Brain (2000), 123, 908919

Clinical and molecular genetic characteristics of

patients with cerebrotendinous xanthomatosis
Aad Verrips,1 Lies H. Hoefsloot,2 Gerry C. H. Steenbergen,3 Joop P. Theelen,2 Ron A. Wevers,3 Fons
J. M. Gabreels,1 Baziel G. M. van Engelen1 and Lambert P. W. J. van den Heuvel3
1Department

of Neurology, 2Department of Human

Genetics and 3Laboratory of Pediatrics and Neurology,
University Hospital Nijmegen, Nijmegen, The Netherlands

Correspondence to: L. P. W. J. van den Heuvel, Laboratory

of Pediatrics and Neurology, University Hospital Nijmegen,
PO Box 9101, 6500 HB Nijmegen, The Netherlands
E-mail: B.vandeHeuvel@ckslkn.azn.nl

Summary
Cerebrotendinous xanthomatosis (CTX) is a lipid storage
disease caused by a deficiency of the mitochondrial enzyme
27-sterol hydroxylase (CYP 27), due to mutations in its
gene. In this study we report on mutations in 58 patients
with CTX out of 32 unrelated families. Eight of these
were novel mutations, two of which were found together
with two already known pathogenic mutations. Twelve
mutations found in this patient group have been described
in the literature. In the patients from 31 families,
mutations were found in both alleles. In the literature, 28
mutations in 67 patients with CTX out of 44 families have
been described. Pooling our patient group and the patients

from the literature together, 37 different mutations in

125 patients out of 74 families were obtained. Identical
mutations have been found in families from different
ethnic backgrounds. In 41% of all the patients, CYP 27
gene mutations are found in the region of exons 68. This
region encodes for adrenodoxin and haem binding sites of
the protein. Of these 125 patients, a genotypephenotype
analysis was done for 79 homozygous patients harbouring
23 different mutations, out of 45 families. The patients
with compound heterozygous mutations were left out of
the genotypephenotype analysis. The genotype
phenotype analysis did not reveal any correlation.

Keywords: cerebrotendinous xanthomatosis; mutations; genotypephenotype correlation; pathogenesis

Abbreviations: CTX cerebrotendinous xanthomatosis; CYP 27 sterol 27-hydroxylase

Introduction
Cerebrotendinous xanthomatosis (CTX) is a rare, autosomal
recessive, lipid storage disease caused by a deficiency of
the mitochondrial enzyme 27-sterol hydroxylase (CYP 27).
Because of this deficiency, large amounts of cholestanol and
cholesterol are produced. These metabolites accumulate in
many tissues, especially eye lenses, the CNS and muscle
tendons. Besides the cholestanol and cholesterol production,
large amounts of bile alcohols are produced in CTX, which
are excreted in urine. Clinical characteristics of CTX are
premature bilateral cataracts, formation of tendon xanthomas
(most often in the Achilles tendons), neurological and
neuropsychiatric abnormalities such as pyramidal and
cerebellar signs, peripheral neuropathy and dementia
(Bjorkhem and Boberg, 1995). Most patients have cerebellar
signs and dementia from the age of 20 years onwards. In
childhood, the combination of bilateral cataracts and diarrhoea
is almost pathognomonic for the disease (Cruysberg et al.,
1991; van Heijst et al., 1996). The biochemical diagnosis is
made by determination of the serum cholestanol level and
Oxford University Press 2000

by the determination of bile alcohol excretion in urine

(Wolthers et al., 1983, 1991).
In 1989, Andersson and colleagues characterized the cDNA
encoding rabbit mitochondrial CYP 27 starting with rabbit
enzyme protein, which is a member of the mitochondrial
cytochrome P-450 enzyme family (Andersson et al., 1989).
In 1991, the cDNA for human CYP 27 was isolated by
hybridizing rabbit cDNA to a liver cDNA library and its
gene was localized on the long arm of chromosome 2 (Cali
and Russell, 1991). The genomic structure of the CYP 27
gene was elucidated in 1993; the gene contains nine exons
and eight introns and spans 18.6 kb of DNA (Leitersdorf
et al., 1993). The mature enzyme consists of 498 amino
acids and contains putative binding sites for adrenodoxin and
haem; these sites are encoded by the region between exons
6 and 8 (Leitersdorf et al., 1993). The enzyme is expressed
in the CNS, liver, lung, duodenum and endothelial cells
(Reiss et al., 1997). In 1991, the first mutations in the
CYP 27 gene were described (Cali et al., 1991).

Genotype and phenotype in CTX

In this paper we present the phenotypes and genotypes
(including eight novel mutations) of 58 patients with CTX
out of 32 families, of which 21 were Dutch families; this is
the largest series ever reported. We have reviewed the
literature and identified 67 additional CTX patients out of
44 families in whom the genotype had been established.
Finally, we have performed a genotypephenotype analysis
for 79 homozygous patients harbouring 23 different mutations, out of 45 families.

Methods
Patients
Between 1983 and 1998 the clinical and biochemical
information of 42 Dutch CTX patients, of which seven were
children, out of 21 families in the Netherlands were collected.
DNA from these 42 patients and from 16 CTX patients out
of 11 families from the UK, Belgium, Spain, Tunisia,
Germany and China was also analysed. All patients had
elevated serum cholestanol levels and an excessive urinary
excretion of bile alcohols, measured according to Wolthers
and colleagues using capillary gas chromatography (Wolthers
et al., 1983, 1991). Informed consent was obtained from
each participating subject or the parents of younger children.
The study was approved by the Ethics Committee of the
University Hospital Nijmegen, The Netherlands.

Mutation analysis
The CYP 27 gene was amplified in four fragments (exons 1,
2, 35 and 69), by PCR (polymerase chain reaction) from
genomic DNA of leucocytes. Exons 39 with their intron
boundaries were subsequently amplified separately, with the
two PCR fragments 35 and 69 as templates (Luyten
et al., 1995). The oligonucleotides used as primers for PCR
amplification and for sequence analysis are those described
by Leitersdorf and colleagues (Leitersdorf et al., 1993).
Human genomic DNA from patients from these 32 families
was screened for mutations in the CYP 27 gene by single
strand conformation polymorphism analysis using the
Pharmacia Phast System (Amersham Pharmacia Biotech.
Ruusendaal, The Netherlands), or were directly sequenced.
Cycle sequencing of the coding and the non-coding strands
was carried out by the Taq Dye Deoxy Terminator method
(Applied Biosystems Inc., Forster City, Calif., USA) using
an ABI 377 DNA sequencer.
To analyse the effects of the splice site mutations, RNA
from cultured fibroblasts was reverse transcribed to cDNA
according to established procedures (Ploos van Amstel et al.,
1996; Verrips et al., 1997). The CTX cDNA, amplified by
PCR, was used for agarose gel electrophoresis and for DNA
sequence analysis, as described above. The segregation of
novel mutations has been studied in members of families 6,
14, 16, 21, 28, 31 and 32 (Table 1). In families 24 and 30
no family members could be evaluated.

