Triosephosphate isomerase deficiency: Facts ... - Wiley Online Library

IUBMB

Life, 58(12): 703 – 715, December 2006

Critical Review Triosephosphate Isomerase Deficiency: Facts and Doubts Ferenc Orosz, Judit Ola´h and Judit Ova´di Institute of Enzymology, Biological Research Center, Hungarian Academy of Sciences, Budapest, Hungary

Summary Many glycolytic enzymopathies have been described that manifest clinically as chronic hemolytic anemia. One of these, triosephosphate isomerase (TPI) deficiency, is unique among the glycolytic enzyme defects since it is associated with progressive neurological dysfunction and frequently with childhood death. The physiological function of TPI is to adjust the rapid equilibrium between dihydroxyacetone phosphate and glyceraldehyde-3-phosphate produced by aldolase in glycolysis, which is interconnected to the pentose phosphate pathway and to lipid metabolism via triosephosphates. The TPI gene is well characterized; structure and function studies suggest that instability of the isomerase due to different mutations of the enzyme may underlie the observed reduced catalytic activity. Patients with various inherited mutations have been identified. The most abundant mutation is a Glu104Asp missense mutation that is found in homozygotes and compound heterozygotes. Two germ-line identical Hungarian compound heterozygote brothers with distinct phenotypes question the exclusive role of the inherited mutations in the etiology of neurodegeneration. This paper: (i) reviews our present understanding of TPI mutation-induced structural alterations and their pathological consequences, (ii) summarizes the consequences of TPI impairment in the Hungarian case at local and system levels, and (iii) raises critical questions regarding the exclusive role of TPI mutations in the development of this human disease. IUBMB Life, 58: 703–715, 2006 Keywords

Human molecular disease; protein structure; cellular glucose metabolism; erythrocytes; neurodegenerative disorders; general bioenergetics.

Abbreviations

AGE, Advanced glycation end product; DHAP, Dihydroxyacetone phosphate; GAP, Glyceraldehyde-3-phosphate; PPP, Penthose phosphate pathway; POP, Prolyl-oligopeptidase; TPI, Triosephosphate isomerase.

Received 13 October 2006; accepted 13 November 2006 Address correspondence to: Judit Ova´di, PhD, DSc, Institute of Enzymology, Biological Research Center, Hungarian Academy of Sciences, Karolina u´t 29., H-1113 Budapest, Hungary. E-mail: [email protected]; Website: http://www.enzim.hu/*ovadi ISSN 1521-6543 print/ISSN 1521-6551 online Ó 2006 IUBMB DOI: 10.1080/15216540601115960

INTRODUCTION The major energy source for many cells is glucose, which is metabolized via the glycolytic pathway. Erythrocytes (red blood cells) consume glucose exclusively to produce the ATP required for their normal functions. Disturbances in their energy metabolism reduce the lifespan of erythrocytes (1). Enzymopathy has been identified in cases of the inherited mutations of almost all the glycolytic enzymes, however, neurological phenotypes have been found to be associated only with specific mutations affecting triosephosphate isomerase (TPI), phosphoglycerate kinase and, in rare cases, glucose6-phosphate isomerase (2). The symptoms of TPI deficiency are generally much more severe than those of any other glycolytic enzyme deficiency. TPI deficiency is a rare autosomal recessive multisystem disorder, characterized by decreased enzyme activity in all tissues, which is accompanied by the elevation of dihydroxyacetone phosphate (DHAP) level in erythrocytes. Homozygotes manifest congenital hemolytic anemia, increased susceptibility to infection, cardiomyopathy and progressive neuromuscular impairment that, in most cases, pursues an inexorable course with fatal outcome in early childhood. Only two TPI deficiency cases have been described with patients free of neurodegenerative symptoms (cf. Table 1). No effective therapy is available for TPI deficiency. There are at least two factors that appear to be relevant to the TPI deficiency as a unique glycolytic enzymopathy coupled with neurodegenerative disorder. The presence of the mutant protein can result in the formation of toxic protein aggregates and/or impairment of energy metabolism. It has been welldocumented that in other neurodegenerative diseases (commonly termed as conformational diseases), unfolded or misfolded proteins form aberrant protein-protein interactions that lead to the formation of toxic protein aggregates (inclusions) triggering neuronal dysfunction. Another problem related to the accumulation of unfolded or misfolded proteins is that these protein species can impair energy metabolism by mechanisms that are not fully understood. Neural dysfunction resulting from misfolded proteins and impaired energetics may

Substrate binding; Transcription; near to dimer nuclear factor interface binding

Yes

Yes

1

Arg189Stop

Yes

No

1

-5-8a/ Met Init Lys

Only one of the patients

Yes

2

Glu145stop/ Phe240Leu

Absolutely

10000

No 6

Yes 19.5 (fibroblast)

Transcription; nuclear factor binding

45e

(continued )

Near to substrate binding site

Mostly

5000 – 8000

Yes 3

High Microcephalic No susceptibility encephalopathy, to infection growth retardation, movement and speech disability No data (413) 12 years No data (48) No (Still alive in adulthood)

No

Yes

2

Ile170Val

Near to substrate Substrate binding site binding; and to dimer flexible interface loop

No

Yes 30

No data (48) No data (48)

Very mildc, yes Yes

2

Cys41Tyr

Absolutely

– NDd

4 days

No time to check due to the early death Very severe

Yes

1

Leu28del 2nt

b

1600

a

In combination with Glu104Asp

1800

Moderate 28

No data (48)

ND

Yes

Yes

Severe

?

Yes

Cardiomyopathy, Similar to increased mitochondrial susceptibility miopathy to infection

1

-5-8-24

1

Val231Met

Most frequent (*20) Yes

Glu104Asp

In most cases in early childhood Thermolability Yes TPI activity 2 – 20 in RBC, % of normal DHAP in RBC, 5000 – 10000 % of normal Amino acid Absolutely conservation Role in TPI Near to dimer function (8) interface

Death at age

Other symptoms

Number of case reports Hemolytic anemia Neurological disorder

Mutation

Homozygous

Compound heterozygous

Table 1 Biochemical, genetic and clinical characteristics of the known TPI-deficiency cases

704 OROSZ ET AL.

(62) (26, 27)

(5) etc

Low activity in white blood cells

Val231Met

(5)

Glu104Asp

(15, 23)

(14)

Mutation in CPE (cap proximal element) and TATA box;

-5-8-24

a

(19)

(19)

Leu28del 2ntb

(10, 24, 57)

(10)

Two unrelated cases. Axonal neuropathy proven by peripheral nerve biopsy

Cys41Tyr

(10, 24)

Two unrelated cases. Increased Km for GAP; Instability expected but not observed (9, 10)

Ile170Val

In combination with Glu104Asp

(6)

(6)

Arg189Stop

Compound heterozygous

(19)

(14)/(23)

Mutation in CPE (cap proximal element)

-5-8a/ Met Init Lys

(7, 51)

Two brothers with the same mutations; Activity of the recombinant Phe240Leu is 30% (29)/(7)

Glu145stop/ Phe240Leu

Nonsense mutations with italics/missense ones with bold a -5-8 and -5-8-24 stand for -5 A 4 G, -8 G 4 A, and -5 A 4 G, -8 G 4 A, -24 T 4 G nucleotide substitutions, respectively. Numbering refers to the transcription initiation site at -38 and takes the additional C into account (Cf. (23)). b L28del 2nt stands for 28 (CTG: Leu 4 C del 2nt: frameshift). c Only neonatal jaundice (57). d TPI activity for L28del 2nt/wild type was 38% of the normal value (515 U/gHb) (19). e Decreased activity due to the second mutation (Met Init Lys). TPI activity for Met Init Lys/wild type was 49% of the normal value (650 U/gHb). TPI activity for -5-8/wild type was normal (1190 – 1600 U/gHb) (19)) or slightly decreased (85% (15)).

