Mutations in TTBK2, encoding
a kinase implicated in tau
phosphorylation, segregate with
spinocerebellar ataxia type 11
Henry Houlden1, Janel Johnson1,2, Christopher Gardner-Thorpe3,
Tammaryn Lashley1, Dena Hernandez2, Paul Worth4,
Andrew B Singleton2, David A Hilton5, Janice Holton1,
Tamas Revesz1, Mary B Davis1, Paolo Giunti1& Nicholas W Wood1
The microtubule-associated protein tau (encoded by MAPT)
and several tau kinases have been implicated in
neurodegeneration, but only MAPT has a proven role in
disease. We identified mutations in the gene encoding tau
tubulin kinase 2 (TTBK2) as the cause of spinocerebellar
ataxia type 11. Affected brain tissue showed substantial
cerebellar degeneration and tau deposition. These data
suggest that TTBK2 is important in the tau cascade and
in spinocerebellar degeneration.
The autosomal dominant spinocerebellar ataxias (SCAs) are a hetero-
geneous group of neurodegenerative disorders characterized by poor
coordination, abnormal eye movements, impairment of speech and
swallowing, and pyramidal signs1. Autosomal dominant pure cerebel-
lar ataxia is the most common form of this disorder2. Trinucleotide
repeat expansions that encode polyglutamine tracts are found in
approximately 50% of families with dominant ataxia. Noncoding
repeats and dominant mutations have been identified in a smaller
number of cases of ataxia, and these have revealed previously
unknown disease pathways1,3,4. A substantial minority of SCA-affected
families have no identifiable genetic defect.
Spinocerebellar ataxia type 11 (SCA11) is a pure progressive
cerebellar ataxia that has been genetically linked to chromosome
15q14–21 (ref. 5). We further localized the SCA11 locus to a 5.6-cM
region containing 134 genes (Supplementary Table 1 online). An
affected family from Devon, on the southwest coast of England,
stretches over eight generations (family 1; Supplementary Fig. 1
online). All affected individuals have progressive cerebellar ataxia,
abnormal eye signs and pyramidal features (Supplementary Table 2
online), as well as cerebellar atrophy visible upon magnetic resonance
imaging (Supplementary Fig. 2 online).
After screening 54 genes in the linked region (all found to be lacking
in pathogenic mutations), we analyzed a candidate gene encoding tau
tubulin kinase-2 (TTBK2) (Supplementary Methods online). Sequen-
cing of TTBK2 in family 1 revealed a one-base insertion of an
adenosine in exon 13 at nucleotide 1329, codon 444 (Fig. 1a). This
created a premature stop site (TGA) in the mRNA at codon 450,
truncating the normal protein from 1,244 to 450 amino acids. We
identified a second family with pure ataxia associated with a frameshift
deletion of two bases (GA) in exon 13 of TTBK2 at nucleotides
1284_1285, codons 428 and 429 (Fig. 1a). This created a premature
stop site (TGA) in the mRNA at codon 449 (Fig. 1a and Supplemen-
tary Fig. 3 online). The second family was of Pakistani ancestry, with
five affected individuals present over three generations (family 2;
Supplementary Fig. 1). Both mutations segregated in the affected
family members, and neither were present in over 400 elderly
European-descended British or in 50 elderly Pakistani control indivi-
duals. We screened for TTBK2 mutations in a total of 40 families with
autosomal dominant pure SCA. Families screened were negative for
the expansions in the genes associated with SCA types 1, 2, 3, 6, 7, 8,
10, 12 and 17. In addition, we sequenced the genes associated with
SCA types 14 and 27 in their entirety.
The TTBK2 mutations we identified mapped to the conserved
serine-rich region of the protein, and each was conserved between
TTBK2 and TTBK1 and in multiple species (Supplementary Fig. 3).
