Long-Term Course and Mutational
Spectrum of spatacsin-Linked Spastic
Paraplegia
Ute Hehr, MD,1 Peter Bauer, MD,2 Beate Winner, MD,3 Rebecca Schule, MD,4 Akguen Olmez, MD,5
Wolfgang Koehler, MD,6 Goekhan Uyanik, MD,3 Anna Engel, MSc,2 Daniela Lenz, MSc,2
Andrea Seibel, MSc,4 Andreas Hehr, PhD,1 Sonja Ploetz, MSc,3 Josep Gamez, MD,7 Arndt Rolfs, MD,8
Joachim Weis, MD,9 Thomas M. Ringer, MD,10 Michael Bonin, PhD,2 Gerhard Schuierer, MD,11
Joerg Marienhagen, MD,12 Ulrich Bogdahn, MD,3 Bernhard H. F. Weber, PhD,1 Haluk Topaloglu, MD,5
Ludger Schols, MD,4 Olaf Riess, MD,2 and Juergen Winkler, MD3

Objective: Hereditary spastic paraplegias (HSPs) comprise a heterogeneous group of neurodegenerative disorders resulting in
progressive spasticity of the lower limbs. One form of autosomal recessive hereditary spastic paraplegia (ARHSP) with thin
corpus callosum (TCC) was linked to chromosomal region 15q13-21 (SPG11) and associated with mutations in the spatacsin
gene. We assessed the long-term course and the mutational spectrum of spatacsin-associated ARHSP with TCC.
Methods: Neurological examination, cerebral magnetic resonance imaging (MRI), 18fluorodeoxyglucose positron emission tomography (PET), nerve biopsy, linkage and mutation analysis are presented.
Results: Spastic paraplegia in patients with spatacsin mutations (n ⫽ 20) developed during the second decade of life. The Spastic
Paraplegia Rating Scale (SPRS) showed severely compromised walking between the second and third decades of life (mean SPRS
score, ⬎30). Impaired cognitive function was associated with severe atrophy of the frontoparietal cortex, TCC, and bilateral
periventricular white matter lesions. Progressive cortical and thalamic hypometabolism in the 18fluorodeoxyglucose PET was
observed. Sural nerve biopsy showed a loss of unmyelinated nerve fibers and accumulation of intraaxonal pleomorphic membranous material. Mutational analysis of spatacsin demonstrated six novel and one previously reported frameshift mutation and
two novel nonsense mutations. Furthermore, we report the first two splice mutations to be associated with SPG11.
Interpretation: We demonstrate that not only frameshift and nonsense mutations but also splice mutations result in SPG11.
Mutations are distributed throughout the spatacsin gene and emerge as major cause for ARHSP with TCC associated with severe
motor and cognitive impairment. The clinical phenotype and the ultrastructural analysis suggest a disturbed axonal transport of
long projecting neurons.
Ann Neurol 2007;62:656 – 665

Hereditary spastic paraplegia (HSP) genetically comprises one of the most heterogeneous groups of monogenetic neurodegenerative disorders. The mode of inheritance is autosomal dominant, autosomal recessive,
or X-linked. To date, 36 distinct HSP loci designated
SPG (spastic paraplegia) have been assigned with 15
disease-associated genes identified so far.1 The diversity
of HSP loci indicates a complex genetic control for the

function of motor neurons and their axonal projections. The HSP genes encode proteins with diverse cellular functions associated with mitochondria (eg,
SPG7, SPG13), axonal transport (eg, SPG3A, SPG4,
SPG10), myelinization (eg, SPG2), or development of
the corticospinal tract (CST; eg, SPG1).1,2 An axonal
dying back of long-projecting motor neurons is one
potential mechanism underlying the CST degenera-

From the 1Department of Human Genetics, University of Regensburg, Regensburg; 2Department of Medical Genetics, University of
Tuebingen, Tuebingen; 3Department of Neurology, University of
Regensburg, Regensburg; 4Research Division for Clinical Neurogenetics, Centre of Neurology and Hertie-Institute for Clinical Brain
Research, University of Tuebingen, Tuebingen, Germany; 5Department of Pediatric Neurology, Hacettepe University, Ankara, Turkey; 6Department of Neurology, Fachkrankenhaus Hubertusburg,
Wermsdorf, Germany; 7Department of Neurology, Hospital General Vall d’Hebron, Barcelona, Spain; 8Department of Neurology,
University of Rostock, Rostock; 9Institute of Neuropathology,
Rheinisch-Westfälische Technische Hochschule of Aachen, Aachen;
10
Department of Neurology, University of Jena, Jena; 11Institute of
Neuroradiology, Bezirksklinikum Regensburg; and 12Department of

Nuclear Medicine, University of Regensburg, Regensburg, Germany.

