13010Med Genet 1997;34:130-132

Exclusion of CAG repeat expansion as the cause of disease in autosomal dominant retinitis J Med Genet: first published as 10.1136/jmg.34.2.130 on 1 February 1997. Downloaded from pigmentosa families

T Jeffrey Keen, Alex G Morris, Chris F Inglehearn

Abstract identified. Of these, six are associated with The involvement of with expanded mutations in known photoreceptor specific tracts of (CAG). in some neurodegenera- genes,9 while the remainder are defined only tive diseases is weli established. Whether by linkage to anonymous loci. In about one genes containing these motifs could also third of autosomal dominant cases, mutations have a role in degenerative diseases affect- in the genes coding for rhodopsin'° or ing the retina, which is also neural in ori- RDS-peripherin" 12 are associated with the gin, is unknown. We investigated (CAG). disease. This leaves the underlying cause in expansions as a cause ofdisease in a panel most cases of dominant RP unaccounted for. of eight autosomal dominant retinitis pig- This is therefore an extremely heterogeneous mentosa (ADRP) pedigrees, including disease with a complex pathology. Since the families known to map to the RP9, RP11, retina is essentially a neural tissue, it would be and RP13 loci, using the technique known reasonable to speculate that some forms of RP as "repeat expansion detection" (RED). could be caused by triplet repeat expansions. An expansion was detected in one of the We therefore investigated the possible in- unlinked families, but it did not segregate volvement of expanded CAG/CTG repeats in a with the disease and was thus non- panel of autosomal dominant retinitis pigmen- pathogenic. Expansions were not detected tosa families using the technique of "repeat in any other families. In conclusion, expansion detection"."3 With this technique the expanded (CAG)n repeats are not the presence of triplet repeat expansions anywhere cause of disease in the families we have in a person's genome is shown as a series of studied, but given the high level of hetero- ligated (CTG),7 oligomers. These are gener- geneity in RP and in retinal degenerations ated in a thermal cycling reaction containing in general they remain strong candidates genomic DNA, the (CTG),, oligomer, and a for involvement in other forms of retinal thermally stable DNA ligase. Extended tracts http://jmg.bmj.com/ dystrophy. of CAG in the genomic DNA allow (CTG),7 (_7Med Genet 1997;34:130-132) oligomers to align themselves adjacent to each other, making them a substrate for the ligase. A Keywords: ; CAG; repeat expansion single ligation product generates a (CTG),4 detection. band, this being the upper limit of non- pathogenic expansions at triplet expansion loci

Expansion of (CAG)n triplet repeats within the associated with dominant disease phenotypes. on September 28, 2021 by guest. Protected copyright. coding sequence of genes cause at least five However, shorter expansions have been shown dominant neurodegenerative diseases, each of to cause X linked spinal and bulbar muscular which is characterised by the loss of specific atrophy.' The (CTG),4 ligation product also subpopulations of neurones.' A common fea- serves as a positive control for the reaction, ture of some neurodegenerative and psychiatric though it is not known if all expanded CAG disorders is a progressive increase in the sever- loci are equally detectable. Larger expansions ity of the condition through successive genera- (to 51, 68, 85, and more CAGs) have been seen tions. This phenomenon, known as anticipa- in a proportion of normal controls in several tion, has been shown to be indicative of the studies. These presumably derive from one or Department of expansion of such repetitive elements, underly- several other (CAG)n tracts in the genome, at Molecular Genetics, ing the pathological process.2' In some triplet which expansions are not in themselves patho- Institute of repeat expansion (TRE) diseases anticipation genic. Ophthalmology, is not observed, but even in these the size of an Eight ADRP families were studied, all of University College which had been screened for rhodopsin or London, Bath Street, expansion is known to correlate well with the London EC1V 9EL, UK clinical severity.4 5 RDS mutations and found negative. In three of T J Keen Retinitis pigmentosa is a progressive, inher- the families studied, the disease mapped to A G Morris ited retinal degeneration characterised by known ADRP loci on 7p C F Inglehearn restriction of the visual field (tunnel vision), (RP9),'4 17p (RP13),"5 and 19q (RP1 1),16 for blindness, and abnormal pigmentation in which the genes involved have not yet been Correspondence to: night Dr Inglehearn. the peripheral retina.6 A range of modes of identified. For the remaining five families, the inheritance have been observed, including chromosomal location of the disease loci was Received 24 June 1996 autosomal dominant, autosomal recessive, and unknown. Among the group of unlinked fami- Revised version accepted for was one in which there was an publication X linked forms,7 as well as more complex lies there appar- 4 September 1996 digenic Rp.8 At least 17 different loci have been ent increase in severity of the disease in the Exclusion of CAG repeat expansion in autosomal dominant retinitis pigmentosa 131

