116
munohistochemical analysis, we demonstrated for the first time that the HCNl isoform is
likely the ion channel that underlies ∕⅛ in both rods and cones. We also performed the tech-
nically challenging experiment of estimating the small single channel conductance of HCNl
channels in photoreceptors (section 3.2.3), and show that it is similar to the conductance of
HCN1 channels in other systems— around 660 fS. These studies, along with molecular and
genetic evidence from amphibians (section 3.2.2) suggest that HCN channels are, like the
voltage-gated sodium channel, a highly conserved building block in physiology. Our stud-
ies on the function of HCN channels in photoreceptors (section 3.3) show how they act to
speed up the light response of rod and cone photoreceptors at the expense of decreased gain
at low temporal frequencies. They demonstrate that HCN channels are more important for
rod function at dimmer light intensities, and more important in cones at brighter light in-
tensities.
Next, we combined the information learned about HCN channels with parameters from
other photoreceptor ion channels to form functional physiology-based models of rod and
cone photoreceptors. Based on the results of the rod model, we demonstrate that the time
dependent changes in h and Kx conductance largely cancel one another (section 4.1.2). As
a result, change in rod membrane impedance during a light response is minimized, while
simultaneously preserving the high-pass filter characteristic important for the rod temporal
response. We propose that the stable membrane impedance may be beneficial for signal
averaging through rod-rod coupling by minimizing distortion in lateral signal propagation
from rod-rod coupling.