JM Polke1, K Mok2, M Poulter3, T Lashley4, J Beck3, NW Wood2, J Hardy2, MG Sweeney1, A Haworth1, H Houlden2

1
Diagnostic testing for the frontotemporal
dementia / amyotrophic lateral sclerosis
(GGGGCC)n expansion in C9orf72
JM Polke1, K Mok2, M Poulter3, T Lashley4, J
Beck3, NW Wood2, J Hardy2, MG Sweeney1, A
Haworth1, H Houlden2
Introduction
PCR Validation and Diagnostic Testing

Sequencing of sizing PCR products from samples
homozygous for 2, 5 and 11 repeats confirmed
amplification of the target sequence (data not
shown). The sizing and RP-PCR protocols were also
validated by a blind test of 28 samples of known
genotype, including 4 expansion positives, all
samples were correctly genotyped. Examples of
sizing and RP-PCR electropherograms are shown in
figure 3.
Frontotemporal dementia (FTD) and Amyotrophic
Lateral Sclerosis (ALS) are both progressive
fatal neurodegenerative disorders. Degeneration
of the frontal and temporal lobes in FTD results
in changes in personality, behaviour and
language, while ALS is traditionally
characterised by degeneration of upper and lower
motor neurones that causes progressive
spasticity, muscle wasting and weakness. However
it is now recognised that these disorders can
represent a clinical spectrum patients can
present with symptoms of both disorders, and both
diseases can be found in autosomal dominant
pedigrees, leading to the description of
frontotemporal dementia and/or amyotrophic
lateral sclerosis (FTD/ALS). A positive family
history is seen in approximately 10 of ALS and
50 of FTD, though until recently mutations in
all the known genes collectively accounted for
25-50 of familial cases. Since 20001 a number of
FTD/ALS linkage and genome-wide association
studies have implicated a region of 9p21, and two
groups recently described a pathogenic (GGGGCC)n
repeat expansion in intron 1 of a transcript of
the C9orf72 gene at this locus (Figure 1)2,3. The
repeat range published in normal individuals to
date is 2-27 GGGGCC repeats (2 repeats being the
most common allele), pathological expansions
consists of 700-1600 repeats, though the
expansion has only been sized in one pedigree
to-date3. We collaborated in a large
cross-sectional study analysing the (GGGGCC)n in
a cohort of 800 patients and 1000 controls from
different populations, the frequencies of
familial and sporadic FTD/ALS were congruent with
other literature4 the expansion accounts for
approximately 23 and 10 of familial and
sporadic cases, respectively. To date there has
been a single report of a non-penetrant expansion
in one pedigree (patient aged 70)5 and worldwide
the expansion has been detected in 5/2500
asymptomatic controls. C9orf72 encodes a gene of
unknown function, it has been speculated that the
expansion may cause disease via
haploinsufficiency or 'toxic-RNA' gain of
function2,3. We have developed diagnostic PCR and
Southern-blot protocols for the (GGGGCC)n C9orf72
repeat expansion. Our experience of the
validation process and the first 4 months of
testing is described below.
(A) Sizing PCRs
(B) RP-PCRs
Figure 3 Sizing PCR and RP_PCRs (A) Sizing
PCRs showing samples heterozygous for 4 and 11
repeats and 10 and 21 repeats. (B) RP-PCRs
showing an expansion positive and a sample with
10 and 21 repeats. The 2-3 rpt peak is a constant
peak in all RP-PCRs generated by primer R1
annealing to 5' end of the repeat sequence. The
18-bp 'clamp' sequence accentuates the
amplification of the largest alleles (within the
non-expanded range).
A total of 12 Diagnostic PCRs have been performed
since May 2012, identifying 3 expansions. All
samples tested to date except one gave concordant
results between sizing and RP-PCRs for unexpanded
alleles. One sample showed allele sizes of 2 and
10 repeats by RP PCR and 2 and 12 repeats by
sizing PCR (figure 4). This indicated a 12-base
pair insertion 3' of the (GGGGCC)n repeat, which
was confirmed by sequencing.
Figure 4 Discrepant Allele Sizes The Sizing
PCR and RP-PCR results for one sample
consistently gave discrepant allele sizes in
several repeat PCRs 2 and 12 repeats by sizing
PCR and 2 and 10 repeats by RP-PCR. The sizing
PCR was repeated and run on an agarose gel, the
larger PCR product was extracted form the gel and
sequenced, confirming a 12 base-pair insertion 3'
of the (GGGGCC)10 repeat.
Southern Blotting
DeJesus-Hernandez et al3 published a Southern
blot protocol using an XbaI restriction digest
that generates a 2.4 kb 'normal' band on Southern
blotting. Blots from somatic tissue expansions
resulted in faint smeared bands. For our
diagnostic testing we have optimised a method
using a BglII restriction digest, which results
in a 12 kb. Bands for expanded alleles are
considerably compressed and easier to identify,
though accurate sizing is not possible (figure
5). Further development is ongoing.
Figure 1 C9orf72 Gene and transcripts C9orf72
spans 27.3 kb. The two main transcripts are
shown, though others are predicted. The (GGGCC)n
repeat is located between the alternatively-splice
d 5'UTR exons 1a and 1b and is transcribed as
part of NM_145005.4. NM_145005.4 and NM_018325.2
encode 222 and 481 amino-acid proteins,
respectively. Figure adapted from reference 3.
Diagnostic PCRs
Figure 5 Southern blot One normal control (N)
and two expansion-positives (E1 and E2) were
digested with BglII (5 ug DNA from blood). After
electrophoresis (0.8 agarose), DNA was
transferred to a membrane and probed with a
c9orf72 locus-specific digoxigenin (DIG)-labelled
probe. Probe hybridisation and detection was
performed according to manufacturer's protocols
(Roche DIG System). Non-expanded bands are
visible at 12 kb. The expanded alleles are
visible at approximately 20 kb. We are continuing
to optimise the protocol using different
electrophoresis/restriction enzyme/ probe
combinations so that we can more accurately size
the expansions.
The diagnostic test involves a 'sizing PCR' and
repeat-primed PCR (RP-PCR) across the
hexanucleotide repeat using published primers3
(figure 2). PCR conditions have been modified
from published protocols, both use the Thermo
Scientific Extensor Hi-Fidelity PCR Master Mix.
The sizing PCR also includes 5 DMSO, 25 Qiagen
Q-solution, 0.2mM 7-deaza-dGTP and 1mM MgCl2. The
RP-PCR includes 1.9 M betaine, and 0.33µM FWD,
0.33µM R1, 0.033µM R2 primers. PCRs are
electrophoresed on an ABI 3730xl capillary
analyser.
Summary and Conclusions

