Allelic Variation of the FRMD7 Gene in Congenital Idiopathic Nystagmus

OPHTHALMIC MOLECULAR GENETICS

SECTION EDITOR: JANEY L. WIGGS, MD, PhD Allelic Variation of the FRMD7 Gene in Congenital Idiopathic Nystagmus

James E. Self, BM, MRCOphth; Fatima Shawkat, PhD; Crispin T. Malpas, BM, MRCOphth; N. Simon Thomas, PhD; Christopher M. Harris, PhD; Peter R. Hodgkins, FRCOphth; Xiaoli Chen, BSc; Dorothy Trump, MD, FRCP; Andrew J. Lotery, MD, FRCOphth

Objectives: To perform a genotype-phenotype corre- onstrated no clear causal link between skewing and vari- lation study in an X-linked congenital idiopathic nystag- able penetrance. mus pedigree (pedigree 1) and to assess the allelic vari- ance of the FRMD7 gene in congenital idiopathic Conclusions: We confirm profound phenotypic varia- nystagmus. tion in X-linked congenital idiopathic nystagmus pedigrees. We demonstrate that other congenital nystagmus Methods: Subjects from pedigree 1 underwent de- genes exist besides FRMD7. We show that the role of X tailed clinical examination including nystagmology. inactivation in variable penetrance is unclear in congen- Screening of FRMD7 was undertaken in pedigree 1 and ital idiopathic nystagmus. in 37 other congenital idiopathic nystagmus probands and controls. Direct sequencing confirmed sequence Clinical Relevance: We demonstrate that phenotypic changes. X-inactivation studies were performed in variation of nystagmus occurs in families with FRMD7 pedigree 1. mutations. While FRMD7 mutations may be found in some cases of X-linked congenital idiopathic nystag- Results: The nystagmus phenotype was extremely vari- mus, the diagnostic yield is low. X-inactivation assays are able in pedigree 1. We identified 2 FRMD7 mutations. unhelpful as a test for carrier status for this disease. However, 80% of X-linked families and 96% of simplex cases showed no mutations. X-inactivation studies dem- Arch Ophthalmol. 2007;125(9):1255-1263

YSTAGMUS IS A DISORDER these pedigrees had causative FRMD7 mu- of oculomotor control tations. An additional 14 small pedigrees and can occur as an iso- with congenital idiopathic nystagmus and Author Affiliations: Clinical lated inherited trait, a an inheritance pattern consistent with X Neurosciences Division, congenital idiopathic linkage were screened and FRMD7 muta- University of Southampton nystagmus,N or as secondary to other vi- tions were found in 8 of the pedigrees (Drs Self and Lotery and sual (sensory deficit nystagmus) or neu- (57%). Forty-two simplex cases were Ms Chen), and Southampton rological (neurological nystagmus) dis- screened for mutations in this gene and Eye Unit, Southampton General eases.1 In all cases, the underlying yielded 3 mutations (7%). Therefore, the Hospital (Drs Self, Shawkat, pathophysiology is poorly understood. overall contribution of FRMD7 mutations Malpas, Hodgkins, and Lotery), Furthermore, it is unclear which charac- to the cause of both X-linked and single- Southampton, England; Wessex Regional Genetics Laboratory, teristics of the nystagmus phenotype, if ton cases remains relatively unexplored. Salisbury District Hospital, any, are reliable as a diagnostic tool to help In X-linked congenital idiopathic nys- Salisbury, England identify underlying etiology. tagmus pedigrees, penetrance among fe- (Dr Thomas); SensoriMotor Congenital idiopathic nystagmus is ge- male obligate carriers has been vari- Laboratory, Centre for netically heterogeneous and has been de- able,3-5,7 ranging from 30% to 100%. Theoretical and Computational scribed as an autosomal dominant,2 auto- Possible mechanisms for this variability in- Neuroscience, University of somal recessive,3 X-linked dominant,3 or clude skewed X inactivation, genetic modi- Plymouth, Plymouth, England X-linked recessive4 trait. X-linked loci have fiers (such as polymorphisms within in- (Dr Harris); and Academic Unit been identified at Xp11.4-p11.34 and Xq26- teracting proteins), and other nongenetic of Medical Genetics, School of q27.3,5 In November 2006, Tarpey et al6 influences (such as environment) on ocu- Medicine and Centre for Molecular Medicine, Faculty of identified 22 nystagmus-causing muta- lomotor development. These factors may Medical and Human Sciences, tions in the FERM domain–containing 7 also explain why X-linked dominant and University of Manchester, (FRMD7) gene, which resides within the recessive pedigrees, with nystagmus or Manchester, England Xq26-q27 interval. Sixteen X-linked fami- other ocular diseases, can both show link- (Dr Trump). lies underwent linkage analysis and 15 of age to the same region.5,8,9