Mutation reference Case reference

Remarks

Mutation

Homozygous

Table 1 (Continued ) TRIOSEPHOSPHATE ISOMERASE DEFICIENCY 705

706

OROSZ ET AL.

both significantly contribute to chronic neurodegenerative disorders such as Alzheimer’s, Parkinson’s or Huntington’s diseases. In the case of TPI deficiency, the role of toxic protein effects and impaired energetics in disease pathogenesis requires further study.

TRIOSEPHOSPHATE ISOMERASE MUTATIONS AND DISEASE A series of reports describing the cloning, sequencing and molecular characterization of the human TPI gene were published by Maquat and co-workers (3, 4). This lab also identified specific mutations from erythrocytes of human patients (5 – 7). Patients with various inherited mutations of the isomerase, including homozygotes and compound heterozygotes, have been identified in laboratories worldwide. These data demonstrate that more than 20 families carried the Glu104Asp mutation while the other known mutations were found in one or two families ((8) and references therein) (Table 1). The repeated occurrence of the 104 mutation in multiple unrelated families was extensively investigated and reviewed by Schneider (8). The CD4 gene lies telomeric to TPI and contains a 162 base pair short tandem repeat sequence and an intronic single nucleotide polymorphism (G to A) at position 2262 of the TPI gene. Both haplotypes were found to be in complete linkage disequilibrium with the TPI Glu104Asp mutation (9, 10). These studies revealed that the CD4 162, TPI 2262A haplotype was present in all Glu104Asp homozygotes, but was found in only 5% of the normal chromosomes, indicating a common origin of these mutations. People carrying the Glu104Asp mutation are probably descendants of a common ancestor of West-European origin. All of the known families with the Glu104Asp mutation were included in studies that suggest the original mutation likely occurred well in excess of 1000 years ago. Since then several other cases have been reported in Italy (11) and Turkey (12). A very recent paper has reported an exceptional case from Spain (13). In this case, although the intronic marker was the same as other known patients with the Glu104Asp mutation, the polymorphic tandem repeat marker in CD4 gene was different. This finding indicates that the Spanish mutation may have a different origin. Mutations in the TPI promoter region were also identified. In 1996, Watanabe and co-workers (14) identified three different mutations within the TPI gene promoter during a population survey. These mutations lie within, or are in close proximity to known cis-active regulatory elements in the TPI gene promoter. They are the -5A/G and the -8G/A mutations within the cap proximal element, and the -24T/G within the TATA box. Other laboratories have reported similar findings (15 – 17). These mutations, especially the -5A/G, occurred with high frequency in populations of Sub-Saharan origin but did not occur in Caucasians. Detailed investigation has revealed the high frequency of -5A/G change in Asian populations as

well (17), suggesting this is single nucleotide polymorphism. This conclusion is supported by the finding of a G at the -5 position in the chimpanzee (18) and the fact that -5G polymorphism has no detectable effect on erythrocyte TPI activity (16). In contrast to the -5A/G substitution, the -8 and -24 substitutions were associated with a modest but significant diminution in enzyme activity (15, 16). This is not surprising taking into account the functional significance of these sites. The cap proximal element sequence, which involves the -8 site, binds a 110 kDa nuclear factor (4). This site is thought to be essential for transcription-initiation (14), however, no difference was found in the affinity of protein binding of the variant haplotypes (16). Homozygotes for the -5A/G, -8G/A allele have a significant (25 – 35%) reduction in TPI activity that does not have any clinical relevance (15, 16). The -24T/G mutation, which was found only in a few cases and as the -5A/ G, -8G/A, -24T/G haplotype, appears not to further reduce the TPI activity (14, 15). Thus, it was concluded that the promoter mutations could not alone cause clinical TPIdeficiency (16), but might contribute to it (8, 15). Indeed, in two compound heterozygote cases (inheritance of two distinct TPI mutations), where mutations occurred in the promoter region and other positions (Glu104Asp/-5,-8,-24 and Met Init. Lys/-5,-8), severe clinical symptoms were diagnosed, which demonstrates the clinical significance of these promoter mutations (15, 19) (cf. Table 1). Interestingly, Met Init. Lys/-5,-8 does not result in hemolytic anemia, which is in accordance with the relatively high TPI activity (45% of the control). However, severe progressive neuromuscular dysfunction and other neurological symptoms were observed (Table 1) suggesting this may result in a neuronal-specific form of TPI deficiency (19).

MAJOR CHARACTERISTICS OF TPI Human TPI is encoded by a single gene located at chromosome 12p13 and is expressed in all tissues. Its amino acid sequence is highly conserved among all known TPI proteins (8, 20). Sequence analysis revealed that there are seven exons and six introns (3, 4). The gene product is a housekeeping glycolytic enzyme, which catalyzes the interconversion of D-glyceraldehyde-3-phosphate (GAP) and DHAP, causing a rapid equilibrium of triosephosphates in favor of DHAP formation. In fact, TPI has been described as a nearly perfect enzyme since the high catalytic capacity is a diffusion-limited process (21). TPI is a stable homodimer of two 27 kDa subunits consisting of 248 amino acids. Three loops of the N-terminal half of the molecule are involved in the intersubunit interactions (20) (Fig. 1). TPI is catalytically active only in its dimeric form. The 3-dimensional structure of human recombinant isomerase obtained by crystallography at a resolution of 0.28 nm (cf. Fig. 1A) showed an asymmetric unit which contained 1 complete dimer with only one of the subunits binding

TRIOSEPHOSPHATE ISOMERASE DEFICIENCY

Figure 1. 3D structure of TPI with marked amino acid substituted by inherited mutation. (A) Schematic ribbon diagram of the crystal structure of recombinant human triosephosphate isomerase (20). Lys 13, His 95 and Glu 165 (all blue) are the active site residues. Amino acid residues substituted by inherited mutations are Cys41, Glu104, Ile170, Val231 and Phe240. (B) Molecular dynamic structure of double mutant TPI (33). The green subunit contains the Phe240Leu missense mutation. The red subunit contains the Glu145stop codon truncated protein. The position of a substrate analogue, 2-phosphoglycolate (PG), is shown in both structures.

2-phosphoglycerate (20, 22). Three residues, Lys13, His95 and Glu165 form the active site. However, it was proposed that residues indispensable for enzyme activity exist throughout the C-terminal region of the protein, with the possible exception of the ultimate few amino acids (6). The importance of C-terminal amino acids 189, 201, 206, 208 – 211, 228, 230 – 236, and 240 is implied by their conservation in the TPI of most of the species that have been characterized (20). The traditional numbering of TPI amino acids assumes the removal of the N-terminal Met, which may not occur. Some papers use a numbering system that includes the initial Met (23).

