Second Messenger-gated Ion channels

1
Second Messenger-gated Ion channels
Dr. Debra Ann Fadool 18 February 2005

• CNG Ion Channels
• Matulef and Zagotta. 2003. Cyclic
  nucleotide-gated
• Ion channels. Annu. Rev. Cell Dev. Biol. 19
  23-44
• Zagotta Laboratory Stoichiometry and Assembly
• Krammer Laboratory Modulation
• Martens Laboratory Lipid Raft Localization

2
Classic Phototransduction Cascade Main Player
is the CNG ion Channel
What events occur to produce the dark current
and subsequent depolarization? What events
change to create light adaptation and
hyperpolarization?
3
Family of CNG Channels
Olfactory CNGA2, CNGA4, CNGB1b (2, 4, b) Rod
CNGA1, CNGB1 Cone CNGA3, CNGB3 Fly CNGL,
CNG-PA C. Elegans TAX-2, -4 Six Family Members
4
Structural Features of CNG ion channels

• Pore
• CNBD
• C-linker
• N-terminal
• Domain
• Post CNBD
• Region
• Modulation

5

• Pore
• Has the TVGYG K channel signature sequence.
• Modeled after KscA, is thought to have
  water-filled
• vestibule intracellular to the selectivity
  filter.
• Cations would be stabilized between the pore
  helices and
• water molecules that hydrate the cation in the
  vestibule.
• Putative inner helix is thought to be S6 that
  exhibits
• conformational changes to open the pore. It is
  thought to
• widen during nucleotide binding but is not the
  physical gate
• that would control permeation.
• Evidence for a intersubunit disulfide bond that
  would
• form spontaneously for rapid closure of the
  channel in the
• absence of sufficient ligand concentration.
• Unlike other channels we have studied, has a
  specific blocker
• that has a higher affinity for the closed channel
  state
• Tetracaine.

6
Secondary Structure motifs Similar in CNBD and
CAP Yellow Identical Green Conserved Blue
others Red the cAMP molecule

• B. CNBD
• Several different types of nucleotide binding
  proteins one of
• which is used as a model for nucleotides binding
  to this domain.
• a. PKA
• b. PKC
• c. CAP catabolite gene activator protein in
  bacteria
• Channel Activation VIA Concerted Allosteric
  Opening Transition
• First described by Monod Model to show that
  independent binding
• of the nucleotides (2) stabilizes a concerted
  opening. Energetics may
• vary according to the number of ligands bound.
• Distinct Order of Specificity Structures differ
  by side groups
• on the purine ring
• cGMP gtgtgtgtgtgtgt cIMPgtgtgtgt cAMP

7
Activity of Nucleotides cGMP, cIMP, and cAMP -
O
C
Amplitude Histogram Area o / Area o Area c
Propen Plot the pA for a number of Vc to derive
the IV relation slope of the relation is the
slope conductance in pS.
All three nucleotides can bind to the CNBD but
the allosteric opening transitions vary as a
reflection in different Propen Propen increases
with increased of ligands bound but the
bound of course is dictated by affinity.
8
The Molecular Basis for Ligand Specificity.?

• T560 is conserved in CNGA1 mutation affects
  cGMP affinity but
• not cAMP.
• But cant be only source for specificity because
  oCNG have
• affinity for cAMP cGMP.
• D604M mutation made the selectivity reverse
  order
• cAMPgtgtgtgtcIMPgtgtgtgtcGMP.
• Model again is taken from CAP crystal structure
  Think that the
• C-helices move toward the B roll of each subunit,
  allowing the D604
• residue to interact with the purine rings of the
  bound cyclic
• nucleotide.

9

• The C Linker
• The residues of the linker are modulated by
  metals.
• Three residues of the linker can affect gating
• R460, I465, and N466.
• D. N-Terminal Domain
• Stabilizes the open state by decreasing the free
  energy (delta G)
• of gating Called the autoexcitatory effect on
  gating.
• Ca/Cam binding to the N-terminal of CNGA2 causes
  a decrease
• in Propen.
• Olfactory adaptation negative feedback of Ca
  (permeant ion)
• for Ca/Cam that inhibits the N-terminal domain
  of the channel.
• N and C terminus interact directly Ca/Cam
  prevents this
• interaction required for gating, and causes
  decrease
• cAMP/cGMP from binding.