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1255

II:1 II:2 II:3 II:4 II:5 II:6 II:7 II:8

III:1 III:2 III:3 III:4 III:5 III:6 III:7 III:8 III:9 III:10 III:11 III:12 III:13 III:14 III:15 III:16 III:17 III:18 III:19 III:20

IV:1 IV:2 IV:3 IV:4 IV:5 IV:6 IV:7 IV:8 IV:9 IV:10 IV:11 IV:12 IV:13 IV:14 IV:15 IV:16 IV:17 IV:18 IV:19 IV:20 IV:21 IV:22 IV:23 IV:24 IV:25

Underwent detailed phenotyping Unaffected male Affected male Unaffected female Affected female Obligate female carrier V:1 V:2 V:3 V:4 V:5 V:6 V:7 V:8 V:9 V:10 V:11

Figure 1. X-linked congenital idiopathic nystagmus pedigree (pedigree 1). This family has 9 affected males, 2 affected females, and 11 obligate female carriers.

This article describes high-resolution phenotyping electrophysiological examination but without detailed eye move- using infrared limbal tracking, electrophysiology, and ment recordings. clinical examination in affected and unaffected mem- International Society for Clinical Electrophysiology of Vision– bers of a single large congenital nystagmus pedigree (pedi- standardized electroretinogram (http://www.iscev.org/) and visual evoked potential tests were completed in 2 affected males gree 1) (Figure 1). The role of FRMD7 mutations in 9 and 2 obligate female carriers. Monocular stimulation and a other X-linked families and 28 simplex cases is investi- 3-channel transoccipital electrode montage were employed for gated and, in addition, studies of X inactivation are per- visual evoked potential recordings to optimize detection of neu- formed to investigate the hypothesis that skewing of X ronal misrouting suggestive of ocular albinism. Informed con- inactivation may be a major contributor to the variable sent was obtained from all subjects for genetic studies, and ge- penetrance seen in X-linked nystagmus pedigrees. nomic DNA was isolated from either blood or Oragene saliva sample kits (DNA Genotek Inc, Ottawa, Ontario, Canada). METHODS GENOTYPING AND The study had the approval of the local and regional ethics LINKAGE ANALYSIS committees and conformed to the tenets of the Declaration of Helsinki. Twenty-nine individuals (7 affected males, 2 af- We had previously shown by conventional linkage mapping techniques that the nystagmus gene in pedigree 1 links to an fected females, 11 obligate female carriers, and 9 unaffected 5 members) in a single congenital idiopathic nystagmus pedi- 8-cM region at Xq24-q26.3. This contains the FRMD7 gene, gree underwent detailed clinical examination (Figure 1), in- which is now known to be a cause of X-linked nystagmus and cluding tests for logarithm of the minimum angle of resolu- was thus investigated in this family. DNA was amplified using tion visual acuity, refraction, color vision, intraocular standard polymerase chain reaction (PCR) protocols. Primers pressure; anterior and posterior segment slitlamp examina- for FRMD7 (complementary DNA sequence NM_194277) were designed to include all 12 exons, splice sites, and both the 5Ј tion, including iris transillumination testing in a darkened Ј room; and orthoptic assessment. Twenty-four of these pa- and 3 untranslated regions in amplimers with fewer than 250 tients had detailed recordings of their nystagmus waveform base pairs (primers available on request). Pedigrees 2 through performed using Skalar IRIS IR Light Eye Tracker equipment 10 were not large enough to provide linkage information in- (Cambridge Research Systems Ltd, Rochester, England). dividually, and to avoid errors in light of known X-linked con- Twenty-four eye movement recordings were completed for genital idiopathic nystagmus locus heterogeneity, these fami- each patient. Binocular and uniocular saccades were recorded lies were not combined. Therefore, linkage information was not to calibrate amplitude measurements at ±10° and ±20° from ascertained for these families and definitive proof of X linkage fixation in the horizontal plain using a 1° red square target was not possible. moving at 500-millisecond intervals. Binocular and uniocular optokinetic nystagmus (OKN) measurements were carried SINGLE-STRAND CONFORMATIONAL out to rightward and leftward drifting gratings measuring 0.2 POLYMORPHISM ANALYSIS cycles per degree at a velocity of 25° per second. Waveforms were analyzed while viewed in 5 positions (primary, 10° right, Single-strand conformational polymorphism (SSCP) was per- 10° left, 20° right, and 20° left of fixation), both binocularly formed using standard techniques.10 Specifically, polyacryl- and uniocularly using a 1° red square target. amide gels with glycerol, 6%, were run for 3.5 hours at 25 W An additional 9 congenital idiopathic nystagmus pedigrees per gel at room temperature. We screened for sequence varia- with inheritance consistent with X linkage were identified tions in 96 female controls (182 control X chromosomes), fol- (Figure 2). These pedigrees had similarly detailed clinical and lowed by 2 affected subjects from the first pedigree, 9 pro-