TPI MUTATIONS: STRUCTURE AND FUNCTION Genetic studies of multiple unrelated families have shown that a single mutation, substitution of aspartate for glutamate at residue 104 (cf. Fig. 1A), results in loss of activity due to instability of the mutant enzyme (24). This amino acid is strictly conserved in all the known TPI sequences (8, 20). The Glu104Asp substitution lies at the subunit interface of the TPI dimer. The X-ray crystal structure of chicken TPI suggests that the loss of a side-chain methylene group (-CH2CH2COOto -CH2COO-) is sufficient to disrupt the counterbalancing of charges that normally exists within a hydrophobic pocket of

707

the native enzyme. The amino acid substitution results in loss of activity due to the instability of the mutant enzyme, which likely results from the dissociation of the active dimers into inactive monomers (24). The instability of this mutant enzyme was demonstrated by assays of TPI activity in cultured fibroblasts from patients and in cultured Chinese hamster ovary (CHO) cells that were stably transformed with mutant alleles (5). The necessity of the dimeric form for catalytic function and stability of TPI was shown by Manfroid et al. (25). They produced recombinant enzymes with a mutation in amino acids Met14 or Arg98, which are located at the dimer interface. Both mutants showed not only a decreased stability, but partial or complete loss of activity due to the increased dissociability of the dimeric enzyme. Since Glu104 is a near neighbor of Arg98 and is also localized at the dimer interface (cf. Fig. 1), the extreme sensitivity of the enzyme with the Glu104Asp mutation is not surprising. Several other causative mutations (missense, nonsense and frameshift) distributed throughout the TPI gene have been described (8, 23) (cf. Table 1). In addition to the Glu104Asp homozygotes, another type of homozygosity (Val231Met) has been identified (cf. Table 1). The amino acid Val at 231 position is conserved from bacteria to man, is known to be involved in substrate binding, and is predicted to significantly alter catalytic activity (8). The Val231Met mutation is associated with decreased isomerase activity, increase in intracellular glycogen, and mitochondrial changes similar to those seen in mitochondrial myopathies (26, 27). TPI deficiency can be the result of homozygosity of missense mutations or due to compound heterozygosity. Compound heterozygosity was demonstrated with the following missense mutation combinations: Cys41Tyr/Glu104Asp and Ile170Val/ Glu104Asp (10). There are three cases of compound heterozygosity that have been described where missense mutations are coupled with mutations producing premature stop codons due to nonsense or frameshift mutations. In one case a patient with the genotype Glu104Asp/Leu28del (2nt frameshift mutation) died after the 4th day of birth, which did not allow detailed investigation (19). Patients with Glu104Asp/ Arg189Stop and Phe240Leu/Glu145stop compound heterozygosity have also been described (6, 7, 29). TPI mRNAs with premature stop codons were the first to be shown to be efficiently targeted by nonsense-mediated mRNA decay, dramatically reducing the cytoplasmic half-life of the TPI mRNA (6, 7, 28). Due to nonsense-mediated mRNA decay, it is not known whether any appreciable truncated proteins are expressed from this class of TPI mutation. The Arg189Stop mutation has been examined and does not posses detectable isomerase activity, although, if it were to express a truncated protein it would contain all three active site amino acids (6). Fig. 1A illustrates the 3D structure of TPI determined by crystallography (20). The positions of the substituted amino

708

OROSZ ET AL.

acids identified in TPI deficiency are indicated in the 3D structure. Schneider suggested the division of these mutations into three classes – substrate binding, flexible loop and subunit contact domains – to evaluate the structure-function relationships (8). Accordingly, mutations (Glu104Asp, Cys41Tyr, Val231Met) at the contact surface of the dimeric TPI were suggested to have led to enzyme instability. While the involvement of amino acid 104 (Glu104Asp) is clearly supported by the 3D structural model, dimer instability of Cys41Tyr and Val231Met is only suggestive (Fig. 1A) (cf. Table 1). Additionally, time-dependent activity decrease at elevated temperature was demonstrated for the Phe240Leu mutation (30), although it is not predicted to affect the dimer interface. Catalytic defects were predicted for the Cys41Tyr, Ile170Val, Val231Met and Phe240Leu mutations due to their localization in or near to the substrate-binding site (8). It was proposed that both residues at positions 170 and 231 are involved in the substrate binding, yet the mutants Ile170Val and Val231Met showed the lowest and highest isomerase activities, respectively, in the hemolysates of the patients (cf. Table 1). It seems that the static 3D model does not capture all the functional relationships between the mutations and catalytic defects in the case of TPI deficiency. There are two plausible explanations for this discrepancy: (i) some mutations may cause local or global conformational changes that are not evident using the static model; and (ii) protein activity is typically assayed in hemolysates of the patients where intracellular associations could further alter either structure or function. Molecular dynamic modeling renders it possible to study in silico the mutation-induced conformational changes at the atomic level. Such types of investigation were carried out with mutations occurring in the Hungarian Phe240Leu/Glu145stop patient. Modeling data have shown that position 240 (Phe or Leu) is not in the active centre of the isomerase. These data suggest the amino acid is not directly part of the active site but do not rule out an essential function in maintaining active site geometry (7). Dynamic modeling has revealed that the 240 mutation induced local conformational changes that do not spread to the contact surface area which could be responsible for the decrease in TPI activity (30, 31). If expressed, the Glu145stop truncated protein would include the contact surface loops, as revealed by molecular dynamic modeling, resulting in a heterodimer of the full-length species and the truncated protein without energetic restriction in the compound heterozygote patient. The functional consequence of the full-length/truncated heterodimer is not known but may result in dramatically reduced catalytic activity. This idea was supported by experimental observations, namely, isomerase activity measured from the hemolysate of a Phe240Leu/Glu145stop patient, which was 10-fold lower than the activity of human recombinant Phe240Leu mutant (30).

HETEROASSOCIATION OF MUTANT TPI Recent studies have revealed that the association of TPI to subcellular particles decreases isomerase activity. Both wild type and the Phe240Leu mutant TPI protein bind to the N-terminus of the band 3 transmembrane protein of erythrocytes. Interestingly, the mutant protein appears to bind more efficiently, suggesting the formation of a heterodimer or the role of other not-yet identified factors, i.e., enhancement or defect of piggy-back binding, microcompartmentation. How relevant is the heteroassociation of TPI detected in erythrocytes to brain physiology? There is no clear answer to this question. However, there are two issues which could help our understanding: (i) heteroassociation of glycolytic enzymes to the cytoskeleton including microtubules, the major constituent of neuronal axon, with purified enzyme and in brain extract have been demonstrated ((32, 33) and references therein); and (ii) the N-terminus of band 3 trasmembrane protein shows high homology with the C-terminus of tubulin, the major binding domain of tubulin exposed to the surface of the microtubules (34). Consequently, experiments performed with human recombinant Phe240Leu TPI or with hemolysates of Hungarian patients and microtubules isolated from brain may model key aspects of brain physiology. Such experiments revealed that both the wild type and the Phe240Leu isomerases bound to microtubules, however, the binding of the mutant enzyme from the hemolysate, where heterodimeric TPI could be formed, was more efficient (30). Since the binding of mutant TPI heterodimers to microtubules is thought to further decreases isomerase activity, such structural and functional alterations may induce metabolic or ultrastructural defects.

