Exp. Eye Res.

(2002) 74, 737±745

doi:10.1006/exer.2002.1169, available online at http://www.idealibrary.com on

Spinocerebellar Ataxia Type 7 (SCA7) Shows a Cone±Rod

Dystrophy Phenotype
TO M A S S . A L E M A N a, A R T U R V. C I D EC I YA N a, N I C H O L A S J. VO L P E a,
G I OVA N N I S T E VA N I N bc, A L E X I S B R I C E bc A N D S A M U E L G . J ACO B S O N a*
a
Department of Ophthalmology, Scheie Eye Institute, University of Pennsylvania, Philadelphia, PA, U.S.A.,
b
INSERM U289, HoÃpital de la SalpeÃtriere, Paris, France and cDeÂpartement de GeÂneÂtique CytogeÂneÂtique et
Embryologie and Federation de Neurologie, Groupe Hospitalier PitieÂ-SalpeÃtrieÁre, Paris, France

(Received St. Louis 5 November 2001 and accepted in revised form 8 January 2002)

Autosomal dominant spinocerebellar ataxia 7 is associated with retinal degeneration. SCA7, the causative
gene, encodes ataxin-7, a ubiquitous 892 amino acid protein of variable sub-cellular localization, and the
disease is due to expansion of an unstable CAG repeat in the coding region of the gene. Recent increases in
understanding of the mechanisms of SCA7-related retinopathy from in vitro and murine model studies
prompted us to perform a detailed study of the retinal phenotype of affected members of a family with
SCA7 mutation (45±47 CAG repeats). There was a spectrum of severity from mild to severe dysfunction.
Early functional abnormalities were at both photoreceptor and post-receptoral levels. When cone and rod
photoreceptor dysfunction was present, it was approximately equal. Regional retinal dysfunction was
evident: there was more dysfunction centrally than peripherally with least effect in the midperiphery. In
vivo cross-sectional retinal images with optical coherence tomography showed an early disease stage of
altered foveal lamination (abnormal area of low re¯ectivity splitting the outer retina-choroidal complex)
accompanied in the parafovea by reduced retinal thickness. Later disease stages showed foveal and
parafoveal retinal thinning. The phenotype in this family with SCA7 is that of a cone±rod dystrophy. These
observations increase interest in a recent hypothesis that ataxin-7 may interfere with the function of CRX
(cone±rod homeobox), a transcription factor regulating photoreceptor genes and a cause of a cone±rod
dystrophy phenotype in man. # 2002 Academic Press
Key words: ataxin; cone; polyglutamine disease; repeat expansions; retinal degeneration; rod; spino-
cerebellar ataxia.

1. Introduction et al., 2000; Einum et al., 2001). It has been postulated

that the expanded allele induces a cytotoxic gain of
Spinocerebellar ataxia type 7 (SCA7) is an autosomal
function in the protein (Mauger et al., 1999). Recent
dominant neurodegenerative disorder characterized by
experimental studies have led to the speci®c hypothesis
progressive neurological manifestations and visual loss
that the retinal phenotype in SCA7 results from
with a retinal basis (Harding, 1982, 1993; David et al.,
interference with the action of the cone±rod homeobox
1998; Giunti et al., 1999). SCA7, previously known
(CRX) gene (La Spada et al., 2001).
as autosomal dominant ataxia type II (ADCAII) or
There have been many reports of the visual
olivopontocerebellar ataxia type III (OPCAIII), belongs
disturbances in hereditary spinocerebellar ataxias in
to a heterogeneous group of neurological disorders
the era before molecular diagnoses (reviewed in
caused by heritable unstable trinucleotide repeats
Harding (1982, 1993)) and more recent study of
(David et al., 1997; Cummings and Zoghbi, 2000).
SCA7-associated retinopathy (Abe et al., 2000). To
The causative gene has been mapped to the short
complement and extend these earlier investigations
arm of chromosome 3 (3p12±13) and encodes
and increase understanding of the role that ataxin-7
ataxin-7, an 892 amino acid protein of unknown
may play in the human retina, we have characterized
function (David et al., 1997, 1998; Del-Favero et al.,
in detail with non-invasive tests the retinal function
1998; Koob et al., 1998). Ataxin-7 shares a short and
and structure of members of a family with mutations
functional motif homologous to the phosphate-binding
in the SCA7 gene.
site of arrestins (Mushegian et al., 2000). It localizes to
the nucleus and/or the cytoplasm depending on the
cell population and is widely expressed in human 2. Materials and Methods
tissues, including the retina (Holmberg et al., 1998;
Subjects
Kaytor et al., 1999; Cancel et al, 2000; Lindenberg
A family with spinocerebellar ataxia was included in
* Address correspondence to: Samuel G. Jacobson, Department of this study (Fig. 1). Three patients had complete ocular
Ophthalmology, Scheie Eye Institute, University of Pennsylvania, 51
N. 39th street, Philadelphia, PA 19104, U.S.A. E-mail: jacobsos@ examinations, Goldmann kinetic perimetry, dark- and
mail.med.upenn.edu light-adapted static threshold perimetry, full ®eld and