656

Received Jul 20, 2007, and in revised form Oct 9. Accepted for
publication Oct 26, 2007.
U.H., P.B., B.W., and R.S. contributed equally to this work.
Published online Dec 7, 2007, in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.21310
Address correspondence to Dr Winkler, Department of Neurology,
University of Regensburg, Universitaetsstrasse 84, D-93053 Regensburg, Germany. E-mail: juergen.winkler@klinik.uni-regensburg.de

© 2007 American Neurological Association
Published by Wiley-Liss, Inc., through Wiley Subscription Services

tion. Neuropathological analysis of the spinal cord,
rarely performed in HSP, showed axonal degeneration
most marked in the distal terminals of CST and fasciculus gracilis projections.3
Recently, progress has been made to establish a rating scale for HSP (Spastic Paraplegia Rating Scale
[SPRS]) as an important clinical tool to assess severity
and natural course of HSP.4 Clinically, pure and complicated forms of HSP are distinguished.5 Pure forms
present with clinical features of pyramidal motor neuron dysfunction such as spastic weakness of the lower
limbs and are often accompanied by urinary urgency
and subtle dorsal column impairment.1 However, clinical and neuroimaging parameters currently fail to distinguish between the different genetic forms of pure
HSP. In contrast, complicated forms are characterized
by additional nonpyramidal signs and symptoms, and
are associated with additional structural abnormalities
of the central nervous system (CNS) such as thin corpus callosum (TCC), white matter (WM) lesions, and
cortical and/or cerebellar atrophy. In addition, neurological symptoms such as neuropathy, amyotrophy,
ataxia, extrapyramidal motor signs, epilepsy, or cognitive
deficits are observed. Furthermore, other organs may
also be affected (eg, cataracts, skeletal abnormalities).1
A clinically distinct form of autosomal recessive hereditary spastic paraplegia (ARHSP) with TCC has
been linked to the long arm of chromosome 15 at
15q13-15 and designated SPG11.6 – 8 SPG11 is characterized by progressive spasticity of the lower limbs,
TCC, cognitive impairment, and severe neuropathy.9,10 We have recently linked four large consanguineous Turkish families to this region.11 Based on the
presence of two recombination events, we refined the
SPG11 critical region to a 2.93 cM interval with a
maximum LOD score of 11.84.11 Within this candidate interval, KIAA1840 was recently identified as the
spatacsin gene associated with SPG11 and appears to be
the most frequent cause for ARHSP with TCC.12
By sequencing the entire coding region of the spatacsin gene, we identified 1 previously reported and 10

novel spatacsin mutations in 3 consanguineous pedigrees from Turkey, 1 from Saudi Arabia, and 1 from
Germany, linked to SPG11, and in 4 sporadic patients
from Germany and Spain. Our data demonstrate that
frameshift and nonsense mutations, as well as splice
mutations, result in the SPG11 phenotype. Furthermore, we assess the phenotype and natural long-term
course of these spatacsin-associated HSP patients.
Patients and Methods
Minimal inclusion criteria for this patient cohort were: (1)
family history compatible with an autosomal recessive inheritance, (2) progressive spasticity of the lower limbs, (3) TCC,
and (4) cognitive impairment in at least one index patient of
each family. Detailed neurological examination of 18 patients from 9 families with suspected SPG11 was performed
applying the SPRS. This scale measures disease severity consisting of a 13-item scale to rate functional impairment.4
Some of these patients were continuously treated for up to
10 years. Diagnostic workup included comprehensive laboratory analysis (cerebrospinal fluid for the presence of oligoclonal bands, blood testing for electrolytes, hormones, very
long chain fatty acids, fats, vitamins, and infectious diseases)
and screening tests for mitochondrial dysfunction (lactate
and pyruvate before and after glucose loading13), and tandem
mass spectra analysis of acylcarnitine (C16:0⫹C18:1/C2 ratio14). Neuroradiological evaluation included long-term
follow-up of cerebral and spinal magnetic resonance imaging
(MRI) in three patients. A single-voxel 1H-proton magnetic
resonance spectroscopy (TE/TR 272/2,000 milliseconds
from frontal and parietooccipital brain regions, centrum
semiovale, and internal capsule) was performed in one patient. Long-term follow-up 18fluorodeoxyglucose positron
emission tomography (FDG-PET) images were obtained for
three patients. Histological and ultrastructural evaluation was
performed for one sural nerve biopsy.
Peripheral blood samples for the genetic analysis were obtained and clinical data included after informed consent of
the patients or parents, respectively, according to bioethical
laws for research on human subjects. DNA was extracted
from whole EDTA blood using standard protocols. The coding exons and the intron exon boundaries of 30 genes
(CHST14, DLL4, GANC, GCHFR, GCP4, CTDSPL2, IT-

Fig 1. (A) Linkage of the consanguineous Saudi-Arabian family (F5) to SPG11 and (B) reverse sequence of exon 10 of index proband P5_3 with homozygous base substitution CGA⬎TGA, resulting in the nonsense mutation R651X.

Hehr et al: Genotype-Phenotype of SPG11

657

Table 1. Summary of Spatacsin Mutations
Location

Familya
IDOrigin

Zygosity

Exon 1

Mutation acc. to
NM_025137 (cDNA)

