Supplementary Information Crystal structure of the UBA domain of p62 and its interaction with ubiquitin

rapidcrimsonMécanique

22 févr. 2014 (il y a 3 années et 1 mois)

77 vue(s)


1

Supplementary Information



Crystal structure of the UBA domain of p62 and its interaction with ubiquitin


Shin Isogai
1
, Daichi Morimoto
1
, Kyohei Arita
1
, Satoru Unzai
3
, Takeshi Tenno
2
,
Jun Hasegawa
4
,
Yu
-
shin Sou
4
,
Masaaki Komatsu
4
, Keiji Tanaka
4
, Masahiro Shirakawa
1
* and Hidehito Tochio
1
*


Department of Molecular Engineering, Graduate School of Engineering, Kyoto University
,

Kyoto
-
Daigaku
Katsura, Kyoto 615
-
8510, Japan
1
,
Division of
Structural Biology, Graduate School of Medicine, Kobe
University
2
, 7
-
5
-
1, Kusunoki
-
cho, Chuo
-
ku, Kobe, Hyogo, 650
-
0017, Japan
2
,
Graduate School of
Nanobioscience, Yokohama City University, 1
-
7
-
29 Suehiro
-
cho, Tsurumi
-
ku, Yokohama 230
-
0045, Japan
3
,
Laborato
ry of Frontier Science, Tokyo Metropolitan Institute of Medical Science, Bunkyo
-
ku, Tokyo
113
-
8613, Japan
4
.



This file includes:

Supplementary
Methods

Supplementary Figure
S
1

Supplementary Figure
S
2

Supplementary Figure
S
3

Supplementary Figure
S
4

Supplementary Figure
S
5

Supplementary Figure
S
6

Supplementary Figure
S7

Supplementary Figure
S8

Supplementary Table
S
1





2

Supplementary

Methods


Analytical

Ultracentrifugation
.


Analytical ultracentrifugation

sedimentation velocity
(AUC/SV)

and equilibrium
experiments
(AUC/SE)

were carried out using an Optimal XL
-
I

analytical
ultracentrifuge (Beckman Coulter). For AUC/SV experiments, cells with a standard
Epon two
-
channel centerpiece and sapphire windows were used. Samples in the buffer
cont
aining 50 mM Tris
-
HCl, pH 8.0, 150 mM NaCl and 1 mM DTT and reference buffer
were loaded into cells. The buffer was supplemented with 20% DMSO for the molecular
weight analysis of
the monomeric form of
wild
-
type p62 UBA. The rotor temperature
was equilibra
ted at 20 °C in the vacuum chamber for 1

2 h prior to start
-
up. Absorbance
(A280) scans were collected at 10
-
min intervals during sedimentation at 50 × 10
3

rpm.
Sample concentrations were 0.3
3
, 0.2
0
, and 0.1
0

mg of protein/ml (wild
-
type, W414F
and I433A mu
tants). Partial specific volume of the protein, solvent density, and solvent
viscosity were calculated from standard tables using the program SEDNTERP, version
1.09 (1). These values for the 20%
DMSO
supplemented buffer were calculated using the
value repo
rted in the previous report (2). The resulting scans were analyzed using the
continuous distribution (c(s)) analysis module in the program Sedfit version 11.0

(3). In
this analysis, a differential sedimentation coefficient distribution ((c(s)) that
deconvo
lutes diffusion effects is determined, based on the direct boundary modeling
with distributions of Lamm equation solutions (4). Sedimentation coefficient increments
of 200 were used in the appropriate range for each sample, and the weight average
frictiona
l ratio (
f/f
o
) was allowed to float during fitting. The weight average
sedimentation coefficient was obtained by integrating the range of sedimentation
coefficients in which peaks were present.
The
s
20,w

value, (sedimentation coefficient
corrected to 20 °C

in pure water)

was calculated from the observed sedimentation
coefficient value using the program SEDNTERP.


3

AUC/SE experiments were also carried out in cells with a six
-
channel centerpiece and
quartz windows. Samples were diluted 0.33, 0.2
0
, and 0.1
0

mg of protein/ml with the
buffer. The absorbance wavelength was set at 280 nm, and data were acquired at 20 °C.
Data were obtained at
18,
25 and 32 × 10
3

rpm

for all the samples
. A total equilibration
time of
16

h was used for each speed, with a scan take
n at
12

h and
14

h to ensure
equilibrium had been reached. The optical baseline was determined by accelerating at
4
2

× 10
3

rpm at the end of data collection. Data analysis was performed by global
analysis of data sets obtained at different loading concentr
ations and rotor speeds using
UltraSpin software (MRC Center for Protein Engineering, Cambridge, UK,
www.mrc
-
cpe.cam.ac.uk/ultraspin)


Assignment of monomer peaks under the DMSO
-
supplemented condition.


