Locus heterogeneity in Friedreich ataxia

2 downloads 0 Views 227KB Size Report
ABSTRACT. Cardiac abnormalities are considered an important diagnostic criterion for FRDA. Friedreich ataxia (FRDA) is the most common form. The disease ...
© 1997 Springer-Verlag

Neurogenetics, Vol. 1, No. 1

43–47

ORIGINAL ARTICLE Locus heterogeneity in Friedreich ataxia Markus Kostrzewa, Thomas Klockgether1, Maxwell S. Damian2 and Ulrich Mu¨ller* Institut fu¨r Humangenetik der Justus-Liebig-Universita¨t, Schlangenzahl 14, D35392 Gießen, Germany, 1Neurologische Klinik der Universitat, Hoppe-Seyler-Straße 3, D72076 Tubingen, Germany and 2Neurologische ¨ ¨ Klinik der Justus-Liebig-Universita¨t, Am Steg 22, 35385 Gießen, Germany Received December 19, 1996; Revised and Accepted January 22, 1997

ABSTRACT Friedreich ataxia (FRDA) is the most common form of autosomal recessive ataxia. The disease locus was assigned to chromosome 9 and the disease gene, STM7/X25, has been isolated. To date most data suggest locus homogeneity in FRDA. We now provide strong evidence of a second FRDA locus. Studying two siblings with FRDA from two families we did not detect a mutation in STM7/X25. Haplotype analysis of the STM7/X25 region of chromosome 9 demonstrated that the relevant portion of chromosome 9 differs in the patients. Although the patients studied had typical FRDA, one sibpair had the uncommon symptom of retained tendon reflexes. In order to investigate whether retained tendon reflexes are characteristic of FRDA caused by the second locus, FRDA2, we studied an unrelated FRDA patient with retained tendon reflexes. The observation of typical mutations in STM7/X25 (GAA expansions) in this patient demonstrates that the two genetically different forms of FRDA cannot be distinguished clinically. Keywords: FRDA, Friedreich ataxia, STM7, X25, locus heterogeneity

INTRODUCTION Friedreich ataxia (FRDA) is an autosomal recessive ataxia that occurs at a frequency of 1/50 000. Onset is usually during child- and early adulthood. Major neurological findings are ataxia, areflexia in the legs, pyramidal weakness, and impaired sense of vibration. The symptoms are progressive and result in wheelchair confinement. Less frequently, optic atrophy, nystagmus, and sensorineural deafness occur. Non-neurological findings include cardiomyopathy in most and diabetes in ~10% of the cases (1). Cardiac symptoms vary widely and range from subtle ECG changes to severe hypertrophic cardiomyopathy.

Cardiac abnormalities are considered an important diagnostic criterion for FRDA. The disease gene was assigned to the long arm of chromosome 9 (9q13-q21.1) and identified by positional cloning (2). It is referred to as X25 and is composed of at least seven exons (59–39: 1,2,3,4, 5a, 5b, 6) that can be alternatively spliced. Exon 6 is untranslated. The most common splicing variant appears to be composed of exons 1–5a, was reported to result in a 1.3 kb transcript that is translated into a polypeptide of 210 amino acids, termed frataxin. Corresponding to the non-neuronal tissues primarily affected in FRDA, the highest levels of the 1.3 kb transcript of X25 were described in the heart, intermediate levels in liver, skeletal muscle and pancreas and low amounts in other tissues including whole brain. Within the central nervous system (CNS) highest levels were reported in the spinal cord, less in the cerebellum and very little in the cerebral cortex. More recent data suggest that X25 is much larger and that it includes the proximally adjacent gene STM 7 which encodes the catalytic domain of a phosphatidylinositol-4-phosphate 5 [PtdIns(4)P-5] kinase (3). According to this finding, the FRDA gene is composed of at least 24 exons, the seven exons of X25 (now exons 18–24) plus 17 exons of STM 7. Expression of STM7/X2 5is complex. There are different splicing variants that may code for different isoforms of PIP kinases. In contrast to the report by Campuzano et al. (2), Carvajal et al. (3) detected no transcript of 1.3 kb that was composed of X25 exons only. More investigations are necessary to clarify these discrepant observations. A polymorphic tandem repeat of the trinucleotide GAA is located in intron 1 of X25 (intron 18 of the STM7/X2 5 gene). The physiological copy number of the repeat ranges from 7 to 22 trinucleotides (2). The repeat is greatly expanded in the majority of patients with copy numbers in the range of 200 to more than 1000 (2,4,5). While GAA expansions occur in both alleles of STM7/X2 5in the great majority of cases, compound heterozygotes have been described. In these cases, a point mutation was observed in one and a GAA expansion in the other allele. To date, no patients with point mutations in both alleles of X25 have been observed. GAA expansions and point