im:, J Med Genet: first published as 10.1136/jmg.34.2.130 on 1 February 1997. Downloaded from

11 -;-M4

III 1 1

IV 12 12 12 11 Figure 1 The pedigree of the ADRPfamily potentially displaying anticipation. Members of thefamily who carry the expanded (CAG), allele are indicated by an asterisk. Numbers below the symbols denote allelesfor D18S499. lower generations (fig 1). Also, with both RP9 and RP1 1 there exists considerable variability 1 2 3 4 5 6 7 8 9 10 11 12 1314 15 16 in the disease phenotype, both in its severity and the age of onset.'7 18 Such variability could Figure 2 Autoradiograph of the ligation products generated in a "repeat expansion detection "experiment. be explained by dynamic variation of an The samples are arranged in pairs of affected subjectsfrom expanding and contracting CAG repeat. ADRPfamilies. Lanes 9 and 10 contain products generated from the two severely affected sibs (IV.2 and IV.3) from the pedigree in fig 1. Methods Affected and normal members of the ADRP families used in this study were assessed as fol- acrylamide gel, then electroblotted onto lows. First an ophthalmic history was taken to Hybond-N+ positively charged nylon mem- include details of visual acuity loss and age of brane. To visualise the RED products, the onset of night blindness, then visual fields were membrane was probed using a (CAG)IO oligo- determined. In addition, a brief general medi- nucleotide end labelled with Ca-32P dATP using cal history and physical examination were terminal deoxynucleotidyl transferase. The undertaken to exclude the possibility of other hybridisation was carried out at 58°C in a defects cosegregating with RP in these families. standard phosphate buffer for 16 hours, Specifically, no members of these families gave followed by two stringency washes in a buffer a history indicating any neurological disease. containing 0.5% SSC and 0.5% SDS at 58°C. http://jmg.bmj.com/ Pupils were then dilated to allow fundus The autoradiograph was exposed overnight at examination. Some patients also underwent -80°C using Kodak X-Omat film and two more detailed psychophysical and electrodiag- intensifying screens. nostic tests. 17 18 In the light of the apparent par- tial penetrance of the disease in the family shown in fig 1, key subjects were re-examined