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1256

Pedigree 6 Pedigree 7 Pedigree 8

Affected male Unaffected male Affected female Unaffected female Obligate female carrier Pedigree 9 Pedigree 10

Figure 2. Nine additional pedigrees screened for FRMD7 mutations.

bands from other X-linked nystagmus pedigrees, and 28 cuts at the promoter site on the active (unmethylated) but not nystagmus simplex cases. on the inactive (methylated) X chromosome. Therefore, by examining the size and peak heights of the amplification SEQUENCING products and by tracking the inheritance of alleles from par- ents to offspring, we ascertained the percentage of cells in Sequencing was performed for samples with either shifts or failed which the maternal or paternal X chromosome had been inac- PCR seen on SSCP gels. Following standard PCR, each frag- tivated. To calculate the X-inactivation ratio, each sample was ment was sequenced using the Big Dye Terminator Cycle Se- set up in duplicate: 1 digest and 1 mock digest without en- quencing kit, version 1.1, and an ABI 3100 Automated Capil- zyme. Both were then amplified by PCR. For heterozygotes, lary DNA Sequencer (Applied Biosystems, Foster City, the ratio of the 2 peaks were compared between the undi- California) following the manufacturer’s protocol. gested and digested samples. Male samples were used for hap- lotype reconstruction. RESTRICTION ENZYME DIGEST RESULTS Cosegregation of an identified mutation with the nystagmus phenotype in pedigree 1 was tested using a restriction enzyme digest following standard PCR. The PstI enzyme was used ac- CLINICAL PHENOTYPING STUDIES cording to the manufacturer’s protocol (Promega, Madison, Wis- consin) and chosen to cut wild type (CTGCAG) but not mu- Flash and pattern electroretinograms and occipital pat- tant sequence (CTGCAAG) mutations in this pedigree. tern visual evoked potential recordings in 2 affected males and 2 obligate female carriers were normal, thus X-INACTIVATION ANALYSIS excluding masquerading eye conditions. Computerized tomographic brain scans were performed on 2 affected Two assays comprising amplification of a short fragment of X individuals at diagnosis with no abnormal findings. chromosome (amplimer) were used to determine the activa- Ophthalmic examination results were also normal ex- tion status of each X chromosome (ie, the maternal or pater- cept for nystagmus in affected patients. The prevalence nal X chromosome) in females. Each amplimer contains a of strabismus was 44% in affected subjects (4 of 9) and polymorphic marker adjacent to the promoter of a gene, which is unmethylated on the active X chromosome and 15.8% in unaffected subjects (3 of 19). Refractive errors methylated on the inactive chromosome. The 2 genes investi- ranged from uniocular logarithm of the minimum angle gated were the androgen receptor Humar11 and ZNF261.12 of resolution acuities of 0.1 to 0.5 in affected subjects. Both genes are in Xq13. Prior to amplification, both assays We have published the detailed results previously5 and employ a methylation-sensitive restriction enzyme, which they are in agreement with previous findings in congen-

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1257

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Table 1. Infrared Limbal Nystagmology Findings in Affected Members of an X-linked Congenital Idiopathic Nystagmus Pedigree (Pedigree 1)

Individual Affected With Congenital Idiopathic Nystagmus

Measure IV:15 IV:18 III:16 V:4 II:5 III:8 III:6 V:8 V:9 III:2 IV:2 OKN present . . . NP Presenta NP NP NP NP NP Low gain Low gainb NP Direction in primary . . . Pen R Pen R Pen Pen R R R R Amplitude in primary . . . 2.5° 0.25° 2.22° 1.7° 3.5° 2.5° 5.4° 0.31° 5.6° 4.2° Frequency in primary, . . . 2.66 1.10 3.92 3.33 3.00 4.44 3.60 1.47 1.40 4.50 cycles/s Waveform in primary . . . Pen D Pen A Pen Pen A Lin D A Direction in L gaze . . . L R L L Pen L L R Pen L Amplitude in L gaze . . . 2.6° 0.3° 2.8° 1.0° 3.2° 3.3° 2.2° 0.2° 2.8° 4.1° Frequency in L gaze, . . . 4.20 0.92 5.50 5.00 2.94 5.50 2.80 2.10 8.50 3.80 cycles/s Waveform in L gaze . . . A Lin Lin Lin Pen Lin D Lin Pen PC Direction in R gaze . . . R R R R R Pen Pen R R R Amplitude in R gaze . . . 2.9 ° 0.3° 2.1° 2.1° 3.1° 1.6° 0.5° 0.2° 0.8° 5.3° Frequency in R gaze, . . . 4.16 1.20 3.00 2.25 3.06 5.00 3.33 1.12 3.56 3.42 cycles/s Waveform in R gaze . . . PC D PC Lin PC Pen Pen Lin Lin PC Waveforms seen any . . . Lin, Pen, Lin, D D, PC, Lin, PC, A, Lin A, PC, D, PC, Lin, D, PC, Lin Lin, Pen, A, Pen, direction A, PC Pen, A Pen Lin, A Pen, A D, A PC Rebound seen, yes/no . . . No No No No No No No No No No Changes in phenotype . . . MLN No MLN No No No No No MLN MLN when viewing monocularly

