引物

on ith Mitochondrion 6 * Corresponding author. Tel.:C98 21 44580390; fax: C98 21 44580399. E-mail address: massoudh@nrcgeb.ac.ir (M. Houshmand). Massoud Houshmand a,*, Mehdi Shafa Shariat Panahi a, Shahriar Nafisi b, Akbar Soltanzadeh b, Fawziah M. Alkandari c a Department of Medical Genetics, National Institute for Genetic Engineering and Biotechnology, P.O. Box 14155-6343, Pajoohesh Blvd., Tehran-Karaj Highway, 17th km, Tehran, Iran b Neurology Department, Shariati Hospital, Tehran, Iran c Kuwait Medical Genetic Centre, Maternity Hospital, Kuwait Received 29 August 2005; received in revised form 18 December 2005; accepted 31 January 2006 Available online 3 April 2006 Abstract Friedreich’s Ataxia (FA) is the commonest genetic cause of ataxia and is associated with the expansion of a GAA repeat in intron 1 of the frataxin gene. Iron accumulation in the mitochondria of patients with FA would result in hypersensitivity to oxidative stress. Mitochondrial DNA (mtDNA) could be considered a candidate modifier factor for FA disease, since mitochondrial oxidative stress is thought to be involved in the pathogenesis of this disease. We studied 25 Iranian patients (16 females and 9 males) from 12 unrelated families. DNA from each patient was extracted and frequency and length of (GAA)n repeat was analyzed using a long-range polymerase chain reaction (PCR) test. Also we investigated impact of GAA size on neurological findings, age of onset and disease development. In order to identify polymorphic sites and genetic background, the sequence of two hypervariable regions (HVR-I and HVR-II) of mtDNA was obtained from FA patients harbouring GAA trinucletide expansions. Alignment was made with the revised cambridge reference sequence (rCRS) and any differences recorded as single base substitution (SBS), insertions and deletions. Homozygous GAA expansion was found in 21 (84%) of all cases. In four cases (16%), no expansion was observed, ruling out the diagnosis of Friedreich’s ataxia. In cases with GAA expansions, ataxia, scoliosis and pes cavus, cardiac abnormalities and some neurological findings occurred more frequently than in our patients without GAA expansion. Molecular analysis was imperative for diagnosis of Friedreich’s ataxia, not only for typical cases, but also for atypical ones. Diagnosis bases only on clinical findings is limited, however, it aids in better screening for suspected cases that should be tested. Our results showed that the rate of D-loop variations was higher in FA patients than control (P!0.05). mtDNA deletions were present in 76% of our patients representing mtDNA damage, which may be due to iron accumulation in mitochondria. q 2006 Elsevier B.V. and Mitochondria Research Society. All rights reserved. Keywords: Friedreich’s ataxia; mtDNA; Haplogroup; D-loop 1. Introduction Friedreich’s Ataxia (FA) is an autosomal recessive neurodegenerative disorder and is the most frequent form of hereditary ataxia (Harding, 1981), with an estimated preva- lence of one in 30,000 (Cossi et al., 1997). Clinical diagnostic autosomal recessive inheritance (c) progressive gait ataxia without remission (d) absence of lower limb tendon reflexes. Campuzano et al. (1996) detected in about 96% of FA patients an expanded GAA trinucleotide repeat in intron 1 of the gene X25 that encodes a 210 amino acid protein, Frataxin, which is a nuclear-encoded protein located within mitochondrial inner Identification and sizing of GAA investigation for D-loop variati in Iranian patients w trinucleotide repeat expansion, s and mitochondrial deletions Friedreich’s ataxia (2006) 87–93 www.elsevier.com/locate/mito mechanism (Rotig et al., 1997). Finding a patient manifesting an inherited recessive ataxia with only one GAA expanded mutant allele raises the question of diagnosis. It is likely that FA is the correct diagnosis, especially if the clinical picture is 1567-7249/$ - see front