909

Results
Patients
The clinical characteristics of 54 patients are listed in Table 1.
In 10 of the 32 families, three patients were present and in
six families two patients. Sixteen patients were sporadic
cases and accurate clinical information could not be obtained
in two families (19 and 29). The phenotypes of families 2
8, 10, 1216, 24 and 31 have been described in previous
publications. In two families (16 and 18) consanguinity was
present: in both families the parents are first cousins. Among
the general signs, bilateral premature cataracts were present
in 90% of the patients, tendon xanthomas in 45% and
intractable diarrhoea in 33%. Among the most frequent
neurological signs were pyramidal (67%) and cerebellar signs
(60%) and low intelligence (57%). Epilepsy and peripheral
neuropathy were both present in 24% of the patients. The
high prevalence of diarrhoea in our patient group in contrast
to reports in the literature may be due to the fact that it is
generally not known that diarrhoea is an important
phenomenon in this mainly neurological disease.
In all patients, excessive amounts of bile alcohols were
found in the urine. In the patients in whom serum cholestanol
levels were determined, elevated levels were found. Cranial
MRI findings were available in 34 patients. In two-thirds of
the patients global atrophy and parenchymal lesions were
seen. In the majority of the patients in whom an EMG was
performed, axonal neuropathy could be established. Evoked
potentials studies (visual evoked potential, brainstem auditory
evoked potential and somatosensory evoked potential)
revealed delayed central conduction times. A diffuse slowing
with paroxysmal discharges were the main EEG findings.

Mutations
Mutations found in the 32 families, together with their
distribution among these families and the mutations described
in the literature are listed in Table 2. The distribution of the
mutations over the CYP 27 gene is depicted in Fig. 1. In the
32 families presented, 20 different mutations were found of
which 12 were discovered in the Dutch CTX patients.
Eight novel mutations were found in the 32 families
presented. A 1151CT transition in exon 6, resulting in the
substitution of proline by leucine in codon 384 (families 5,
14, 16 and 26), and a 1213CT transition in exon 7, resulting
in replacement of arginine by tryptophan in codon 405
(family 6), were observed. A missense mutation in exon 6
was found in the same allele as a C insertion in exon 1 on
position 56, an already known mutation (Segev et al., 1995)
(Table 2). We found these two mutations on the same allele
in families 5, 14, 16 and 26. In one allele from the patients
of family 16 and in both alleles in their mother, no mutation
in the CYP 27 gene was found, indicating a co-segregation
of these two mutations. In families 24 and 32, a 776AG
transition in exon 4, resulting in substitution of lysine by
arginine in codon 259, was found. On the other allele in

34

EMG axonal neuropathy

EMG demyelinating
neuropathy
EMG myopathy
BAEP dcc
VEP dcc
SSEP dcc median nerve
SSEP dcc sural nerve
EEG diffuse slowing
EEG paroxysmal
discharges

.
.
.
.
.
.
.

.
.

.
.

.
.
.

.
.
.
.
.
.
.

.
.

.
.

.
.
.
.
.
.
.

.
.

.
.

.
.

2.4

33

.
.

35
41
35
35
35
27

30

35
44

5b
27
44

51

32

2.1

26

.
.

20
16

20
35

16
16

16
33

4a

NA 36
.

3.1

4.4

.
.

50

.
.

35

?
35

104 36

.
.

MRI (age)
39
MRI supratentorial

atrophy
MRI infratentorial

atrophy
MRI parenchymal lesions
Periventricular

white matter
Globus pallidus/

capsula interna
Cerebral peduncles

Dentate region

Cerebellar white

matter

Biochemistry
Cholestanol
(3.312.5 mol/l)
Cholesterol
(2.65.2 mmol/l)

Therapy
Simvastatin
Pravastatin
Lovastatin

30

30

30

10

34

Cataracts (years)
Tendon xanthomas
(years)
Diarrhoea (years)

35
35
35
35

10
30

Age of onset (years)

34
Age of diagnosis (years) 37

Neurology start (years)

Dementia (years)
Pyramidal (years)
Cerebellar (years)
Epilepsy (years)
Peripheral neuropathy
(years)

Patient:

Family:

.
.

.
.

44

3.9

83

.
.

37
37
37

36
37

36
37

3.9

46

.
.

36

36

36

36
37

.
.
.
.
.
.
.

.
.
.

.
.

.
N

.
.

NA 36
.

4.4

80

.
.

40
40
40
40

40

40
40

40
41

4
9

.
N
N
N
N

N
.

3.5

45

.
.

7
7

2
7

.
N
N
N
N

N
.

15

4.3

63

.
.

12
12

2
12

10

.
N
.
N
N

.
.

13

2.7

58

.
.

7
7
7
7

2
13

.
N
.
.
.

.
.

20

4.2

74

.
.

20

20
20

20

2
20

11c,d,f 12

.
N
.
.
.

.
.

17

3.4

42

.
.

17

17

17

2
17

13

37

NA

NA

.
.

33

33

27
27

27
33

.
.

35

30
40

20

17

17
46

16

.
.

.
.
.

.
.

.
.
.
.
.
.
.

.
.

.
.
.

.
.

NA NA
.
.

NA 5.2

NA 19

.
.

4
4
37

22

4
38

14c,e,g 15

10

61

.
.

12

12

18

1
24

.
.

40

.
.
.
.
.
.
.
.
.

.
.

55

4.5

71

.
.

19
42
19

5
55

.
.

24

NA 2.7

88

.
.

16
33
16
16

20

18

16
37

17c,g 18e 19a

Table 1 Clinical, biochemical, radiological and neurophysiological characteristics of 54 CTX patients from 30 families

.
.

5
7

21

.
.

30
15
30
30

30

10

4
15

4
41

.
.

15
10
15
15

15

10

9
10

9
34

22c,h 23

12

.
.

15
15
30
30

30

9
32

9
42

24

.
.
.
.
.
.
.

.
.

.
.
.

.
.

.
.
.
.
.
.
.

.
.

.
.
.

.
.

.
.

NA NA 55
.
.

NA NA 5

.
.

.
.

.
.

48

.
.
.
.
.
.
.

.
.
.

.
.

NA
.

5.5

NA NA 229 299 311

.
.

5
5

5
9

20

11

.
.

N
N

58

5.3

124

.
.

10
10
20
25
45

30
15

?
46

.
.
.
.
.
.
.

N
.

54

5.1

59

.
.

47
39
47
47
39

47
39

39
47

25 c,h 26

13

.
.
N
N
N
N
N

N
.

45

55

.
.

10

10
35

27

.
.
.
.
.
.
.

.
.

48

4.4

66

.
.

28

28
28

30
30

28
32

.
.

10
10
25
17

.
.
.
.
.
.
.

.
.

.
.
.

.
.

.
.
.
.
.

.
.

.
.
.

.
.

NA NA
.
.

3.2 4.2

64 126

.
.

26
16
27
27

16

27 25
18

16 10
36 26

28h,i 29 30

14

910
A. Verrips et al.

.
.

Neurology start (years)

Dementia (years)
Pyramidal (years)
Cerebellar (years)
Epilepsy (years)
Peripheral neuropathy

Therapy
Simvastatin
Pravastatin
Lovastatin

.
.

.
.

.
.

?
40 39

34h 35

.
.

31

36

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.

NA NA
.
.
.
.
.
.
.
.
.

.
.

26

26

10
36
10

10

39e

18
41

.
.