0014-4835/02/ $35.00/0 # 2002 Academic Press

738
F IG . 1. Pedigree of the family with spinocerebellar ataxia caused by SCA7 mutation. Filled symbol, affected; open symbol,
unaffected; slashed symbol, deceased member.
focal electroretinography (ERG) and optical coherence
tomography (OCT). Details of previous neurological
and ocular examinations were obtained from medical
records with patient's permission. Informed consent
for all procedures was obtained and the research
procedures were in accordance with institutional
guidelines and the Declaration of Helsinki.
Molecular Analyses
DNA from family members was analyzed for SCA7
gene mutation according to published methods
(Stevanin et al., 1998).
Evaluation of Phenotype
Psychophysics. Dark- and light-adapted static
threshold perimetry was performed using a modi®ed
automated perimeter to determine rod and L/M (long/
middle wavelength) cone function. Techniques and
normal results have been described previously (Jacob-
son et al., 1986, 1998).
Electroretinography. Full ®eld ERGs were performed
with a standard protocol. Details of the methods and
normal data have been published (Jacobson et al.,
1989, 1998). Rod- and cone-isolated ERG photore-
sponses to high-energy stimuli were recorded and
analyzed with a biochemically based model of photo-
transduction (Cideciyan and Jacobson, 1996; Cideci-
yan et al., 1998; Jacobson et al., 1998). Focal ERGs
were elicited under visualization (hand-held indirect
ophthalmoscope; LKC Technologies, Guithersburg,
MD, U.S.A) with a 48, 30 Hz ¯ickering stimulus
projected within a steady brighter (30 cd m 2)
annulus (Jacobson et al., 1997).
Optical coherence tomography. Multiple horizontal
and vertical cross-sectional images of the central retina
were obtained with a commercially available OCT
instrument (Zeiss Humphrey Instruments, Dublin, CA,
U.S.A.) using methods described elsewhere (Huang
et al., 1991; Hee et al., 1995; Pulia®to et al., 1996;
Jacobson et al., 1998, 2000). Scans were 4.5 and
2 mm long. Each scan, formed by a series of
T. S . A L E M A N E T A L .
longitudinal re¯ectivity pro®les (LRPs), was analyzed
using a customized computer program (Huang et al.,
1998, 2000; Jacobson et al., 1998, 2000). Retinal
thickness at ®xation and in the paracentral retina was
measured with a computer algorithm (Hee et al., 1996)
and averaged results from at least three central scans
were used for the data analyses.
3. Results
Clinical and Molecular Characteristics
A pedigree of English±Irish origin had members
with progressive ataxia and visual loss (Fig. 1). Three
members were deceased (I-1, II-4, II-5); the three
living members (II-1, III-1, III-2) were studied for
retinal phenotype. Molecular analysis demonstrated
an abnormally expanded CAG repeat within the SCA7
gene (45±47 repeats; Table I) present in one allele
in all three patients. Table I provides some clinical
characteristics in Patients II-1, III-1 and III-2 and
further historical details are given as follows.
Patient II-1 had progressive gait unsteadiness and
gradual loss of visual acuity (VA) beginning at about
47 years of age. Visual acuities at this age were 20/
100 in the right eye (RE) and 20/40 in the left eye (LE);
ophthalmoscopic examination was normal. Neuro-
logical examination showed positive tandem walking
and brisk tendon re¯exes. At age 55, VA declined to
20/400 both eyes (BE) and there was ophthalmoplegia
and cerebellar ataxia. There was an ophthalmoscopi-
cally normal-appearing macula, but small RPE win-
dow defects on ¯uorescein angiography were noted
in both maculae. At age 66, there was pigmentary
maculopathy and temporal pallor of the optic nerve;
VA was worse than 20/400. By age 69, there was
central geographic atrophy.
Patient III-1 reported some dif®culties with reading
small print at age 35. VA was 20/30 in the RE and
20/40 in the LE. There was subtle pigment mottling in
the macula. No signs of cerebellar ataxia were present.
At age 37, VA had decreased in the RE to 20/70 and in
the LE to 20/80; ophthalmoscopic examination was
unchanged. Neurological examination showed some
instability in tandem walking. At age 39, the clinical
S C A 7 I S A C O N E -- R O D DY S T R O P H Y 739

TABLE I
Clinical and molecular characteristics of the patients

Patient Number Age at onset of Visual acuity{ Refraction

number/ of CAG neurological Age at ocular
sex repeats symptoms (years) exam* (years) RE LE RE LE

II-1/F 45 47 66 3/200 6/200 3.00±0.50  90 1.50±0.50  90

III-1/F 45 37 39 20/100 20/100 0.50 sphere 0.25 sphere
III-2/F 47 32 44 20/200 20/200 1.00±0.50  90 1.00±0.50  90

RE, right eye; LE, left eye.

*When retinal phenotype studies were performed.
{
Best corrected.

®ndings were unchanged; VA had further declined to
20/100.
Patient III-2's earliest symptoms were at age 32;
the patient described decreased clarity of vision and
some dif®culty with walking. VA was 20/25 in BE. No
abnormalities were noted on ophthalmoscopy. At age
34, the patient showed brisk tendon re¯exes in the
lower extremities. At age 44, VA was reduced to 20/
200 in BE. On ophthalmoscopy, central RPE changes
were present.
Peripheral Retinal Function
Standard ERGs from Patients III-1, III-2 and II-1,
with similar CAG repeat sizes, are compared with
those of a normal subject (Fig. 2). The patients show a

F IG . 2. Standard ERGs in patients with SCA7 mutations. (A) Representative rod, mixed cone±rod and cone ERGs from the
three patients compared with a normal subject. (B) ERG parameters in the three patients (symbols). Rectangles are + 2S.D. from
mean normal amplitude (A) and timing (T) parameters.
740 T. S . A L E M A N E T A L .