matter q 2006 Elsevier B.V. and Mitochondria Research Society. All rights reserved. doi:10.1016/j.mito.2006.01.005 criteria for typical cases basically include (a) early age of onset !20 (Geffroy et al., 1976) or 25 years (Harding, 1981) (b) membrane and crests (Babcock et al., 1997). A defect on mitochondrial iron metabolism has postulated as a pathogenic itoc the classic one. This discovery allows for further delineation of the disease, especially in areas of controversy atypical forms such as late onset FA (LOFA) (Klockgether et al., 1993), Friedreich’s ataxia with retained reflexses (FARR) (Panolfo and Montermmi, 1998), unusual presentation such as cardiac symptoms only (Geffroy et al., 1976) or pseudodominant inheritance (Harding and Zilkha, 1981). An inverse relation- ship has been demonstrated between GAA repeat size in smaller allele and the age of onset and severity (Filla et al., 1996). Mitochondrial genome is exclusively maternally inherited, does exhibit high mutation and mutation fixation rates (Anderson et al., 1981). Modifier genes are defined on the basis of their ability to modulate the clinical phenotype of individuals with monogenic and multigenic disease. Mito- chondrial DNA (mtDNA) could be considered a candidate modifier factor for neurodegenerative disorders, since mito- chondrial oxidative stress is thought to be involved in the pathogenesis of degenerative disorders (Orth and Schapira, 2001). The mutation rate for mtDNA isw10 times higher than that of nuclear genomic DNA. The displacment loop (D-loop), which is 1124 bp in size (positions 16,024–576), is a non- coding region, and acts as a promotor for both heavy and light strands of mtDNA, and contains essential transcription and replication elements. D-loop region is a hot spot for mtDNA alterations, and it contains two hypervariable regions (HVR-I at positions 16,024–16,383 and HVR-II at positions 57–372) (Anderson et al., 1981). Despite its functional importance, this region is believed to be the most rapidly evolving part of the molecule (Upholt and David, 1977). Nucleotide substitutions accumulate in the mitochondrial genome with a considerably higher rate than for single-copy nuclear DNA (Brown et al., 1979). This is most probably due to the lower efficiency of DNA repair as well as to a higher frequency of DNA replication errors in mitochondrial DNA (Wilson et al., 1993). Consequently, mtDNA and in particular the non-coding region, is highly polymorphic. We, therefore, reviewed a group of 25 patients with the clinical diagnosis of FA, supplementing clinical and laboratory details with identification and sizing of X25 trinucleotide expansion. We also investigated mtDNA D-loop variations within Friedreich’s Ataxia patients with nucleotide sequencing of D-loop region. Taking the D-loop background into account may help understand unknown molecular mechanism of phenotypic modifiers for FA and also the difference of haplotypes between patients. The mitochondrial deletions were also examined in our FA patients. Considering mtDNA as a modifier factor in FA, probable correlation between frataxin small allele size with D-Loop variations were also evaluated. 2. Patients and methods 2.1. Patients We studied 25 Iranian patients (16 females and 9 males) F.M. Alkandari et al. / M88 from 12 unrelated families (because of the maternal inheritance of mtDNA) with diagnosis of FA regarding their clinical aspects. We basically adopted the clinical criteria of Harding (1981) and Geffroy et al. (1976). We also chose 73 healthy controls (48 females and 25 males) matched for age, sex and ethnicity. Control subjects had no significant signs of FA when enrolled in the study. All of the patients and control group were informed on the aims of the study and gave their informed consents to the genetic analysis. The age range of patients and controls were 10–22 and 12–25, respectively. Mean age of onset