10

• Post-CNBD Region
• Also mediates the Ca/Cam modulation/inhibition.
• Important for trafficking and heteromeric
  assembly
• RP truncated mutation in this region.
• Types of Modulation
• Ca/Cam
• Metals
• PKC
• DAG
• Na/Ca K exchangers in protein-protein
  interatactions
• Role of circadian rhythms

11
Paper 1 Zheng and Zagotta. 2004. Stoichiometry
and Assembly of Olfactory Cyclic
Nucleotide-gated Channels. Neuron 42 411-421.
Key Finding CNG channels in olfactory neurons
are tetramers of a fixed, non-random (precise)
stoichiometry 211 (CNGA2, CNGA4, and
CNGB1b). Secondary Finding CNGA4 and CNGB1b
have a higher affinity for CNGA2 than for self
assembly Conclusion Extramembranous
intersubunit interactions promote assembly from
the N-C or C-linker interactions.
12

• Background
• K channel assembly of subunits is random, whereas
  that for
• ligand-gated ion channel (AChR) was known to be
  fixed to promote
• ligand affinity.
• Knew that CNGA2 (2), CNGA4 (4), and CNGB1b (b)
  were expressed
• in native olfactory neurons but did not know
  ratio.
• New that 2 could form functional homomeric
  channels and that
• 4 and b would only express if also with 2.
• 4. 2 by itself did not exhibit native biophysical
  properties of the
• CNG olfactory channel.
• Properties they analyzed
• Functional expression
• Activation by cAMP
• Block by ditiazem
• Ca/Cam modulation

13
4 Biophysical Properties
Functional
Ligand
Blocker
Modulator
Combinations All 2 2 4 2 b 2 4 b
(closest to native)
Red Low cAMP Black High cAMP Green Plus
Blocker
Time course of current inhibition due to
Ca/Cam modulation fit with single exponential.
Fast Modulation most like Native
14
Fluorescence Intensity Ratio Method to Determine
Stoichiometry FIR
Using Rod CNG as a control

• 458 nm CFP / 488 mm YFP
• Assumption that unassembled
• channels would not be membrane
• inserted.
• C-terminal tagged constructs
• where CFP green and YFP yellow
• Since two dye molecules so
• close in proximity must subtract
• any FRET-induced decrease in the
• intensity of the donor dye (green)
• Show linear regression of actual
• data (red line), FRET correction is
• computed to show very little
• difference (green line), and solid lines
• (black) are predicted based upon
• logical potential ratios.

GREEN
YELLOW
15
Now are using the olfactory combinations after
optimization of the protocol. A vs. B
switching green and yellow tagged constructs for
2 and 4 demonstrates in really 31. C Fit
mathematical models for different subunit ratios
based upon RNA ratios.
green
yellow
16
Same experimental protocol Now 2 and b instead
of 2 and 4. Same 31 results BUTgtgtgtgtgtgt Ratio
in the membrane does not Necessarily The
Ratio of subunits in the physical channel
therefore. Must use FRET to determine
interaction distance and channel stoichiometry.
17
Fluorescence Resonance Energy Transfer (FRET)

• Is a distance-dependent interaction between the
  electronic excited states of two dye molecules in
  which excitation is transferred from a donor
  molecule to an acceptor molecule without emission
  of a photon.
• The efficiency of FRET is dependent on the
  inverse sixth power of the intermolecular
  separation, making it useful over distances
  comparable with the dimensions of biological
  macromolecules.
• Spatial resolution beyond the limits of
  conventional optical microscopy.