• We have developed a robust PCR method for
  detection of the (GGGCC)n expansion in C9orf72.
  Development of our Southern blot protocol is
  ongoing.
• During PCR validation one sample was identified
  with an unreported 12-bp insertion adjacent to
  the (GGGGCC) repeat, indicating that other
  mutations may be detected at this locus.
• This test will greatly benefit ALS/FTD patients,
  families and clinicians. The optimised Southern
  blot protocol will provide invaluable data on the
  variation of pathogenic (GGGGCC)n repeat sizes.

Figure 2. Sizing and RP-PCRs Sizing and RP-PCRs
use a common forward primer. For a 2-repeat
allele, the sizing PCR generates a 132 bp
fragment, the RP-PCR generates a 116 bp fragment.
The RP-PCR reverse primer 'R1' includes 3 x
GGGGCC repeats which allow it to bind anywhere in
the repeat tract, plus an 18 bp 'clamp' sequence
which binds adjacent to the repeat. The clamp
sequence encourages amplification of the entire
repeat.
1 Hosler et al (2000). JAMA. 2841664-9 2
Renton et al (2011) Neuron. 72257-68 3
DeJesus-Hernandez et al (2011) Neuron. 72
245-56 4 Majounie et al (2012) Lancet
Neurology. 11 32330 5 Cooper-Knock et al
(2012) Brain 135 75164
Refs

1. Neurogenetics Unit, National Hospital for
   Neurology and Neurosurgery, Queen Square, London
2. Department of Molecular Neuroscience, Institute
   of Neurology, UCL, Queen Square, London
3. MRC Prion Unit, Institute of Neurology,
   University College London, Queen Square, London
4. Queen Square Brain Bank, UCL Institute of
   Neurology, 1 Wakefield Street, London