Amino Acid
Substitution

Reference

c.118C⬎T

p.Q40X

12

Exon 2

F1G

Heterozygous

c. 267G⬎A

p.W89X

This report

Intron 2

F2T

Homozygous

c.442⫹1g⬎c

splice mutation

This report

c.529_533delATATT

p.I177_I178delfs179X

12

Exon 3
Exon 4

S

F3

Heterozygous

Exon 6
Intron 6
Exon 9
Exon 10

F1G, F4G

p.M245fs246X

12, this report

p.K401fsX

12

Heterozygous

c.1757-2a⬎g

p.E486fs508X

This report

G

Heterozygous

c.1845_1846delGT

p.S616fs618X

This report

SA

Homozygous

F4
F5

Exon 11
Exon 14

c.733_734delAT
c.1203delA

F3S

Heterozygous

Exon 16

c.1951C⬎T

p.R651X

This report

c.2198T⬎G

p.L733X

12

c.2472insT

p.K825fs837X

This report

c.2842_2843insG

p.V948fsX

12

Exon 17

T

F6

Homozygous

c.3076insA

p.R1026fs1029X

This report

Exon 17

F7T

Homozygous

c.3145_3146insCA

p.T1048fs1053X

This report

Exon 30

G

Heterozygous

c.5255delT

p.F1752fs1765X

This report

G

Heterozygous

c.5798delC

p.A1933fs1951X

This report

Exon 30

F8
F9

Exon 31

c.5974C⬎T

p.R1992X

12

Exon 32

c.6100C⬎T

p.R2034X

12

Exon 34

c.6451delG

p.A2151fsX

12

Exon 39

c.7029_7030insT

p.V2344fsX

12

a

Origin: superscript G indicates Germany; superscript T indicates Turkey, superscript S indicates Spain; superscript SA indicates Saudi
Arabia.
cDNA ⫽ complementary DNA.

PKA, LTK, MAP1A, MFAP1, MGA, FRMD5, EXDL1,
NDUFAF1, NUSAP1, OIP5, PLDN, RHOV, TRIM69,
SNAP23, KIAA1840, SPATA5L1, SPTBN5, SQRDL, TYRO3, UBR1, VPS18, ZFP106, ZFYVE19, ZSCAN29, according to the human genome nomenclature database)15
were amplified by polymerase chain reaction (PCR). PCR
products were cycle sequenced using the ABI Prism BigDye
Terminator Cycle Sequencing Kit (version 1.1) and analyzed
on an ABI 3100 Avant sequencer (Applied Biosystems, Foster City, CA). Each mutation identified within the 40 exons
of the KIAA1840 gene and their absence in any of 100 control chromosomes of Caucasian and/or Turkish origin was
confirmed by restriction fragment length polymorphism
when possible, or by sequencing or SNaPshot analysis (Applied Biosystems) of a second independent PCR product.
Likewise, compound heterozygosity and cosegregation of the
mutations with the phenotype in the respective families were
demonstrated. To determine the effects of the splice-site mutation in Patients P1 and P4, we isolated messenger RNA
from peripheral blood (PAXgene) of Patient P1 according to
the manufacturers’ recommendations. First-strand complementary DNA was synthesized using Superscript II (Invitrogen, La Jolla, CA). Subsequently, PCR products were gener-

658

Annals of Neurology

Vol 62

No 6

December 2007

ated with exonic primers 5F and 8R cloned into a pGem
vector (Promega, Madison, WI); eight clones were sequenced.

Results
Mutational Analysis
In addition to previous microsatellite analysis, we confirmed linkage to the refined candidate region of
SPG11 for the consanguineous Saudi-Arabian pedigree
with ARHSP with TCC (Family F5; Fig 1A).9,11 Following a positional candidate gene approach, no
disease-associated mutations were detected in 30 genes
tested from the SPG11 candidate region (see Patients
and Methods). Finally, a total of 11 causal mutations
were identified in the KIAA1840 coding sequence (now
designated spatacsin) in 3 consanguineous pedigrees
from Turkey, 1 family from Saudi Arabia, 1 from Germany, and 4 sporadic patients of German or Spanish
origin (Table 1). All mutations identified represent
truncating mutations and include one previously reported and six novel frameshift mutations, as well as
two novel nonsense mutations. Furthermore, we found

Table 2. Summary of the Clinical Data
Patient Data
Patient
ID

Family
ID(reference)

P1
P2_1
P2_2
P2_3
P3
P4
P5_1
P5_2
P5_3
P6_1
P6_2
P6_3
P6_4
P7_1
P7_2
P7_3
P7_4
P8
P9_1
P9_2

F110
F211

F3
F4
F5

F611

F711

F8
F99

Position in
Pedigree

II-4 fam 3
II-1 fam 3
II-2 fam 3

II-4
II-6
II-7
II-1 fam 2
II-2 fam 2
II-3 fam 2
II-5 fam 2
IV-3 fam 1
IV-10 fam 1
IV-12 fam 1
IV-13 fam 1
II-9
II-10

MRI

SPRS

Sex

Age at the
Time of
Examination
(yr)

Age
at
Onset
(yr)

MRI/CT

TCC

WML

Cortical
Atrophy

Motor
Function
(total)

Spasticity
and
Weakness
(total)

Total
Score

F
F
F
F
M
M
M
F
F
F
F
F
M
M
M
F
F
F
F
F

38
40
22
20
19
33
26
19

31
13
13
15
11
8
17
17

MRI
np
MRI
CT
MRI
MRI
MRI
np

⫹

⫹

⫹

⫹

⫹

⫹

⫹
⫹
⫹

⫹
⫺
⫹

⫹
⫹
⫹

17
24
16
16
13
16
17

7
13
11
10
3
8
11

24
39
29
28
21
24
29

30
28
26
17
31
26
23
21
25
34
29

16
17
13
15
13
12
13
13
23
25
25

MRI
np
np
np
np
np
np
MRI
MRI
MRI
MRI

⫹

⫹

⫹

23
21

12
12

38
36

24
24
21

13
13
12

37
40
36

9
18
16

4
11
10

16
40
28

⫹
⫹
⫹
⫹

⫹
⫹
⫹
⫹

⫹
⫺
⫹
⫹

Pedigrees linked to SPG11: P110; P2, P6, and P711; P99; and P5 (see Fig 1).
MRI ⫽ magnetic resonance imaging; SPRS ⫽ Spastic Paraplegia Rating Scale; CT ⫽ computed tomography; TCC ⫽ thin corpus
callosum; WML ⫽ white matter lesion; np ⫽ not performed.