CONSEQUENCES OF TPI IMPAIRMENT AT THE SYSTEM LEVEL Most of the results published in relation to TPI deficiency were obtained with erythrocytes of patients. These data have provided key insight into the disease pathogenesis, however, erythrocytes are unique cells, and we should be cautious about inferences to neuromuscular aspects of the disease. Specifically, in mature erythrocytes DHAP is metabolized exclusively by glycolysis; TPI adjusts the rapid equilibrium of triosephosphates, and GAP is metabolized by the downstream enzyme, glyceraldehyde-3-phosphate dehydrogenase. Therefore, impairment of this function results in extensive elevation of DHAP concentration, more extensively than in those cells in which DHAP is consumed by other pathways, i.e., by glycerol3-phosphate dehydrogenase, an oxidoreductase, which accepts NADH or NADPH as co-substrates depending on the supply. Consistent with these data, platelets carrying the Phe240Leu/ Glu145stop mutation have a 2-fold increase in DHAP concentration, whereas, in erythrocytes a 60-fold increase was reported (30). The product of glycerol-3-phosphate


dehydrogenase catalyzed reaction is glycerol-3-phosphate, a common substrate for formation of glycerolipids such as phospholipids and triacylglycerides, suggesting this is not a likely toxic end product. Although there is no indication that DHAP itself is toxic, it decomposes nonenzymatically to methylglyoxal. Methylglyoxal is a highly reactive a-oxo-aldehyde that can modify both proteins and DNA to form advanced glycation end products (AGEs) (35). The accumulation of glycation adducts results in deleterious consequences including oxidative stress, DNA damage, and apoptosis. Also, methylglyoxal is toxic to neurons and may contribute to the neurological symptoms observed in patients (36). Under oxidative stress conditions glyoxalases cannot efficiently detoxify methylglyoxal, which may underlie the associated neurodegeneration (37). DHAP is also a precursor of acyl-DHAP, which is an intermediary in the synthesis of glycerol ether lipids (alkyl glycerol ethers and plasmalogens) as well as being converted into non-ether glycerolipids (38). Decreased plasmalogen levels were detected in the membranes of erythrocytes and lymphocytes of TPI deficient Glu104Asp (39) and Phe240Leu/ Glu145stop (38, 40) patients. Decreased plasmalogen levels can influence many membrane-related processes and impair oxidative stress protective mechanisms (40). Research on TPI deficiency seems to be entering a new stage by adapting new model systems, such as Drosophila and yeast strains, which can improve our understanding of the pathological basis of this disease. The high degree of TPI sequence conservation from bacteria to human and the high degree of structural similarity observed in TPI crystals (20) suggests that information obtained from flies and yeast will be directly relevant to humans. Mutant yeast was constructed with a loss-of-function tpi1 allele that has provided insight into TPI deficiency (41). In yeast, the Asn65Lys tpi1 mutation completely abolished TPI enzyme activity and led to a 30-fold increase in the intracellular DHAP concentration. The tpi1 mutant was unable to grow in the absence of inositol and exhibited the ‘inositol-less death’ phenotype. DHAP competitively inhibits the activity of myo-inositol-3-phosphate synthase, the rate-limiting enzyme in de novo inositol biosynthesis, suggesting inositol depletion may be an important contributor to disease pathogenesis. Defects in brain inositol concentration have been reported in a number of pathological conditions including bipolar disorder, Down Syndrome and diabetic peripheral neuropathy (42), suggesting this may be especially important to neurological impairment associated with TPI deficiency. Recently, Drosophila mutant models have been identified to study human TPI deficiency. The architecture of the fly nervous system is not unlike that of mammals, with areas that separate specialized functions as vision, learning and memory, suggesting the fly might enable the study of disease progression and prove especially important to the study of neuropathogensis. Using forward genetic screens, two independent

709

research groups identified a TPI-deficient fly mutant with phenotypes showing analogous symptoms to human TPI deficiency, including: progressive locomotor impairment, temperaturesensitivity, vacuolar neuropathology, and severely reduced lifespan (43, 44). A recessive hypomorphic TPI mutant of a fly carrying the Met80Thr mutation (which corresponds to Met82 in human TPI) was identified that features the most important characteristics of TPI-deficiency (43, 44). This mutation affects a conserved methionine residue that resides at the TPI dimer interface. Mutation at the dimer interface results in the instability of the isomerase coupled with decreased activity, leading to the proposal that dimer stability might underlie the temperature-sensitivity observed in the flies (43). Therefore, not surprisingly, the above mentioned characteristic features of the mutant fly show the best parallel with the Glu104Asp human mutation, namely a progressive neural degeneration associated with temperature-sensitivity. Biochemical studies on mutant Drosophila models demonstrated that the phenotypes were not the result of impaired bioenergetics, suggesting an alternative cause of enzymopathy associated with TPI impairment (43, 44). Interestingly, bioenergetic impairment was not detected in Drosophila TPI mutants despite markedly reduced lactic acid levels (43). It was hypothesized that the mutations in TPI may lead to an accumulation of methylglyoxal and the consequent enhanced production of AGEs. These suggestions are in agreement with those published so far for the human disease.

EFFECT OF TPI MUTATIONS ON THE ENERGY METABOLISM Erythrocyte metabolism is well-established and the catalytic properties of the glycolytic enzymes have been extensively characterized. Accordingly, mathematical models have been developed for the glycolytic and related metabolic fluxes of human erythrocytes that accurately fit experimental data in normal cells (45 – 47). The models failed to simulate the increase in DHAP concentration measured experimentally in TPI-deficient cells unless a very low TPI activity was introduced, which was not supported by experimental data (45, 47). Different explanations were provided for this striking discrepancy including the possibility that the in vivo TPI activity is significantly lower than predicted from in vitro data, which may result from instability of the mutant protein (2, 45, 47) or the binding of mutant TPI to intracellular components (48). As an alternative explanation, the role of reticulocytosis was suspected. Due to reticulocytosis, enzyme activities determined in erythrocytes may be often overestimated (47). However, this does not seem sufficient to account for all of the data, since there are instances where only a slight increase in reticulocyte measurement (3%) was detected (7) and lactate dehydrogenase and enolase activity were no different between control and TPI deficient cells.

710

OROSZ ET AL.

The metabolic effect of TPI deficiency has been measured experimentally in hemolysates from patients, and the concentrations of several important compounds were determined (50, 51). Since DHAP can be further metabolized only by the TPI catalyzed reaction in erythrocytes (cf. Fig 2), it was expected that reduced TPI function will decrease the rate of glycolysis, resulting in a reduction of the rate of glycolytic ATP synthesis and an accumulation of DHAP. In addition, alterations in the concentrations of other metabolites, both upstream and downstream of TPI in glycolysis, were assumed. However, alterations in glucose utilization, ATP and lactate production and in glycolytic intermediates were found to be unimpressive. The apparent discrepancy of elevated DHAP and normal ATP levels associated with impaired TPI activity in erythrocytes has been reconciled by a new mathematical model (52). The mathematical model is based on well-established kinetic equations and the experimentally determined kinetic parameters of the glycolytic enzymes in the hemolysate of the Phe240Leu/Glu145stop compound heterozygote Hungarian patient. It was found that the low TPI activity in the mutant cells was associated with an increase in activity of the other glycolytic enzymes, including glycolytic kinases, which are known to predominantly control glycolytic flux. The biosimulation data (cf. http://www.BiochemJ.org/bj/ 392/bj3920675add.htm) showed that the rate of glucose conversion into pyruvate or lactate was higher for the mutant cell, and the steady-state flux was increased 2.5-fold due to the enhanced activities of hexokinase, phosphofructokinase, glyceraldehyde-3-phosphate dehydrogenase and pyruvate kinase (Fig. 3). The model predicted significant enhancement both in the DHAP and fructose 1,6-bisphosphate levels (52). In accordance to the simulation, the experimentally determined concentrations of DHAP and fructose 1,6-bisphosphate increased 40- and 5-fold, respectively, in the erythrocytes of the patient as compared to the control. The major conclusion of this analysis was that equilibrium of triosephosphates was not achieved in the mutant cells causing DHAP to remain at a significantly higher concentration than in the control. These perturbations led to the conclusion that the energy state of the cells was not abnormal and ATP concentration was maintained at normal level due to the activation of key regulatory enzymes. However, it is unclear whether the activation of glycolytic enzymes is unique to the Phe240Leu/Glu145stop mutation or if this is general feature of TPI deficiency. It has also been suggested that activation of pentose phosphate pathway (PPP) may be a compensatory mechanism allowing normal ATP level despite impaired TPI activity (8). PPP is interconnected with glycolysis via the intermediates, glucose-6-phosphate and GAP (Fig. 2). It was suggested that the shunt functions at near-maximal level during TPI deficiency, which could explain the increased glucose utilization. This is quite possible since activation of glucose-6-phosphate