of FA in our patients was 14.2G3.7 (SD). All of patients had gait and limb ataxia, abnormalities in tendon reflexes, impairment of position and vibratory sense and dysarthria. Foot deformity was present in 82% of patients. Also 91% and 29% of cases had scoliosis and hearing loss, respectively. Patients were referred for assessment by consultant neurologists in Iran. 3. Methods 3.1. Identification and sizing of GAA repeats Peripheral blood samples were obtained and DNA was purified after lyses of white blood cells by use of DNA extraction kit. (Diatom DNA extraction Kit-Genfanavaran, Tehran, Iran) A long range PCR technique was used to amplify the region of the X25 gene reported to contain the intronic GAA triplet repeats expansion. The Roche Diagnostics Expand Long template PCR System kit was used to set up the reactions, and standard conditions were used as suggested by the manufacturer with primers Bam (5 0-GGA GGG ATC CGT CTG GGC AAA GG-3 0) and 2500 F (5 0-CAA TCC AGG ACA GTC AGG GCT TT-3 0). These primers generated a 1.5 kb normal fragment. Amplification were performed in a MWG-Biotech Primus PCR machine (Germany) and conducted with the long PCR protocol, in 20 cycles composed of the following steps: 94 8C for 20 s, 68 8C for 2.5 min, followed by 17 cycles in which the length of the 68 8C step was increased by15 s/cycle. Allele sizes were independently assessed by two of investigators blinded to the patients’ clinical details. Nonparametric statistical methods were used, with the spearman rank correlation coefficient to assess the relationship between allele size and age of onset of clinical symptoms. 3.2. MtDNA D-loop haplogroup analysis To investigate association of HVS-I and HVS-II substi- tutions with FA, the nucleotide sequence of these two regions was determined in 21 patients affected with FA and in control group. PCR amplification was carried out in a final volume of 50 ml. Primers were as follows: primer pair 1; ONPF 38 (1–20 nt) 5 0-GAT CAC AGG TCT ATC ACC CT-3 0, ONPR 79 (780–761 nt) 5 0-GAG CTG CAT TGC TGC GTG CT-3 0. Primers pair 2; ONPF206 (15,340–15,360 nt) 5 0-ATC CTT GCA CGA AAC GGG ATC -3 0, ONPR 77 (110–91 nt) 5 0-GCT CGG GCT CCA GCG CTC CG-3 0. These primers amplified a hondrion 6 (2006) 87–93 780 and 1366 bp, respectively, encompassing two HVRs in the D-Loop of the mtDNA to fetch the 359 bp sequence (16,024– itoch F.M. Alkandari et al. / M 16,383 nt) and 315 bp sequence (57–372 nt) for HVS I and HVS II. The nucleotide sequence of the amplicon was directly determined by automated sequencing 3700 ABI machine, using primer ONPF 38 and ONPR 77 (Macrogene Seoul, South Korea). The obtained mtDNA sequences were aligned with a multiple sequence alignment interface CLUSTAL_X with comparison to rCRS. (http:/www.gen.emory.edu/mitomap/ mitoseq.html) RFLP (Restriction Fragment Length Poly- morphism) method was applied to investigate homoplasmy of D-Loop variations. 3.3. MtDNA deletions analysis The PCR reactions were performed for 35 cycles composed of the following steps: 94 8C for 1 0 (min), 55 8C for 1 0 (min) and 72 8C for 35 00 s. Primers used for the analysis of mtDNA deletions and the investigated regions of mtDNA are illustrated in Fig. 1. The deletion-prone region between 5461 of light strand and 15,000 of heavy strand was investigated in all Fig. 1. A scheme illustrating the strategy to determine multiple mtDNA deletions usi of mtDNA deletions are presented. ondrion 6 (2006) 87–93 89 patients using the primers ONP 86, ONP 89, ONP 10, ONP 74, ONP 25 and ONP 99. The distances between the primers were long enough to allow amplification only if a part of the DNA between respective primers was deleted. Primer pair ONP 86, 89 was used to amplify a normal internal mtDNA fragment in a region, which is seldom afflicted by deletions, and served as a control of the preparation and PCR analysis. Deletion break points in some samples were analyzed by direct sequencing of DNA fragments amplified by the PCR reactions using ABI 3700 capillary sequencer. 