.
.

6
7
21 11
6
19 11

40

20
43

13 23
13
30 23
30 23
13
30 23

5 23
31 28
18 30
10 25
5

42

21

24

45e 46j

23

26

47e 48

25
49

27

.
.

.
.

.
.

28
9 35

10
28
35

43

9
28

.
.

.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

21 11

35

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

NA 28
.

51

52k 53

30 31

.
.

7
7

.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.

N
.
.
.
.

.
.
.
.
.

.
.
N

.
.

.
.
.
.
.
.
.
.

.
.
.
N
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.

29
29
29
33

1
36
19

54

32

.
.
.

20
14
29
27
10
28

12
35
18
20
9

Mean

.
.
.
.

.
.

N
N
.
.
.
N
N

.
.
.
.
.
.
.
.
.

.
.
.
.
.

35 .
.
.

85
65 121
6.8 NA
5

NA 41 39 43 42 36 NA 37 41 45
.

55 70
6.1NA

.
.

30 .

30
30

20 10
7
32 36 41 45

25 19
7

20

50

28

31

10
10
31
31

28 16 10 20 10
34 41 39 43 42 36

28 24 10 20 14
?

36

20

44

22

55 240
100
219 141
NA NA
NA 63 NA NA 59 101 NA
4.7 6.1
4.6
3.7 2.7
5.6 4.3 NA NA NA NA NA NA NA

.
.

?
41 39

37h 38

17

.
.

.
.
.

.
.

19
2
1
4
12
6
9
19
10

23
20
23
28
19

34
21
23

33
39
35
14
13

52
27
19

Total

Due to the absence of clinical data, families 19 and 29 are not included. The onset of symptoms and signs (when available) is given in years. ? unknown; present;
absent; N normal; NA not available; dcc delayed central conduction time; SSEP somatosensory evoked potential; BAEP brainstem auditory evoked potential;
VEP visual evoked potential. Some patients have been described before (aVerrips et al., 1996; bVerrips et al., 1997; cCruysberg et al., 1995; dWevers et al., 1992; eVerrips
et al., 1999a; fvan Heijst et al., 1998; gvan Hellenberg Hubar et al., 1992; hWaterreus et al., 1987; ide Jong et al., 1977; jSiebner et al., 1996; kChakraverty et al., 1995).

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.

EMG axonal neuropathy

EMG demyelinating neuropathy
EMG myopathy
BAEP dcc
VEP dcc
SSEP dcc median nerve
SSEP dcc sural nerve
EEG diffuse slowing
EEG paroxysmal discharges

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

NA NA NA
.
.
.
.
.
.

NA NA NA
.
.
.
.
.
.

64 62 53
244 237 252
4.9 4.1 4.1
6.6 4.9 5.6

.
.

35

33

16

MRI (age)
MRI supratentorial atrophy
MRI infratentorial atrophy
MRI parenchymal lesions
Periventricular white matter
Globus pallidus/capsula interna
Cerebral peduncles
Dentate region
Cerebellar white matter

Biochemistry
Cholestanol (3.312.5 mol/l)
Cholesterol (2.65.2 mmol/l)

?
41 39

31h 32

15

Age of onset (years)

Age of diagnosis (years)
Cataracts
Tendon xanthomas
Diarrhoea

Patient:

Family:

Table 1 continued

Genotype and phenotype in CTX

911

1183CA

1184GA

(24) E6

(25) E6

(21) E6
(22) E6
(23) E6

850AT
1016CT
China,
Netherlands
1061AG
1151CT
1183CT
Netherlands,
Germany

(19) E5
(20) E5

56insC
355delC
379CT
380GA
409CT
435GT

4461ga
475CT
525/526delG
646GC
691CT
745CT
776AG
779GA
808CT
819delT
8441ga
845-1ga

E1
E2
E2
E2
E2
E2

New*

1205GA

1204CA

1082AG
1172CT
1204CT
15,17,19

871AT
1037CT
4,9,11,29#

4671ga
496CT
546/547delG
667GC
712CT
766CT
797AG
800GA
829CT
840delT
8651ga
866-1ga

26/27insC
376delC
400CT
401GA
430CT
456GT

Old

Nucleotide change

(7) I2
(8) E3
(9) E3
(10) E3
(11) E4
(12) E4
(13) E4
(14) E4
(15) E4
(16) E4
(17) I4
(18) I4

(1)
(2)
(3)
(4)
(5)
(6)

Mutation
number

Aberrant splicing

R395H

R395S

D354G
P384L
R395C
1,7,10,18,26,32

K284X
T339M
7,8,13,23,25

Aberrant splicing?
Q159X
T175 5aaptc
A216P
R231X
Q249X
K259R
W260X
R270X
L27212aaptc
Deletion of exon 4
Aberrant splicing?

Met 1178aaptc
Met 11823aaptc
R127W
R127Q
R137W
G145G

New

Amino acid change

Lys 251 (ptc)

Thr 306 Met
Brautbar et al., 1983;
Reshef et al., 1994
Asp 321 Gly
Pro 351 Leu
Arg 362 Cys
Harlan and
Still, 1968; Cali
et al., 1991
Arg 362 Ser
Val 332 deleted
(From cDNA
1116, 89 bpdel
upstream end exon 6)
Arg 362 His (full
length)
Val 332 deleted
(From
cDNA 1116, 89 bp
del upstream end
exon 6) Exon 6
skipped

Met33178aaptc
Met 8523aaptc
Arg 94 Trp
Arg 94 Gln
Arg 104 Trp
Gly 112 Gly
Tyr 111 deleted
Exon 2 skipped
Aberrant splicing?
Gln 126 (ptc)
Thr 1425aaptc
Ala 183 Pro
Arg 198 (ptc)
Gln 216 (ptc)
Lys 226 Arg
Trp 227 (ptc)
Arg 237 (ptc)
Leu 23912aaptc
(66 aa)
Aberrant splicing?

Old
14,16,26

25,28
21

21
10

31
24,32
24
14

18,22

2
20

Japan

Japan

Turkey
Netherlands, Belgium
USA, Belgium,

30
14,16,26

Heterozygous

Homozygous

Mutation distribution

S. Africa, Netherlands
Sephardic Jews,

France, Netherlands
Israeli Druze
Netherlands, Belgium
UK, Spain
Japan

Japan
Spain
Netherlands
Surinam Creole
Italy, Tunisia
Italy
UK
Germany
Germany
Pakistan, Netherlands
Morocco, Jews
Netherlands
Morocco, Jews

Ethnic origin

Table 2 Mutation distribution in 32 families, listed together with those described in the literature

Chen et al., 1996, 1998a

Chen et al., 1998c

This study
This study

Segev et al., 1995

Leitersdorf et al., 1994
Verrips et al., 1999a
Watts et al., 1996
Nakashima et al., 1994

Chen et al., 1998b

This study
Verrips et al., 1996
Verrips et al., 1996
Garuti et al., 1996b
Garuti et al., 1997
This study
This study
This study
Ahmed et al., 1997
Leitersdorf et al., 1993
Verrips et al., 1997
Leitersdorf et al., 1993;
Meiner et al., 1994
Meiner et al., 1994