*To whom correspondence should be addressed. Tel: 149 641 9941600; Fax: 149 641 9941609; Email: [email protected]

44

Neurogenetics, 1997, Vol. 1, No. 1

Figure 2. Examination of GAA expansion in FRDA patients and controls. Primer pair GAAF and GAAR (2) were used. Identical results were obtained with primer pair Bam/2500F (not shown). Lanes 1,2 show affected persons of family 1, and lanes 3,4 affecteds of family 2. No expansion of the repeat was detected in patients of either family. 5 demonstrates the GAA repeat expansions of both alleles in the FRDA patient with retained tendon reflexes and 7 depicts a homozygous GAA repeat expansion in the FRDA patient with signs of cerebellar atrophy. In lanes 6 and 8 findings in unaffected controls are given.

FRDA (6,7). We have tested this hypothesis in two FRDA families with two affected members each. MATERIALS AND METHODS

Figure 1. Pedigrees of the two core families studied. Haplotype reconstructions are given for 9q markers. In family 1 (a) the affected individuals have inherited different maternal alleles. In family 2 (b) the haplotypes of both homologous chromosomes are different in the affected persons. The following allele sizes were determined: D9S1845, 1 5 273 bp, 2 5 275 bp, 3 5 277 bp, 4 5 281, 5 5 287; D9S1859, 1 5 101 bp, 2 5 103 bp, 3 5 105 bp, 4 5 107 bp, 5 5 119 bp; D9S1862, 1 5 183 bp, 2 5 185 bp, 3 5 187 bp, 4 5 189 bp, 5 5 191 bp; D9S889, 1 5 120 bp, 2 5 122 bp, 3 5 124 bp, 4 5 126 bp; D9S887, 1 5 158 bp, 2 5 160 bp, 3 5 162 bp, 4 5 166 bp; D9S888, 1 5 147 bp, 2 5 149 bp, 3 5 172 bp; ‘FR6’, 1 5 163 bp, 2 5 165 bp, 3 5 169 bp, 4 5 171 bp; D9S886, 1 5 162 bp, 2 5 164 bp, 3 5 170 bp, 4 5 171 bp, 5 5 172 bp; D9S202, 1 5 196 bp, 2 5 206 bp, 3 5 208 bp, 4 5 212 bp. ‘n.i.’ is ‘not informative’

mutations are thought to result in a loss of gene function. The trinucleotide repeat expansions may interfere with processing of the transcript and the point mutations may inactivate the gene by well established mechanisms such as the introduction of stop codons, amino acid changes, alteration of splice sites, etc. At least two of the STM7/X25 splicing variants terminate upstream of the GAA repeat (3). Others, however, include X25. Given that all point mutations detected in FRDA so far are located within the X25 portion of the gene (2,5), it is most likely that a kinase(s) is altered in FRDA that is encoded by a splicing variant of exons of both STM7 and X25. Linkage analyses have suggested locus homogeneity in

The GAA repeat was amplified using both primer pairs (GAAF1GAAR and Bam12500F) described by Campuzano et al. (2). In brief, PCR was performed in a total volume of 20 µl containing 200 ng of DNA, 15 pmol of each primer, Taq polymerase and Taq extender (Stratagene). For primer pair GAAF/GAAR, a prerun at 94°C was followed by 34 cycles at 94°C for 40 s, 68.5°C for 30 s, 72°C for 4 min, and a final extension at 72°C for 5 min. PCR conditions for primer pair Bam/2500F were: prerun at 94°C for 3 min, 16 cycles at 94°C for 30 s and at 68°C for 2 min 30 s, 17 cycles with a stepwise increase of annealing/elongation by 15 s during each cycle, final extension at 72°C for 10 min. For SSCP analysis, the six translated exons of X25 (exons 18–23 of STM7/X25) were amplified with published primers (2). The amplification products were separated on 0.53 MDE gels (AT Biochem) at 4°C/ 5 W and at room temperature at 5 W. Chromosome 9-specific STRPs surrounding STM7/X25 were D9S202 (FR1), D9S886 (FR2), D9S887 (FR7), D9S888 (FR8), D9S889 (FR5), D9S1845, D9S1859, D9S1862, and FR6 (8). A prerun of 94°C for 3 min was followed by 30 cycles at 94°C for 30 s, annealing for 1 min, 72°C for 30 s (15 s, D9S1859), and a final extension for 5 min at 72°C. Annealing temperatures were 53°C (D9S202), 55°C (D9S1859, D9S1862, D9S1845), 56°C (D9S886), 58°C (D9S887, FR6), 61°C (D9S888), and 63°C (D9S889). PCR was performed in the presence of [32P]dCTP, amplification products were separated on 6% polyacrylamide gels and autoradiographed. CLINICAL DESCRIPTION Two siblings, a boy and a girl, were affected in each family. They were the only affected individuals in their families. The