spectrum of severity of dysfunction. Patient III-1, with

a 2-year neurological disease duration, has borderline
rod- and cone-speci®c ERGs with a mixed cone±rod
ERG that falls within normal limits. Patients III-2 and
II-1 have reduced ERG amplitudes and delayed timing
after 12 and 19 years of duration of neurological
disease, respectively. Rod b-wave maximum amplitude
(Vmax) was reduced in Patients II-1 and III-2; semi-
saturation constant (K) was abnormal in Patient II-1
(data not shown). Cone ERG (1 Hz) b-wave shape
in Patients III-2 and II-1 was not simply a reduced
amplitude version of normal. Oscillatory potentials,
analyzed by digitally ®ltering (100±300 Hz) the light-
adapted 1 Hz responses, are abnormally reduced in
amplitude in Patient III-1 and absent in Patients III-2
and II-1. The cone response to 29 Hz is also abnormal
in waveform shape, with two discernible peaks.
Quantitative analysis of amplitude and timing of the
responses are shown relative to normal (Fig. 2(B)).
Photoresponses evoked by high-energy stimuli in
the dark-adapted state are shown for a representative
normal subject and the three patients (Fig. 3(A)).
A phototransduction activation model is ®t to the
leading edge of photoresponses and maximum ampli-
tude and sensitivity parameters of rod and cone
components of the model are estimated. Patient III-1,
with the mildest disease expression, had rod (298 mV)
and cone (61 mV) maximum amplitudes reduced
to approximately 70 % of mean normal (rod mean +
S.D. ˆ 451 + 51 mV; cone mean + S.D. ˆ 81 + 12
mV); Patient III-2, with more advanced disease,
showed further reductions in rod (169 mV) and cone
(41 mV) maximum amplitudes. Patient II-1, the most
severely affected individual, had photoresponse ampli-
tudes that were approximately 20 % of mean normal.
Rod and cone sensitivities were normal for Patients III-
1 and III-2, but were abnormally reduced in Patient II- F IG . 3. ERG photoresponses in SCA7. (A) Dark-adapted
1. The relationship between rod and cone maximum ERG photoresponses (thin noisy lines) in a representative
amplitudes indicated near-equal rod and cone photo- normal subject and the three patients evoked by a red
(middle trace, 2.5 log scot td sec or 3.6 log phot td sec) and
receptor dysfunction (Fig. 3(B)). In this small sample,
two blue (top and bottom traces, 4.6 and 2.3 log scot td sec,
the degree of retinal dysfunction appeared to be related respectively) ¯ash stimuli. Leading edges of the waveforms
to length of neurological disease duration. were ®tted with a phototransduction activation model (thick
Peripheral extent of visual ®eld with kinetic peri- smooth lines) that is the sum of rod and cone components.
metry using the V-4e target was normal in Patients Cone component is also explicitly shown (thick dashed
lines). (B) Plot of cone vs rod maximum amplitude,
III-1 and III-2. Patient II-1 had slightly reduced
normalized by respective mean normal values. Hypothetical
peripheral extent of ®eld; there was also a 58 diameter loci of equal rod and cone reduction lie on a diagonal, the
central scotoma. With the I-4e target, the expanse of uncertainty of which is de®ned by the dashed lines.
®eld was normal in Patient III-1, but there is decreased
extent of peripheral ®eld in Patients III-2 and II-1
Central Retinal Function and Structure
(Fig. 4). Relative central scotomas are present in
Patients III-2 and II-1. Light- and dark-adapted static Light- and dark-adapted sensitivities across the
threshold perimetry, displayed as grayscale maps of central 308 of Patient III-1 are shown (Fig. 5(A)).
cone and rod sensitivity loss, shows a spectrum of There is reduced L/M cone sensitivity in the central
severity of dysfunction from mild cone and rod few degrees; rod sensitivity falls within normal limits
sensitivity losses (Patient III-2) to more pronounced except at one locus. Focal cone ERGs were not
dysfunction (Patient II-1). These maps suggest a detectable in this patient; results of three normal
regional variation of disease with apparently greater subjects are shown for comparison (Fig. 5(B)).
abnormalities centrally and peripherally than at In vivo cross-sectional images across the central
midperipheral loci (Fig. 4). retina using OCT in a normal subject and the three
S C A 7 I S A C O N E -- R O D DY S T R O P H Y 741

F IG . 4. Kinetic and static visual ®elds in patients with SCA7 mutations. Kinetic perimetry with V-4e and I-4e targets from one
eye of each patient at the speci®ed ages. Relative scotomas to the I-4e target are displayed in gray; physiological blind spot is
shown in black. Static threshold perimetry using monochromatic stimuli in the light (with a 600 nm stimulus) and dark (with
a 500 nm stimulus) adapted states displayed as gray scale maps of cone and rod sensitivity loss. Scale has 16 levels of gray,
representing 0 (white) to 3 (black) log units. Black square at 128 in the temporal ®eld represents the physiological blind spot. N,
nasal; T, temporal; I, inferior; S, superior ®eld.