3.4. Statistical analysis Fisher’s exact probability test was used to examine the association between two groups. Values of P!0.05 were regarded as statistically significant. Spearman rank correlation coefficient was also used to assess the relationship between small allele size and number of D-Loop mutations. ng multiplex PCR technique. Also oligonucleotide primers used for the analysis 4. Results We found homozygous GAA expansion repeats in intron 1 of the gene X25 in 21 out of 25 ataxia patients (84%)—all typical cases. Four patients whose clinical diagnosis had been FA did not present any GAA expansion (16%). Therefore, they did not have FA. This group was comprised of three atypical cases (two FARR cases and one with cardiac symptoms only) and one typical case. The (GAA)n repeats on FA patients were observed in both alleles, ranging from 265 to 947 GAA motifs. Thus, we have not observed only one expanded alleles on one chromosome with the other presenting a normal range of expansion repeats. The mean value of expanded alleles was 571 repeats for allele 1 (Smaller allele) and 725 for allele 2 (Larger allele). Patients with GAA expansion repeats, when compared with those without GAA expansion, more frequently showed abnormalities in: (a) deep tendon reflexes (b) postural and vibratory sense, (c) feet (Pes Cavus), and (d) ECG findings. None of our control group was homozygote for GAA expansion repeats. Only two cases were heterozygote for one normal allele and one GAA expansion. The nucleotide sequence of 780 bp for HVS-II and 1366 bp for HVS-I were determined in 21 FA patients. Alterations in the two hypervariable D-loop regions and the patients carrying variants are summarized in Table 1. Sequence comparison with the rCRS led to identification of 23 mtDNA types within D-loop with 12 polymorphic positions in HVS-I and 11 in HVS-II. Of these 23 mitotypes, 14 were observed only in one individual. All samples contained apparent mutations differing in the sequence shown in the above reference. Our results showed that D-loop mutation rate in FA samples was higher than normal controls. (P!0.05) Most of the mutations were single base substitutions and most of them were transitions (95.0%) rather than transversions (5.0%). Sequence comparison showed also one A–G transition at position np263 and one T–C at position np310 in all of patients (100%), one T–C transition at position np146 in (66.6%), two T–C transition at positions np16304 and np16319 in (57.1%) of patients, one A–G transition at position np73 in (47.6%), one T–C transition at position np152 in (33%), one G–A transition at np16274 in (28.5%) and one G–A transition at position np185 in (19%). One polymorphisms (C16176 A) was newly identified in this study population, not recorded in the human genome database. [Mitomap database] Our results showed that all of the D-Loop variations in FA patients were homoplasmic. mtDNA deletions were present in 16 patients out of 21 (76%). The sizes of deletion were 8.6 and 9.0 kb. We found also a 10 kb deletion in three patients with FA. Deletions mostly occurred Table 1 Variations of mitochondrial D-loop in 21 patients with FA and 73 healthy controls V p 7 1 1 1 1 1 2 2 2 3 4 1 1 1 1 1 1 1 1 1 1 1 1 F.M. Alkandari et al. / Mitochondrion 6 (2006) 87–9390 Map locus Shorthand Map position (np) Description MTHV2 HVS-II 57–372 Hypervariable sequence- 2 MTOHR OH 110–441 H-Strand origin MTCSB1 CSB1 213–235 Conserved sequence block 1 MTTFX TFX 233–260 mtTF1 binding site MTCSB2 CSB2 299–315 Conserved sequence block 2 MTTFL – 418–445 Mt TF1 binding site MT7sDNA 7sDNA 16,106–16,191 7s DNA MTTAS TAS 16,157–16,172 Termination associated sequence MTMT5 mt5 16,194–16,208 Control element MTMT3L mt3L 16,499–16,506 L-Strand control element a Significant, number in parenthesis shows