Reference

912
A. Verrips et al.

Del 1.9 kb

1202CG
1213CT
1214GA
12631ga

12635gt

1264-1ga

1420CT
Kim et al.,
1994;
Nagai et al.,
1996
1415GC
1421GA

1435CT

(27) I6

E7
E7
E7
I7

(28)
(29)
(30)
(31)

(32) I7

(33) I7

(34) E8

(35) E8
(36) E8

(37) E8

387AC
873GA

1456CT

1436GC
1442GA

1441CT

1285-1ga

12845gt

1223CG
1234CT
1235GA
12841ga

Del 1.9kb

12051ga

Old

G122G
K284K

R479C

G472A
R474Q

R474W

Aberrant splicing

Aberrant splicing

P401R
R405W
R405Q
Aberrant splicing

Del exon 79

Aberrant splicing

New

Amino acid change

Gly 89 Gly
Lys 251 Lys

Arg 446 Cys

Gly 439 Ala

Arg 441 Gln

Pro 368 Arg

Arg 372 Trp
Arg 372 Gln
Skipping exon 7,
Arg 36228 aaptc
Skipping exon 7,
Arg 36228 aaptc
Asn 38857 aaptc
Del of Thr 389 and
Gln 390
Arg 441 Trp

Del 29 aa, continue

with
Val 33216 aaptc
Del 56 aa, continue
with Thr 30716
aaptcDel 82 aa,
Thr 307 continue
with Asn 388
Del exon 79

Old

27

Homozygous

Japan
Morocco, Jews

Canada

China
Japan

Japan

Italy
Italy

29#

Dotti et al., 1995;

Garuti et al., 1996a, 1997
Okuyama et al., 1996
This study
Chen et al., 1997a

Garuti et al., 1997

Reference

Nakashima et al., 1994

Leitersdorf et al., 1993

This study
Kuriyama et al., 1991;
Kim et al., 1994;
Kuwabara et al., 1996;
Okuyama et al., 1996
Pastershank et al., 1974;
Cali et al., 1991

Kuriyama et al., 1991;

Garuti et al., 1997

Garuti et al., 1997

1,6,13,23,31,22 Garuti et al., 1996b, 1997

28,30

Heterozygous

Mutation distribution

Italy, UK, Netherlands 12

Japan
Netherlands
Japan

Italy

Turkey

Italy, UK, Belgium

Ethnic origin

The 32 families in whom these mutations are present (homozygous or compound heterozygous) are listed according to the numbering in Table 1. Nucleotides given in capital
letters are present in exons and those in lower case in introns.Nucleotide and amino acid numbering are in the old (Cali and Russell, 1991) and the recently proposed new
nomenclature (Antonarakis, 1998). *Nucleotide numbering: the A of the first ATG is number 1 (Antonarakis, 1998); nucleotide numbering: the G of the first GCA is number
1 (Cali and Russell, 1991); amino-acid numbering: the first methionine of the translated frame is number 1 (Antonarakis, 1998); amino-acid numbering: the first methionine
of the translated frame is number 33, alanine (number 1) is the first amino acid (Cali and Russell, 1991). 56ins C and 1151CT on one allele. #1016CT and 1415GC
on one allele. aa amino acid; ptc premature termination codon; ins insertion; del deletion.

Polymorphisms (putative)
(38) E2
366AC
(39) E5
852GA

11841ga

New*

Nucleotide change

(26) I6

Mutation
number

Table 2 continued

Genotype and phenotype in CTX

913

914

A. Verrips et al.

Fig. 1 Schematic diagram of the mutations in the CYP 27 gene. The length of the exons is given proportional to their size. The numbers
within the mutation symbols correspond with those in Table 2. Ten of the 16 missense mutations are found in the region of exons 68.
Deletions and insertions are found in exons 1, 2, 3 and 4. Nonsense mutations are found in exons 3 (one), 4 (four) and 5 (one). Seven of
the 11 mutations affecting pre-mRNA splicing are found in the region of exons 68. Mutations 6, 24 and 25 are nucleotide substitutions
in exons affecting pre-mRNA splicing, resulting in aberrant splice products (Chen et al., 1996, 1998a, b, c).

family 24, a novel nonsense mutation was present: a 779GA

transition in exon 4, changing codon 260 into an opal
termination codon. Another novel nonsense mutation was
found in family 31: a 745CT transition in exon 4 changing
codon 249 into an amber termination codon, together with a
splice site mutation on the other allele were present. In family
21, patients were compound heterozygotes for a novel splice
site mutation: a 4461ga transition in the splice donor
site in intron 2. On the other allele a missense mutation in
exon 2 was found (already known) (Watts et al., 1996). In
family 30, a 1061AG transition leading to a replacement
of asparagine by glycine in codon 354 was found, together
with an already known splice site mutation on the other allele
(Garuti et al., 1997). A 1415GA homozygous transition in
exon 8, resulting in the substitution of glycine by alanine in
codon 472 was found in family 29, together with the known
1016CT transition in exon 5 (Reshef et al., 1994). None
of the novel mutations mentioned above were found in any
of the 50 controls (100 alleles).

Overview of mutations in the Dutch CTX

patients and in the world literature
In the Dutch CTX patients, three mutations were most
frequent, being found in almost two-thirds of the alleles.
These were the 1016CT transition in exon 5 (Reshef et al.,
1994), a 12631g a transition in intron 7, leading to exon
skipping and a frameshift (Garuti et al., 1996b), and finally
the 56 C insertion in exon 1 (Segev et al., 1995), together
with the exon 6 1151CT missense mutation. Since these
recurrent mutations were identified in patients from different
geographical origins, it is likely that they are ancient variants
occurring frequently in CTX. An additional 42 families (67
patients) in whom genotyping has been done were identified
from the literature. These, taken together with our 32 families,
gave a total of 45 families which were homozygous for one
mutation, and 29 families which were compound
heterozygous.
Mutations 2, 5, 6, 9, 16, 18, 24, 25, 34, 35 and 37 were
only found in homozygous patients. Mutations 1, 4, 6, 10,

Genotype and phenotype in CTX

915

Fig. 2 Allele frequencies in CTX mutations. In 74 families (a total of 148 alleles), 37 mutations were found. Except for one family in
which only one mutant allele has been identified, mutations were found on both alleles. The mutation numbers on the horizontal axis
correspond with those in Fig. 1 and Table 2. Cross-hatched columns The Netherlands; black columns abroad.

15, 17, 19, 20, 23, 26, 27, 30, 31 and 36 were found in both
homozygous and compound heterozygous families. Mutations
3, 7, 8, 1114, 21, 22, 2830, 32 and 33 were only found in
compound heterozygous patients.
Overall, five mutations were found in eight or more alleles
(Fig. 2). These are mutations 2 (in Israeli Druze patients
only), 20 (in Sephardic Jews, Chinese and Dutch patients),
23 (USA, Belgium, The Netherlands, Germany), 31 (Italy,
UK, the Netherlands), and 30, 34 and 36 (only in the Japanese
patients).
Of all point mutations identified in the CYP 27 gene, eight
were transversions and 26 were transitions. There were 13
point mutations in CpG dinucleotides; eight CT transitions
together with four GA transitions were found in the same
codon. These mutations led to a substitution of arginine by
another amino acid. Codon 395 was affected in three patients
by a point mutation. Several non-CpG point mutations were
found within 10 bp up- or downstream of mutation sites
harbouring tetra- and trinucleotide motifs. These gene
structures are hypothesized to be hotspots for point mutations
or deletions (Cooper et al., 1995). Two CTTT tetranucleotide
motifs were present in exon 4 in the vicinity of mutations
17 and 18. A TTTG motif was present in exon 3 (mutation
10) and in exon 8 in the neighbourhood of mutation 35.
Trinucleotide motifs were CTT (mutations 10) and TGA
(mutation 21). Tetra- and trinucleotide motifs within a 10 bp
region of deletions were AAGT (mutation 2), CTTT (mutation
9), TTGG and GAA (mutation 16).

