Binding Studies on Trafficking Proteins

1
Binding Studies on Trafficking Proteins Using
Microcalorimetry   McMahon lab Neurobiology
Division Laboratory of Molecular Biology
Cambridge
2
Clathrin Mediated Endocytosis
Receptor
Ligand
AP adaptor complex
Regulatory adaptor
Clathrin
Dynamin
3
Receptor Mediated Endocytosis
a
b
c
Replica of the inner membrane surface (Heuser
and Anderson 1989)
a) Yolk protein (Gilbert und Perry 1979) b) Low
Density Lipoprotein (Anderson et al. 1977) c)
Virus particle (Matlin et al. 1981)
4
AP Adaptor Complex
Appendage binds regulators Hinge binds
clathrin Trunk binds lipids and membrane
proteins
AP-1 (?) TGN / Endosome AP-2 (?) Plasma
membrane AP-3 (?) Lysosome AP-4 (?) TGN
5
AP Trafficking Pathways
Plasma membrane
AP-2
Endosome
Lysosome
AP-3
Lysosome-related Organelle
AP-1
AP-4
AP-3
GGA
Trans-Golgi-Network
6
AP Appendage Domains
DP(F/W)
FxDxF DP(F/W)
DPW
FxxF
?-Adaptin
?-Adaptin
?-Adaptin
Owen et al. 2000
Owen et al. 1999 Brett et al. 2002
Kent et al. 2002 Nogi et al. 2002
7
Regulatory Adaptors
Epsin1
EpsinR
AP180
Dab2
Amphiphysin1
DOMAIN PARTNER SH3 PxxPxR PTB Receptor and
Lipids ANTH/ENTH Lipids BAR Lipids EH NPF Clathin-
Box Clathrin DxF/W ?- and ?-Adaptin NPF EH PxxPxR
SH3
8
Interactions in Trafficking
Amphiphysin
Receptor
Lipids
FEDNF
Yxx? or LL
AP-Complex
LLDLD
DxF or FxxF
DxF or FxxF
LLDLD
NPF
Epsin1 EpsinR AP180 Dab2
Eps15
Clathrin
LLDLD
9
Determination of Binding Constants
Definition of Association and Dissociation
Constants k1 Pfree conc. of free
protein For a binding reaction at equilibrium P
L PL Lfree conc. of free ligand
k-1 PL conc. of PA complex k1 rate
constant for formation of PL k-1 rate
constant for breakdown of PL The rate of
formation of PL is k1 Pfree Lfree, where k1
is a second order rate constant with units of
l/mol-1s-1. The rate of breakdown of PL is k-1
PL, where k-1 is a first order rate constant
with units of s-1. At equilibrium, the rate of
formation of PL equals the rate of its
breakdown, so k1 PfreeLfree k-1 PL. Also
recall that KD k-1
/ k1 Pfree Lfree/ PL 1 / KA KD is
given in units of concentration (e.g., mol/l) Or,
in terms of fraction of protein binding sites
occupied (y), which is often convenient to
measure y
PL / (Pfree PL) Use PL KA Pfree
Lfree Divide through by KA Replace KA
by 1 / KD Lfree / (Lfree KD)

10
Determination of Binding Constants

• Special cases
• y Lfree /
  (Lfree KD)
• For Lfree 0 y 0 nothing bound
• For Lfree ?? y 1 full occupancy
• For Lfree KD y 0.5 half occupancy
• Two possible ways to determine binding constants
• Measure bound and free ligand at equilibrium as a
  function of concentration
• Measure association and dissociation rate
  constants and use these to calculate binding
  constants

11
Methods to determine Binding Constants
Signal Information Advantage Disadvantage Spectro
scopy change of absorption KD (10-4-10-11M) in
solution probe needed (Fluorescence, UV/Vis,
CD) or emission of light Microcalorimetry heat
of binding KD (10-3-10-11M) no labels, large
sample ?H, ?S, n in solution direct access
to ?H direct access to n Surface Plasmon
Resonance change of refractive KD
(10-3-10-13M) small sample, surface
coupled, index due to mass k1,
k-1 automated ligand must have large
mass Stopped-Flow coupled to spectroscopy KD
(10-3-10-12M) fast probe needed k1,
k-1 Analytical Ultracentrifugation absorption at
different KD (10-3-10-8M) good for slow radii
for different times homomeric interactions N
uclear Magnetic Resonance shift of magnetic KD
(10-3-10-6M) in solution, slow, resonance
frequency structural large sample, information
expensive Binding Assays various, e.g.
SDS-PAGE, KD (10-3-10-15M) can be
most sometimes densitometry, radio- sensitive in
accurate activity