Neurogenetics, 1997, Vol. 1, No. 1

45

Table 1. Symptoms in FRDA patients Patient

FA1.1

FA1.2

FA2.1

FA2.2

FA3.1

FA8.1

Sex Age Age of onset Progressive ataxia Tendon reflexes Vibration and positional sense Dysarthria Nystagmus Optic atrophy Pes cavus Scoliosis

F 27 11 1 1 A

M 37 19 1 1 A

F 36 16 1 – A

M 33 20 1 – A

F 37 17 1 1 A

F 34 6 – – A

1 1 1 1 1

1 1 1 1 1 (mild)

1 1 – 1 –

1 1 – 1 –

1 1 1 1 1

1 1 – 1 1

1 – P

– – P

– – P

– – P

– – P

1 1 P

SEP: P AEP: N cerebellar atrophy

SEP: P AEP: P n.d.

SEP: P AEP: N n.d.

SEP: P AEP: N spinal atrophy

SEP: P AEP: P n.d.

spinal atrophy – N

spinal atrophy – N

n.d.

cerebellar atrophy 1 n.d.

Cardiomyopathy by a) Electrocardiography b) Echocardiography Electroneurogram (sensory conduction) Evoked potentials

MRI

n.d.

SEP: P AEP: N cerebellar atropy (mild) n.d.

Diabetes Vitamin E value

– N

– N

CT

– n.d.

N, normal; P, pathological; A, affected; n.d., not determined; 1, presence of symptom; –, absence of symptom; SEP, sensory evoked potentials; AEP, auditory evoked potentials.

pedigrees of the two core families studied are given in Figure 1. The parents of siblings of both families came from the same small villages and might thus have common ancestors. The findings in the families are consistent with autosomal recessive inheritance of the disorder. The patients had FRDA by clinical criteria. Symptoms observed are summarized in Table 1. ECG abnormalities suggest cardiac involvement in the male patient of family 1. Affected members of family 1 (FA1.1 and FA1.2) had retained lower tendon reflexes. In addition, MRI investigations provided evidence of cerebellar atrophy in family 1. Vitamin E levels were within the normal range in all patients. Two additional Friedreich cases were studied as controls. One (FA3.1) had retained tendon reflexes, the other (FA8.1) did not. MRI studies in the second patient suggested cerebellar atrophy. Clinical findings in these patients are also listed in Table 1. The clinical picture of the FRDA case with retained tendon reflexes has been described before (9).

We then tested, whether the FRDA gene was at all involved in the disease in these individuals, and analysed STRP loci surrounding STM7/X25. The two siblings carried different alleles at all informative loci. Alleles detected at loci D9S888, (FR8) and D9S886 (FR2) in the two families are given in Figure 3a, b. D9S888 lies within STM7 and D9S886 is less than 100 kb distal to X25. At both loci different alleles were detected in the affected individuals of both families. Different alleles were also found in one family at least at loci FR6 (9) and D9S202 (FR1) distal to STM7/X25, at the intragenic loci D9S887 (FR7) and D9S889 (FR5) and at the proximal loci D9S1862, D9S1859, and D9S1845. The haplotypes are given in Figure 1. The findings clearly demonstrate that different chromosomes 9 or portions thereof are present in each patient of the two families. DISCUSSION

RESULTS Analysis of the GAA repeat within STM7/X25 in the affected siblings of the two families did not reveal an expansion. The copy number was within the normal range (Fig. 2a). In the two controls, large expansions of the repeat were found on both chromosomes. The trinucleotide repeat copy numbers were 550 and 950 in the patient with retained tendon reflexes and 900 (homozygous) in the second patient who had signs of cerebellar atrophy on MRI (Fig. 2b). In order to determine whether the two sib pairs had mutations other than GAA expansions within X25, SSCP analyses of exons 1–5b of X25 were performed. No band shifts were detected (not shown).