Genotypephenotype correlations
Genotypical and phenotypical characteristics of 125 patients
out of 74 families identified in the literature, together with
the presented cohort were determined. Forty-six patients out

of 29 families were compound heterozygous, 79 patients (45

families) were homozygous for 23 different mutations. In
these homozygous patients we examined possible differences
in sex, age of onset, diagnosis, biochemical characteristics
and the presence of signs and symptoms with respect to
mutation site (exon 15 versus exon 69) and mutation type
(missense versus other types of mutations, frameshift or
mutations resulting in a premature termination codon versus
other types of mutations). No specific genotypephenotype
correlation could be established.
Apart from the different phenotypes displayed between
patients from different families, there is also a striking
intrafamilial phenotypic variability in CTX (Dotti et al.,
1996; Nagai et al., 1996) (Table 1).

Discussion
The diagnosis of all the CTX patients was made on clinical
grounds. The biochemical diagnosis was made by the
determination of the bile alcohols in urine and of the
cholestanol in serum, rather than by measurement of the
CYP 27 enzyme activity. As mutations in the CYP 27 gene
were found in 63 of the 64 alleles in all of these patients, it
is highly unlikely that another gene is involved in the
pathogenesis of CTX.
Including the novel mutations presented in the current
patient cohort, 37 different mutations have been described in
the CYP 27 gene in CTX patients. They consist of 16
missense mutations (resulting in amino acid replacements),
three mutations in the last nucleotides of exons (resulting in
both amino acid replacements and affecting pre-mRNA
splicing), three deletions, one insertion, eight splice site and
six nonsense mutations (Cali et al., 1991; Leitersdorf et al.,
1993, 1994; Kim et al., 1994; Meiner et al., 1994; Nakashima

916

A. Verrips et al.

Table 3 Evolutionary conservation of CYP 27 residues substituted in patients with CTX.

CYP 27 amino
acid substitution
Human mutated:
Human wild-type:
Rabbit wild-type:
Rat wild-type:
Mouse wild-type:
CYP 27 amino
acid substitution
Human mutated:
Human wild-type:
Rabbit wild-type:
Rat wild-type:

Mouse wild-type:

K259R

D354G

P384L

R
G
L
QNSLYATFLPKWTRPVLPFWK---LTWALYHLSKDPEIQEALHEE---VPQHKDFAHMPLLKAVLKETL
QNS.Y.TFLPKWTRP.LPFWK---LTWALYHLSK.PEIQ.AL..E---VPQHKDFAHMPLLKAVLKETL
.NS.Y.TFLPKW.RP.LPFWK---LTWALYHLSK.PEIQEALH.E---VPQ.KDFAHMPLLKAV.KETL
.........PKWTR..L..W.---L.W.LY.LS..P..Q.ALH.E---..........PLLKA.LKETL
R405W

G472A

W
A
RLYPVVPTNSRIIEKEIEVDG---HPFGSVPFGYGVRACLGRRIA
RLYPV.P.NSRI..KEIEV.G---HPFGSVPFGYGVRACLGRRIA
RLYPVVPTNSRII........---HPFGSVPFGYGVR.CLGRRIA
RLYPVVP.NSR.....I.V..---HPF.S.PFG.G.R.C.GRR.A

The amino acid sequence alignment of human CYP 27 with that from rabbit, rat and mouse. Points indicate the amino acid residues that
were different from the human CYP 27 sequence. Evolutionarily conserved (wild-type) amino acids are indicated in bold.

et al., 1994; Reshef et al., 1994; Segev et al., 1995; Garuti

et al., 1996a, b, 1997; Okuyama et al., 1996; Watts et al.,
1996; Verrips et al., 1996, 1997, 1999a; Ahmed et al., 1997;
Chen et al., 1996, 1997, 1998a, b, c).
In 19 of the 32 families in this study, mutations were
located in the region of exons 68 of the CYP 27 gene.
Although the mutations were distributed throughout the whole
gene, 15 of the 37 mutations (41%) were found in this region
that comprises 28.4% of the nucleotides of the CYP 27 gene.
This finding may indicate that this conserved part of the
gene, coding for adrenodoxin and haem binding sites, plays
a pivotal role in the function of the enzyme. The pathogenicity
of individual CTX mutations is based on their predicted
effect on the CYP 27 protein and on segregation in families.
None of the novel mutations were found during the analysis
of 50 control chromosomes. Among the CYP 27 gene
mutations identified in our CTX cohort are variants which
are likely to have deleterious effects on the function of the
CYP 27 protein. Thus, the deletion/insertion (mutations 1
and 9), and nonsense mutations (mutations 8, 12, 14) will
cause premature termination of translation and result in
truncated CYP 27 proteins. The different splice site mutations
(mutations 7, 17, 26, 31) will lead to incorrect splicing
and exon skipping, resulting in incorrect CYP 27 proteins.
However, the majority of the mutations detected in the CTX
cohort are amino acid substitutions. Our data, together with
the results of previous studies, indicate that 16 out of 37 are
missense mutations of which five are novel to the described
patient cohort. These mutations are inferred to be pathogenic
when they substitute amino acids, which, in view of their
conservation through evolution, are presumed to be of
functional importance. Except for the 1061AG transition,
which leads to a replacement of asparagine by glycine in
codon 354, all novel missense mutations detected in our
CTX cohort are substitutions of strongly conserved amino
acids by non-conservative ones (Table 3). Three novel
missense mutations (mutations 3, 13, 22) were found in

unrelated CTX families. Except for the presence of two

mutations together on the same allele in families 5, 14, 16
and 26, screening of the remaining exons of the CYP 27
gene in patients with missense mutations did not reveal other
mutations. Therefore, it is likely that these missense mutations
are indeed pathogenic mutations and not innocuous
polymorphisms. However, the ultimate proof that these amino
acid substitutions can indeed result in impairment of CYP
27 function are expression studies which are currently being
conducted in our laboratory. In one allele of the patients
from family 16, no mutations were found. In the mother of
these patients the absence of mutations in the CYP 27 gene
indicated co-segregation of mutations 1 and 22. It is possible
that in this family, a mutation is present in the promoter
region of the CYP 27 gene or at a branch point within one
of the introns.
It is remarkable that the amino acid arginine is frequently
involved in missense mutations. In 12 missense mutations,
there is involvement of CpG dinucleotides leading to the
replacement of arginine by another amino acid. Arginine can
form two hydrogen bonds which may be of major importance
for the conservation of tertiary structure or may play a role
in substrate binding. Both of these aspects could influence
the catalytic activity of the enzyme.
Mutations 1, 3, 4, 10, 15, 19, 20, 22, 23, 26 and 31 were
found in CTX patients from different ethnic backgrounds
(Table 2). The mutations may be ancient variants frequently
occurring in CTX. In order to determine whether these
mutations are introduced into the population by a single
founder, it is necessary to study the chromosome 2 haplotypes
of the patients carrying these identical mutations.
In a large series of 58 CTX patients out of 32 unrelated
families, we found 21 different mutations and a striking
phenotypic heterogeneity, even within families. No genotype
phenotype correlation could be established with these patients
taken together with all CTX patients reported in the literature.
Several authors have stressed the marked phenotypic