Abbreviations: A, accelerating slow phase; D, decelerating slow phase; L, left; Lin, linear slow phase; MLN, manifest-latent nystagmus; NP, not present (disrupted); OKN, optokinetic nystagmus; Pen, pendular; PC, pseudocycloid waveform; R, right; (. . .) data not recorded. aOptokinetic nystagmus observed with monocular nasotemporal asymmetry. bLow-gain OKN, but only in the direction of the nystagmus.

ital nystagmus pedigrees.3,13,14 Eye movement recording In an affected male from X-linked pedigree 2 and an un- results are presented in Table 1 for all patients with related singleton female, a shift was seen in an amplimer findings positive for nystagmus. Examples of nystag- containing exon 4. Sequencing identified a point muta- mus waveforms seen in this pedigree are presented in tion in the first intronic base after exon 4 in both subjects, Figure 3. Eighteen unaffected subjects, including 6 designated IVS4ϩ1G→A. This sequence had been previ- obligate female carriers, had no abnormalities on any of ously described by Tarpey et al6 and the change is pre- the 24 recordings and so were not included in Table 1. dicted to cause a splice recognition site mutation leading Also included in Table 1 is a female carrier with nystag- to continued transcription into intron 4-5 resulting in a pre- mus, possibly secondary to a cochlear implant (III:16), mature stop codon after 9 amino acids, thereby truncat- and a male subject (III:8) who is affected but was ex- ing the FRMD7 protein or leading to nonsense-mediated cluded owing to a history of congenital cataract. The re- decay of the messenger RNA. These mutations were not sults also exclude an affected subject (IV:15) for whom seen in controls and are summarized in Table 2. eye movement recordings were not collected. RESTRICTION ENZYME DIGEST RESULTS SSCP AND SEQUENCING RESULTS FOR PEDIGREE 1

Thirteen gel shifts were found in 9 amplimers in both The results of PstI restriction enzyme digest of the exon the control and affected probands and these were 9 PCR amplimer for individuals in pedigree 1 are illus- checked by sequencing and corresponded to known trated in Figure 4, along with a condensed pedigree to single nucleotide polymorphisms. A shift was found for aid subject identification. Mutation 880insA, 293fs co- both samples from pedigree 1 in an amplimer contain- segregates with all affected and female carriers but not ing part of exon 9. Sequencing revealed this was caused with unaffected patients. Notably, affected individual III:8, by a single base insertion (“A”) after base 880 (comple- who was excluded from initial work because of a his- mentary DNA sequence NM_194277) designated tory of congenital cataract, is also hemizygous for the 880insA, 293fs. This frameshift mutation is predicted to 880insA, 293fs mutation. cause a premature stop codon at amino acid position 301 (protein sequence NP_919253), thereby truncating X-INACTIVATION STUDIES the FRMD7 protein (Lasergene, version 6.1.3; DNASTAR Inc, Madison) or leading to nonsense-medi- Results for the X-inactivation assays are in Table 3. ated decay of the messenger RNA. Ten of the 16 females who were tested demonstrated at

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1258

5° 5° 1 s 5 s

Left Left

A E

B F

C G

D H

Figure 3. Infrared eye movement recordings from a pedigree with congenital idiopathic nystagmus (pedigree 1). Each recording is binocular with the left eye trace on top of the right eye trace. A, Normal recording from individual IV:17 with a normal finding of a square-wave jerk. B, Accelerating slow-phase nystagmus typical of congenital idiopathic nystagmus from individual IV:18. C, Decelerating slow-phase nystagmus atypical for congenital idiopathic nystagmus from individual III:2. D, Subtle nystagmus waveforms missed by clinical examination from individual V:9. E, Normal optokinetic nystagmus from individual IV:18. F, Disrupted (absent) optokinetic nystagmus (on a pendular nystagmus background) typical of congenital idiopathic nystagmus from individual III:6. G, Low-gain optokinetic nystagmus unusual for congenital idiopathic nystagmus from individual V:9. H, Pseudocyclic nystagmus (a combination of pendular and accelerating slow phase) common in congenital idiopathic nystagmus from individual V:9.