18
Requirements 1. Donor and acceptor molecules
must be in close proximity (typically 10100 Å).
2. The absorption spectrum of the acceptor
must overlap the fluorescence emission spectrum
of the donor (J). 3. Donor and acceptor
transition dipole orientations must be
approximately parallel.
4. The distance at which energy transfer is 50
efficient (i.e., 50 of excited donors are
deactivated by FRET) is defined by the Förster
radius (Ro). The magnitude of Ro is dependent on
the spectral properties of the donor and acceptor
dyes
Donor 458 CFP green Acceptor 488 YFP
yellow Ratio A Excitation of YFP (488) by CFP
(458) at 458 nm laser
19
FRET between 4 or b in the Presence of 2
subunits.
FRET between the 2 subunits in presence of 4 or
b.
Ratio Ao excitation of acceptor YFP in control
oocytes when only 2 YFP and no b subunits A
Ao FRET efficiency greater it is means
closer together in physical distance.
Donor 458 CFP green Acceptor 488 YFP
yellow Ratio A Excitation of YFP (488) by CFP
(458) at 458 nm laser
20
In this example 2 that is CFP green, a 2 that
is YFP yellow, and then add in an unlabeled b.
Black the total spectrum Red 458 green
(donor) Blue Control Background Without
b. Green difference spectra Ratio A
green/black
Take Home Concept! If High Ratio A Ratio Ao,
then occurrence of FRET indicates two or more
copies of a particular subunit of the channel or
an interaction across the subunits.
21
31 Stoichiometry if you add either 4 or b to 2.
211 Stoichiometry if you add both 4 and b to 2.
22

• Testing Functional Expression of Homomeric
  Channels
• In terms of current (A) and surface expression
  (B)
• Another demonstration that only 2 shows FRET with
  itself,
• homomeric, whereas 4 and b do not.

23
Intersubunit Interactions Promote Assembly in the
Absence of a T1
Between 2 and either 4 or b high affinity
binding (see dark lines). Therefore
decreases frequency of 4 and b binding as a dimer.
Two stage First 2-4 and 2-b dimers form Then
the dimers assemble into the heteromeric
channels Uses a head-tail arrangement using the
N and C terminal interacting domains
24
Paper 2 Krajewski et al. 2003. Tyrosine
Phosphory- Lation of Rod Cyclic
Nucleotide-gated Channels Switches off Ca/Cam
Inhibition. JNS 23(31) 10100-10106.
Key Finding Y498 in the CNGA1 is the
phosphorylation Site responsible for Ca/Cam
Inhibition. Secondary Finding Y Phosphorylation
of CNGA1 on the C- terminus can cause an
uncoupling of the N-terminus of CNGB1 so that
there is no Ca/Cam modulation. Conclusion Y
Phosphorylation decreases Propen whereas
Dephosphorylation increases Propen.
25

• Background
• It was known that Ca/Cam binds with high affinity
  to the N-terminus
• of the CNGB1 subunit to weaken the intramolecular
  interaction
• Between the N and C termini of CNGB1 and CNGA1.
• Y498 is on CNGA1 and Y1097 is on CNGB1 either
  mutation causes
• a decreased affinity for cGMP to decrease gating.
• IGF causes dephosphorylaton of these sites to
  increase
• CNG sensitivity.
• What the authors wish to address..do
  phosphorylation and
• Ca/Cam act as separate, independent modulators of
  the channel or is
• there a common mechanism?
• 6. I-O patches of transfected oocytes,
  spontaneously dephosphorylate
• the channels, so the Propen increases over a 5-10
  minute period.
• 7. During dePhos K1/2 decreases approximately
  2x change,
• reflecting an increase in cGMP affinity.

26

• Pervanadate (to keep phosphory-
• lated) and non-pervanadate
• conditions - demonstrates that
• more cGMP is needed to get same
• response when the channel is
• phosphorylated.
• In the absence of Pervanadate
• (no phosphorylation), Ca/CaM
• need more cGMP to get same response.