fundoscopically three years after their initial Results on September 28, 2021 by guest. Protected copyright. assessment to look for signs of disease progres- The repeat expansion detection method sion, but they remained normal on re- (RED) was carried out as previously examination. described'3 "9 (fig 2). Two affected subjects Genomic DNA was prepared from periph- were studied from each of the eight families. In eral blood lymphocytes of RP patients and the families where the disease penetrance was their relatives. The RED technique was then variable (RP9 and RP1 1), severely affected performed essentially as it was originally subjects were studied to maximise the chance described'3 incorporating some slight of observing an expansion if one was present. modifications."' The ligation mixture consisted Ligation products of 255 bp (five oligos) were of 50 ng of 5'-phosphorylated (CTG)17,4 U PYu seen in two sibs from the family that appeared ligase (Stratagene), and 5 jg of genomic DNA to display anticipation. On the strength of this in 20 mmol/l Tris-HCl (pH 7.5), 20 mmol/l result the rest of the available members of the KC1, 10 mmol/l MgCl,, 0.1 mmol/l ATP, 1 pedigree were tested. However, subsequent mmol/l DTT, and 0.1% NP40 made up to a analysis showed that the source of the ex- final volume of 20 gl. Thermal cycling of the panded allele in the family was a normal person mixture was carried out in a Perkin Elmer who had married into it (fig 1). The triplet GeneAmp PCR System 9600 at an annealing expansion had been inherited by both normal ligation temperature of 77°C for 45 seconds and affected descendants. Ligation products and a denaturing temperature of 95°C for 10 that would be indicative of a CAG expansion seconds over 198 cycles. At that point each were not seen in any of the other families, reaction was replenished with 4 U Pfu ligase, 5 though in all cases the single ligation product jg genomic DNA, and subjected to a further was present indicating that the oligomer 198 thermal cycles. The reaction products ligation reaction had occurred with that DNA were fractionated on a denaturing, 6% poly- sample. 132 Keen, Morris, Inglehearn

Discussion bipolar affective disorder.22 It might thus also If triplet repeat expansion was the underlying be appropriate to study the relative occurrence