FRDA is well defined by clinical criteria which include areflexia and onset before age 25. Molecular studies, however, have shown that the clinical spectrum of FRDA is broader than originally assumed (4,5,8,10,11). Thus linkage of anonymous DNA markers to the FRDA locus on chromosome 9 was found in some patients with atypical features such as retained tendon reflexes or late disease onset. In agreement with an earlier study (4) the present investigation demonstrates that mutations within FRDA, i.e. GAA expansions, do indeed occur in FRDA with retained tendon reflexes (Fig. 2a). The clinical findings in the two families studied were thus compatible with the clinical diagnosis of FRDA. In family 1, cerebellar atrophy was observed. Although this is not a common symptom in

46

Neurogenetics, 1997, Vol. 1, No. 1

Figure 4. Phosphatidylinositol-dependent function and metabolism. PtdIns(3)P 5 phosphatidylinositol-3-phosphate; PITP 5 phosphatidylinositol transfer protein; PtdIns(3,4)P2 5 phosphatidylinositol-3,4-bisphosphate; PtdIns(4)P 5 phosphatidylinositol-4-phosphate; Rho 5 small GTP-binding protein ρ; PtdIns(4,5)P2 5 phosphatidylinositol-4,5-bisphosphate; PtdIns(3,4,5)P3 5 phosphatidylinositol-3,4,5-trisphosphate; Ins(1,4,5)P3 5 inositol-1,4,5-trisphosphate; DAG 5 diacylglycerol; PKC 5 protein kinase C (modified from ref. 15).

Figure 3. Alleles at locus D9S888 (a) and D9886 (b) in family 1 (left) and in family 2 (right). Note different allele sizes in the affected siblings of both families.

FRDA, it was also found in one of the controls with the typical GAA repeat expansion (Fig. 2b). This demonstrates that cerebellar atrophy as well as retained tendon reflexes can be a symptom in FRDA. Thus the four familial cases studied did not have any symptom that suggests another clinical diagnosis than FRDA. The finding of different alleles at STRP loci surrounding STM7/X25 strongly suggests that the FRDA locus on chromosome 9 is not involved in disease in these patients. Given that the markers are closely spaced within and flanking STM7/X25, the findings cannot be explained by the occurrence of a double crossover which would have to have occurred twice independently in two cases. Previous linkage analyses in FRDA patients have provided evidence of a second FRDA locus in one family only (12). However, in this study no markers proximal to STM7 were analysed. The present study now demonstrates beyond doubt the existence of a second locus in FRDA. We therefore propose to refer to the locus on chromosome 9 as FRDA1 and to the second locus as FRDA2. It is apparent that mutations at FRDA2 occur much less frequently in FRDA than mutations at FRDA1. Presently, the location of FRDA2 is unknown. Given that the vitamin E levels were within the normal range in the two sibpairs, the findings cannot be explained by mutations in the α-tocopherol transfer protein gene on chromosome 8 (8q13) that cause ataxia with isolated vitamin E deficiency (13).

The nature of FRDA2 may be deduced from the function of the STM7/X25 gene product, a PtdIns(4)P-5-kinase (Fig. 4). This enzyme phosphorylates PtdIns(4)P. The resulting PtdIns(4,5)bisphosphate plays a central role in phosphatidylinositid function (reviewed in refs 14,15). It is required for intracellular vesicular traffic, secretion and function of the cytoskeleton. In addition, it activates phospholipase D and its metabolite, PtdIns(3,4,5)trisphosphate, regulates endocytosis, lysosomal traffic and is involved in the normal function of the cytoskeleton. PtdIns(4,5)bisphosphate is also a substrate for phospholipase C which generates the second messengers inositol(1,4,5)trisphosphate and diacylglycerol. These in turn mobilize Ca11, and activate protein kinase C. Candidates of FRDA2 are thus all those genes that are involved in these pathways. It might be significant that a PI-3 kinase encoded by chromosome 11 (11q22–q23) is mutated in another ataxia, ataxia telangiectasia (16). To date, the role of the phosphoinositid pathway in the function of the nervous system is mostly unknown. There is one interesting finding, however, demonstrating that absence of the inositol 1,4,5-trisphosphate receptor, type 1 results in the development of severe ataxia and epileptic seizures in knock-out mice (17).

