CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 6, July 2013
234
AFRICA
dependence of the
I
Ks
channel activation, the p.A344V mutation
increased the sensitivity of the channel for bupivacaine (a local
anaesthetic).
34
KCNQ1:p.T587M was found to result in haplo-
insufficiency as a consequence of being transport deficient.
35,36
Furushima
et al
. (2010) identified a case of foetal atrio-
ventricular block and unmasked maternal QT prolongation in the
postpartum period in a mother and baby carrying this mutation.
37
KCNH2 mutations
KCNH2:c.917-3T
>
C has not been previously reported; however,
splice-site analysis suggests that the tentative consequence of
this variant is disruption the splice acceptor of exon 5. Such
a splice-site disruption would be expected to cause aberrant
mRNA splicing and result in either the synthesis of a truncated
protein or, more likely, haplo-insufficiency. In both cases, the
result is likely to be a reduced amount of Kv11.1 protein and
therefore reduced repolarisation capacity.
KCNH2:p.R100W was identified in two of 2 500 patients
reported by Kapplinger
et al
. in 2009, however, no specific
clinical information is provided.
5
A mutation at residue 100
(p.R100G) has been reported in a 41-year-old French woman
who had suffered an aborted cardiac arrest,
38
which may indicate
that the R100 residue has a particular functional relevance.
The p.G572S mutation has been reported in LQTS cohorts
around the world
4,5,23,25,39-44
but has not been identified in an exome
sequencing project to date. Zhao
et al
. in 2009 determined
that p.G572S causes a dominant negative trafficking defect.
44
KCNH2:p.F627L was first identified by Splawski
et al
. in 2000
as part of his LQTS screen of 262 probands
45
and later was
reported to be the cause of LQTS with foetal onset of atrio-
ventricular block and ventricular tachycardia.
46
Double heterozygotes
In addition to the KCNQ1:p.A341Vmutation, we have confirmed
the presence of the KCNE1:p.D91E variant in a single proband.
The clinical significance of this variant is unclear. It is a very rare
variant, identified in a single African-American individual in an
exome analysis of 4 406 African-American individuals (http://
evs.gs.washington.edu/EVS/) (date accessed July 2012).
A more extensive analysis of the phenotypes in the family,
where the clinical effect of individual mutations can be assessed
as the
KCNQ1
mutation is located on chromosome 11 and the
KCNE1
-encoded MinK variant is located on chromosome 2,
14
is
necessary to establish the clinical significance of the compound
heterozygozity. Due to the rarity of the MinK variant and the
known association between mutations in the N-terminus of
MinK and adverse drug effects, we consider it a significant
finding.
In a second proband, we identified both the KCNQ1:p.A341V
mutation and KCNH2:p.R328C. The p.R328C variant was first
reported by Chevalier
et al
. (2001) in an acquired LQTS cohort.
47
Grunnet
et al
. (2005) described a cLQTS patient harbouring
double mutations (KCNQ1:pR591H; KCNH2:p.R328C). They
determined that p.R328C did not produce a functional phenotype
and that KCNQ1:p.R591H was sufficient to explain disease.
48
Further functional assessment of KCNH2:p.R328C by
Anderson
et al
. (2006) determined the Kv11.1 channels carrying
p.R328Cwere normally trafficked to themembrane and conferred
no functional phenotype.
39
However, a subsequent report by
Chevalier
et al
. (2007) suggested that p.R328C had a dominant
negative effect on
I
Kr
. Finally, Kapa
et al
. (2009) identified
KCNH2:p.R328C in a control individual.
40
The p.R328C variant
is, in all likelihood, a rare polymorphism; however, we cannot
exclude that it has a functional effect in the presence of certain
drugs or conditions, e.g. hypokalaemia.
49
Apart from the variants described above and reported in
Table 2, a number of synonymous (i.e. not having any effect on
the amino acid sequence) variants as well as variants that have
been found with a relatively high frequency (i.e.
>
1%) in other
populations were found during the screening of the 44 probands
(data not shown). Such polymorphisms are not disease causing,
but they may modify the phenotype.
4
These potential
forme fruste
mutations, i.e. mutations that do not appear to cause disease in
isolation, should be noted.
Numerous non-synonymous polymorphisms have been
reported to be associated with an effect in cardiac repolarisation
currents.
50
KCNE1:p.D85N has been implicated in drug-induced
LQTS
30
and KCNH2:p.R1047L has been reported to reduce
I
Kr
in
a mammalian cell-based system.
51
Occasionally, the presence of very rare variants in the normal
population (e.g. the KCNE1: p.D91E mentioned above) that
have not previously been associated with disease and have a
population frequency less than 1:4 000 may raise questions
about their possible role in disease causation. In such a case,
family studies, which may not be possible in small families, and
electrophysiological analysis, which is costly and time consuming,
may be necessary to ascribe disease causation with certainty.
Consequently, novel mutations where the disease association
has not been established should be evaluated carefully and the
uncertainty should be considered when deciding the appropriate
management of the family, particularly as prophylactic treatment
with beta-blockers or implantation of an ICD unit can have
adverse effects.
To establish a genetic diagnosis in a cLQTS family is in most
cases complicated, as described above, and is best done as a
collaborative effort involving cardiologists, clinical geneticists
and molecular geneticists. The findings reported here suggest
that most South African cLQTS cases will be caused by
mutations already described in other populations, as we only
found two novel mutations. However, this probably reflects the
fact that most of the patients included have a European ancestry
and that to date no black South African LQTS patients have been
encountered for enrolment in the study.
Furthermore, no other African-based research groups have
reported the occurrence of LQTS patients within their black
African populations.
52
It must be expected, due to the large
genetic diversity of African populations, as is evident from the
Exome Sequencing Project (
),
that comprehensive genetic screening of at least five genes will
be necessary to provide an adequate genetic diagnosis in these
populations.
A limitation of this study was a failure to detect potential gene
rearrangements, such as large duplications or deletions. Multi-
CE-SSCP can only identify point mutations or small insertions
and deletions in coding regions or at splice junctions. Koopman
et al
. detected a large duplication in the
KCNH2
gene, which
they determined to be responsible for cLQTS in a Dutch family.
53
No mutation has so far been identified in approximately 10–20%