two different splice mutations (Table 1, Fig 1B) associated with the SPG11 phenotype (Figs 2A–C). Mutations are distributed throughout the entire spatacsin
gene without obvious clustering in mutational hotspots
(see Table 1).
The frameshift mutation c.733_734delAT, previously reported in a French patient, was observed in
heterozygous state in a sporadic patient from Spain
(Patient P3).12 Two German patients of nonconsanguineous parents (Patients P1 and P4) were heterozygous for the novel base substitution at position ⫺2 of
the splice acceptor site of intron 6 (c.1757-2a⬎g; see
Figs 2A, B). Another novel homozygous splice mutation of guanine to cytosine at position ⫹1 of the splice
donor site of intron 2 (c.442⫹1g⬎c) was identified in
a consanguineous Turkish family (Family F2; see Fig
2C). For both splice mutations, the wild-type nucleotide at this position is highly conserved in the respective consensus sequence of mammalian splice sites. To
confirm the consequences of the splice mutation
c.1757-2a⬎g, we sequenced mature messenger RNA
transcripts isolated from peripheral blood of Patient
P1. Our data indicate for the mutant allele a skipping
of exon 7, followed by the regular splicing of exon 8
sequences, resulting in a truncated protein
p.E486fs508X (see Figs 2A, B). Sequence analysis of
the entire spatacsin coding region and the flanking
splice sites did demonstrate only one heterozygous mutation in Patient P8 (c.5255delT) and in Patient P9_1

and her sister P9_2 (c.5798delC), suggesting a heterozygous larger intragenic deletion or a heterozygous
mutation in the regulatory sequence as a second
disease-causing mutation in these patients. Both affected sisters of Family 9 share an identical haplotype.9
Clinical Characterization, Imaging, and
Histopathological Findings
Detailed clinical information was available for 18 patients with identified spatacsin mutations (see Table 2).
All patients fulfilled our inclusion criteria, that is, spasticity of the lower extremities, hyperreflexia, and positive extensor plantar responses. The mean age at onset
of walking impairment as recognized by the patients
was 16 years (range, 8 –31 years). Detailed neurological
examination showed predominantly bilateral proximal
pronounced paresis of the lower limbs with a profound
proximal spasticity and hyperreflexia of the lower
limbs. All patients had extensor plantar responses. The
gait was slow, spastic, and slightly ataxic. Interestingly,
dysarthria was also observed as a common feature in
SPG11 patients (85%). Two patients developed dysphagia and subsequent weight loss. Furthermore, a
prominent bilateral amyotrophy of the hypothenar and
thenar muscles was commonly present. Applying the
SPRS to different cohorts with an approximate similar
disease duration, we observed a faster progression of
the SPRS score (mean, 31.2) compared with both a
nonselected HSP cohort (mean, 20.0)4 and a group of

Hehr et al: Genotype-Phenotype of SPG11

659

Table 2. Summary of the Clinical Data (continued)
Clinical Features
Patient
ID

Cognitive
Deficit

GazeEvoked
Nystagmus

Dysmetria/
Ataxia

Extrapyramidal
Motor Signs

Muscle
Wasting
(upper limbs)

Muscle
Wasting
(lower limbs)