patients are shown as pseudocolor scans (Fig. 6, left)
and as foveal and parafoveal LRPs (Fig. 6, right). In
the fovea, there are OCT abnormalities in all three
patients. The pseudocolor image in Patient III-1
shows an altered foveal lamination pattern with an
abnormal re¯ection (Fig. 6, white arrow) interposed
between the re¯ections representing the vitreoretinal
interface and the outer retina. The other two patients
have reduced foveal thickness and Patient II-1 has
increased backscatter sclerad to the outer retinal
re¯ection. Inspection of the scans suggests thinning of
the outer retinal re¯ection across the scanned region
and reduced total retinal thickness outside the fovea
in all three patients.
LRPs reveal further details of the abnormalities in
signal components of the OCTs. The normal LRP at
both foveal and parafoveal loci has two prominent
peaks: one at the vitreoretinal interface and another
in the outer retina, termed the outer retina-choroidal
complex or ORCC. The ORCC is normally double
peaked, but in all patients and at both loci the vitread
peak of the ORCC is reduced or absent. The foveal LRP
in Patient III-1 shows an abnormal re¯ection (black
arrow) and a trough of low re¯ectivity vitread to the
ORCC. Patients III-2 and II-1 do not show this feature
at the fovea but only the single-peaked ORCC and
severely reduced retinal thickness. At the parafoveal
locus, OCT data appear to support a `missing compo-
nent' hypothesis (Jacobson et al., 2000). Speci®cally,
Patient III-1 shows reduction in the vitreal peak of the
ORCC and absence of the valley component vitread to
this peak; Patients III-2 and II-1 show absence of both
LRP features.
Retinal thickness is severely reduced at the fovea in
Patient III-2 (57 mm) and II-1 (29 mm) compared to
normal (mean + S.D. ˆ 166.9 + 14.7 mm, n ˆ 13).
Patient III-1 has normal retinal thickness (168 mm),
a measurement complicated by the abnormal lamin-
ation. At the parafovea, all three patients have reduced
retinal thickness (normal mean + S.D. ˆ 268 +
13 mm, n ˆ 13). Patient III-1 shows the least reduc-
tion in thickness (218 mm); there is greater thinning in
Patient III-2 (168 mm). Patient II-1 is missing almost
one-third of retinal thickness (162 mm).
742
F IG . 5. Central retinal function in an SCA7 patient. (A)
Static perimetric pro®les across the horizontal meridian
(central 608) under light (600 nm stimulus) and dark
(500 nm stimulus) adapted states in Patient III-1. Filled
symbols are patient data; hatching marks physiologic blind
spot; dashed lines are normal limits (+ 2S.D.). (B) Focal cone
ERGs in normal subjects (overlaid) and in Patient III-1.
4. Discussion
Expansion of unstable trinucleotide repeats has been
associated with an increasing number of autosomal
(usually dominant) or X-linked neurodegenerative
disorders. Among the diseases known to share this
molecular mechanism are the Fragile X syndromes,
Huntington disease, myotonic dystrophy, Friedreich
ataxia, dentatorubral-pallidoluysian atrophy, spino-
bulbar muscular atrophy, and several of the spinocer-
ebellar ataxias. Although the diseases have different
clinical features, there is the shared phenomenon of
anticipation (decreasing age of onset and increasing
T. S . A L E M A N E T A L .
severity with successive generations) and this may
be more exaggerated depending on the sex of the
transmitting parent. Two molecular subdivisions of
these disorders have been proposed based on whether
the trinucleotide repeats are located in non-coding
(untranslated) or coding (exonic) sequences of the
respective genes (reviewed in Cummings and Zoghbi
(2000)).
The spinocerebellar ataxias are a group of autosomal
dominant neurological conditions with a complex
clinical and neuropathological nosology (reviewed in
Harding (1982, 1993) and Stevanin, Durr and Brice
(2000)) that has now been clari®ed by molecular
genetics (Subramony and Filla, 2001). One subtype of
dominant spinocerebellar ataxia is associated with
retinal disease; names previously given to this syn-
drome are ADCAII and OPCAIII. Families with this
clinically de®ned entity have now been shown to have
CAG repeats in the coding sequence of the SCA7 gene
on chromosome 3p (David et al., 1998; Del-Favero
et al., 1998; Gouw et al., 1998; Koob et al., 1998;
Stevanin et al., 1998; Giunti et al., 1999).
What is the exact retinal disease expression in
patients with SCA7? In the era preceding molecular
diagnosis, the hereditary spinocerebellar ataxias with
visual disturbances were well investigated. The associ-
ated ocular diseases were most often described as
macular dystrophies and less often as involving the
peripheral retina (reviewed in Harding (1982, 1993)).
If full ®eld ERGs were performed, the results could be
normal or abnormal. If rod and cone selective stimuli
were used in the ERG studies, abnormalities tended to
be more prominent in cone-mediated responses or in
both cone and rod signals (BjoÈrk 1956; Jampel 1961;
de Jong et al., 1980; Hamilton et al., 1990; Neetens
et al., 1990; To et al., 1993). It can be assumed that
many of these families had SCA7.
The present study is of a small family with mole-
cularly proven SCA7; the number of pathological
alleles (45±47 CAGs) in these patients are at the
lower end of repeats observed in some earlier work (for
example, Benton et al., 1998; Einum et al., 2001), but
near the median determined in large numbers of
SCA7 patients (David et al., 1998; Gouw et al., 1998;
Stevanin et al., 2000). The limited number of patients
in this study precludes strong commentary on features
associated with trinucleotide repeat disorders, such as
anticipation and parental origin of the disease allele
(Cummings and Zoghbi, 2000). Repeat length, how-
ever, showed little or no increase between the two
generations, which is consistent with reports that
suggest maternal transmission tends to produce
smaller repeat expansions than paternal transmission
(Benton et al., 1998; David et al., 1998; Gouw et al.,
1998).
We provide evidence through detailed retinal studies
at relatively early disease stages that the retinopathy
can be considered a cone±rod dystrophy. Retina-wide
static perimetry maps of photoreceptor-mediated
S C A 7 I S A C O N E -- R O D DY S T R O P H Y 743