cause of RP at the RP9, 11, or 13 loci or in a of TREs in a multifactorial retinal disease such J Med Genet: first published as 10.1136/jmg.34.2.130 on 1 February 1997. Downloaded from substantial proportion of the unmappedADRP as age related macular dystrophy. families tested, we would have expected to find that all affected members of such a family We thank the Wellcome Trust (grant numbers 035535/Z/92 and would have a repeat expansion above the 037331/Z/93) for funding our work and two anonymous reviewers for their comments. normal range, consistent with a dominant helpful mode of inheritance. The negative results from this study therefore suggest that expanded 1 Ross CA. When more is less: pathogenesis of glutamine repeat neurodegenerative diseases. Neuron 1995;15:493-6. (CAG)n repeats are not the cause of disease in 2 Huntington's Disease Collaborative Research Group. A this group of ADRP families. The detection of novel containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromo- a non-pathogenic expansion makes us confi- somes. Cell 1993;72:971-83. dent that we would have detected expansions 3 Koide R, Ikeuchi T, Onodera 0, et al. Unstable expansion of CAG repeat in hereditary dentatorubral-pallidoluysian had they been present. However, while (CAG)n atrophy (DRPLA). Nat Genet 1994;6:9-13. expansions do not cause RP in the families we 4 Orr HT, Chung MY, Banfi S, et al. Expansion of an unstable trinucleotide CAG repeat in spinocerebellar ataxia type 1. have studied, they remain strong candidates to Nat Genet 1993;4:221-6. underlie some forms of inherited retinal 5 La-Spada AR, Wilson EM, Lubahn DB, Harding AE, Fischbeck KH. Androgen receptor gene mutations in degeneration. X-linked spinal and bulbar muscular atrophy. Nature 1991; Expansions beyond the level occur 352:77-9. (CAG)34 6 Bird AC. Retinal photoreceptor dystrophies. Edward in the normal population in about 25% of Jackson Memorial Lecture. Am J Ophthalmol 1995;119: people.'9 In the original report describing the 543-62. 7 Rosenfeld PJ, McKusick VA, Amberger JS, Dryja TP. RED technique, an anonymous (CAG)n expan- Recent advances in the gene map of inherited eye sion was linked in a disorders: primary hereditary diseases of the retina, (RED-1) genetically choroid, and vitreous. J Med Genet 1994;31:903-15. CEPH family to a microsatellite marker on 8 Kajiwara K, Berson EL, Dryja TP. Digenic retinitis pigmen- 18. of this marker tosa due to mutations at the unlinked peripherin/RDS and Analysis ROMI loci. Science 1994;264:1604-8. (DI 8S499) in the family described here shows 9 Dryja TP, Li T. Molecular genetics of retinitis pigmentosa. at least one crossover between the marker and Hum Mol Genet 1995;4:1739-43. 10 Dryja TP, McGee TL, Reichel E, et al. A point mutation of the expansion. Subjects III.3 and III.4 inher- the rhodopsin gene in one form of retinitis pigmentosa. ited different alleles from the father carrying Nature 1990;343:364-6. 11 Kajiwara K, Hahn LB, Mukai S, Travis GH, Berson EL, the expansion, yet neither inherited the expan- Dryja TP. Mutations in the human retinal degeneration sion. While this result argues against the slow gene in autosomal dominant retinitis pigmentosa. Nature 1991 ;354:480-3. expansion observed here being at the same 12 Farrar GJ, Kenna P, Jordan SA, et al. A three-base-pair dele- location as RED-1, such tentative data cannot tion in the peripherin-RDS gene in one form of retinitis pigmentosa. Nature 1991;354:478-80. be said to have excluded the RED-I locus. It 13 Schalling M, Hudson TJ, Buetow KH, Housman DE. remains to be seen whether there is a single Direct detection of novel expanded trinucleotide repeats in the . Nat Genet 1993;4:135-9. active locus where non-pathogenic expansions 14 Inglehearn CF, Carter SA, Keen TJ, et al. A new locus for occur or if such expansion events occur autosomal dominant retinitis pigmentosa on chromosome 7p. Nat Genet 1993;4:51-3. http://jmg.bmj.com/ throughout the genome. 15 Bardien S, Ebenezer N, Greenberg J, et al. An eighth locus One further limitation in the interpretation for autosomal dominant retinitis pigmentosa is linked to chromosome 17q. Hum Mol Genet 1995;4:1459-62. of this study is that it was not possible to ana- 16 Al-Maghtheh M, Inglehearn CF, Keen TJ, et al. Identifica- lyse DNA from the tissue affected, namely the tion of a sixth locus for autosomal dominant retinitis pigmentosa on chromosome 19. Hum Mol Genet 1994;3: retina. It has recently been suggested that tissue 351-4. specific variation in the extent of repeat expan- 17 Evans K, Al-Maghtheh M, Fitzke FW, et al. Bimodal expres- sivity in dominant retinitis pigmentosa genetically linked to sion can affect the presentation of the pheno- chromosome 19q. Br3' Ophthalmol 1995;79:841-6. type. In the case of dentatorubral- 18 Kim RY, Fitzke FW, Moore AT, et al. Autosomal dominant on September 28, 2021 by guest. Protected copyright. retinitis pigmentosa mapping to chromosome 7p exhibits pallidoluysian atrophy it appears that there is a variable expression. BrJ' Ophthalmol 1995;79:23-7. correlation between the degree of variation of 19 Morris AG, Gaitonde E, McKenna PJ, Mollon JD, Hunt DM. CAG repeat expansions and schizophrenia: associ- repeat size and age at death, though an expan- ation with disease in females and with early age-at-onset. sion was observed in all tissues examined.20 Hum Mol Genet 1995;4:1957-61. 20 Takano H, Onodera 0, Takahashi H, et al. Somatic In this study we have assumed that fully mosaicism of expanded CAG repeats in brains of patients dominant expression would be the most likely with dentatorubral-pallidoluysian atrophy: Cellular population-dependent dynamics of mitotic instability. Am J model for the effect of an expanded (CAG)n Hum Genet 1996;58:1212-22. repeat in a retinal gene. While this has been the 21 O'Donovan MC, Guy C, Craddock N, et al. Expanded CAG repeats in schizophrenia and bipolar disorder. Nat mode of expression in the neurodegenerative Genet 1995;10:380-1. diseases that have been characterised so far, 22 Lindblad K, Nylander PO, De Bruyn A, et al. Detection of expanded CAG repeats in bipolar affective disorder using recent reports have also implicated (CAG)n the repeat expansion detection (RED) method. Neurobiol repeats as risk factors in schizophrenia'9 21 and Dis 1995;2:55-62.