Genotype and phenotype in CTX

heterogeneity between CTX patients, even between patients
with the same mutation (Dotti et al., 1996; Nagai et al., 1996).
Since the phenotype varies between patients independent of
their biochemical characteristics, other features must be
responsible for these clinical differences and it has been
suggested that environmental factors are responsible (Chen
et al., 1996; Nagai et al., 1996; Garuti et al., 1997). In CTX
the same mutation may result in different phenotypes, or
mutations at different sites of the CYP 27 gene may result
in the same or in different phenotypes. Recently, we described
a spinal variant of CTX, spinal xanthomatosis, that has a
relatively mild clinical course compared with the classic form
of CTX, which shows cerebellar involvement, dementia,
tendon xanthoma formation and peripheral neuropathy early
in the disease process. Mutation analysis in these patients
revealed missense mutations, predominantly in exons 5 and
6 of the gene, that were also found in the classical form of
CTX (Verrips et al., 1999a). This polyphenotypy in CTX is
the result of a complex pathophysiology. The CYP 27
deficiency, caused by mutations in the CYP 27 gene, leads
to several, different cascades of metabolic derangement, such
as excessive production of cholestanol and bile alcohols
(Batta et al., 1987; Bjo rkhem and Boberg, 1995). The
contribution of each of these pathological metabolic processes
to the phenotype is poorly understood at the present time.
The excessive cholestanol production and its accumulation
within many tissues, particularly the CNS, play a major role
in the disease process. Recently it was shown that in rats
fed a cholestanol enriched diet, cholestanol accumulated in
Purkinje cells, resulting in apoptosis (Inoue et al., 1999).
However, in patients with sitosterolaemia, a rare lipid storage
disorder, serum hypercholestanolaemia is also found. These
patients do not develop neurological disease (Bjo rkhem and
Boberg, 1995).
In 1987 it was reported that a defect of the bloodbrain
barrier was present in CTX, which was reflected by an
elevation of the CSF/serum albumin quotient (Salen et al.,
1987). In the CSF, high amounts of apolipoprotein B, the
protein component of low-density lipoproteins and a carrier
of cholestanol, were present. This finding suggested an
increased influx of sterols from the blood into the CNS. The
defect in the bloodbrain barrier disappears after several
months of chenodeoxycholic acid therapy, so the deficiency
of CYP 27 itself cannot be responsible for this dysfunction.
It can be hypothesized that the phenotype in CTX is the
result of a primary membrane dysfunction, followed by an
increased influx of sterols into the eye lens (bloodlens
barrier), the CNS (bloodbrain barrier), peripheral nerves
(bloodnerve barrier), and vessel wall (endothelial cell
membrane) leading to accelerated arteriosclerosis.
It is unlikely that use of an animal model will clarify
the complex genotypephenotype relationship. Mice with a
disrupted CYP 27 gene, which resulted in a markedly reduced
synthesis of bile acids, had normal plasma levels of cholesterol
and cholestanol. In bile and in faeces of these CYP 27 /
mice, only traces of bile alcohols were found. There was

917

no cholestanol accumulation or CTX-related pathological

abnormalities (Rosen et al., 1998).
Since 1975, chenodeoxycholic acid has been commonly
used as a therapy for CTX (Salen et al., 1975) and has
proven to be effective (Berginer et al., 1984). With this
therapy there is a considerable decrease in the serum
cholestanol level and a sharp decline in the excretion of bile
alcohols in the urine (Wolthers et al., 1983; Batta et al.,
1985). Perhaps the most effective inhibitor of cholestanol
production is a combination of chenodeoxycholic acid with
a -HMG-CoA reductase inhibitor, resulting in a further
lowering of an already normal serum cholestanol level
(Verrips et al., 1999b) and facilitating the long-term washout
of cholestanol from the CNS. Finally, as therapy is available,
the early recognition of CTX is important. Because of the
phenotypic heterogeneity, in all siblings of novel CTX patients
determination of the genotype must be done to exclude or
confirm the diagnosis.

Acknowledgements
The authors wish to thank F. Barkhof, Department of
Diagnostic Radiology, Free University Hospital, Amsterdam,
The Netherlands, for the evaluation of the MRIs, and
R. P. Kleyweg (family 25), Department of Neurology, Hospital
Dordrecht; M. de Visser (family 18), B. M. van Geel (family
23) and J. J. P. Kastelein (family 2), Academic Medical
Centre, Amsterdam; J. W. B. Moll (family 10), University
Hospital Rotterdam; J. H. J. Wokke (family 3), Department
of Neurology, University Hospital Utrecht, Utrecht, The
Netherlands; S. Tam (family 29), Department of Clinical
Biochemistry, Queen Mary Hospital, Hong Kong, China;
R. Denays (family 26), Department of Neurology, New
Paul Brien Centre, Brussels, Belgium; P. Hart (family 27),
Neurology Department, Atkinson Morleys Hospital, London,
UK; R. Pe rez Moyano (family 21), Almeria, Spain; T. J. Walls
(family 31), Regional Neurosciences Centre, Newcastle upon
Tyne, UK; J. Mebis (family 28), Algemeen Ziekenhuis
Middelheim, Department of Internal Medicine, Antwerpen,
Belgium; N. Miladi (family 20), Institut National de
Neurologie, Tunis, Tunisia; L. van Malderghem (family 19),
Institute de Pathologie et de Ge ne tique, Gerpinnes (Loverval),
Belgium; R. Kimmelre (families 30 and 32), Heinrich Heine
Universita t Du sseldorf, Klinik fu r Stoffwechselkrankheiten
und Erna hrung, Du sseldorf, Germany; and S. Berndt (family
24), Department of Neurology, Paderborn, Germany for
referring the patients.