To assign monomer
-
derived peaks, we used DMSO
-
supplem
ented buffer that
increases monomer populations
even
at higher
protein
concentrations (F
ig
. S1A).
C
oncentration of DMSO required to completely dissociate the dimer
depended on the
concentration of the p62 UBA domain. F
or
example,
20% DMSO was required

to
c
ompletely dissociate the dimer
in the case of
100

M
15
N
-
labeled p62 UBA in buffer
containing 20 mM sodium potassium phosphate (pH 6.5), 5 mM KCl, 1 mM EDTA and
10% D
2
O.
In order t
o
confirm

the
DMSO induced
NMR
spectral change
(Fig. S1A)
was
indeed

due to the
dissociation

of the dimer
,
we conducted a molecular weight analysis
of
the p62 UBA domain in the presence of DMSO
by analytical ultracentrifugation
.
Sedimentation v
elocity and sedimentation equilibrium
experiments
both indicated that
the molecu
lar weight of
the
p62 UBA in the presence of 20% DMSO is around 7 kDa,
which is much less than the value observed under buffer conditions without DMSO (~11
kDa).

Chemical shift

value
s

of backbone amide
1
H and
15
N nu
cl
ei
of

the p62 UBA
domain


4

with DMSO supplementation were
obtained

by performing

three
-
dimensional (3D)
triple resonance
NMR experiments

(HNCO, HNCACB, CBCA(CO)NH and HN(CA)CO)
on
300
μ
M
13
C/
15
N doubly labeled p62 UBA dissolved in 35% DMSO buffer. The chemical
shift
assignments
obta
ined in this way
were

then
transferred to the
1
H
-
15
N correlation
spectr
um

of
monomer
ic
p62 UBA

observed under
low concentration (3

M)
in
the

standard

buffer

by comparing the series of spectra under different DMSO
concentrations (Fig. S1B).

Two

NMR
titration experiments
were
further
performed
for
15
N
-
labeled p62 UBA
to
examine the effect of DMSO on the
activity of the UBA domain.

In the
first

experiment,
15
N
-
labeled
p62 UBA
was titrated
with
ubiquitin
in the presence of 20% DMSO
(Fig.
S1C)
, in which
the
cross
-
peaks
of residues in ubiquitin biding sites
shifted upon addition
of ubiquitin
.

This observation
indicat
es

that
the p62 UBA domain can
still
bind
ubiquitin
even
in the presence of 20% DMSO. I
n the second

experiment
,
15
N
-
labeled
p62
UBA
was titrated
with
DMSO

in the presence of
six

equivalents of
ubiquitin (Fig. S1D)
.

Positions of s
everal

c
ross
-
peaks
changed

upon addition of DMSO
,
but those cross
-
peaks
were not
of residues
in the ubiquitin binding sites. Thus, the
observed
changes

were

li
kely due to the solvent effect of DMSO.
It should be
also
noted that

t
he magnitude of
the
changes

was much smaller than that was seen in the similar DMSO titration
experiment performed for

free p62 UBA
domain
(Fig. S1A).
This
substantial difference

supports
the idea
that DMSO has a
n

impact on dimer dissociation

of the p62 UBA
domain
, while
its effect
on

the UBA fold itself

is

negligible
.



References

1.

Laue, T. M., Shah, B. D., Ridgeway, T. M., & Pelletier, S. L. (1992)
Royal Society of

5

Chemistry,

Cambridge, UK,

pp. 19
-
125.

2.

Cowie, J. M. & Toporowski, P. M. (1961)
Can J Chem,

39
(11)
:
2240
-
&.

3.

Schuck, P., Perugini, M. A., Gonzales, N. R., Howlett, G. J., & Schubert, D. (2002)
Biophys
J
,

82
(2)
:
1096
-
1111.

4.

Schuck, P. (2000)
Biophys
J
,

78
(3)
:
1606
-
1619.







6

Supplementary

Figures





Supplementary
Fig
ure

S
1

(A) Overlay of
1
H
-
15
N HSQC spectra of
15
N
-
labeled p62 UBA
at various concentrations of DMSO. (B) Assignments of
15
N
-
labeled p62 UBA at 20%
DMSO. Amino acid residues derived from the expression tag are shown in blue (residues
3
-
5). (C) Overlay of
15
N
-
labeled p62 UBA spectra with various concentrations of
ubiquitin with 20% DMSO. (D) Overlay of
15
N
-
labeled p62 UBA spect
ra in the presence
of six molar equivalents of ubiquitin at various concentrations of DMSO.




7



Supplementary
Fig
ure

S
2
. ITC experiments for dimer dissociation and ubiquitin
binding.

(A) An ITC thermogram for dimer dissociation (top) and the plot of corrected
heat values are shown (bottom). The line in the plot indicates the best fit of the data to a
simple 1:1 binding model. The experiment was performed by titrating concentrated p62
U
BA in a syringe to the blank buffer in the cell. (B, C, D) ITC thermograms for ubiquitin
binding of both wild type and mutants (W414F and I433A) (top) and the plots of
corrected heat values (bottom) are shown. Concentrated ubiquitin was titrated from the
s
yringe to the wild
-
type, W414F and I433A p62 UBA mutants in the cell. Each
measurement was repeated at least twice.