Table 2. Mutations Identified in FRMD7a

Sample Pedigree Class Mutation Origin XN3P X-linked (pedigree 1) Truncating 880insA, 293fs England NRD2 Singleton Truncating IVS4 ϩ 1G→A England XNTW1 X-linked (pedigree 2) Truncating IVS4 ϩ 1G→A England

a The reference complementary DNA sequence NM_194277 is used as a basis for numbering the nucleotide of the mutation. All mutations are located relativeto the A of the first coding ATG at position 179. The reference protein sequence NP_919253 is used as the basis for numbering the amino acid variation starting from the first methionine at position 1.

least moderate skewing, including both affected fe- and the X carrying the FRMD7 mutation was active in males, and 3 cases showed skewing of more than 95%. 73% of cells. For informative unaffected carriers, the X However, as illustrated, it is not always possible to tell chromosome carrying the FRMD7 mutation was active which X chromosome is inactivated. Mutation status of in 6% to 80% of cells. Therefore, there was no clear-cut each X chromosome was known for 1 affected female difference in the pattern of X inactivation between af-

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1259

∗

267-bp Product 133/134-bp Products

Affected male Unaffected male Affected female Unaffected female Obligate female carrier

Figure 4. Restriction enzyme digest of DNA samples from an X-linked congenital idiopathic nystagmus pedigree (pedigree 1) using the enzyme Pst I. The mutant (uncut) allele generates a 267–base pair (bp) fragment, while the wild-type allele (cut) produces fragments of 133 bp and 134 bp. Control lanes are included. Asterisk indicates unaffected male III:17 with no Pst I enzyme; and triangle, unaffected male III:17 with Pst I enzyme.

fected and unaffected carriers of the mutations. Inter- waveform characteristics seen in individuals of this pedi- estingly, for individual IV:17, assays were performed gree would suggest wholly differing etiologies. These re- for both blood- and saliva-extracted DNA with identical sults support the opinion that waveforms alone should results. be interpreted with caution when employed as a diagnostic tool in congenital idiopathic nystagmus. How- COMMENT ever, detailed eye movement recordings may be neces- sary to assign affection status in congenital idiopathic We have shown that congenital idiopathic nystagmus pa- nystagmus pedigrees and the families of congenital id- tients in a single family with this novel 880insA, 293fs– iopathic nystagmus singleton cases. An example is indi- truncating FRMD7 mutation have extremely variable nys- vidual V:9 from pedigree 1 who had been previously di- tagmus phenotypes. Classically, the hallmarks of the agnosed as unaffected by an experienced ophthalmologist congenital idiopathic nystagmus phenotype are acceler- but was found to have subtle nystagmus only after ex- ating slow phases, loss of OKN, and no changes in phe- amination with infrared oculography (and was subse- notype on monocular viewing.1 In pedigree 1, however, quently found to be hemizygous for the causative mu- 7 of 9 patients with congenital idiopathic nystagmus had tation in this family). This has significant implications 3 or more waveforms, including prolonged periods of de- for diagnosis and genetic counseling for this individual celerating or linear slow phases, which are more com- and other currently undiagnosed individuals from con- monly seen in patients with nystagmus of neurologic ori- genital idiopathic nystagmus pedigrees. gin. Similarly, OKN was preserved to some degree in 2 We confirm that FRMD7 mutations are a cause of X- patients, which is considered to be rare in congenital id- linked congenital idiopathic nystagmus and describe a novel iopathic nystagmus. Interestingly, in subject III:16 (fe- protein-truncating mutation. By screening all exons and male carrier but also known to have a cochlear im- splice sites of the FRMD7 gene, we identified mutations in plant), OKN was preserved, the nystagmus amplitude was 20% (2 of 10) of apparent X-linked pedigrees. This per- very low, and no accelerating slow phases were seen (also centage is significantly smaller than that found by Tarpey rare in congenital idiopathic nystagmus). These find- et al,6 who identified mutations in 8 of 14 (57%) apparent ings might suggest that her nystagmus is secondary to X-linked pedigrees and 15 of 16 (94%) proven X-linked the cochlear implant and not congenital idiopathic nys- pedigrees. This may be because we used SSCP as our ini- tagmus. Subject III:8, who has nystagmus but was ex- tial method of mutation detection. In our experience, SSCP cluded owing to a history of congenital cataract, has nys- has a sensitivity of 89% (95% confidence interval, 79%- tagmus waveforms that are very similar to the affected 96%) compared with direct DNA sequencing.15 There- subjects, including disrupted OKN and accelerating slow fore, our detection rate is likely to be lower than that found phases. These findings would be more suggestive of con- by direct DNA sequencing. However, additionally, the co- genital idiopathic nystagmus than sensory deficit nys- hort used by Tarpey et al6 had been previously used for link- tagmus1 due to congenital cataract. This subject was sub- age analysis to identify the causative gene inevitably lead- sequently found to be hemizygous for the causative ing to an ascertainment bias. We also identified FRMD7 mutation in this family. mutations in 1 of 28 (3.6%) singleton cases, which is simi- The clinical phenotyping demonstrated that in some lar to the 3 of 42 (7%) singleton cases found by Tarpey et cases waveform characteristics were indicative of under- al.6 Therefore, according to our results, most (80% in this lying etiology in this family; but conversely, many of the study) unselected X-linked families and 96.4% of single-