Neuropathy
Sensory

Neuropathy
Motoric

Amyotrophy

Pes
Cavus

P1
P2_1
P2_2
P2_3
P3
P4
P5_1
P5_2
P5_3
P6_1
P6_2⫺
P6_3
P6_4
P7_1
P7_2
P7_3
P7_4
P8
P9_1
P9_2

⫹
⫹
⫹
⫹
⫹
⫹
⫹

⫹
⫺
⫺
⫺
⫹
⫹
⫺

⫹
⫹
⫺
⫹
⫹
⫹
⫹

⫺
⫺
⫺
⫺
⫺
⫺
⫺

⫹
⫹
⫺
⫺
⫺
⫺
⫹

⫹
⫹
⫺
⫺
⫺
⫹
⫹

⫹
⫹
⫺
⫺
⫺
⫺
⫹

⫹
⫹
⫹
⫺
⫹
⫺
⫹

⫹
⫹
⫺
⫺
⫺
⫺
⫹

⫺
⫹
⫺
⫺
⫹
⫹
⫹

⫺
⫺
⫺
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹

⫺
⫹

⫹
⫺
⫹
⫺
⫹
⫹
⫹
⫹
⫺
⫹
⫹

⫺
⫺

⫹
⫺

⫺
⫺

⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫹
⫹

⫺
⫺
⫺
⫺
⫹
⫺
⫹
⫹
⫺
⫹
⫹

⫹
⫺
⫺
⫺
⫹
⫺
⫹
⫹
⫺
⫹
⫹

⫺

⫺
⫺
⫺
⫺
⫹
⫺

⫺
⫺
⫺

⫺
⫺
⫺

⫺
⫺
⫺

⫺
⫹
⫺

⫺
⫺
⫺

⫺
⫹
⫹

autosomal dominant HSP with confirmed SPG4 (Spastin) mutations (mean, 17.2; GeNeMove databank; Fig
3). This reflects a more severe motor impairment in
this SPG11 cohort. In particular, the gait-related items
(items 1– 6) were severely increased with a median total
score of 18.1 points determined after about 12 years of
disease duration. General mental impairment of varying degree was noted in 83% of our cohort and confirmed in a subset of white patients as progressive with
severe attentional deficits with intact long-term memory (Hamburger–Wechsler Intelligence Test for adults
score, ⬍70). MRI was performed for at least 1 affected
member of each family (total: n ⫽ 10 patients) showing TCC associated with bilateral periventricular WM
lesions and frontal cortical atrophy (Fig 4). Corpus callosum (CC) atrophy was pronounced in the rostral
part in all patients with spatacsin mutations. In addition, progressive hypometabolism of the frontal cortex
and thalamus was detected using 18fluorodeoxyglucose
positron emission tomography (P9_1; Fig 5). Furthermore, severe amyotrophy of thenar muscles was present
in 45% of our patients associated with a mixed axonal
demyelinating peripheral neuropathy (Figs 6A, B).
Neuropathological analysis of the sural nerve (P9_1)
showed hypomyelinization of large nerve fibers and loss
of unmyelinated nerve fibers. In addition, accumulation of membranous material was noted in an unmyelinated axon (see Figs 6C, D). Single-voxel 1H-proton
magnetic resonance spectroscopy showed mild reduc-

660

Annals of Neurology

Vol 62

No 6

December 2007

⫺
⫺
⫹
⫺
⫹
⫹
⫺
⫺
⫺

tion of N-acetylaspartate acid in frontal WM, suggesting axonal loss with an increase of choline compatible
with WM demyelinization in Patient P5_1. In addition, increased lactate level was observed both in frontal and parietooccipital cerebral regions, suggesting mitochondrial dysfunction (data not shown).
Natural Long-Term Course of One Characteristic
SPG11 Patient
Patient P9_1 of German descent was first seen at the
age of 26 and managed for more than 10 years. Initially, gait difficulties were noted when walking long
distances associated with painful cramps. Maximum
walking distance was about 1,000 m. Five years later,
she was urged to retire from her position as a nurse
assistant because of progressive motor impairment.
Currently, the walking distance (item 1, SPRS, walking
distance without pause) is less than 50 m. The walking
speed is reduced to 180 seconds for 10 m (item 3,
SPRS, maximum gait speed). Frequently, she is forced
to use the wheelchair. In addition, urinary urge incontinence is increasingly prominent. The initial increased
body mass index of 36 declined during the last year to
24 because of dysphagia. Currently, she receives antispastic treatment with baclofen (60 mg daily) and
physiotherapy. During the 10-year period, the SPRS
score increased from 32 to 40 points based on a reevaluation of the obtained measures for the motor function. Neurological examination showed the classic SPG

Fig 2. Identification of two spatacsin splice mutations. (A) Messenger RNA (mRNA) analysis demonstrates the skipping of exon 7
in Patient P1 because of the splice-site mutation c.1757-2a⬎g. Electrophoretic separation of polymerase chain reaction (PCR) products obtained with primers located in exons 5 and 8 demonstrates the presence of the regular 724bp fragment and an additional
smaller band of 578bp in the index case when compared with a control individual. (B) Sequence analysis of cloned PCR products
after amplification of complementary DNA from a control (top lane) and Patient P1 confirming the skipping of exon 7 (bottom
lane); schematic diagram of the genomic region and relative location of the exonic primers used for reverse transcriptase polymerase
chain reaction and of the mutation (red arrow). (C) Homozygous base substitution c.442⫹1g⬎c at the donor site of intron 2
identified in Family F2 (bottom lane) when compared with a wild-type sequence (bottom lane).

11 features: bilateral, proximal pronounced weakness of
the lower limbs, profound proximal spasticity, hyperreflexia of the lower limbs, positive extensor plantar responses, gaze-evoked nystagmus, dysarthria, dysphagia,
and palmar amyotrophy. Clinical MRI confirmed a
thinning of the rostral part of the CC, which appeared

as a thin band in the midsagittal T1-weighted images
without obvious change of CC volume (see Figs 4A, C,
E). Moreover, symmetric WM lesions were present in
the axial fluid-attenuated inversion recovery sequences
(see Figs 4D, F). Frontal lobe functions deteriorated
and are paralleled by the progressive frontal cortical at-

Fig 3. Comparison of Spastic Paraplegia Rating Scale (SPRS) scores between SPG4 (white bar, n ⫽ 43), SPG11 (black bar, n ⫽
15), and GeNeMove spastic paraplegia (SP) outpatients (gray bar, n ⫽ 165). The data are presented as SPRS score in relation to
disease duration. Percentages of the patients/bin are presented. The SPRS for SPG11 is shifted toward higher scores, indicating a
more severe motor impairment compared with SPG4 and the GeNeMove SP outpatient cohort.

Hehr et al: Genotype-Phenotype of SPG11

661

rophy (see Figs 4B, D, F) with a severe hypometabolism in the frontal cortex and the thalamus (see Fig 5).
Discussion
In patients with spatacsin-linked ARHSP with TCC,
the neurodegenerative process is not restricted to primary motor neurons and the CST, but also affects major projections, in particular, mediating corticocortical
connections via the CC. Additional cerebral regions
such as the thalamus, periventricular WM, and fasciculus gracilis are affected. The presence of a severe palmar amyotrophy with mixed axonal demyelinating
neuropathy indicates a degenerative process not solely
restricted to the CNS.
Currently, about 60 patients linked to the SPG11

Fig 5. Frontoparietal cortical and thalamic hypometabolism in
18
fluorodeoxyglucose positron emission tomography (FDGPET). FDG-PET in the course of 10 years in Patient P9_1.
PET images show a decrease in the thalamic metabolism (arrows in A, C, E). Moreover, metabolic decrease in the mediofrontal polar cortex (arrows in B, D, F) and in the cingulum
is present. PET images were acquired using PET ECAT Exact
47 (1997, 2001) and Biograph 16 (2007; both Siemens,
Erlangen, Germany).

