Biomedical Engineering Reference
In-Depth Information
pacemakers to the site of injection by cell fusion to cardiomyocytes. Therefore, the
cell fusion approach allows us to create a biological pacing at a specific site by design
rather than by chance.
6 Gene Transfer of a Synthetic Pacemaker Channel into the Heart
HCN family of channel genes figure prominently in physiological automaticity [18],
and transfer of such genes into quiescent heart is the most obvious way of creating a
biological pacemaker. However, use of HCN genes may be confounded by
unpredictable consequences of hetero-multimerization with multiple endogenous
HCN family members in the target cell [8,72]. As
I
f
is expressed in ventricular
myocytes and can contribute to arrhythmogenesis [13, 29], HCN gene transfer in vivo
may have unpredicted consequences. Similarly, little flexibility with regard to
frequency tuning would be achieved if the engineered pacemaker channel co-
assemble with wild-type channels upon transduction. A synthetic pacemaker channel
(SPC) with no affinity to co-assemble with HCN channels would circumvent these
limitations inherent with HCN gene transfer. To this end, we exploited accumulated
knowledge in the biophysical properties of Shaker K
+
channels. First, depolarization-
activated Shaker K
+
channels had been shown to convert to a hyperpolarization-
activated inward rectifier by mutating three amino acid residues in the voltage sensor
(S4) of the channel [49]. Furthermore, amino acid residues in the selectivity filter of
Shaker K
+
channels were found to lose its specificity and conduct Na
+
as well as K
+
when mutated to certain residues [25]. We combined the lessons from the two studies
on Shaker K
+
channels and applied them to the human homologue, Kv1.4 channels
[34]. By targeted mutagenesis involving <1% of the protein sequence, we were able to
convert the depolarization-activated, potassium-selective human Kv1.4 channel into a
hyperpolarization-activated, nonselective cation channel suitable for biological pacing
applications. These mutations are comprised of three point mutations (R447N,
L448A, and R453I) in the S4 segment and a single mutation (G528S) in the signature
sequences of the pore's selectivity filter (Fig. 4a). The SPC channel activation by
hyperpolarization and permeation by both K
+
and Na
+
ions were remarkably similar to
the gating and selectivity properties of HCN channels (Fig. 4b).
An absence of heteromultimeric interactions with endogenous HCN-channels is a
prerequisite to the use of such SPCs. As for the wild-type, Kv1.4 channels have
previously been reported not to multimerize with the HCN channels [80]. In addition,
when co-transfected into HEK293 cells, SPC did not multimerize with HCN1 as
assayed by reversal potential measurements.
Next, we sought to examine in vivo pacemaker capability of SPC by creating
bicistronic adenovirus expressing SPC and GFP as a reporter (AdSPC) and injected
it epicardially into a guinea-pig heart. Whole-cell voltage-clamp recordings of
isolated myocytes transduced with AdSPC revealed robust hyperpolarization-
activated, inward currents in the presence of 0.5 mM BaCl
2
to eliminate
I
K1
(Fig. 4c
left panel). The myocytes transduced with control adenovirus (GFP-alone) produced
little inward current under identical conditions. More importantly, action potential
recordings clearly demonstrated spontaneous action potentials from the AdSPC-
transduced myocytes (Fig. 4c right panel, n = 7 out of 14), but not from the
