Engineering Ion Channels for Selective Neuronal Activation and Silencing

1
Henry Lester June 2009
Engineering Ion Channels for Selective Neuronal
Activation and Silencing
2
Neuronal Engineering with Cys-loop Receptor
Channels
Goal develop a general technique to selectively
and reversibly silence or activate specific
sets of neurons in vivo.
Ideal approach would Have on- and off- kinetics
on a time scale of minutes Have simple activation
(ie, via drug injected or in animals diet) Avoid
nonspecific effects in animal Maintain target
neurons healthy in an off-state for a few days
without morphological/other changes Silence or
activate diffuse molecularly defined sets of
neurons, not just spatially defined groups
The chosen channel Cys-loop receptor (like
nicotinic receptors)
Heteropentamer a2ß3 or a3ß2 subunits.
This feature allows one to intersect two
promoters, to enhance cellular specificity
3
The channelohm is 2 of the human genome,
and many other organisms expand the repertoire
Voltage (actually, ?E 107 V/m) External
transmitter Internal transmitter Light Temperature
Force/ stretch/ movement Blockers
1/r 0.1 100 pS
Nernst potential for Na, K, Cl-, Ca2, H
Invertebrate glutamate-gated Cl- channel . At
this resolution, resembles nicotinic
acetylcholine receptor
4
The drugs avermectins

• IVM Lactone originally isolated from
  Streptomyces avermitilis
• AVMs are used as antiparasitics in animals and
  humans (River blindness / Heartgard)
• IVM is probably an allosteric activator of GluCl
  channels
• Also modulates GABA, 5HT3, P2X, and nicotinic
  channels, at much higher doses

(IVM)
5
IVM-induced silencing in GluCl-expressing
cultured rat hippocampal neurons
5 nm IVM
500 nm IVM
50 nm IVM
6
We eliminated glutamate sensitivity in GluCl by
mutating a conserved aromatic residue
Colored by subunit (chain)
7
Excessive variability among culture dishes
8
Optimized constructs optGluCla,b AVMR-Cl

• Binding site
• subunit unmutated b Tyr182Phe (cation-p site)
• suppresses endogenous glutamate sensitivity
• M3-M4 intracellular loop a YFP b CFP
• allows visualization
• Coding region codons adapted for mammalian
  expression
• 10-fold greater expression

9
AAV-2 constructs injected into mouse striatum
slice experiments Single neurons correlation
between IVM-induced conductance AP silencing
Lerchner et al, 2007 (collaboration with D. J.
Anderson at Caltech)
10
Plans to extend the AVMR system
Transfer AVM sensitivity to mammalian glycine
receptor ? no immune response
Tighter AVM binding ? increased AVM sensitivity
M2 mutations ? increased AVM sensitivity

• Na-permeable
• selective neuronal activation

• Ca2-permeable
• manipulate signal transduction

Increased single-channel current ? increased AVM
sensitivity
Optimize ER exit and trafficking ? increased
surface expression
M3-M4 loop
11
Very slow (several hr) AVM reversibility is
puzzling
GluCl-?? heteromer
GluCl-? homomer
No potentiation
GluCl-? homomer
(Etter et al., JBC 1996)
12
Location of the AVM binding site is unknown
Likely distinct from the glutamate binding site
Covalent binding interaction?
At the ECD-TMD interface?
Within the cavity of the TMD?
(where other anesthetics are bind)
Yoav Paas, BIU
McCammon Lab, UCSD
Radioligand binding experiments with 3H-IVM on
C. elegans membrane preps IVM binding sites
exhibit high affinity binding (KD 0.11 nM) IVM
does dissociate from its receptor, with a rate
constant of 0.005-0.006/min
(Cully Paress, 1991)
13
The first AVMR-Na
GluCl ? WT ? WT
Muscle nAChR
(10 nM IVM)
GluCl ? P(-2)?/A(-1)E ? WT
(200 nM IVM)
14
Many AVMRs remain in intracellular compartments,
but are chaperoned by IVM

(GluClaYFP)GluClß 24 h incubation (control
solution) The intensity ratio,
peripheral/whole cell, is 0.86 0.07 in control
and 1.51 0.10 in IVM-treated cells (SEM
(1 µM IVM)
Confocal TIRF
15
First tests HEK cells
16
Fluorescent Labels in the M3-M4 loop, function is
retained
A
a, YFP b, CFP
ab
a ab
(FRET shows that the subunits co-assemble)
17
b Y182F eliminates glutamate responses but
retains IVM responses
1 mm Glu 1 mM IVM