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1260

Humar ZNF261

Allele 1 Allele 2 Allele 1 Allele 2

Size, Size, Size, Size, Degree Sample DNA Sex Status base pairs Activity, % base pairs Activity, % base pairs Activity, % base pairs Activity, % of Skewing II:3 Saliva F Carrier 233 44 241 56 264 52 272 48 Randomb Failed Failed 53 47 II:5 Blood F Affected 233 5 241 95 264 98 272 2 Very skewedc 4 96 98 2 III:2 Saliva M Affected 233 264 III:6 Saliva M Affected 233 272 III:16 Saliva F Carrier 233 X 256 14 272 86 Skewedd 15 85 III:17 Blood M Normal 233 268 III:18 Blood F Normal 233 X 256 98 272 2 Very skewedc 98 2 III:19 Blood F Carrier 233 X 256 88 272 12 Skewedd 87 13 IV:1 Saliva F Carrier 241 X 264 73 270 27 Skewedd 69 31 IV:2 Saliva F Affected 236 27 241 73 264 X Skewedd 29 71 IV:6 Saliva F Carrier 227 39 233 61 270 42 272 58 Randomb 40 60 42 58 IV:7 Saliva F Carrier 230 31 233 69 262 23 272 77 Skewedd 27 73 27 73 IV:14 Blood F Normal 233 73 256 27 264 18 266 82 Skewedd 71 29 18 82 IV:15 Blood M Affected 233 272 IV:17 Saliva F Carrier 233 X 268 51 272 49 Randomb 52 48 IV:17 Blood F Carrier 233 X 268 54 272 46 Randomb 57 43 IV:18 Blood M Affected 233 272 IV:19 Blood F Normal 221 20 233 80 262 44 270 56 Skewedd 30 70 43 57 IV:21 Blood F Carrier 215 94 233 6 256 4 274 96 Very skewedc 91 9 5 95 V:2 Blood F Carrier 233 63 256 37 264 21 272 79 Skewedd 63 37 20 80 V:6 Saliva F Carrier 221 53 233 47 270 55 272 45 Randomb 67 33 Failed Failed V:7 Saliva F Normal 221 56 233 44 256 41 264 59 Randomb 56 44 33 67

aTwo activity values are shown for each assay, as all experiments were performed twice. Empty cells show that there is no second allele. X denotes that a percentage calculation was not possible, as the patient was homozygous for this marker. bMean result of skewing from 50% to 70%. cMean result of skewing greater than 90%. dMean result of skewing between 70% and 90%.

ton congenital idiopathic nystagmus cases do not have tions in FRMD7 may cause nystagmus by altering the FRMD7 mutations as a cause. This suggests the existence neurite length and degree of branching of neurons as of other prevalent nystagmus genes on the X chromo- they develop in the midbrain, cerebellum, and retina.6 some. Alternatively, mutations in as yet uncharacterized The mutations identified in our study are also clustered regulatory elements for FRMD7, such as promoters, silenc- around these domains and thus our results support the ers, or enhancers, may also exist but were not detected by hypothesis that mutations in this region of FRMD7 are our assays. particularly important for nystagmus. FARP2 and other FRMD7 mutations found by Tarpey et al6 are clus- homologues may also be potential candidates for nys- tered around the B41 and FERM-C domains.6 The tagmus genes. FARP2 gene on chromosome 2 shares significant ho- We had previously proposed that the likeliest expla- mology with a large portion of FRMD7, including these nation for the variable phenotype in female carriers in domains. It is also known that FARP2 alters neurite pedigree 1 was variability in the pattern of X inactiva- length and degree of sprouting in rat embryonic cortical tion.5,18 Skewed X inactivation (significant deviation away neurons.16,17 This has led to the hypothesis that muta- from the expected 50:50 contribution of each X chro-