12
Isothermal Titration Calorimetry (ITC)
13
Isothermal Titration Calorimetry (ITC)
Taken from Micro Cal website
14
Isothermal Titration Calorimetry (ITC)

Review of Free Energies, Enthalpies, and
Entropies of Binding ?Gbind RT lnKD (where R
1.98 cal mol1 K-1 T 273.2 K, and RT 0.62
kcal/mol at 37C) Note log relationship between
free energy and binding constants Recall that
?Gbind is relative to standard conditions
(typically 1M reactants, 25 C, standard salt) A
convenient rule of thumb is that a 10-fold change
in binding constant corresponds to 1.4 kcal /
mol. ??GA1-A2 RT ln(KDA1 / KDA2) (0.62 kcal /
mol)ln(10-8 M / 10-7M) -1.4 kcal / mol How
many kcal / mol change in free energy do you need
to change KD 100-fold?
15
Isothermal Titration Calorimetry (ITC)

Review of Free Energies, Enthalpies, and
Entropies of Binding ?Gbind RT lnKD (where R
1.98 cal mol1 K-1 T 273.2 K, and RT 0.62
kcal/mol at 37C) Note log relationship between
free energy and binding constants Recall that
?Gbind is relative to standard conditions
(typically 1M reactants, 25 C, standard salt) A
convenient rule of thumb is that a 10-fold change
in binding constant corresponds to 1.4 kcal /
mol. ??GA1-A2 RT ln(KDA1 / KDA2) (0.62 kcal /
mol)ln(10-8 M / 10-7M) -1.4 kcal / mol How
many kcal / mol change in free energy do you need
to change KD 100-fold? ? - 2.8 kcal / mol
16
Isothermal Titration Calorimetry (ITC)