dehydrogenase was detected in hemolysates of the Hungarian Glu145stop/Phe240Leu patient (31). Consequently, the PPP shunt can be activated by increased glucose-6-phosphate dehydrogenase activity, the first enzyme of the shunt, which predominantly controls the pathway. Including PPP in the new mathematical model using experimentally determined parameters of the enzymes resulted in a predicted modest reduction in the steady-state rate of pyruvate formation in hemolysates of both the normal control and the patient (Fig. 3). This finding suggests that the TPI mutation resulting in decreased activity is compensated exclusively by the activation of the other glycolytic enzymes, and that PPP activation does not contribute to the maintaining of normal ATP level (Ola´h et al., unpublished result). The mathematical model may be appropriate for biosimulation in other tissues where experimentally-determined data are sufficient to set the parameters (cf. Fig. 3). The model was also used with lymphocytes, where the impairment of TPI activity is much less significant (20%). The mutant enzyme is predicted to be able to adjust the rapid equilibrium of triosephosphates and no compensatory mechanism is needed.

UNIQUE CHARACTERISTICS OF TPI DEFICIENCY IN A HUNGARIAN FAMILY A unique Hungarian pedigree was identified by Susan Hollań in 1990 in which two germ-line identical but phenotypically different compound heterozygote brothers inherited two independent mutations, Phe240Leu and Glu145stop codon, from their mother and father, respectively. The mutations result in severely decreased TPI activity, however, only the younger sibling (affected brother) manifests a neurologic disorder. The entire genomic TPI locus (exons, introns and promoter) was sequenced and found to be identical in the two compound heterozygote brothers (7, 29). Both brothers have the same well-compensated level of nonspherocytic hemolytic anemia and both the heterozygous mother and father are symptom-free (Table 2). Studies on this unique Hungarian pedigree continue to provide an excellent opportunity to study the pathomechanism of TPI deficiency. The reduction in TPI activity and increase in DHAP are similar for the two brothers. The differences in features for the healthy and the affected brother may provide key information about the etiology of neurodegenerative symptoms associated with TPI deficiency. The following differences were observed that may prove important to disease pathogenesis: (1) Polyethylene glycol enhances macromolecular associations (30, 48, 53) and enhances binding of TPI to microtubules or to red cell membrane in normal control samples and in the healthy Hungarian brother. However, polyethylene glycol did not enhance binding in samples from the affected Hungarian patient or a British patient


711

Figure 2. Glycolysis and related pathways in erythrocyte and in brain. Glycolysis is a central metabolic machinery in which one molecule of glucose is catabolysed to two molecules of pyruvate, NADH and ATP. Under aerobic conditions, pyruvate is further oxidized by the mitochondrial system. Glycolysis and pentose phosphate pathway (PPP) are interconnected via fructose-6-P and glyceraldehyde-3-P. A high level of NADPH favors lipid synthesis via the pentose phosphate shunt. In erythrocytes DHAP is a dead-end product, however, in brain it can be diverted into lipid synthesis. a-GDH catalyses the reduction of DHAP to glycerol-3-P coupled with NADH to NADþ conversion; glycerol-3-P is a substrate for the biosynthesis of glycerolipids (i.e., phospholipids, triacylglyceride). At a low level of NADPH, DHAP can be transformed to plasmalogens or AGE via methylglyoxal. Enzymes abbreviations are as follows: Bisphosphoglycerate mutase, BPGM; Bisphosphoglycerate phosphatase, BPGP; Enolase, ENO1; Glucose-6-phosphate dehydrogenase, G6PDH; Glucose-6phosphate isomerase, GPI; Glutathione, GSH; Glyceraldehyde-3-phosphate dehydrogenase, GAPDH; Glycerol-3-phosphate dehydrogenase, GDH; Hexokinase, HK; Oxidative phosphorylation, O.P.; Phosphofructokinase, PFK; Phosphoglycerate kinase, PGK; Phosphoglycerate mutase, PGM; Pyruvate kinase, PK; and Tricarboxylic acid cycle, TCA.

712

OROSZ ET AL.

similarly low in both brothers) and DHAP was near normal in both. It was suggested that the altered gene expression of TPI may originate from additional inherited genetic or epigenetic factors (30). (6) Significant differences between the mRNA levels of nitric oxide synthase and prolyl-oligopeptidase (POP) were found in the lymphocytes of the two Hungarian brothers (52). These two enzymes have been suggested to be indirectly involved in neurodegenerative diseases (54, 55) and nitration of TPI has been documented in Alzheimer’s disease. A 15-fold increase in urinary 3-nitrotyrosine was found in the affected brother (37). POP activity in the cells of the affected brother was reduced by about 40% with respect to the control (and the healthy brother). Since POP plays a key role in neurotransmission, in addition to other functions, it has been suggested that the reduction of POP activity in brain tissues as an early response to cellular stress, may contribute to the development of the neurodegenerative process in Alzheimer’s, Parkinson’s and Huntington’s diseases (55). Figure 3. Biosimulation of red blood cell glycolysis coupled with or without pentosephosphate pathway (PPP). The new mathematical model (http://www.BiochemJ.org/bj/392/ bj3920675add.htm) was used to compute the fluxes of glucose metabolism in the case of normal control (solid line) and TPI deficient patient (dotted line). The simulated fluxes are also shown in the case of normal control (bold solid line) and of TPI deficient patient (bold dotted line) assuming glycolysis was coupled to PPP. The parameters for the model were determined experimentally.

(2)

(3)

(4)

(5)

carrying the Glu104Asp mutation (30). This finding may indicate that an altered structure or protein-protein interactions underlie neurological symptoms. Plasmalogen was found to be lower in the neurologically affected Hungarian brother than in the healthy sibling (40). The lower plasmalogen level can influence membrane-related processes and protection against oxidative stress. The neurologically affected sibling had an increase in oxidized glutathione, glutathione S-transferase activity, D-lactate and a more pronounced decrease in alphatocopherol (1). This indicates an imbalance of the prooxidant/antioxidant homeostasis that may lead to the shortened lifespan of erythrocytes and likely has important neurological consequences. The activities of acetylcholinesterase and calmodulininduced Ca2þ-ATPase were significantly enhanced in erythrocytes from the affected brother (38). Both enzymes are crucially involved in the function of nerve cells. In lymphocytes TPI mRNA expression is higher in the affected brother. However, this did not result in a difference in TPI protein or activity (TPI activity was

These studies on this unique Hungarian pedigree have provided many compelling insights that have opened new avenues of research that may elucidate the mechanisms of TPI impairment and the associated neurological dysfunction. It will be helpful to perform the assays outlined above on TPI deficiency patients carrying different mutations to compare with the data from Phe240Leu/Glu145stop patients.