Both cause a premature stop codon in the mRNA, leading either to the
production of a truncated protein and/or to nonsense-mediated decay
(NMD) of the mutant transcript. We carried out semiquantitative
PCR with reverse transcription (RT-PCR) on RNA extracted from
lymphoblasts of affected individuals in families 1 and 2, and found
that each mutation was associated with an approximate 50%
reduction in total TTBK2 mRNA relative to lymphoblast mRNA
from unaffected individuals. Treating lymphoblasts from affected
individuals of both families with cycloheximide, a known inhibitor
of NMD, resulted in increased total TTBK2 mRNA and a selective
increase in the abundance of the mutant mRNA, which became
roughly equal to that of the normal transcript (Fig. 1b). Some mutant
transcript was detected on fragment analysis, which suggests that a
proportion of the mutant mRNA escapes NMD (Fig. 1b and Supple-
mentary Fig. 4 online). We obtained a result similar to that shown in
Figure 1b by quantifying, by densitometry, TTBK2 RT-PCR signal
from the lymphoblasts of affected individuals relative to GAPDH
signal (Supplementary Fig. 4). Quantitative PCR carried out on
affected and control lymphoblasts also revealed an approximately
50% reduction of TTBK2 mRNA in affected individuals, which was
prevented by cycloheximide treatment (Fig. 1c)6.
TTBK2 produces a 5.6-kb transcript in which the longest open
reading frame is 3,732 nucleotides, encoding a protein of 1,244 amino
Received 22 June; accepted 14 September; published online 25 November 2007; doi:10.1038/ng.2007.43
1Departments of Molecular Neuroscience, Institute of Neurology and The National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK.
2Molecular Genetics Unit, Laboratory of Neurogenetics, National Institute on Aging, National Institutes of Health, Bethesda, Maryland 20892, USA.3Royal Devon
and Exeter Hospital (Wonford), Barrack Road, Exeter, Devon EX2 5DW, UK.4Department of Neurology, Norfolk and Norwich University Hospital National Health
Service Trust, Norwich NR4 7UY, UK.5Department of Histopathology, Derriford Hospital, Plymouth PL6 8DH, UK. Correspondence should be addressed to
H.H. (email@example.com) or N.W.W. (firstname.lastname@example.org).
1434VOLUME 39 [ NUMBER 12 [ DECEMBER 2007 NATURE GENETICS
© 2007 Nature Publishing Group http://www.nature.com/naturegenetics
acids. The gene is alternatively spliced, with ubiquitous expression in
human adult and fetal tissues (Supplementary Figs. 4 and 5). The
N terminus of TTBK2 encodes a serine-threonine-tyrosine kinase
domain, and the C-terminal region shows homology to TTBK1.
In situ hybridization showed that TTBK2 was expressed in all
brain regions in human, rat and mouse tissue. There was particularly
high expression in the cerebellum Purkinje cells, granular cell layer,
hippocampus, midbrain and substantia nigra (Fig. 2). Lower expres-
sion was seen in the cortex of human, rat and mouse brains
(Supplementary Fig. 5).
We carried out neuropathological examination of archival paraffin
brain blocks from one affected individual from family 1. Spinal cord
tissue was not available. Macroscopically, the brain showed gross
atrophy of the cerebellum (Supplementary Fig. 2). Microscopically,
there was severe and near-complete loss of Purkinje cells, with a
marked preservation of the baskets, and a significant loss of cerebellar
granule cells (Fig. 2). We did not find neuronal inclusions using
hematoxylin and eosin, p62, ubiquitin or 1C2 immunohistochemistry.
Neurofibrillary tangles, neuropil threads and tau-positive neurites
werevisiblein the medullary
tegmentum and tectum and putamen (Fig. 2). Pre-tangles and
neurofibrillary tangles were seen in the globus pallidus, as were
oligodendroglial tau-positive filamentous inclusions. There were also
changes consistent with pathological aging, with sparse tangles
entorhinal and insular cortex. Plaques were
not present in the cerebellum.