Fig 4. Long-term follow-up of cerebral magnetic resonance
images in Patient P9_1. Sagittal (sag) T1-weighted images
depict pronounced thinning of the rostral part of an atrophic
corpus callosum (CC). Note that the atrophy of the CC is not
progressive (A, 1997; C, 2001; E, 2007; arrows indicate
CC). Moreover, axial T1 (B) and axial fluid-attenuated inversion recovery (FLAIR) images (D, F) demonstrate progressive frontal atrophy with enlargement of the lateral ventricles
and broadening of the interhemispheric sulcus; the latter (D,
F) also demonstrate white matter lesions.

662

Annals of Neurology

Vol 62

No 6

December 2007

locus (15q13-15) have been described worldwide (see
Winner and colleagues’ review10). In addition, sporadic
cases with genetically confirmed SPG4,16,17 SPG21,18
or SPG5a19 also show TCC. However, the majority of
families with ARHSP and TCC are associated with
SPG11, probably accounting for 35 to 50% of all
forms of ARHSP.1,12,20 A clustering of genes associated
with structural anomalies of the CC and/or motor and
sensory neuropathy occurs in chromosomal region
15q14-q21.1 within an interval of 14Mb. Truncating
and missense mutations in the gene for the potassium
chloride cotransporter SLC12A6 (KCC3) in 15q14
cause the autosomal recessive Andermann syndrome

(MIM 218000), which is clinically characterized by
early-onset and progressive peripheral neuropathy with
agenesis of the CC.21,22 ALS5 (MIM 602099), the major locus for autosomal recessive amyotrophic lateral
sclerosis with a juvenile onset, has also been linked to
this region.23 Furthermore, based on linkage analysis of
eight families from the United States and Europe, the
SPG11 locus was originally defined between markers
D15S1007 and D15S1012 in 15q14.6 According to
the current version of the gene map (ensemble current
release 45, June 2007),24 the spatacsin gene resides outside and approximately 6Mb distal to this originally
defined SPG11 region. We speculated in our previous
study refining the SPG11 locus11 that this discrepancy
may result from greater stringency of applied clinical
inclusion criteria. A recombination event in a family
with Andermann syndrome may have contributed to
the initial definition of the distal border for the original SPG11 region.6 Spatacsin is expressed ubiquitously
in the rat nervous system but most prominently in the
cerebellum, cerebral cortex, hippocampus, and pineal
gland.12 Its biological function is still unknown. The
presence of at least one transmembrane domain suggests a function as receptor or transporter. Furthermore, overexpression of a green fluorescent proteinspatacsin fusion protein in COS-7 cells resulted in
fluorescence even after methanol fixation, indicating an
association of the protein to membranes.12 Gene ontology analysis suggests for the spatacsin protein to
confer both a catalytic (GO:0003824) and a ferric
iron–binding activity (GO:0008199).25 In silico analysis for functional protein domains is predicting the
presence of a coiled coil domain (between amino acids
1563 and 1588; SMART),26 a tightly associated bundle of long ␣ helices, and the presence of an O-glycosyl
hydrolase domain (between amino acids 482 and 490).
These enzymes hydrolyze the glycosidic bond between
two or more carbohydrates, or between a carbohydrate
and a noncarbohydrate moiety (InterPro: IPR001360).27
Besides spatacsin, glycoside hydrolase family 1 contains
other proteins known to be involved in neurodegenerative disorders. For instance, mutations in the gene for
6. Severe SPG11 associated neuropathy in Patient P1.
Š Fig
Note the amyotrophy of the thenar (arrow) and hypothenar
muscle (see arrows in A, B). Light (C) and electron microscopic (D) analysis of sural nerve biopsy. Several large myelinated nerve fibers (red arrows) show disproportionately thin
myelin sheaths (C; semithin resin section, toluidine blue). Electron microscopy showed numerous nonmyelinating Schwann
cells encircling bundles of collagen fibers (white arrows) indicative of a loss of unmyelinated nerve fibers. There is a prominent accumulation of pleomorphic membranous material in an
unmyelinated axon (black arrow) compatible with disturbed
axonal transport (D; ultrathin section). Scale bars ⫽ 20 ␮m
(C); 2 ␮m (D).