Note D/R curves are fit with the Hill
coefficient that indicates slope of cGMP
molecules binding remains as two (no
slope change)
27

• ATP gamma S is non-hydrolyzable
• therefore do not get spontaneous
• dephosphorylation shift over time.
• Now if try and modulate with
• Ca/CaM.fails to have an affect if
• retained phosphorylation.

28
Are these repetitive experiments? Why or Why
Not? What do they add?
29
Look at Table 1 K1/2 values for cGMP activation
what pairs of values give the greatest clues
about site-directed mutation function?
Now in Native rod CNG channels When
Phosphorylated, Ca/CaM has NE Why do they use
Population Histograms?
What do these values tell you ?
30

• Mechanism of Dual Modulation by Phosphorylation
  and
• by binding Ca/Cam
• A. Ca/CaM binding to the B1 to break interaction
  of N-C
• B. Phosphorylation of A1 pulls C terminus away
  so that
• there is no N-C interaction for Ca/Cam to disrupt
  when it
• binds to the B1 N terminus.
• Phosphorylation now on B1 C-terminus, can still
  allow
• Ca/CaM to bind B1 on N-terminus to have
  inhibition.

31
Paper 3 Brady et al., 2003. Functional Role of
Lipid Raft Microdomains in Cyclic
Nucleotide- Gated Channel Activation. Mol.
Pharm. 65 503-511.
Key Finding Movement of CNGA2 into lipid raft
domains can change function of the channel by
increasing affinity of cAMP. Secondary Finding
Heterologously expressed and native CNGA2 in
olfactory tissue are expressed as a fraction in
the lipid raft domains. Conclusion Lipid
lowering drugs could affect olfaction
via functionally altering the biophysics of the
CNGA2 (but they do not have the correct
stoichiometry..?)
32

• Background on Lipid Rafts
• 1. Rich in sphingolipids and cholesterol
• Act to concentrate certain membrane
• proteins, signal transduction cascades,
• and ion channels.
• Many channelopathies are attributed
• to improper trafficking to the membrane
• therefore rafts are important to assemble
• the correct signalling molecules in a
• spatially confined manner for
• efficient transduction.
• 4. Lipid rafts have good resistance to
• solubilization with nonionic detergents
• (like Triton X-100) and therefore
• proteins are retained in the pellet.

33
Control vs. High Salt (KI) To interrupt any p-p
interactions
Used fractionation of CNGA2 transfected HEK 293
cells Sucrose Density Gradient
Centrifugation. Track migration of the Flag
epitope tagged channel by comparison with other
raft associated (caveolin, flotillin) molecules
but not with one that is not generally associated
(transferrin R).

34

• A 1 wt
• 2 Flag tagged
• 3 CFP tagged
• 4 mock transfection
• B Can treat with enzymes to digest
• the suspected glycosylation (lose
• the upper band but what is at
• 114 kDa?)
• C Why are all the CNGA2 fractions
• at 114 kDa and not 81 kDa like in the
• control panel of A?
• Demonstrates is a fraction that
• co-migrates with caveolin but also
• expression that co-migrates with the
• transferrin R

35

• Despite the co-migration of Caveolin
• and CNGA2 using sucrose density gradient,
• Protein-protein interaction is not supported
• No ability to co-immunoprecipitate
• (reciprocal pull down).
• Confocal does not support co-localization
• at the cell membrane.

36
Using drugs to deplete cholesterol (CD), there
is a Reduced Buoyancy of CNGA2 in the raft.
37

• Typically PGE1 stimulates CNG channel
• activity unless there is CD pretreatment.
• Ca influx into the open CNG channels
• causes a decrease in delta F so they
• decided to invert their spectrographic
• curves to denote a positive direction
• increase in Ca influx.
• Even though PGE stimulation evoked
• Increase in Ca influx..
• 2 interpretations
• More cAMP (wasnt according to their
• Elisa Assay).
• Different biophysical property of the
• CNG (turned out to be the later with
• single channel analysis.)

38
Cholesterol Depletion (CD) increases the
Concentration of cAMP neededto achieve the same
I/Imax in terms of current calculated from
single channel data I N po i