As the signs of pathological aging, with tau
and amyloid lesions, are a near-universal
finding in the cortical regions of elderly
individuals7, interpreting the tangle patho-
logy is difficult, although in this case tau
was also found in the basal ganglia, midbrain
and medulla. To further confirm the associa-
tion of tau pathology in individuals with
TTBK2 mutations, it will be important to
analyze further brain tissue from affected
individuals. Notably, the degree of TTBK2
gene expression did not correlate with pathology: the gene was highly
expressed in the cerebellum, where we observed cell loss but no other
visible pathology. This is perhaps due to the high resistance of the
cerebellum to forming tangles8,9.
The role of tau phosphorylation in Alzheimer’s disease is complex.
Considerable work has been invested in identifying the specific sites
and timing of phosphorylation and their relationship to neurodegene-
ration10. TTBK2 encodes a member of the casein kinase (CK1) group
of eukaryotic protein kinases, as indicated by its ability to phosphor-
ylate tau and tubulin in vitro11–13. TTBK1 has been implicated in
Alzheimer’s disease and in tangle formation12. The two TTBK2
phosphorylation sites in tau (Ser208 and Ser210) are priming sites
for the phosphorylation of tau by GSK-3b, which is known to
influence tau pathology14.
Further evidence for the role of TTBK2 in the tau pathway has been
reported from RNA interference experiments in Caenorhabditis
elegans15. The authors identified genes whose knockdown in the
presence of an FTDP-17 mutant tau transgene enhanced the un-
coordinated movement (UNC) phenotype. Treating the worms with
short interfering RNA against a homolog of TTBK2 increased the
UNC phenotype, as measured by a behavioral assay.
This is the first time that a genetic defect affecting a tau kinase has
been shown to cause neurodegeneration, and also the first time that
spinocerebellar ataxia has been linked to defects in a gene associated
Family 1: 1329InsA
Family 2: 1284_1285DelAG
+ 2 h
+ 2 h
+ 8 h
+ 8 h
Affected family 1
TTBK2 mRNA level (%)
Figure 1 Identification and characterization of
TTBK2 mutations. (a) Chromatograms of TTBK2
exon 11 sequence showing frameshift mutations
in family 1 (c.1329InsA, codon 444) and family
2 (c.1284_1285, codon 428–429) next to
sequences from an unaffected member of each
family. Arrows indicate the positions of the
mutations. (b) RT-PCR fragment analysis across
intron 13 of TTBK2 in lymphoblastoid cells from
affected individuals shows that the abundances
of the mutant mRNAs (family 1,286 base pairs
(bp); family 2,283 bp) are reduced (arrows) to
approximately 55% those of the wild-type
transcripts (normal allele, 285 bp). Treatment
with cycloheximide (CHX, 500 mM) results in a
selective increase in the abundance of the mutant
mRNAs. (c) Quantitative RT-PCR using SYBR.
Expression of the mutant TTBK2 transcript is
lower in an affected individual from family
1 than in an unaffected control individual
(P o 0.05, t-test, n ¼ 3 experiments), and
expression is increased by cycloheximide
NATURE GENETICS VOLUME 39 [ NUMBER 12 [ DECEMBER 20071435
© 2007 Nature Publishing Group http://www.nature.com/naturegenetics
with tau regulation or deposition. Given the enhanced tau pathology Download full-text
seen in individuals with TTBK2 mutations and the C. elegans knock-
down data, it is likely that TTBK2 has a role in preventing tau toxicity.
It will be interesting to identify further TTBK2 mutations, especially
missense or truncating mutations, to see whether they have a similar
clinical and pathological phenotype.
The mechanism of tau toxicity is unknown, but these data seem to
suggest an indirect interaction between TTBK2 and tau, supporting
the existence of a common pathway that involves TTBK2, GSK-3b and
tau. Our data also suggest that defects in genes associated with this
pathway may lead to different clinical and pathological phenotypes of
neurodegeneration, including spinocerebellar ataxia.
Note: Supplementary information is available on the Nature Genetics website.
We are grateful to the UK Medical Research Council (MRC) for their support
of the entire project: H.H. holds an MRC clinician scientist fellowship. We also
thank the European Commission (EUROSCA) for supporting the initial work.