BEYOND OUR UNDERSTANDING Classic interpretation of TPI deficiency is based on the early observation that inherited mutations decrease TPI activity and prevent the normal rapid adjustment of triosephosphate equilibrium. In erythrocytes, reduced TPI catalytic activity is coupled with an extensive increase of DHAP concentration (cf. Table 1). These biochemical alterations are typically associated with hemolytic anemia and neurological dysfunction: hallmarks of TPI deficiency enzymopathy. Varying severity of the syndrome may be well associated with different degrees of enzyme deficiency, however, biochemical activitygenotype comparisons for the specific mutations are not straightforward. Past research has provided examples of cases where TPI deficiency did not cause both hemolytic anemia and a neurological disorder (e.g., promoter region mutations causing only anemia and the Ile170Val mutation causing only neurological symptoms). In addition, the unique Hungarian pedigree, where two brothers carry the same inherited mutations (resulting in nearly the same decrease of TPI activity and increase in DHAP level) yet only the younger brother is affected neurologically, exemplifies that fact that we do not fully understand the nature of this disease. The major hurdle to elucidating the pathomechanism of TPI deficiency is the lack of brain tissues available for


713

Table 2 Distinct characteristics of the Hungarian compound heterozygote brothers

TPI activity in

DHAP level in TPI amount in mRNA in lymphocyte Binding to microtubule Binding to membrane POP in lymphocyte

Acetylcholinesterase Calmodulin-induced Ca2þ-ATPase NADPH D-lactate Methylglyoxal Urinary 3-nitrotyrosine Neurological disorder

RBC, U/g Hb Lymphocyte, U/mg protein Platelet, U/1010 cells RBC, mM Platelet, nmol/1010 cells RBC, mg/g Hb Lymphocyte, mg/mg protein molecules/cell Bound TPI, % þPEG Bound TPI, % þPEG mRNA, molecules/cell Activity, nmol substrate hydrolysed hour71 (g protein)71 Activity, % Relative amount of protein, % Vmax nmol of Ca2þ per mg of protein/min mM/liter Plasma, nM/ml Urine, mmol/24 hour pmole/ml RBC nmol/mmol creatinine

experimental purpose. Postmortem examinations of TPI deficiency brain samples are rare (56). Recently an Australian compound heterozygote (Glu104Asp/Cys41Tyr) patient has been diagnosed with chronic axonal neuropathy (57). Since most TPI patient studies have utilized erythrocytes, our understanding of brain-related symptoms is less complete and our understanding of peripheral neuropathy associated with TPI-impairment is in its infancy. Lymphocytes, which share some characteristics with nerve cells, may provide some insight into neurological symptoms of TPI deficiency (52, 58). In lymphocytes 20% residual activity of mutant TPI is enough to adjust the rapid equilibrium of triosephosphates resulting in near normal DHAP level. Nevertheless, one can assume that further decrease of TPI activity could take place in vivo due to the formation of heterodimeric species and/or rearrangement of macromolecular assemblies of the glycolytic enzymes, which was suggested in the postsynaptic density of brain tissue (59). These alterations apparently did not result in accumulation of DHAP, however, subtle changes to the rate of triosephosphate conversion could result in the production of toxic compounds (AGEs) or abnormal lipid metabolism

Affected brother

Healthy brother

Control

33 3.0 40 701 267 145 0.95 108 11 11 4.2 5.3 30 0.865

41 3.3 40 547 163 119 0.88 47 5 12 6.5 10.2 22 1.14

1400 15 181 11 136 330 1.75 48 1.5 3.1 0.53 1.1 20 1.45

140 – 155 290 5.4 55.0 44.8 119 724 35.6 Yes

110 – 118 120 3.2 42.9 35.3 49 625 2.80 No

100 100 3.2 37.1 12.5 28 455 5.16 No

(decreased plasmalogen level) that could have important chronic neurological affects. Studies of DHAP-related metabolite pattern from post-mortem brain tissue of patients would accelerate our understanding of the pathological consequences of the TPI deficiency. Conformational diseases result from unstructured or misfolded proteins, like a-synuclein, tau or mutant huntingtin protein. These proteins enter into aberrant protein-protein interactions leading to pathological protein aggregates (for review see (33)). Protein aggregates could also secondarily affect energy and other physiological processes, such as ion homeostasis, protein synthesis, transport and degradation, any of which could contribute to neuropathogenesis (60, 61). We have little information on the conformation of the TPI substituted in different amino acids. One crucial remaining question is whether TPI deficiency is a metabolic disease or a conformational disease. The second possibility has arisen because no bioenergy deficit (no ATP deficiency) has been demonstrated in any cases of TPI mutation. This was also true in the Drosophila animal model of TPI deficiency (43, 44). Further investigations the recently developed Drosophila model of TPI deficiency will hopefully

714

OROSZ ET AL.

contribute to our understanding of the mechanisms responsible for the symptoms of TPI deficiency in humans.

ACKNOWLEDGEMENTS We would like to express our appreciation to Professor Susan Hollań of the Central Institute of Blood Transfusion Service, Budapest, for inviting us to collaborate on the TPI deficiency project in relation to her Hungarian patients. We thank Dr Michael Palladino of the University of Pittsburgh for the critical reading and revision of the manuscript. This work was supported by Hungarian National Scientific Research Fund Grants OTKA T-046071 and TS-044730 to J. Ova´di and T049247 to F. O.; FP6 – 2003-LIFESCIHEALTH-I: Bio-Sim and NKFP-MediChem2 1/A/005/2004 to J. Ova´di.

REFERENCES 1. Karg, E., Ne´meth, I., Horańyi, M., Pinte´r, S., Većsei, L., and Hollań, S. (2000) Diminished blood levels of reduced glutathione and alphatocopherol in triosephosphate isomerase deficient brothers. Blood Cells Mol. Dis. 26, 91 – 100. 2. Tanaka, K. R., and Zerez, C. R. (1990) Red cell enzymopathies of the glycolytic pathway. Semin. Hematol. 27, 165 – 185. 3. Maquat, L., Chilcote, R., and Ryan, P. (1985) Human triosephosphate isomerase cDNA and protein structure: studies of triosephosphate isomerase deficiency in man. J. Biol. Chem. 260, 3748 – 3753. 4. Boyer, T. G., and Maquat, L. E. (1990) Minimal sequence and factor requirements for the initiation of transcription from an atypical, TATATAA box-containing housekeeping promoter. J. Biol. Chem. 265, 20524 – 20532. 5. Daar, I. O., Artymiuk, P. J., Phillips, D. C., and Maquat, L. E. (1986) Human triose-phosphate isomerase deficiency: a single amino acid substitution results in a thermolabile enzyme. Proc. Natl. Acad. Sci. USA 83, 7903 – 7907. 6. Daar, I. O., and Maquat, L. E. (1988) Premature translation termination mediates triosephosphate isomerase mRNA degradation Mol. Cell. Biol. 8, 802 – 813. 7. Chang, M.-L., Artymiuk, P. J., Wu, X., Hollań, S., Lammi, A., and Maquat, L. E. (1993) Human triosephosphate isomerase deficiency resulting from mutation of Phe-240. Am. J. Hum. Genet. 52, 1260 – 1269. 8. Schneider, A. S. (2000) Triosephosphate isomerase deficiency: historical perspectives and molecular aspects. Baillie´re’s Clin. Haematol. 13, 119 – 140. 9. Schneider, A., Westwood, B., Yim, C., Cohen-Solal, M., Rosa, R., Labotka, R., Eber, S., Wolf, R., Lammi, A., and Beutler, E. (1996) The 1591C mutation in triosephosphate isomerase (TPI) deficiency. Tightly linked polymorphisms and a common haplotype in all known families. Blood Cells Mol. Dis. 22, 115 – 125. 10. Arya, R., Lalloz, M. R., Bellingham, A. J., and Layton, D. M. (1997) Evidence for founder effect of the Glu104Asp substitution and identification of new mutations in triosephosphate isomerase deficiency. Hum. Mutat. 10, 290 – 294. 11. Baronciani, L., Bianchi, P., Bredi, E., Pelissero, G., Iurlo, A., and Zanella, A. (1996) Triosephosphate isomerase deficiency: two Italian cases associated to mutation 104 (315G-C, 105 Glu-Asp). Blood. 88, 4b. 12. Yenicesu, I., Kalayci, O., Semizel, E., Kavak, U., Gumruk, F., Ferec, C., and Beutler, E. (2000) Triosephosphate isomerase deficiency with elevated sweat chloride test: report of a case. Turk. J. Pediatr. 42, 319 – 321.