number of patients harbouring mutati b Variations not found previously. ariation osition Variation Patients with variation Control with variation P-value 3 A–G 47.6% (10) 17.8% (13) 0.008a 46 T–C 66.6% (14) 26.0% (19) 0.001a 50 C–T 4.7% (1) 4.1% (3) 1.0 52 T–C 33% (7) 8.2% (6) 0.007a 85 G–A 19% (4) 8.2% (6) 0.222 95 T–C 4.7% (1) 5.4% (4) 1.0 22 C–T 4.7% (1) 2.7% (2) 0.536 28 G–A 4.7% (1) 4.1% (3) 1.0 63 A–G 100% (21) 49.3% (36) 0.0005a 10 T–C 100% (21) 58.9% (43) 0.0001a 97 C–T 4.7% (1) – 6,126 T–C 4.7% (1) 2.7% (2) 0.536 6,145 G–A 4.7% (1) – 6,167 C–T 4.7% (1) – 6,176 C–Ab 4.7% (1) – 6,183 A–C 4.7% (1) – 6,274 G–A 28.5% (6) – 6,304 T–C 57.1% (12) 12.3% (9) 0.0006a 6,311 T–C 4.7% (1) – 6,319 G–A 57.1% (12) 9.5% (7) 0.0001a 6,362 T–C 4.7% (1) – 6,390 G–A 4.7% (1) – 6,519 T–C 4.7% (1) – on. itoch in the region between np 5461 and np 15,000. (Primers ONP86 and ONP10) Healthy controls showed no deletions in their mtDNA. None of our patients had multiple deletions, so we could not correlate the number of mtDNA deletions with small allele size, but the number of D-Loop mutations in patients showed a significant direct correlation with small allele size. 5. Discussion The expansion of trinucleotide repeat sequences is the underlying cause of many neurodegenerative diseases, including myotonic dystrophy, fragile X syndrome, Hunting- ton disease, FA, and several spinocerebellar ataxias. FA has interesting points that should be emphasized once new boundaries were established: (a) a trinucleotide repeat disease with a recessive pattern of inheritance; (b) this is the first time that these dynamic expansions have been detected in a intronic region and the triplet involved is comprised of GAA repeats; and finally (c) the DNA molecule in FA assumes triplex structure (Klockgether et al., 1993). Our 21 patients that were homozygous for an expanded (GAA)n repeat in X25 gene had: (a) onset of disease before the age of 24 years, (b) progressive ataxia of gait and limbs, (c) lower limb areflexia and (d) abnormalities in postural and vibratory sense. The mean age at onset and becoming wheelchair-bound were 14.2G3.7 (SD) and 18G4.1 (SD), respectively. The size of GAA expansion observed in our patients, ranging from 265 to 947, reflects the instability of this expansion during transmission. We analyzed the size of alleles in relation to (a) age of onset; (b) the age at which patients became wheelchair-bound. A statistical significance could be detected only for the difference between the size of allele 1 and allele 2. By applying the clinical criteria described by Geffroy et al. (1976) and Harding (1981), We detected a homozygous GAA expansion in (21/25) of all patients who met the criteria. We ruled out FA in the three atypical and one typical patients without GAA expansion, so the possibility of vitamin E deficiency should be considered in these cases. Molecular analyses is essential for confirming the diagnosis of FA, not only in typical cases, but especially in atypical ones, contributing to adequate genetic counselling for the recessive and sporadic cerebellar ataxias. We demonstrate that mutation rate in D-loop of patient group was higher than control. (P!0.005) Our finding revealed that all of the D-Loop variations in FA patients were homoplasmic. There may be also other heteroplasmic mutations beside ones mentioned in Table 1. Heteroplasmic mutations in FA are acute mutation events. Applying only fluorescent sequencing for mutation detection can miss up to 30% of mutations, so the mutation rate in the D-Loop of our FA patients may be even higher than the present results. Polymorphisms at np73, 150, 152, 195 and 263 specify part of ethnic-specific haplotypes. (HV, D5, M, W, H, respect- ively) Patients with FA have a lower than normal rate of F.M. Alkandari et al. / M mitochondrial ATP production and decreased oxidation activity shows a strong negative correlation with t