F IG . 6. In vivo cross-sectional retinal images with OCT in a representative normal subject and three patients. Left: pseudocolor
images across central 158 of retina. Scale represents re¯ectivity: red and white are high re¯ectivity; blue and black are low
re¯ectivity. T, temporal retina; N, nasal retina. Right: OCTs are analyzed with LRPs at two locations, the fovea and 48 in parafoveal
temporal retina. Raw scans are in gray for normals and black for patients. For the parafoveal location, the patient LRPs were `cut'
and overlaid on the normal LRP; alignment was with the vitreoretinal interface and the sclerad peak of the ORCC.

function showed the pattern seen commonly in cone±
rod dystrophies: greater dysfunction at central and
peripheral retinal regions than in the midperiphery
(Yagasaki and Jacobson, 1989). The photoreceptor
was found to be a site of disease by ERG photoresponse
recordings. Relatively equal loss of cone and rod
ERG responses, as we found in the SCA7 patients, is a
feature of many cone±rod dystrophies (for example,
Yagasaki and Jacobson, 1989; Birch and Anderson,
1990; Szlyk et al., 1993; Jacobson et al., 1998). The
abnormality in oscillatory potentials suggests inner
retinal dysfunction; whether this is a primary disease
site or secondary to the photoreceptor disease is
uncertain. Another recent study of retinal function
in molecularly proven SCA7 disease affecting two
Japanese families emphasized the macular degener-
ation in the patients; a cone worse than rod ERG
abnormality in one subject was found (Abe et al.,
2000).
Cross-sectional retinal imaging with OCT provided
an in vivo view of histopathology of SCA7 retinopathy.
At the earliest stage studied, represented by Patient III-
1, the unusual foveal OCT abnormality may be the
equivalent of a subretinal deposit described previously
in postmortem retinal tissue from a young patient with
OPCAIII (To et al., 1993). OCT of the fovea at later
disease stages was consistent with foveal cone and
Henle's ®ber layer losses and abnormalities of the
retinal pigment epithelium. The paracentral retina of
Patients III-2 and II-1 showed reduced retinal thick-
ness. LRP analyses of these OCTs suggested loss of both
outer and inner retinal elements. These interpretations
are based on previous histological-OCT comparisons in
avian and mammalian retinas (Huang et al., 1998,
2000). Postmortem histopathological studies of
patients with ADCAII or OPCAIII have shown both
outer and inner retinal abnormalities (Ryan et al.,
1975; Traboulsi et al., 1988; Martin et al., 1994).
Mechanisms of the retinal disease in SCA7 have
recently been explored in transgenic mouse models
(Yvert et al., 2000; La Spada et al., 2001). In a study
using the rhodopsin promoter to generate rod photo-
receptor overexpression of human ataxin-7, mice with
a (CAG)90 repeat expansion developed outer retinal
degeneration and ERG photoreceptor (a-wave)
abnormalities (Yvert et al., 2000). An unexpected
result was the extension of disease beyond the
photoreceptors to inner retinal neurons, attributed to
transneuronal degeneration. As has been noted in
human pathological tissue from patients with SCA7
(Mauger et al., 1999; Cancel et al., 2000; Einum et al.,
2001), there were ataxin-7 immunoreactive intra-
nuclear inclusions. What role these intranuclear
inclusions or ataxin-7 staining in the cytoplasm of
neurons play in disease pathogenesis is not clear
(Cancel et al., 2000; Cummings and Zoghbi, 2000).
A more recent study generated transgenic mice by
inserting the ataxin-7 coding region into the murine
prion promoter with the goal of achieving more
generalized neuronal expression (La Spada et al.,
2001). Mice with a (CAG)92 insertion showed patchy
retinal degeneration thought to be mainly affecting
cone photoreceptors. ERGs in mild and moderate
phenotypes showed both cone and rod a- and b-wave
abnormalities. The most severe phenotype had non-
detectable signals, although histopathology was not
744
profoundly altered. Nuclear inclusions of ataxin-7 in
the three nuclear layers of the retina were observed. Of
major interest to the present study were in vitro
experiments that showed interaction of ataxin-7 and
CRX, a transcription factor regulating photoreceptor