References
Ahmed MS, Afsar S, Hentati A, Ahmad A, Pasha J, Juneja T, et al.
A novel mutation in the sterol 27-hydroxylase gene of a Pakistani
family with autosomal recessive cerebrotendinous xanthomatosis.
Neurology 1997; 48: 25860.
Andersson S, Davis DL, Dahlback H, Jornvall H, Russell DW.
Cloning, structure, and expression of the mitochondrial cytochrome

918

A. Verrips et al.

P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J

Biol Chem 1989; 264: 82229.
Antonarakis SE. Recommendations for a nomenclature system for
human gene mutations. Nomenclature Working Group. Hum Mutat
1998; 11: 13.
Batta AK, Shefer S, Batta M, Salen G. Effect of chenodeoxycholic
acid on biliary and urinary bile acids and bile alcohols in
cerebrotendinous xanthomatosis; monitoring by high performance
liquid chromatography. J Lipid Res 1985; 26: 6908.
Batta AK, Salen G, Shefer S, Tint GS, Batta M. Increased plasma
bile alcohol glucuronides in patients with cerebrotendinous
xanthomatosis: effect of chenodeoxycholic acid. J Lipid Res 1987;
28: 100612.
Berginer VM, Salen G, Shefer S. Long-term treatment of
cerebrotendinous xanthomatosis with chenodeoxycholic acid. N
Engl J Med 1984; 311: 164952.
Bjo rkhem I, Boberg KM. Inborn errors in bile acid biosynthesis
and storage of sterols other than cholesterol. In: Scriver CR, Beaudet
AL, Sly WS, Valle D, editors. The metabolic and molecular bases
of inherited disease. 7th ed. New York: McGraw-Hill; 1995.
p. 207399.
Brautbar C, Yehuda O, Eisenberg S, Cohen N, Amar A, Sharon R,
et al. Study of a family with cerebrotendinous xanthomatosis. No
HLA linkage, but an informative recombination between HLA-B
and Bf. Tissue Antigens 1983; 21: 2337.
Cali JJ, Russell DW. Characterization of human sterol 27hydroxylase. J Biol Chem 1991; 266: 77748.
Cali JJ, Hsieh CL, Francke U, Russell DW. Mutations in the bile
acid biosynthetic enzyme sterol 27-hydroxylase underlie
cerebrotendinous xanthomatosis. J Biol Chem 1991; 266: 777983.
Chakraverty S, Griffiths PD, Walls TJ, McAllister VL.
Cerebrotendinous xanthomatosis in two sisters: case reports and
MR imaging. Clin Radiol 1995; 50: 1179.
Chen W, Kubota S, Nishimura Y, Nozaki S, Yamashita S, Nakagawa
T, et al. Genetic analysis of a Japanese cerebrotendinous
xanthomatosis family: identification of a novel mutation in the
adrenodoxin binding region of the CYP 27 gene. Biochim Biophys
Acta 1996; 1317: 11926.
Chen W, Kubota S, Kim KS, Cheng J, Kuriyama M, Eggertsen
G, et al. Novel homozygous and compound heterozygous mutations
of sterol 27-hydroxylase gene (CYP27) cause cerebrotendinous
xanthomatosis in three Japanese patients from two unrelated families.
J Lipid Res 1997; 38: 8709.
Chen W, Kubota S, Seyama Y. Alternative pre-mRNA splicing of
the sterol 27-hydroxylase gene (CYP 27) caused by a G to
A mutation at the last nucleotide of exon 6 in a patient with
cerebrotendinous xanthomatosis (CTX). J Lipid Res 1998a; 39:
50917.
Chen W, Kubota S, Teramoto T, Nishimura Y, Yonemoto K, Seyama
Y. Silent nucleotide substitution in the sterol 27-hydroxylase gene
(CYP 27) leads to alternative pre-mRNA splicing by activating a
cryptic 5 splice site at the mutant codon in cerebrotendinous
xanthomatosis patients. Biochemistry 1998b; 37: 44208.
Chen W, Kubota S, Ujike H, Ishihara T, Seyama Y. A novel

Arg362Ser mutation in the sterol 27-hydroxylase gene (CYP 27):

its effects on pre-mRNA splicing and enzyme activity. Biochemistry
1998c; 37: 150506.
Cooper DN, Krawczak M, Antonarakis SE. The nature and
mechanisms of human gene mutation. In: Scriver CR, Beaudet AL,
Sly WS, Valle D, editors. The metabolic and molecular bases of
inherited disease. 7th ed. New York: McGraw-Hill; 1995. p. 25991.
Cruysberg JR, Wevers RA, Tolboom JJ. Juvenile cataract associated
with chronic diarrhea in pediatric cerebrotendinous xanthomatosis
[letter]. Am J Ophthalmol 1991; 112: 6067.
Cruysberg JR, Wevers RA, van Engelen BG, Pinckers A, van
Spreeken A, Tolboom JJ. Ocular and systemic manifestations of
cerebrotendinous xanthomatosis. Am J Ophthalmol 1995; 120:
597604.
de Jong JG, van Gent CM, Delleman JW. Cerebrotendinous
cholestanolosis in relation to other cerebral xanthomatoses. Clin
Neurol Neurosurg 1977; 79: 25372.
Dotti MT, Manneschi L, Federico A. Mitochondrial enzyme
deficiency in cerebrotendinous xanthomatosis. J Neurol Sci 1995;
129: 1068.
Dotti MT, Garuti R, Calandra S, Federico A. Clinical and genetic
variability in CTX [abstract]. Eur J Neurol 1996; 3 Suppl 5: 12.
Garuti R, Lelli N, Barozzini M, Dotti MT, Federico A, Bertolini S,
et al. Partial deletion of the gene encoding sterol 27-hydroxylase in
a subject with cerebrotendinous xanthomatosis. J Lipid Res 1996a;
37: 66272.
Garuti R, Lelli N, Barozzini M, Tiozzo R, Dotti MT, Federico A,
et al. Cerebrotendinous xanthomatosis caused by two new mutations
of the sterol-27-hydroxylase gene that disrupt mRNA splicing.
J Lipid Res 1996b; 37: 145967.
Garuti R, Croce A, Tiozzo R, Dotti MT, Federico A, Bertolini S,
et al. Four novel mutations of sterol 27-hydroxylase gene in Italian
patients with cerebrotendinous xanthomatosis. J Lipid Res 1997;
38: 232234.
Harlan WRJ, Still WJ. Hereditary tendinous and tuberous
xanthomatosis without hyperlipidemia. A new lipid-storage disorder.
N Engl J Med 1968; 278: 41622.
Inoue K, Kubota S, Seyama Y. Cholestanol induces apoptosis of
cerebellar neuronal cells. Biochem Biophys Res Commun 1999;
256: 198203.
Kim KS, Kubota S, Kuriyama M, Fujiyama J, Bjo rkhem I, Eggertsen
G, et al. Identification of new mutations in sterol 27-hydroxylase
gene in Japanese patients with cerebrotendinous xanthomatosis
(CTX). J Lipid Res 1994; 35: 10319.
Kuriyama M, Fujiyama J, Yoshidome H, Takenaga S, Matsumuro
K, Kasama T, et al. Cerebrotendinous xanthomatosis: clinical and
biochemical evaluation of eight patients and review of the literature.
[Review]. J Neurol Sci 1991; 102: 22532.
Kuwabara K, Hitoshi S, Nukina N, Ishii K, Momose T, Kubota S,
et al. PET analysis of a case of cerebrotendinous xanthomatosis
presenting hemiparkinsonism. J Neurol Sci 1996; 138: 1459.
Leitersdorf E, Reshef A, Meiner V, Levitzki R, Schwartz SP, Dann
EJ, et al. Frameshift and splice-junction mutations in the sterol 27-