Review of Free Energies, Enthalpies, and
Entropies of Binding ?Gbind RT lnKD (where R
1.98 cal mol1 K-1 T 273.2 K, and RT 0.62
kcal/mol at 37C) Note log relationship between
free energy and binding constants Recall that
?Gbind is relative to standard conditions
(typically 1M reactants, 25 C, standard salt) A
convenient rule of thumb is that a 10-fold change
in binding constant corresponds to 1.4 kcal /
mol. ??GA1-A2 RT ln(KDA1 / KDA2) (0.62 kcal /
mol)ln(10-8 M / 10-7M) -1.4 kcal / mol How
many kcal / mol change in free energy do you need
to change KD 100-fold? ? - 2.8 kcal / mol
Recall also that free energy has enthalpy and
entropy components ?G ?H -T ?S (and
therefore) RTlnKA ?H -T ?S When is an
interaction strong? ?G must be large and
negative ? ?H must be large and negative (gain
new bonds) ? ?S must be large and positive
(gain more entropy)
17
Isothermal Titration Calorimetry (ITC)
18
Isothermal Titration Calorimetry (ITC)
19
Binding Specificity a-Adaptin and Amphiphysin
Amph1 1-372 DNF-SGA DPF-SGA DNFDPF-SGA Extrac
t
?-Adaptin ?-Adaptin
Praefcke et al. 2004 Olesen et al. 2007
20
Binding Specificity a-Adaptin and Amphiphysin
DxF Peptide Sequence KD (?M) DNF 7mer
FEDNFVP 21 DNF to RNF 7mer
FERNFVP no binding DNF 8mer FEDNFVPE
28 DNF 12mer INFFEDNFVPEI 2.5 DNF to DPF
12mer INFFEDPFVPEI 120 DNF to DAF
12mer INFFEDAFVPEI 21 DNF FE-change INFEFDNFVP
EI 180 DPF 12mer LDLDFDPFKPDV 190 DPF to
DNF-12mer LDLDFDNFKPDV no binding
Praefcke et al. 2004 Olesen et al. 2007
21
Binding Specificity a-Adaptin and Amphiphysin
DxF Peptide Sequence KD (?M) DNF 7mer
FEDNFVP 21 DNF to RNF 7mer
FERNFVP no binding DNF 8mer FEDNFVPE
28 DNF 12mer INFFEDNFVPEI 2.5 DNF to DPF
12mer INFFEDPFVPEI 120 DNF to DAF
12mer INFFEDAFVPEI 21 DNF FE-change INFEFDNFVP
EI 180 DPF 12mer LDLDFDPFKPDV 190 DPF to
DNF-12mer LDLDFDNFKPDV no binding Synaptojanin L
DGFEDNFDLQS 4.5 HIP1 DNKFDDIFGSSF
100 Dab2 QSNFLDLFKGNA no binding
DNF-site is 80 fold stronger than DPF-site Very
good correlation between Western Blots and
ITC Residue at position 4 in FxDxF is important
(NgtSgtAgtIgtPgtL) Prediction for other proteins
possible
Praefcke et al. 2004 Olesen et al. 2007
22
Lipid Binding Epsin1 ENTH domain
Ford et al. 2002
23
Lipid Binding
Lipid Binding Epsin1 ENTH domain
Time (min)
KD
(?M) ? Ins(1,4)P2 gt1,000 ? Ins(1,5)P2
gt1,000 ? Ins(1,3,5)P3 120 ? Ins(1,4,5)P3
3.6 ? Ins(1,3,4,5)P4 4.1 ? InsP6
0.55 diC8PtdIns(4,5)P2
0.85
?cal/s
kcal/mol InsPx
InsPx / Epsin1 ENTH
Good correlation between ITC and other binding
assays Head groups are a good model for the
lipid molecules
Ford et al. 2002
24
Lipid Binding
Lipid Binding Epsin1 ENTH domain
Time (min)
?cal/s
kcal/mol Protein
Disabled2
Epsin1
Protein / PI(4,5)P2 in outer leaflet
Data for Epsin1-ENTH with liposomes is different
from control protein ? ITC reveals tubulation of
liposomes by the ENTH domain
Ford et al. 2002
25
291-429 291-397 291-379 291-345 291-334 D325R
D328R D349R D371R E391R D422R
Multiple Binding Sites EpsinR and ?-Adaptin
Truncations Point Mutations
Clathrin ?-Adaptin
291-625
291-426
lt325
lt328
lt349
lt334
lt345
(291)AHYTGDKASPDQNASTHTPQSSVKTSVPSSKSSGDLVDLFDGTSQ
STGGSADLFGGFADFGSAAASGS FPSQVTATSGNGDFGDWSAFNQA
PSGPVASSGEFFGSASQPAVELVSGSQSALGPPPAASNSSDLFDL(426)

lt371
lt391
lt422
lt397
lt379
Mills et al. 2003
26
Multiple Binding Sites
Multiple Binding Sites EpsinR and ?-Adaptin
Time (min)
?cal/s
kcal/mol EpsinR
EpsinR 291-426 / ?-Adaptin-Appendage
One Site Model N KD (?M) 0.61 3.8 Two Site
Model N1 KD (?M) 1.2 0.26 N2 KD (?M) 2.4 9.3
Mills et al. 2003
27
Multiple Binding Sites
Multiple Binding Sites EpsinR and ?-Adaptin
Time (min)
Time (min)
?cal/s
?cal/s
kcal/mol EpsinR
kcal/mol ?-Adaptin
EpsinR 291-426 / ?-Adaptin-Appendage
?-Adaptin-Appendage / EpsinR 291-426
One Site Model N KD (?M) 1.3 19 Two Site
Model N1 KD (?M) 0.90 0.72 N2 KD (?M) 0.84 51
One Site Model N KD (?M) 0.61 3.8 Two Site
Model N1 KD (?M) 1.2 0.26 N2 KD (?M) 2.4 9.3
swap cell and syringe content
Mills et al. 2003
28
Multiple Binding Sites
Multiple Binding Sites EpsinR and ?-Adaptin
Time (min)
Peptide KD (?M) EpsinR ?-Adaptin
P1-SGDLVDLFDGTS no binding P2-TGGSADLFGG
FA 230 P3-SADLFGGFADFG
110 P4-FGGFADFGSAAA gt
220 P5-TSGNGDFGDWSA 48