genes (Freund et al., 1997; Swain et al., 1997), and
possible interference of ataxin-7 with CRX-regulated
genes.
CRX mutations, like SCA7, can cause a cone±rod
dystrophy phenotype in man (Jacobson et al., 1998).
The hypothesis that the SCA7 retinal disease may
involve an inappropriate interaction between the
expanded ataxin-7 and CRX leads to similar pheno-
types (La Spada et al., 2001) is worth testing further
with in vivo and in vitro experimental studies and
careful characterization of the human phenotype of
known SCA7 and CRX genotypes. Our experience to
date in small subsets of molecularly de®ned patients is
that there may be similarities in retinal disease
expression in SCA7 and CRX heterozygotes (Jacobson
et al., 1998).
Acknowledgements
This work was supported by National Institutes of Health
(EY-05627; EY-13203); Foundation Fighting Blindness,
Inc.; Macula Vision Research Foundation; Association
FrancËaise Retinitis Pigmentosa; F.M. Kirby Foundation;
The Macular Disease Foundation; the Institut de la SanteÂ
et de la Recherche MeÂdicale (France) and the Mackall Trust.
S.G.J. is an RPB Senior Scienti®c Investigator; A.V.C. is an
RPB Special Scholar. We thank Mr D. Hanna, Ms L.
Gardner, Dr Y. Huang, Mr D. Marks, Ms K. Mejia, Dr J.
Huang, Ms R.-J. Chen and Ms E. Dale for help with data
collection and analyses.
References
Abe, T., Tsuda, T., Yoshida, M., Wada, Y., Kano, T.,
Itoyama, Y. and Tamai, M. (2000). Macular degener-
ation associated with aberrant expansion of trinucleo-
tide repeat of the SCA7 gene in 2 Japanese families.
Arch. Ophthalmol. 118, 1415±21.
Benton, C. S., de Silva, R., Rutledge, S. L., Bohlega, S.,
Ashizawa, T. and Zoghbi, H. Y. (1998). Molecular and
clinical studies in SCA-7 de®ne a broad clinical
spectrum and the infantile phenotype. Neurology 51,
1081±6.
Birch, D. G. and Anderson, J. L. (1990). Rod visual ®elds in
cone±rod degeneration: comparisons to retinitis pig-
mentosa. Invest. Opthalmol. Vis. Sci. 31, 2288±99.
BjoÈrk, A., Lindblom, U. and Wadensten, L. (1956). Retinal
degeneration in hereditary ataxia. J. Neurol. Neurosurg.
Psychiatry 19, 186±93.
Cancel, G., Duyckaerts, C., Holmberg, M., Zander, C.,
Yvert, G., Lebre, A.-S., Ruberg, M., Faucheux, B., Agid,
Y., Hirsch, E. and Brice, A. (2000). Distribution of
ataxin-7 in normal human brain and retina. Brain
123, 2519±30.
Cideciyan, A. V., Hood, D. C., Huang, Y., Banin, E., Li, Z.-Y.,
Stone, E. M., Milam, A. H. and Jacobson, S. G. (1998).
Disease sequence from mutant rhodopsin allele to rod
and cone photoreceptor degeneration in man. Proc. Nat.
Acad. Sci. U.S.A. 95, 7103±8.
T. S . A L E M A N E T A L .
Cideciyan, A. V. and Jacobson, S. G. (1996). An alternative
phototransduction model for human rod and cone ERG
a-waves: normal parameters and variation with age.
Vis. Res. 36, 2609±21.
Cummings, C. J. and Zoghbi, H. Y. (2000). Trinucleotide
repeats: mechanisms and pathophysiology. Ann. Rev.
Genomics Hum. Genet. 1, 281±328.
David, G., Abbas, N., Stevanin, G., DuÈrr, A., Yvert, G., Cancel,
G., Weber, C., Imbert, G., Saudou, F., Antoniou, E.,
Drabkin, H., Gemmile, R., Giunti, P., Benomar, A., Wood,
N., Ruberg, M., Agid, Y., Mandel, J.-L. and Brice, A.
(1997). Cloning of the SCA7 gene reveals a highly
unstable CAG repeat expansion. Nat. Genet. 17, 65±70.
David, G., DuÈrr, A., Stevanin, G., Cancel, G., Abbas, N.,
Benomar, A., Belal, S., Lebre, A.-S., Abada-Bendib, M.,
Grid, D., Holmberg, M., Yahyaoui, M., Hentati, F.,
Chkili, T., Agid, Y. and Brice, A. (1998). Molecular and
clinical correlations in autosomal dominant cerebellar
ataxia with progressive macular dystrophy (SCA7).
Hum. Mol. Genet. 7, 165±70.
de Jong, P. T. V. M., de Jong, J. G. Y., de Jong-Ten Doeschate,
J. M. M. and Delleman, J. W. (1980). Olivopontocer-
ebellar atrophy with visual disturbances. An ophthal-
mologic investigation into four generations.
Ophthalmology 87, 793±804.
Del-Favero, J., Krols, L. and Michalik, A. et al. (1998).
Molecular genetic analysis of autosomal dominant
cerebellar ataxia with retinal degeneration (ADCA
type II) caused by CAG triplet repeat expansion. Hum.
Mol. Genet. 7, 177±86.
Einum, D. D., Townsend, J. J., Ptacek, L. J. and Fu, Y. H.
(2001). Ataxin-7 expression analysis in controls and
spinocerebellar ataxia type 7 patients. Neurogenetics 3,