This work was undertaken at University College London Hospital/University
College London, which received a proportion of funding from the Department
of Health’s National Institute for Health Research Biomedical Research Centers
funding scheme. This work was supported in part by the Intramural Program of
the US National Institute on Aging, National Institutes of Health, Department
of Health and Human Services. We thank S. Schorge for her helpful comments
on the paper as well as the members of the affected families that we studied
and the organization Ataxia UK for their continued support and assistance
with our work.
H.H. planned and supervised the project. J.J. carried out the candidate gene
sequencing and genetic analysis. C.G.-T., P.W., P.G. and N.W.W. performed
phenotypic assessment of family members. P.W. and M.B.D. carried out the
genetic linkage analysis. T.L. and H.H. carried out the in situ hybridization.
T.L. prepared the pathology material to be analyzed by D.A.H., J.H. and
T.R. D.H. and A.B.S. contributed to the writing of the manuscript and
carried out whole-genome arrays. H.H. and N.W.W. wrote the manuscript.
Published online at http://www.nature.com/naturegenetics
Reprints and permissions information is available online at http://npg.nature.com/
1. Schols, L. et al. Lancet Neurol. 3, 291–304 (2004).
2. Harding, A.E. Brain 105, 1–28 (1982).
3. Ranum, L.P. & Day, J.W. Curr. Opin. Genet. Dev. 12, 266–271 (2002).
4. Orr, H.T. & Zoghbi, H.Y. Annu. Rev. Neurosci. 30, 575–621 (2007).
5. Worth, P.F. et al. Am. J. Hum. Genet. 65, 420–426 (1999).
6. Baker, M. et al. Nature 442, 916–919 (2006).
7. Riudavets, M.A. et al. J. Neuropathol. Exp. Neurol. 65, 1143–1148 (2006).
8. Mann, D.M. et al. Neuropathol. Appl. Neurobiol. 16, 17–25 (1990).
9. Mukaetova-Ladinska, E.B. et al. Am. J. Pathol. 143, 565–578 (1993).
10. Ballatore, C., Lee, V.M. & Trojanowski, J.Q. Nat. Rev. Neurosci. 8, 663–672
11. Takahashi, M. et al. FEBS Lett. 372, 59–64 (1995).
12. Sato, S. et al. J. Neurochem. 98, 1573–1584 (2006).
13. Kitano-Takahashi, M. et al. Acta Crystallograph. Sect. F Struct. Biol. Cryst. Commun.
63, 602–604 (2007).
14. Noble, W. et al. Proc. Natl. Acad. Sci. USA 102, 6990–6995 (2005).
15. Kraemer, B.C. et al. Hum. Mol. Genet. 15, 1483–1496 (2006).
Figure 2 In situ hybridization and immunohistochemistry. (a–c) In situ
hybridization using TTBK2 antisense probes shows high expression in
normal human cerebellum (?10) (a), hippocampus (?10) (b) and substantia
nigra (?20) (c) (light microscopy). (d–m) Immunohistochemistry data for
affected brain tissue of individuals from family 1, showing severe cerebellar
atrophy with loss of Purkinje cells and granule cells (d) but preservation
of the basket fibers (f), and also severe nerve cell loss in the dentate
nucleus (h). For comparison, the same regions from an age-matched
control individual are shown in e,g,i. d,e,h,i: hematoxylin and eosin;
f,g: neurofilament immunohistochemistry. Bar in e represents 50 mm on
d–i. Neurofibrillary tangles, neuropil threads and sparse tau-positive grains
are seen in the CA1 hippocampal subregion (j). Scattered neurofibrillary
tangles are also seen in the substantia nigra (k) and medullary tegmentum
(l), and tau-positive neuropil threads are seen in the medullary tegmentum
(m). j–m: tau immunohistochemistry (with AT8 antibody). Bar, 50 mm
on j,l,m and 15 mm on k.
1436 VOLUME 39 [ NUMBER 12 [ DECEMBER 2007 NATURE GENETICS
© 2007 Nature Publishing Group http://www.nature.com/naturegenetics