Genotype and phenotype in CTX

hydroxylase gene cause cerebrotendinous xanthomatosis in Jews or
Moroccan origin. J Clin Invest 1993; 91: 248896.
Leitersdorf E, Safadi R, Meiner V, Reshef A, Bjo rkhem I, Friedlander
Y, et al. Cerebrotendinous xanthomatosis in the Israeli Druze:
molecular genetics and phenotypic characteristics. Am J Hum Genet
1994; 55: 90715.
Luyten JA, Wenink PW, Steenbergen-Spanjers GC, Wevers RA, Ploos
van Amstel HK, de Jong JG, et al. Metachromatic leukodystrophy: a
12-bp deletion in exon 2 of the arylsulfatase A gene in a late infantile
variant. Hum Genet 1995; 96: 35760.
Meiner V, Meiner Z, Reshef A, Bjo rkhem I, Leitersdorf E.
Cerebrotendinous xanthomatosis: molecular diagnosis enables
presymptomatic detection of a treatable disease. Neurology 1994; 44:
28890.
Nagai Y, Hirano M, Mori T, Takakura Y, Tamai S, Ueno S. Japanese
triplets with cerebrotendinous xanthomatosis are homozygous for
a mutant gene coding for the sterol 27-hydroxylase (Arg441Trp).
Neurology 1996; 46: 5714.
Nakashima N, Sakai Y, Sakai H, Yanase T, Haji M, Umeda F, et al.
A point mutation in the bile acid biosynthetic enzyme sterol 27hydroxylase in a family with cerebrotendinous xanthomatosis. J Lipid
Res 1994; 35: 6638.
Okuyama E, Tomita S, Takeuchi H, Ichikawa Y. A novel mutation in
the cytochrome P450(27) (CYP27) gene caused cerebrotendinous
xanthomatosis in a Japanese family. J Lipid Res 1996; 37: 6319.
Pastershank SP, Yip S, Sodhi HS. Cerebrotendinous xanthomatosis.
J Can Assoc Radiol 1974; 25: 2826.
Ploos van Amstel JK, Bergman AJIW, van Beurden EACM, Roijers
JFM, Peelen T, van den Berg IET, et al. Hereditary tyrosinemia type
1: novel missense, nonsense and splice consensus mutations in the
human fumarylacetoacetate hydrolase gene; variability of the
genotype-phenotype relationship. Hum Genet 1996; 97: 519.
Reiss AB, Martin KO, Rojer DE, Iyer S, Grossi EA, Galloway AC,
et al. Sterol 27-hydroxylase: expression in human arterial
endothelium. J Lipid Res 1997; 38: 125460.
Reshef A, Meiner V, Berginer VM, Leitersdorf E. Molecular genetics
of cerebrotendinous xanthomatosis in Jews of north African origin.
J Lipid Res 1994; 35: 47883.
Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L,
et al. Markedly reduced bile acid synthesis but maintained levels of
cholesterol and vitamin D metabolites in mice with disrupted sterol
27-hydroxylase gene. J Biol Chem 1998; 273: 1480512.
Salen G, Meriwether TW, Nicolau G. Chenodeoxycholic acid inhibits
increased cholesterol and cholestanol synthesis in patients with
cerebrotendinous xanthomatosis. Biochem Med 1975; 14: 5774.
Salen G, Berginer V, Shore V, Horak I, Horak E, Tint GS, et al.
Increased concentrations of cholestanol and apolipoprotein B in the
cerebrospinal fluid of patients with cerebrotendinous xanthomatosis.
Effect of chenodeoxycholic acid. N Engl J Med 1987; 316: 12338.
Segev H, Reshef A, Clavey V, Delbart C, Routier G, Leitersdorf E.
Premature termination codon at the sterol 27-hydroxylase gene causes
cerebrotendinous xanthomatosis in a French family. Hum Genet 1995;
95: 23840.

919

Siebner HR, Berndt S, Conrad B. Cerebrotendinous xanthomatosis

without tendon xanthomas mimicking Marinesco-Sjoegren
syndrome: a case report. J Neurol Neurosurg Psychiatry 1996; 60:
5825.
van Heijst AF, Wevers RA, Tangerman A, Cruysberg JR, Renier Wo,
Tolboom JJ. Chronic diarrhoea as a dominating symptom in two
children with cerebrotendinous xanthomatosis. Acta Paediatr 1996;
85: 9326.
van Heijst AF, Verrips A, Wevers RA, Cruysberg JR, Renier WO,
Tolboom JJM. Treatment and follow-up of children with
cerebrotendinous xanthomatosis. Eur J Pediatr 1998; 157: 3136.
van Hellenberg Hubar JL, Joosten EM, Wevers RA. Cerebrotendinous
xanthomatosis. Clin Neurol Neurosurg 1992; 94 Suppl: S1657.
Verrips A, Steenbergen-Spanjers GC, Luyten JA, van den Heuvel LP,
Keyser A, Gabree ls FJ, et al. Two new mutations in the sterol 27hydroxylase gene in two families lead to cerebrotendinous
xanthomatosis. Hum Genet 1996; 98: 7357.
Verrips A, Steenbergen-Spanjers GC, Luyten JA, Wevers RA, Wokke
JH, Gabree ls FJ, et al. Exon skipping in the sterol 27-hydroxylase
gene leads to cerebrotendinous xanthomatosis. Hum Genet 1997; 100:
2846.
Verrips A, Lycklama a` Nijeholt GJ, Barkhof F, Van Engelen BG,
Wesseling P, Luyten JAFM, et al. Spinal xanthomatosis: a variant of
cerebrotendinous xanthomatosis. Brain 1999a; 122: 158995.
Verrips A, Wevers RA, Van Engelen BG, Keyser A, Wolthers BG,
Barkhof F, et al. Effect of simvastatin in addition to chenodeoxycholic
acid in patients with cerebrotendinous xanthomatosis. Metabolism
1999b; 48: 2338.
Waterreus RJ, Koopman BJ, Wolthers BG, Oosterhuis HJ.
Cerebrotendinous xanthomatosis (CTX): a clinical survey of the
patient population in The Netherlands. Clin Neurol Neurosurg 1987;
89: 16975.
Watts GF, Mitchell WD, Bending JJ, Reshef A, Leitersdorf E.
Cerebrotendinous xanthomatosis: a family study of sterol 27hydroxylase mutations and pharmacotherapy. Q J Med 1996; 89:
5563.
Wevers RA, Cruysberg JR, Van Heijst AF, Janssen-Zijlstra FS, Renier
WO, Van Engelen BG, et al. Paediatric cerebrotendinous
xanthomatosis. J Inherit Metab Dis 1992; 15: 3746.
Wolthers BG, Volmer M, van der Molen J, Koopman BJ, de Jager
AE, Waterreus RJ. Diagnosis of cerebrotendinous xanthomatosis
(CTX) and effect of chenodeoxycholic acid therapy by analysis of
urine using capillary gas chromatography. Clin Chim Acta 1983; 131:
5365.
Wolthers BG, Walrecht HT, van der Molen JC, Nagel GT, Van
Doormaal JJ, Wijnandts PN. Use of determinations of 7-lathosterol
(5 alpha-cholest-7-en-3 beta-ol) and other cholesterol precursors in
serum in the study and treatment of disturbances of sterol metabolism,
particularly cerebrotendinous xanthomatosis. J Lipid Res 1991; 32:
60312.

Received July 15, 1999. Revised October 12, 1999.

Accepted November 11, 1999