P3
P5
?cal/s
kcal/mol Peptide
P3
P5
EpsinR Peptide / Adaptin-Appendage
291(AHY)TGDKASPDQNASTHTPQSSVKTSVPSSKSSGDLVDLFDGTSQ
STGGSADLFGGFADFGSAAASGS FPSQVTATSGNGDFGDWSAFNQA
PSGPVASSGEFFGSASQPAVELVSGSQSALGPPPAASNSSDLFDL(426)

Mills et al. 2003
29
Multiple Binding Sites
Multiple Binding Sites EpsinR and ?-Adaptin
Time (min)
Peptide KD (?M) EpsinR ?-Adaptin
P1-SGDLVDLFDGTS no binding P2-TGGSADLFGG
FA 230 P3-SADLFGGFADFG
110 P4-FGGFADFGSAAA gt
220 P5-TSGNGDFGDWSA 48
?-Synergin PEEDDFQDFQDA
13 Eps15 SFGDGFADFSTL
180 Epsin1 EPDEFSDFDRLR
200 EF-hand NEDDFGDFGDFG 8
P3
P5
?cal/s
?Sy
kcal/mol Peptide
P3
P5
?Sy
EpsinR Peptide / Adaptin-Appendage
lt349
291(AHY)TGDKASPDQNASTHTPQSSVKTSVPSSKSSGDLVDLFDGTSQ
STGGSADLFGGFADFGSAAASGS FPSQVTATSGNGDFGDWSAFNQA
PSGPVASSGEFFGSASQPAVELVSGSQSALGPPPAASNSSDLFDL(426)

P3
lt371
P5
Mills et al. 2003
30
Multiple Binding Sites
Multiple Binding Sites EpsinR and ?-Adaptin
Time (min)
Peptide KD (?M) EpsinR ?-Adaptin
P1-SGDLVDLFDGTS no binding P2-TGGSADLFGG
FA 230 P3-SADLFGGFADFG
110 P4-FGGFADFGSAAA gt
220 P5-TSGNGDFGDWSA 48
?-Synergin PEEDDFQDFQDA
13 Eps15 SFGDGFADFSTL
180 Epsin1 EPDEFSDFDRLR
200 EF-hand NEDDFGDFGDFG 8
P3
P5
?cal/s
?Sy
kcal/mol Peptide
P3
P5
?Sy
EpsinR Peptide / Adaptin-Appendage
EpsinR contains two binding sites for ?-Adaptin
Identification of consensus motif using
peptides Motif is also present in other
trafficking proteins
31

Exothermic
Decrease in Entropy Except in..
Mills et al. 2003
32
Temperature Dependence Synaptotagmin C2A domain
and Calcium
Time (min)
10C 25C N 1.8 2.1 KD (?M) 450 340 ?H
(cal/mol) 3080 1830
10 C
25 C
?cal/s
kcal/mol Ca2
Ca2 / Synaptotagmin C2A
Two calcium binding sites per C2A domain No
robust fit for two site model
33
Temperature Dependence Synaptotagmin C2A domain
and Calcium
Time (min)
10C 25C 37C N1 1.8 2.1 0.9 KD1
(?M) 450 340 103 ?H1 (cal/mol) 3080 1830 -530 N
2 0.9 KD2 (?M) 410 ?H2 (cal/mol) 3770
10 C
25 C
?cal/s
37 C
kcal/mol Ca2
Ca2 / Synaptotagmin C2A
At higher temperature the reaction is more
exothermic At 37C the two sites can be fitted
and resolved
34
Summary
Microcalorimetry is a versatile technique to
study biological interactions in solution is
applicable to ligands such as proteins, peptides,
lipids, liposomes, DNA, ions, gives direct
access to all thermodynamic parameters from one
single experiment allows for the precise
determination of stochiometry of binding
reactions