13. Repiso, A., Corrons, J. L., Vulliamy, T., Killeen, N., Layton, M., Carreras, J., and Climent, F. (2005) New haplotype for the Glu104Asp mutation in triose-phosphate isomerase deficiency and prenatal diagnosis in a Spanish family. J. Inherit. Metab. Dis. 28, 807 – 809. 14. Watanabe, M., Zingg, B. C., and Mohrenweiser, H. W. (1996) Molecular analysis of a series of alleles in humans with reduced activity at the triosephosphate isomerase locus. Am. J. Hum. Genet. 58, 308 – 316. 15. Schneider, A., Forman, L., Westwood, B., Yim, C., Lin, J., Singh, S., and Beutler, E. (1998) The relationship of the -5, -8, and -24 variant alleles in African Americans to triosephosphate isomerase (TPI) enzyme activity and to TPI deficiency. Blood 92, 2959 – 2962. 16. Humphries, A., Ationu, A., Wild, B., and Layton, D. M. (1999) The consequence of nucleotide substitutions in the triosephosphate isomerase (TPI) gene promoter. Blood Cells Mol. Dis. 25, 210 – 217. 17. Humphries, A., Ationu, A., Lalloz, M. R., and Layton, D. M. (1999) Ancestral origin of variation in the triosephosphate isomerase gene promoter. Hum. Genet. 104, 486 – 491. 18. Craig, L. C., Pirtle, I. L., Gracy, R. W., and Pirtle, R. M. (1991) Characterization of the transcription unit and two processed pseudogenes of chimpanzee triosephosphate isomerase (TPI). Gene 99, 217 – 227. 19. Valentin, C., Pissard, S., Martin, J., Heron, D., Labrune, P., Livet, M. O., Mayer, M., Gelbart, T., Schneider, A., Max-Audit, I., and Cohen-Solal, M. (2000) Triose phosphate isomerase deficiency in 3 French families: two novel null alleles, a frameshift mutation (TPI Alfortville) and an alteration in the initiation codon (TPI Paris). Blood 96, 1130 – 1135. 20. Mande, S. C., Mainfroid, V., Kalk, K. H., Goraj, K., Martial, J. A., and Hol, W. G. J. (1994) Crystal structure of recombinant human triosephosphate isomerase at 2.8 A˚ resolution. Triosephosphate isomerase-related human genetic disorders and comparison with the trypanosomal enzyme. Protein Sci. 3, 810 – 821. 21. Albery, W., and Knowles, J. (1976) Evolution of enzyme function and the development of catalytic efficiency. Biochemistry 15, 5631 – 5640. 22. Lu, H. S., Yuan, P. M., and Gracy, R. W. (1984) Primary structure of human triosephosphate isomerase. J. Biol. Chem. 259, 11958 – 11968. 23. Schneider, A., and Cohen-Solal, M. (1996) Hematologically important mutations: triosephosphate isomerase. Blood Cells Mol. Dis. 22, 82 – 84. 24. Ationu, A., Humphries, A., Lalloz, M. R. A., Arya, R., Wild, B., Warrilow, J., Morgan, J., Bellingham, J., and Layton, D. M. (1999) Reversal of metabolic block in glycolysis by enzyme replacement in triosephosphate isomerase-deficient cells. Blood 94, 3193 – 3198. 25. Mainfroid, V., Terpstra, P., Beauregard, M., Fre´re, J.-M., Mande, S., Hol, W. G. J., Martial, J., and Goraj, K. (1996) Three hTIM mutants that provide new insights on why TIM is a dimer. J. Mol. Biol. 257, 441 – 456. 26. Bardosi, A., Eber, S. W., Hendrys, M., and Pekrun, A. (1990) Myopathy with altered mitochondria due to a triosephosphate isomerase (TPI) deficiency. Acta Neuropathol. 79, 387 – 394. 27. Eber, S. W., Pekrun, A., Bardosi, A., Gahr, M., Krietsch, W. K. G., Kru¨ger, J., Matthei, R., and Schro¨ter, W. (1991) Triosephosphate isomerase deficiency: haemolytic anaemia, myopathy with altered mitochondria and mental retardation due to a new variant with accelerated enzyme catabolism and diminished specific activity. Eur. J. Pediatr. 150, 761 – 766. 28. Cheng, J., and Maquat, L. (1993) Nonsense codons can reduce the abundance of nuclear mRNA without affecting the abundance of premRNA or the half-life of cytoplasmic mRNA. Mol. Cell Biol. 13, 1892 – 1902.


