measurements exhibited moderately increased values when compared to those in controls, being higher in the propositus (40% flux increase). In contrast, when DHAP was an intermediate of the pathway and 1,3-bisphosphoglycerate was considered the final product, a two-fold increase in the flux was observed in the propositus when compared to her parents and to controls.
In order to determine whether the changes of the glycolytic fluxes observed in the TPI deficient hemolysate could be a consequence of the decreased TPI activity, a kinetic model was constructed using experimental kinetic parameters and rate equations that allow prediction of the steady-state fluxes. The computer modelling showed (Figure 1) that the decrease of TPI activity to about 8% of normal in the propositus does not explain our experimental results and that is necessary to consider both the decrease in TPI activiy and the increase in aldolase activity to account for the changes in the observed experimentally fluxes. Therefore, it was concluded that TPI deficiency does not change the glycolytic flux in the patient's RBCs.
DNA analysis
Sequencing the TPI gene of the propositus demonstrated a homozygous 315 GC replacement at the third exon leading to a replacement of a glutamic acid by aspartate at 104 position (Glu104Asp). DNA sequencing of the TPI gene in both parents confirmed the homozygous characteristic of the mutation in the propositus.
Discussion
At present 14 mutations of the TPI gene producing enzyme deficiency have been reported. The most frequent is the 315 GC transversion that causes the Glu 104 Asp substitution.3 The finding that this mutation is also present in Spain confirms its world wide geographical distribution and of reinforces the need for prenatal diagnosis procedures.
In agreement with previous reports4,19,20 our results show that the Glu 104 Asp mutation increases the sensitivity of the deficient TPI molecule to heat denaturation but does not affect its kinetic properties (Km values for GAP and for DHAP, and response to several competitive inhibitors). The Glu 104 residue lies between two a-helical regions in a short stretch of random coil next to the TPI subunit interface. The side chain of Asp104 in the mutated enzyme is identical in charge but shortened by a methylene group relative to the side chain of Glu 104 in the normal enzyme. This would disrupt the counterbalancing of charges that normally exists and lower the stability of the enzyme, promoting unfolding. In addition, it could indirectly perturb the local structure of the active site.4,21
The mechanism of the deleterious effect of TPI deficiency on RBCs remains speculative. In the present study accumulation of DHAP and to a lesser extent of FBP are the major metabolic abnormalities observed in the TPI deficient RBCs. Furthermore, we have found that TPI deficiency does not significantly affect the flux in the upper part of glycolysis. Similar metabolic abnormalities have been described in distinct TPI mutations.3,22 Therefore, it has been suggested that the accumulation of DHAP itself or of some derivative product might be injurious to cells. Alternatively, it has been postulated that diversion toward the synthesis of a-glycerophosphate might lead to lipid abnormalities of pathogenic significance.3
One interesting aspect of the present study is that the relationship between the decrease of TPI activity and the increase of DHAP concentration observed in the TPI-deficient RBCs does not agree with the theoretical calculations predicted from a mathematical model developed by Schuster and Holzhütter.13 This model predicts that DHAP concentration in RBCs increases significantly only when the residual TPI activity is less than 1%. However, in our case a marked increase of DHAP is observed at higher residual TPI activity (about 8% in the hemolysate). This apparent discrepancy could be explained by the increased proportion of reticulocytes (about 20%) in the patient's peripheral blood, which would increase both DHAP concentration and TPI activity in the hemolysate. Furthermore, as a consequence of the greater instability of the mutant TPI could exist an increased population of RBCs next or beyond the critical threshold of cell integrity. This population, with very low TPI activity and very high DHAP concentration,13 would also produce an increase of the levels of this metabolite in the hemolysate. Orosz et al.22,23 in three TPI-deficient patients with RBC residual TPI activity less than 5%, have found an accumulation of DHAP similar to that observed in the present case. In addition to the possible explanations given above, they suggested an eventual decrease of TPI activity in vivo due to the binding of the enzyme to the RBC membrane, and to differences in the microcompartmentalization of the normal and the mutant TPI. The finding that TPI activity decreases after binding to RBC ghosts and that the mutant TPI exhibited higher binding capacity than the normal enzyme23,24 gives support to this hypothesis.
Acknowledgments
We thank Drs J.Ovadi and F.Orosz for fruitful discussions and technical support. This work was supported by MEC (grant PP99-0040) and FISS (Grants 00/0584 and 50/1120).
References