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1261

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 mosome) has been described in other ocular diseases8,19 status for nystagmus caused by FRMD7 mutations. This and may explain why other X-linked nystagmus pedi- is important from a diagnostic viewpoint, as X inactiva- grees linked to the locus containing FRMD7 have shown tion is used as a surrogate test for carrier status in other either dominant or recessive inheritance patterns.3-6 The X-linked conditions.26,27 vast majority of genes on the long arm of the X chromo- In summary, we have demonstrated the variability of some are subject to X inactivation, including those im- the congenital idiopathic nystagmus phenotype in pa- mediately flanking FRMD7, like MST4, MBNL3, and tients with FRMD7 mutations. We report a novel FRMD7 RAP2C.20 Because genes that are subject to or escape from mutation and confirm that mutations in this gene may X inactivation tend to be clustered into domains, it is very cause nystagmus. This work suggests that other nystag- likely that FRMD7 is inactivated. Only a very small pro- mus genes exist, that FRMD7 is rarely mutated in sim- portion of healthy females show significant levels of plex nystagmus cases, and that routine FRMD7 muta- skewed inactivation.21 However, our results suggest an tion screening would not have a sufficiently high detection excess of skewing in pedigree 1 (10 of 16 females includ- rate to warrant routine use in congenital idiopathic nys- ing 3 with Ͼ95% skewing). Interestingly, of the 10 fe- tagmus cases or suspected carriers. males with skewing, 7 are younger than 40 years; for 3 of these patients, the assays were performed on saliva samples. This suggests that the observed skewing in pedi- Submitted for Publication: December 13, 2006; final re- gree 1 is not because of age-related secondary skewing11 vision received February 22, 2007; accepted February 25, or specific to blood. 2007. There are 3 possible interpretations of the X- Correspondence: Andrew J. Lotery, MD, FRCOphth, inactivation results from blood or saliva. First, the effect Southampton Eye Unit, Southampton General Hospi- of X inactivation on penetrance may be subtle, rather than tal, Tremona Road, Southampton SO16 6YD, England all or nothing, as neither unaffected nor affected female ([email protected]). carriers have complete skewing. It is possible that the pro- Author Contributions: Dr Self had full access to all the portion of cells in which the mutation-carrying X is ac- data in the study and takes responsibility for the integ- tive may be higher in the crucial tissues in the affected rity of the data and the accuracy of the data analysis. cases than in carrier females. Many X-linked disorders Financial Disclosure: None reported. do not fit classic dominant or recessive modes of inher- Funding/Support: Grant support was provided by the itance and female carriers display variable penetrance.22 Medical Research Council, British Eye Research Foun- For example, mutations in the dystonia-deafness pep- dation, Mason Medical Research Foundation, Southamp- tide gene cause incomplete penetrance in females with ton Eye Unit League of Friends, and the Gift of Sight. variable X inactivation ratios in blood from 50:50 to more Additional Contributions: We thank the individuals from than 95:5.23 Thus, mutations in X-linked genes can cause pedigree 1 for their enthusiasm and participation in this partial cell selection and incompletely skewed X inacti- study. We also thank Angela Cree, BSc, and Helen Grif- vation. This could explain variable clinical expression fiths of the Gift of Sight Laboratory for technical and ad- within the same family. visory support. Second, the results obtained from lymphocytes and saliva samples may not necessarily reflect the patterns of REFERENCES X inactivation in the tissue whose pathophysiology causes nystagmus (as yet unknown). In healthy females, X- 1. Casteels I, Harris CM, Shawkat F, Taylor D. Nystagmus in infancy. Br J Ophthalmol. inactivation ratios are normally similar between tis- 1992;76(7):434-437. 11 2. Kerrison JB, Arnould VJ, Barmada MM, Koenekoop RK, Schmeckpeper BJ, sues. However, for certain X-linked conditions, com- Maumenee IH. A gene for autosomal dominant congenital nystagmus localizes plete skewing may be restricted to specific cell lineages, to 6p12. Genomics. 1996;33(3):523-526. and different patterns of X inactivation are seen in all other 3. Kerrison JB, Vagefi MR, Barmada MM, Maumenee IH. Congenital motor nys- tissues.24 Selection against cells that have inactivated the tagmus linked to Xq26-q27. Am J Hum Genet. 1999;64(2):600-607. normal X may be more pronounced in tissues where most 4. Cabot A, Rozet JM, Gerber S, et al. A gene for X-linked idiopathic congenital nystagmus (NYS1) maps to chromosome Xp11.4-p11.3. Am J Hum Genet. 1999; FRMD7 expression occurs, possibly in parts of the de- 64(4):1141-1146. 6 veloping brain and retina. 5. Self JE, Ennis S, Collins A, et al. Fine mapping of the X-linked recessive con- Third, variable penetrance may be at least in part be- genital idiopathic nystagmus locus at Xq24-q26.3. Mol Vis. 2006;12:1211- cause of other genetic factors. The results of the X- 1216. 6. Tarpey P, Thomas S, Sarvananthan N, et al. Mutations in FRMD7, a newly iden- inactivation studies give a clear overlap in the patterns tified member of the FERM family, cause X-linked idiopathic congenital nystagmus. of X inactivation between unaffected and affected fe- Nat Genet. 2006;38(11):1242-1244. male carriers. Individual IV:18, who is negative for the 7. Zhang B, Xia K, Ding M, et al. Confirmation and refinement of a genetic locus of FRMD7 mutation, is one of the females with severe skew- congenital motor nystagmus in Xq26.3-q27.1 in a Chinese family. Hum Genet. ing. It is possible that the higher frequency of skewed X 2005;116(1-2):128-131. 8. McMullan TF, Collins AR, Tyers AG, Robinson DO. A novel X-linked dominant inactivation seen in this family may be unrelated to nys- condition: X-linked congenital isolated ptosis. Am J Hum Genet. 2000;66(4): tagmus, as we know that skewed X inactivation per se 1455-1460. may run in families.25 9. Guo X, Li S, Jia X, Xiao X, Wang P, Zhang Q. Linkage analysis of two families Importantly, by showing that there is no definitive pat- with X-linked recessive congenital motor nystagmus. J Hum Genet. 2006;51 (1):76-80. tern of X inactivation in carrier or manifesting females, 10. Sheffield VC, Beck JS, Kwitek AE, Sandstrom DW, Stone EM. The sensitivity of we have shown that determining the X-inactivation pat- single-strand conformation polymorphism analysis for the detection of single tern in blood or saliva cannot be used to identify carrier base substitutions. Genomics. 1993;16(2):325-332.