Hehr et al: Genotype-Phenotype of SPG11

663

the lysosomal enzyme ␤-glucosidase cause the autosomal
recessive disorder Gaucher syndrome.28 In particular,
the subacute neuropathic Gaucher syndrome type III
(MIM 231000) is associated with a spastic paraparesis,
cognitive deterioration, and seizures. Mutations in the
gene for the lysosomal ␤-mannosidase have been associated with mental retardation and a progressive demyelinating peripheral neuropathy (MIM 248510).29 In addition, spatacsin is predicted to contain an Intradiol
ring-cleavage dioxygenase domain between amino acids
2104 and 2381. Dioxygenases mediate incorporation of
both atoms of molecular oxygen into substrates, with Intradiol dioxygenases being involved in the degradation of
aromatic compounds by cleaving the aromatic ring between two hydroxyl groups (InterPro: IPR000627).27
The most characteristic MRI feature of spatacsinlinked SPG11 is the marked thinning of the anterior
part of the CC mostly sparing the splenium. The CC,
the major commissure connecting both cerebral hemispheres, is formed between the 8th and 12th week of
development, but MRI-based evidence indicates a continued myelinization of the axonal cytoskeleton up to
the second decade of life.30 The anatomic subregions
of the CC (genu, body, isthmus, and splenium) are
topographically organized belonging to the highest
order-latest maturing neural network of the brain.31 A
strong myelinization is predominantly observed in the
medial and posterior CC portion, to a lesser degree in
the anterior CC portion, connecting higher cortical
centers such as prefrontal or posterior parietal cortex.32
ARHSP with TCC is characterized by a pronounced
thinning of the anterior part of the CC, mostly paralleled by the progressive frontal cortical atrophy. In
conjunction with recent studies, our data indicate no
further progression of the CC thinning in advanced
stages of the disease, but rather a progressive atrophy of
the mediofrontal cortex.33 In contrast, a progression of
CC thinning was observed in a genetically mixed HSP
population using MRI volumetry.20
In addition, bilateral periventricular WM changes
are a characteristic MRI feature in the present
spatacsin-linked SPG11 cohort. Notably, these subtle
or diffuse signal alterations in supratentorial WM regions with increased signal intensity in T2-weighted
images still require the exclusion of leukodystrophies
such as adrenoleukodystrophy. Furthermore, the involvement of the thalamus in SPG11 has been frequently observed and may contribute to the cognitive
deficit.9,34 Our positron emission tomography studies
in conjunction with the one 1H-proton magnetic resonance spectroscopy indicate a progression of the metabolic dysfunction within the thalamic and frontal lobe
regions. Another hallmark of spatacsin-linked SPG11 is
a severe motor neuropathy leading to an amyotrophy
in half of our patients. Sural nerve biopsy depicted a
loss of unmyelinated nerve fibers and accumulation of

664

Annals of Neurology

Vol 62

No 6

December 2007

pleomorphic membranous material in unmyelinated
axons. These neuropathological alterations are unspecific but suggestive for a disturbed axonal transport.
Spatacsin may play a role in the maintenance of both
central and peripheral axons secondarily leading to loss
of Betz and anterior horn cells. We hypothesize that
the SPG11 phenotype results from the combined degeneration of central and peripheral axons associated
with a neuronal loss within cortical and thalamic regions and in the spinal cord. Using the SPRS, we determined that the motor function is more severely affected in spatacsin-linked SPG11 compared with an
SPG4-linked HSP cohort. This may be because of the
involvement of the CNS and peripheral nervous system
in this complicated HSP form. We propose that disturbed axonal transport processes may be the underlying mechanism for the neurodegenerative process in
SPG11 involving the CST, CC, periventricular WM,
and peripheral nerves. Future functional studies are
needed to unravel the distinct role of spatacsin for
maintaining long-projecting neurons in the CNS and
peripheral nervous system.

The authors wish to dedicate this article to L. J. Thal, one of the
pioneers in the field of neurodegenerative disorders.
This work was supported by the Tom Wahlig Foundation (Jena,
Germany), the “Regensburger Forschungsförderung der Medizinischen Fakultät” (ReForM; University of Regensburg, Germany), the
German Ministry of Education and Research (BMBF; B.W., G.U.,
J.W., U.H.) to GeNeMove (01GM0603), the German Leucodystrophy Network (LEUKONET; R.S., L.S., P.B., O.R.), the German
academic exchange program (A.O., DAAD), and the Fondo Investigaciones Sanitarias (J.G., 02/0648, A.O., 07/0390).
We are grateful to the patients and their families for their participation in this study. We thank C. Mai, F. Renner, A. Milenkovic,
B. Grigo, and A. Schuller for excellent technical assistance, and W.
Schulte-Mattler, R. Luerding, E. Beck and H. Mader for neurophysiological and neuropsychological assessment of the German patients.

References
1. Fink JK. Hereditary spastic paraplegia. In: Rimoin DL, Emery
AEH, eds. Emery & Rimoin’s Principles and Practice of Medical Genetics. 5th ed. Philadelphia: Churchill Livingstone
Elsevier, 2007; 2771–2801.
2. Reid E. Science in motion: common molecular pathological
themes emerge in the hereditary spastic paraplegias. J Med
Genet 2003;40:81– 86.
3. Behan WM, Maia M. Strümpell’s familial spastic paraplegia:
genetics and neuropathology. J Neurol Neurosurg Psychiatry
1974;37:8 –20.
4. Schule R, Holland-Letz T, Klimpe S et al. The Spastic Paraplegia Rating Scale (SPRS): a reliable and valid measure of disease severity. Neurology 2006;67:430 – 434.
5. Harding AE. Classification of the hereditary ataxias and
paraplegias. Lancet 1983;1:1151–1155.
6. Martinez Murillo F, Kobayashi H, Pegoraro E, et al. Genetic
localization of a new locus for recessive familial spastic paraparesis to 15q13–15. Neurology 1999;53:50 –56.