83±90.
Freund, C. L., Gregory-Evans, C. Y., Furukawa, T.,
Papaioannou, M., Looser, J., Ploder, L., Bellingham, J.,
Ng, D., Herbrick, J.- A. S., Duncan, A., Scherer, S. W.,
Tsui, L.-C., Loutradis-Anagnostou, A., Jacobson, S. G.,
Cepko, C. L., Bhattacharya, S. S. and McInnes, R. R.
(1997). Cone±rod dystrophy due to mutations in a
novel photoreceptor-speci®c homeobox gene (CRX)
essential for maintenance of the photoreceptor. Cell
91, 1±20.
Giunti, P., Stevanin, G., Worth, P. F., David, G., Brice, A. and
Wood, N. W. (1999). Molecular and clinical study of 18
families with ADCA type II: evidence for genetic
heterogeneity and de novo mutation. Am. J. Hum.
Genet. 64, 1594±603.
Gouw, L. G., Castaneda, M A and McKenna, C K. et al.
(1998). Analysis of the dynamic mutation in the SCA7
gene shows marked parental effects on CAG repeat
transmission. Hum. Mol. Genet. 7, 525±32.
Hamilton, S. R., Chatrian, G.-E., Mills, R. P., Kalina, R. E.
and Bird, T. D. (1990). Cone dysfunction in a subgroup
of patients with autosomal dominant cerebellar ataxia.
Arch. Ophthalmol. 108, 551±6.
Harding, A. E. (1982). The clinical features and classi®-
cation of the late onset autosomal dominant cerebellar
ataxias. Brain 105, 1±28.
Harding, A. E. (1993). Clinical features and classi®cation of
inherited ataxias. Adv. Neurol. 61, 1±14.
Hee, M. R., Baumal, C. R., Pulia®to, C. A., Duker, J. S.,
Reichel, E., Wilkins, J. R., Coker, J. G. and Schuman, J. S.
(1996). Optical coherence tomography of age-related
macular degeneration and choroidal neovasculariza-
tion. Ophthalmology 103, 1260±70.
Hee, M. R., Izatt, J. A., Swanson, E. A., Huang, D.,
Schuman, J. S., Lin, C. P., Pulia®to, C. A. and Fujimoto,
J. G. (1995). Optical coherence tomography of the
human retina. Arch. Ophthalmol. 113, 325±32.
S C A 7 I S A C O N E -- R O D DY S T R O P H Y
Holmberg, M., Duyckaerts, C., DuÈrr, A., Cancel, G.,
Gour®nkel-An, I., Damier, P., Faucheux, B., Trottier, Y.,
Hirsch, E. C., Agid, Y. and Brice, A. (1998). Spinocer-
ebellar ataxia type 7 (SCA7): a neurodegenerative
disorder with neuronal intranuclear inclusions. Hum.
Mol. Genet. 7, 913±8.
Huang, Y., Cideciyan, A. V., Aleman, T. S., Banin, E., Huang,
J., Syed, N. A., Petters, R. M., Wong, F., Milam, A. H. and
Jacobson, S. G. (2000). Optical coherence tomography
(OCT) abnormalities in rhodopsin mutant transgenic
swine with retinal degeneration. Exp. Eye Res. 70,
247±51.
Huang, Y., Cideciyan, A. V., Papastergiou, G. I., Banin, E.,
Semple-Rowland, S. L., Milam, A. H. and Jacobson, S. G.
(1998). Relation of optical coherence tomography to
microanatomy in normal and rd chickens. Invest.
Ophthalmol. Vis. Sci. 39, 2405±16.
Huang, D., Swanson, E. A., Lin, C. P., Schuman, J. S., Stinson,
W. G., Chang, W., Hee, M. R., Flotte, T., Gregory, K.,
Pulia®to, C. A. and Fujimoto, J. G. (1991). Optical
coherence tomography. Science 254, 1178±81.
Jacobson, S. G., Buraczynska, M., Milam, A. H., Chen, C.,
Jarvalainen, M., Fujita, R., Wu, W., Huang, Y.,
Cideciyan, A. V. and Swaroop, A. (1997). Disease
expression in X-linked retinitis pigmentosa caused by a
putative null mutation in the RPGR gene. Invest.
Ophthalmol. Vis. Sci. 38, 1983±97.
Jacobson, S. G., Cideciyan, A. V., Huang, Y., Hanna, D. B.,
Freund, C. L., Affatigato, L., Carr, R. E., Zack, D. J.,
Stone, E. M. and McInnes, R. (1998). Retinal degener-
ations with truncation mutations in the cone±rod
homeobox (CRX) gene. Invest. Ophthalmol. Vis. Sci. 39,
2417±26.
Jacobson, S. G., Cideciyan, A. V., Iannaccone, A., Weleber,
R. G., Fishman, G. A., Maguire, A. M., Affatigato, L. M.,
Bennett, J., Pierce, E. A., Danciger, M., Farber, D. B. and
Stone, E. M. (2000). Disease expression of RP1
mutations causing autosomal dominant retinitis pig-
mentosa. Invest. Ophthalmol. Vis. Sci. 41, 1898±908.
Jacobson, S. G., Voigt, W. J., Parel, J. M., ApaÂthy, P. P.,
Nghiem-Phu, L., Myers, S. W. and Patella, V. M. (1986).
Automated light- and dark-adapted perimetry for
evaluating retinitis pigmentosa. Ophthalmology 93,
1604±11.
Jacobson, S. G., Yagasaki, K., Feuer, W. J. and Roman, A. J.
(1989). Interocular asymmetry of visual function in
heterozygotes of X-linked retinitis pigmentosa. Exp. Eye
Res. 48, 679±91.
Jampel, R. S., Okazaki, H. and Bernstein, H. (1961).