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1262

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 11. Sharp A, Robinson D, Jacobs P. Age- and tissue-specific variation of X chro- 20. Carrel L, Willard HF. X-inactivation profile reveals extensive variability in X- mosome inactivation ratios in normal women. Hum Genet. 2000;107(4):343- linked gene expression in females. Nature. 2005;434(7031):400-404. 349. 21. Amos-Landgraf JM, Cottle A, Plenge RM, et al. X chromosome-inactivation pat- 12. Beever C, Lai BP, Baldry SE, et al. Methylation of ZNF261 as an assay for deter- terns of 1,005 phenotypically unaffected females. Am J Hum Genet. 2006;79 mining X chromosome inactivation patterns. Am J Med Genet A. 2003;120 (3):493-499. (3):439-441. 22. Dobyns WB, Filauro A, Tomson BN, et al. Inheritance of most X-linked traits is 13. Kerrison JB, Koenekoop RK, Arnould VJ, Zee D, Maumenee IH. Clinical features not dominant or recessive, just X-linked. Am J Med Genet A. 2004;129(2): of autosomal dominant congenital nystagmus linked to chromosome 6p12. Am 136-143. J Ophthalmol. 1998;125(1):64-70. 23. Plenge RM, Tranebjaerg L, Jensen PK, Schwartz C, Willard HF. Evidence that mu- 14. Klein C, Vieregge P, Heide W, et al. Exclusion of chromosome regions 6p12 and tations in the X-linked DDP gene cause incompletely penetrant and variable skewed 15q11, but not chromosome region 7p11, in a German family with autosomal X inactivation. Am J Hum Genet. 1999;64(3):759-767. dominant congenital nystagmus. Genomics. 1998;54(1):176-177. 24. Belmont JW. Insights into lymphocyte development from X-linked immune 15. Webster AR, Heon E, Lotery AJ, et al. An analysis of allelic variation in the ABCA4 deficiencies. Trends Genet. 1995;11(3):112-116. gene. Invest Ophthalmol Vis Sci. 2001;42(6):1179-1189. 25. Naumova AK, Plenge RM, Bird LM, et al. Heritability of X chromosome– 16. Kubo T, Yamashita T, Yamaguchi A, Sumimoto H, Hosokawa K, Tohyama M. inactivation phenotype in a large family. Am J Hum Genet. 1996;58(6):1111- A novel FERM domain including guanine nucleotide exchange factor is involved 1119. in Rac signaling and regulates neurite remodeling. J Neurosci. 2002;22(19): 26. Schmucker B, Meindl A, Achatz H, et al. A PCR based X-chromosome inactiva- 8504-8513. tion assay for carrier detection in X-linked immunodeficiencies using differen- 17. Toyofuku T, Yoshida J, Sugimoto T, et al. FARP2 triggers signals for Sema3A- tial methylation of the androgen receptor gene. Immunodeficiency. 1995;5 mediated axonal repulsion. Nat Neurosci. 2005;8(12):1712-1719. (3):187-192. 18. Self J, Lotery A. The molecular genetics of congenital idiopathic nystagmus. Se- 27. Greer WL, Kwong PC, Peacocke M, Ip P, Rubin LA, Siminovitch KA. X- min Ophthalmol. 2006;21(2):87-90. chromosome inactivation in the Wiskott-Aldrich syndrome: a marker for detec- 19. Lyon MF. X-chromosome inactivation: pinpointing the centre. Nature. 1996;379 tion of the carrier state and identification of cell lineages expressing the gene (6561):116-117. defect. Genomics. 1989;4(1):60-67.

From the Archives of the Archives

Retained foreign bodies were found in 731 of 3,882 eyes of soldiers studied at the Army Institute of Pathology dur- ing World War II. All globes with penetrating wounds were examined roentgenologically and were searched for foreign bodies. In many instances the particles were so small or so deeply embedded in organizing hemorrhage or inflammatory membrane that they were not recov- ered from the gross specimens and became visible only on microscopic examination. These could not be sub- jected to the magnet test, but sections containing them were stained with prussian blue. Reference: Wilder HC. Foreign bodies of the globe: a symposium, pathologic aspects. Arch Ophthalmol. 1948;39(3):424.

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 9), SEP 2007 WWW.ARCHOPHTHALMOL.COM 1263