7. Shibasaki Y, Tanaka H, Iwabuchi K, et al. Linkage of autosomal recessive hereditary spastic paraplegia with mental impairment and thin corpus callosum to chromosome 15q13–15. Ann
Neurol 2000;48:108 –112.
8. Casali C, Valente EM, Bertini E, et al. Clinical and genetic
studies in hereditary spastic paraplegia with thin corpus callosum. Neurology 2004;62:262–268.
9. Winner B, Uyanik G, Gross C, et al. Clinical progression and
genetic analysis in hereditary spastic paraplegia with thin corpus
callosum in spastic gait gene 11 (SPG11). Arch Neurol 2004;
61:117–121.
10. Winner B, Gross C, Uyanik G, et al. Thin corpus callosum and
amyotrophy in spastic paraplegia—case report and review of literature. Clin Neurol Neurosurg 2006;108:692– 698.
11. Olmez A, Uyanik G, Ozgul RK, et al. Further clinical and
genetic characterization of SPG11: hereditary spastic paraplegia with thin corpus callosum. Neuropediatrics 2006;37:
59 – 66.
12. Stevanin G, Santorelli FM, Azzedine H, et al. Mutations in
SPG11, encoding spatacsin, are a major cause of spastic paraplegia with thin corpus callosum. Nat Genet 2007;39:366 –372.
13. Chi CS, Mak SC, Schian WJ, Chen CH. Oral Glucose lactate
stimulation test in mitochondrial disease. Pediatr Neurol 1992;
8:445– 449.
14. Gempel K, Kiechl S, Hofmann S, et al. Screening for carnitine
palmitoyltransferase deficiency by tandem mass spectrometry.
J Inherit Metab Dis 2002;25:17–27.
15. Wain HM, Lush MJ, Ducluzeau F, et al. Genew: the Human
Gene Nomenclature Database, 2004 updates. Nucleic Acids
Res 2004;32:D255–D257.
16. Orlacchio A, Kawarai T, Totaro A, et al. Hereditary spastic
paraplegia: clinical genetic study of 15 families. Arch Neurol
2004;61:849 – 855.
17. Alber B, Pernauer M, Schwan A, et al. Spastin related hereditary spastic paraplegia with dysplastic corpus callosum. J Neurol
Sci 2005;236:9 –12.
18. Simpson MA, Cross H, Proukakis C, et al. Maspardin is mutated in mast syndrome, a complicated form of hereditary spastic paraplegia associated with dementia. Am J Hum Genet
2003;73:1147–1156.
19. Al-Yahyaee S, Al-Gazali LI, De Jonghe P, et al. A novel locus
for hereditary spastic paraplegia with thin corpus callosum and
epilepsy. Neurology 2006;66:1230 –1234.

20. França MC, D’Abreu A, Maurer-Morelli CV, et al. Prospective
neuroimaging study in hereditary spastic paraplegia with thin
corpus callosum. Mov Disord 2007;22:1556 –1562.
21. Howard HC, Mount DB, Rochefort D, et al. The K-Cl cotransporter KCC3 is mutant in a severe peripheral neuropathy
associated with agenesis of the corpus callosum. Nat Genet
2002;32:384 –392.
22. Uyanik G, Elcioglu N, Penzien J, et al. Novel truncating and
missense mutations of the KCC3 gene associated with Andermann syndrome. Neurology 2004;66:1044 –1048.
23. Hentati A, Ouahchi K, Pericak-Vance MA, et al. Linkage of a
commoner form of recessive amyotrophic lateral sclerosis to chromosome 15q15–q22 markers. Neurogenetics 1998;2:55– 60.
24. Hubbard TJP, Aken BL, Beal K, et al. Ensembl 2007. Nucleic
Acids Res 2007;35:D610 –D617.
25. The Gene Ontology Consortium. Gene Ontology: tool for the
unification of biology. Nat Genet 2000;25:25–29.
26. Letunic I, Copley RR, Pils B, et al. SMART 5: domains in the
context of genomes and networks. Nucleic Acids Res 2006;34:
D257–D260.
27. Mulder NJ, Apweiler R, Attwood TK, et al. New developments in
the InterPro database. Nucleic Acids Res 2007;35:D224 –D228.
28. Jmoudiak M, Futerman AH. Gaucher disease: pathological
mechanisms and modern management. Br J Haemat 2005;129:
178 –188.
29. Kleijer WJ, Hu P, Thoomes R, et al. Beta-mannosidase
deficiency: heterogeneous manifestation in the first female patient and her brother. J Inherit Metab Dis 1990;13:867– 872.
30. Keshavan M, Diwadkar VA, DeBellis M, et al. Development of
the corpus callosum in childhood and adolescence and early
adulthood. Life Sci 2002;70:1909 –1992.
31. Pujol J, Vendrell P, Junque C, et al. When does the human
brain development end? Evidence if the corpus callosum growth
up to adulthood. Ann Neurol 1993;34:71–75.
32. von Richthofen S, Tabrizian S, Grabe HJ, Meyer B-U. Interhemispheric transfer and its implications for neurology and psychiatry. Fortschr Neurol Psychiat 2003;71:449 – 457.
33. Okubo S, Ueda M, Kamiya T, et al. Neurological and neuroradiological progression in hereditary spastic paraplegia with a
thin corpus callosum. Acta Neurol Scand 2000;102:196 –199.
34. Ohnishi J, Tomoda Y, Yokoyama K. Neuroradiological findings
in hereditary spastic paraplegia with a thin corpus callosum.
Acta Neurol Scand 2001;104:191–192.

Hehr et al: Genotype-Phenotype of SPG11

665