Ophthalmoplegia and retinal degeneration associated
with spinocerebellar ataxia. Arch. Ophthalmol. 66,
247±59.
Kaytor, M. D., Duvick, L. A., Skinner, P. J., Koob, M. D.,
Ranum, L. P. W. and Orr, H. T. (1999). Nuclear
localization of the spinocerebellar ataxia type 7 protein
ataxin-7. Hum. Mol. Genet. 8, 1657±64.
Koob, M. D., Benzow, K. A., Bird, T. D., Day, J. W., Moseley,
M. L. and Ranum, L. P. (1998). Rapid cloning of
expanded trinucleotide repeat sequences from genomic
DNA. Nat. Genet. 18, 72±5.
La Spada, A. R., Fu, Y.-H., Sopher, B. L., Libby, R. T.,
Wang, X., Li, L. Y., Einum, D. D., Huang, J., Possin,
D. E., Smith, A. C., Martinez, R. A., Koszdin, K. L.,
Treuting, P. M., Ware, C. B., Hurley, J. B., PtaÂcÏek, L. J.
and Chen, S. (2001). Polyglutamine-expanded ataxin-
7 antagonizes CRX function and induces cone±rod
dystrophy in a mouse model of SCA7. Neuron 31,
913±27.
745
Lindenberg, K. S., Yvert, G., Muller, K. and Landwhermeyer,
G. B. (2000). Expression analysis of ataxin-7 mRNA
and protein in human brain: evidence for a widespread
distribution and focal protein accumulation. Brain
Pathol. 10, 385±94.
Martin, J.-J., Van Regemorter, N., Krols, L., Brucher, J. M.,
de Barsy, T., Szliwowski, H., Evrard, P., Ceuterick, C.,
Tassignon, M. J., Smet-Dieleman, H., Hayez-Delatte, F.,
Willems, P. J. and Van Broeckhoven, C. (1994). On an
autosomal dominant form of retinal±cerebellar degener-
ation: an autopsy study of ®ve patients in one family.
Acta Neuropathol. 88, 277±86.
Mauger, C., Del-Favero, J., Ceuterick, C., LuÈbke, U.,
Van Broeckhoven, C. and Martin, J.-J. (1999). Identi®-
cation and localization of ataxin-7 in brain and retina of
a patient with cerebellar ataxia type II using anti-
peptide antibody. Mol. Brain Res. 74, 35±43.
Mushegian, A. R., Vishnivetskiy, S. A. and Gurevich, V. V.
(2000). Conserved phosphoprotein interaction motif is
functionally interchangeable between ataxin-7 and
arrestins. Biochemistry 39, 6809±13.
Neetens, A., Martin, J.-J., Libert, J. and van den Ende, P.
(1990). Autosomal dominant cone dystrophy±cerebel-
lar atrophy (ADCoCA) (modi®ed ADCA Harding II.
Neuro-ophthalmology 10, 261±75.
Pulia®to, C. A., Hee, M. R., Schuman, J. S. and Fujimoto, J. G.
(1996). Optical Coherence Tomography of Ocular Diseases.
Slack Inc.: Thorofare, NJ.
Ryan, S. J., Jr., Knox, D. L., Green, W. R. and Konigsmark,
B. W. (1975). Olivopontocerebellar degeneration. Clin-
icopathologic correlation of the associated retinopathy.
Arch. Ophthalmol. 93, 169±72.
Stevanin, G., Durr, A. and Brice, A. (2000). Clinical and
molecular advances in autosomal dominant cerebellar
ataxias: from genotype to phenotype and physiopathol-
ogy. Eur. J. Hum. Genet. 8, 4±18.
Stevanin, G., Giunti, P., Belal, G. D., Durr, A., Ruberg, M.,
Wood, N., and Brice, A. (1998). De novo expansion of
intermediate alleles in spinocerebellar ataxia 7. Hum.
Mol. Genet. 7, 1809±13.
Subramony, S. H. and Filla, A. (2001). Autosomal dominant
spinocerebellar ataxias ad in®nitum?. Neurology 56,
287±9.
Swain, P. K., Chen, S., Wang, Q.-L., Affatigato, L. M., Coats,
C. L., Brady, K. D., Fishman, G. A., Jacobson, S. G., Stone,
E. M., Sieving, P. A. and Zack, D. J. (1997). Mutations in
the cone±rod homeobox gene are associated with the
cone±rod dystrophy photoreceptor degeneration. Neuron
19, 1329±36.
Szlyk, J. P., Fishman, G. A., Alexander, K. R., Peachey, N. S.
and Derlacki, D. J. (1993). Clinical subtypes of cone±rod
dystrophy. Arch. Ophthalmol. 111, 781±8.
To, K. W., Adamian, M., Jakobiec, F. A. and Berson, E. L.
(1993). Olivopontocerebellar atrophy with retinal
degeneration. Ophthalmology 100, 15±23.
Traboulsi, E. I., Maumenee, I. H., Green, W. R., Freimer, M. L.
and Moser, H. (1988). Olivopontocerebellar atrophy
with retinal degeneration. A clinical and ocular histo-
pathologic study. Arch. Ophthalmol. 106, 801±6.
Yagasaki, K. and Jacobson, S. G. (1989). Cone±rod
dystrophy. Phenotypic diversity by retinal function
testing. Arch. Ophthalmol. 107, 701±8.
Yvert, G., Lindenberg, K. S., Picaud, S., Landwehrmeyer,
G. B., Sahel, J. A. and Mandel, J. L. (2000). Expanded
polyglutamines induce neurodegeneration and trans-
neuronal alterations in cerebellum and retina of SCA7
transgenic mice. Hum. Mol. Genet. 9, 2491±506.