29. Valentin, C., Cohen-Solal, M., Maquat, L., Horańyi, M., Inselt-Kovaćs, M., and Hollań, S. (2000) Identical germ-line mutations in the triosephosphate isomerase alleles of two brothers associated with distinct clinical phenotypes. C R Acad. Sci. III 323, 245 – 250. 30. Orosz, F., Ola´h, J., Alvarez, M., Keseru¨, G. M., Szabo´, B., Wa´gner, G., Kova´ri, Z., Horańyi, M., Baro´ti, K., Martial, J. A., Hollań, S., and Ova´di, J. (2001) Distinct behavior of mutant triosephosphate isomerase in hemolysate and in isolated form: molecular basis of enzyme deficiency. Blood 98, 3106 – 3112. 31. Ola´h, J., Orosz, F., Keseru¨, G. M., Kova´ri, Z., Kovaćs, J., Hollań, S., and Ova´di, J. (2002) Triosephosphate isomerase deficiency: a neurodegenerative misfolding disease. Biochem. Soc. Trans. 30, 30 – 38. 32. Ova´di, J., and Orosz, F. (1996) New concept for control of glycolysis. In Channelling in Intermediary Metabolism (Agius, L. and Sherratt, H. S. A. eds). pp. 237 – 268, Portland Press: London. 33. Ova´di, J., Orosz, F., and Hollań, S. (2004) Functional aspects of cellular microcompartmentation in the development of neurodegeneration: mutation induced aberrant protein-protein associations. Mol. Cell. Biochem. 256 – 257, 83 – 93. 34. Knull, H. R., and Walsh, J. L. (1992) Association of glycolytic enzymes with the cytoskeleton. Curr. Top. Cell. Regul. 33, 15 – 30. 35. Lo, T. W., Westwood, M., McLellan, A., Selwood, T., and Thornalley, P. (1994) Binding and modification of proteins by methylglyoxal under physiological conditions. A kinetic and mechanistic study with N alphaacetylarginine, N alpha-acetylcysteine, and N alpha-acetyllysine, and bovine serum albumin. J. Biol. Chem. 269, 32299 – 32305. 36. Kikuchi, S., Shinpo, K., Takeuchi, M., Yamagishi, S., Makita, Z., Sasaki, N., and Tashiro, K. (2003) Glycation – a sweet tempter for neuronal death. Brain Res. Brain Res. Rev. 41, 306 – 323. 37. Ahmed, N., Battah, S., Karachalias, N., Babaei-Jadidi, R., Horańyi, M., Baro´ti, K., Hollań, S., and Thornalley, P. J. (2003) Increased formation of methylglyoxal and protein glycation, oxidation and nitrosation in triosephosphate isomerase deficiency. Biochim. Biophys. Acta 1639, 121 – 132. 38. Hollań, S., Magoćsi, M., Fodor, E., Horańyi, M., Harsańyi, V., and Farkas, T. (1997) Search for the pathogenesis of the differing phenotype in two compound heterozygote Hungarian brothers with the same genotypic triosephosphate isomerase deficiency. Proc. Natl. Acad. Sci. USA 94, 10362 – 10366. 39. Hollań, S., Layton, M., Horańyi, M., Arya, R. and Farkas, T. (1998) Decrease in plasmalogen content in triosephosphate isomerase (TPI) deficiency. Blood 92, 6b. 40. Hollań, S., Dey, I., Szolla´r, L., Horańyi, M., Magoćsi, M., Harsańyi, V., and Farkas, T. (1995) Erythrocyte lipids in triose-phosphate isomerase deficiency. Proc. Natl. Acad. Sci. USA 92, 268 – 271. 41. Shi, Y., Vaden, D. L., Ju, S., Ding, D., Geiger, J. H., and Greenberg, M. L. (2005) Genetic perturbation of glycolysis results in inhibition of de novo inositol biosynthesis. J. Biol. Chem. 280, 41805 – 41810. 42. Fisher, S. K., Novak, J. E., and Agranoff, B. W. (2002) Inositol and higher inositol phosphates in neural tissues: homeostasis, metabolism and functional significance. J. Neurochem. 82, 736 – 754. 43. Celotto, A., Frank, A., Seigle, J., and Palladino, M. (2006) Drosophila model of human inherited TPI deficiency glycolytic enzymopathy. Genetics 174, 1237 – 1246. 44. Gnerer, J., Kreber, R., and Ganetzky, B. (2006) Wasted away, a Drosophila mutation in triosephosphate isomerase, causes paralysis, neurodegeneration, and early death. Proc. Natl. Acad. Sci. USA 103, 14987 – 14993. 45. Schuster, R., and Holzhu¨tter, H.-G. (1995) Use of mathematical models for predicting the metabolic effect of large-scale enzyme activity alterations. Application to enzyme deficiencies of red blood cells. Eur. J. Biochem. 229, 403 – 418.

715

46. Mulquiney, P. J., and Kuchel, P. W. (1999) Model of 2,3-bisphosphoglycerate metabolism in the human erythrocyte based on detailed enzyme kinetic equations: equations and parameter refinement. Biochem. J. 342, 581 – 596. 47. Martinov, M. V., Plotnikov, A. G., Vitvitsky, V. M., and Ataullakhanov, F. I. (2000) Deficiencies of glycolytic enzymes as a possible cause of hemolytic anemia. Biochim. Biophys. Acta 1474, 75 – 87. 48. Orosz, F., Wa´gner, G., Liliom, K., Kovaćs, J., Baro´ti, K., Horańyi, M., Farkas, T., Hollań, S., and Ova´di, J. (2000) Enhanced association of mutant triosephosphate isomerase to red cell membranes and to brain microtubules. Proc. Natl. Acad. Sci. USA 97, 1026 – 1031. 49. Zanella, A., Mariani, M., Colombo, M. B., Borgna-Pignatti, C., de Stefano, P., Morgese, G., and Sirchia, G. (1985) Triosephosphate isomerase deficiency: 2 new cases. Scand. J. Haematol. 34, 417 – 424. 50. Schneider, A. S., Valentine, W. N., Baughan, M. A., Paglia, D. E., Shore, N. A., and Heins, H. L. (1968) Triosephosphate isomerase deficiency. A multi-system inherited enzyme disorder: clinical and genetic aspects. In Hereditary Disorders of Erythrocyte Metabolism (Beutler, E. ed.). pp. 265 – 272, Grune and Stratton, New York. 51. Hollań, S., Fujii, H., Hirono, A., Hirono, K., Karro, H., Miwa, S., Harsańyi, V., Gyo´di, E., and Inselt-Kovaćs, M. (1993) Hereditary triosephosphate isomerase (TPI) deficiency: two severely affected brothers one with and one without neurological symptoms. Hum. Genet. 92, 486 – 490. 52. Ola´h, J., Orosz, F., Puska´s, L. G., Hackler, L. Jr., Horańyi, M., Polga´r, L., Hollań, S., and Ova´di, J. (2005) Triosephosphate isomerase deficiency: consequences of an inherited mutation at mRNA, protein and metabolic levels. Biochem. J. 392, 675 – 683. 53. Minton, A. P., and Wilf, J. (1981) Effect of macromolecular crowding upon the structure and function of an enzyme: glyceraldehyde-3phosphate dehydrogenase. Biochemistry 20, 4821 – 4826. 54. Gibson, A. M., Edwardson, J. A., and McDermott, J. R. (1991) Post mortem levels of some brain peptidases in Alzheimer’s disease: reduction in proline endopeptidase activity in cerebral cortex. Neurosci. Res. Commun. 9, 73 – 81. 55. Mantle, D., Falkous, G., Ishiura, S., Blanchard, P. J., and Perry, E. K. (1996) Comparison of proline endopeptidase activity in brain tissue from normal cases and cases with Alzheimer’s disease, Lewy body dementia, Parkinson’s disease and Huntington’s disease. Clin. Chim. Acta 249, 129 – 139. 56. Clay, S. A., Shore, N. A., and Landing, B. H. (1982) Triosephosphate isomerase deficiency. A case report with neuropathological findings. Am. J. Dis. Child. 136, 800 – 802. 57. Wilmshurst, J. M., Wise, G. A., Pollard, J. M., and Ouvrier, R. A. (2004) Chronic axonal neuropathy with triosephosphate isomerase deficiency. Ped. Neurol. 30, 146 – 148. 58. Hollań, S., Fodor, E., Horańyi, M., Inselt-Kovaćs, M., and Farkas, T. (1997) Decrease in non-bilayer forming lipid composition of lympocytes in triosephosphate isomerase (TPI) deficiency. Blood 90, 58b. 59. Wu, K., Aoki, C., Elste, A., Rogalski-Wilk, A., and Siekevitz, P. (1997) The synthesis of ATP by glycolytic enzymes in the postsynaptic density and the effect of endogenously generated nitric oxide. Proc. Natl. Acad. Sci. USA 94, 13273 – 13278. 60. Castegna, A., Thongboonkerd, V., Klein, J. B., Lynn, B., Markesbery, W. R., and Butterfield, D. A. (2003) Proteomic identification of nitrated proteins in Alzheimer’s disease brain. J. Neurochem. 85, 1394 – 1401. 61. Hoyer, S. (2004) Causes and consequences of disturbances of cerebral glucose metabolism in sporadic Alzheimer disease: therapeutic implications. Adv. Exp. Med. Biol. 541, 135 – 152. 62. Neubauer, B. A., Pekrun, A., Eber, S. W., Lakomek, M., and Schro¨ter, W. (1992) Relation between genetic defect, altered protein structure, and enzyme function in triose-phosphate isomerase (TPI) deficiency. Eur. J. Pediatr. 151, 232.