REFERENCES 1. Bundey, S. (1992) Genetics and Neurology. Churchill Livingstone, pp. 325–326. 2. Campuzano, V. et al. (1996) Friedreich’s ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion. Science, 271, 1423–1427. 3. Carvajal, J.J., Pook, M.A., dos Santos, M., Doudney, K., Hillermann, R., Minogue, S., Williamson, R., Hsuan, J.J. and Chamberlain, S. (1996) The

Neurogenetics, 1997, Vol. 1, No. 1

4.

5.

6.

7.

8. 9.

10.

11.

12. 13.

14. 15. 16. 17.

Friedreich’s ataxia gene encodes a novel phosphatidylinositol-4-phosphate 5-kinase. Nature Genet., 14, 157–162. Du¨rr, A., Cossee, M., Agid, Y., Campuzano, V., Mignard, C., Penet, C., Mandel, J.-L., Brice, A. and Koenig, M. (1996) Clinical and genetic abnormalities in patients with Friedreich’s ataxia. N. Engl. J. Med., 335, 1169–1175. Filla, A., De Michele, G., Cavalcanti, F., Pianese, L., Monticelli, A., Campanella, G. and Cocozza, S. (1996) The relationship between trinucleotide (GAA) repeat length and clinical features in Friedreich ataxia. Am. J. Hum. Genet., 59, 554–560. Chamberlain, S., Shaw, J., Wallis, J., Rowland, A., Chow, L., Farrall, M., Keats, B., Richter, A., Roy, M., Melancon, S., Deufel, T., Berciano, J. and Williamson, R. (1989) Genetic homogeneity in Friedreich’s ataxia locus on chromosome 9. Am. J. Hum. Genet., 52, 99–109. Fujita, R., Agid, Y., Trouillas, P., Seck, A., Tommasi-Davenas, C., Driesel, A.J., Olek, K., Grzeschik, K.H., Nakamura, Y. and Mandel, J.-L. (1989) Confirmation of linkage of Friedreich’s ataxia to chromosome 9 and identification of a new closely linked marker. Genomics, 4, 909–914. Montermini, L. et al. (1995) The Friedreich ataxia critical region spans a 150 kb interval on chromosome 9q13. Am. J. Hum. Genet., 57, 1061–1067. Klockgether, T., Zu¨hlke, C., Schulz, J.B., Bu¨rk, K., Fetter, M., Dittmann, H., Skalej, M. and Dichgans, J. (1996) Friedreich’s ataxia with retained tendon reflexes: Molecular genetics, clinical neurophysiology, and magnetic resonance imaging. Neurology, 46, 118–121. Richter, A., Poirier, J., Mercier, J., Julien, D., Morgan, K., Roy, M., Gosselin, F., Bouchard, J.-P. and Melancon, S.B. (1996) Friedreich ataxia in Acadian families from Eastern Canada: Clinical diversity with conserved haplotypes. Am. J. Med. Genet., 64, 594–601. Klockgether, T., Chamberlain, S., Wu¨llner, U., Fetter, M., Dittmann, H., Petersen, D. and Dichgans, J. (1993) Late-onset Friedreich’s ataxia. Molecular genetics, clinical neurophysiology and magnetic resonance imaging. Arch. Neurol., 50, 803–806. Smeyers, P., Monro´s, E., Vı´lchez, J., Lopez-Arlandis, J., Prieto, F. and Palau, F. (1996) A family segregating a Friedreich ataxia phenotype that is not linked to the FRDA locus. Hum. Genet., 97, 824–828. Ouahchi, K., Arita, M., Kayden, H., Hentati, F., Hamida, M.B., Sokol, R., Arai, H., Inoue, K., Mandel, J.-L. and Koenig, M. (1995) Ataxia with isolated vitamin E deficiency is caused by mutations in the α-tocopherol transfer protein. Nature Genet., 9, 141–145. Liscovitch, M. and Cantley, L.C. (1995) Signal transduction and membrane traffic: The PITP/Phosphoinositide connection. Cell, 81, 659–662. De Camilli, P., Emr, S.D., McPherson, P.S. and Novick, P. (1996) Phosphoinositides as regulators in membrane traffic. Science, 271, 1533–1539. Savitsky, K. et al. (1995) A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science, 268, 1749–1753. Matsumoto, M. et al. (1996) Ataxia and epileptic seizures in mice lacking type 1 inositol 1,4,5-trisphosphate receptor. Nature, 379, 168–171.

47