8



Supplementary
Fig
ure

S
3


{
1
H}
-
15
N

heteronuclear

steady
-
state

NOE

values

of

the

main chain

of

the
p62 UBA
domain
in its free

dimer
ic

and ubiquitin
-
bound

monomer
ic

s
t
ate
s.


Note that the hetero
nuclear

NOE values of the N
-
terminal protease recognition
sequence is not shown in the graph.


9



10



Supplementary
Fig
ure

S
4

(A)
1
H
-
15
N correlation spectra of the p62 UBA

(Y435*) mutant
.
The inset schematically shows
the spectral
patterns

around the glycine cross
-
peaks

of
the
free

dimer
ic

(red) and free

mono
m
er
ic

(green)
forms
of
the
p62 UBA domain. The
subregion is similar to

that of the free

monomer. (B)
1
H
-
15
N correlation spectra of t
he
p62 UBA

(Y435*)

mutant
in
the presence of
various molar ratios of
ubiquitin.
Amino
acid residues used to estimate the dissociation constant are indicated. Note that X3, X15
and X40 indicate
unassigned

cross
-
peaks. (C)
NMR chemical shift titration curves for
the
indicated
residues
.

The normalized chemical shift changes
(see Methods and
Materials

section
) are plotted as a function of molar equivalent of added ubiquitin.
The
red
lines drawn through the points are best
-
fit
curves to the standard
1:1
binding
equation. T
he dissociation constant

was

~7
μ
M. Note that X3, X15 and X40 indicate
unassigned

cross
-
peaks.


11



Supplementary
Fig
ure

S
5

GST
-
pulldown experiment of the full
-
length mouse p62 mutants
with polyubiquitins.

GST
-
f
us
ed

full
-
length p62 proteins baring various mutations were expressed
and pull
-
downed with K48
-
linked and K63
-
linked polyubiquitins. GST
-
fus
ed

full
-
length p62 mutants
were immobilized on 25 ml of glutathione beads and incubated with 100 ml of 20 mM polyubi
quitin
solution in TBS (pH 8.0) buffer on ice. Beads were washed with TBS buffer supplemented with 0.1%
Triton
-
X 100 and visualized with Anti
-
K48
-
linked polyubiquitin antibody (Apu2) and Anti
-
Ubiquitin
(P4D1) for K48
-
linked polyubiquitin and K63
-
linked pol
yubiquitin respectively. Bands in the upper
black boxes represent input GST
-
p62 mutants on the beads. The figure shows the representative
experiment of three individual experiments.








12

(A)





(B)


(C)



13

Supplementary Figure S
6

Overlay of Dsk2 UBA
and

p62 UBA structures are shown.
Tyr
435 of free dimer UBA
compensates the stabilizing effect of LL motif
-
like configuration observed in the
Ub
-
bound form UBA.

(A) Free dimer form p62 UBA and Dsk2 UBA. (B) UB
-
bound p62
UBA and Dsk2 UBA. Dsk2 UBA is shown i
n pink, free dimer UBA is shown in cyan and
Ub
-
bound UBA is shown in green. Dsk2 UBA LL (L
eu
368 and L
eu
369) motif and
following
Asn
370 side chains are shown in sticks. Corresponding residues in p62 UBAs
are also shown in sticks. Note that the italicized re
sidues are from Dsk2 UBA. (C)
Residues important for dimer formation are shown. The figure was generated by the
program LIGPLOT using dimer interaction analysis mode. Gln434 and Met406 were
forming a hydrogen bonding pair.


14




Supplementary
Fig
ure

S
7

Interaction between
the
isolated PB1 domain and
the
isolated
UBA domain
of p62
was tested using
solution
NMR.
(Left) The
1
H
-
15
N HSQC
s
pectra
of

100 μM
15
N
-
labeled WT p62 UBA domain

recorded in the presence (red) or
absence (green) of four equivalents of a
PB1 mutant
, in which Asp
69,
Asp
71

and Asp
73

were replaced with alanines. (Right) The
1
H
-
15
N HSQC s
pectra for 100 μM
15
N
-
labeled
WT p62 UBA domain

recorded with or without four equivalents of a
PB1 mutant
, in
which Lys
7,
Arg
68

and

Lys
91

were replaced with alanines.


It should be noted that since the p62 PB1 domain heavily aggregates via the interaction
between the acidic (Asp
69,
Asp
71

and Asp
73
) and the basic (Lys
7,
Arg
68

and

Lys
91
)
surface in a head
-
to
-
tail manner (Saio T., et al.,
J B
iomol NMR
,

45
,
335
-
41
(
2009
)), we
disturbed

either of the surfaces by the alanine replacement. The aggregation of both of
the PB1 mutants substantially reduced.





15


(A)



(B)



Supplementary Figure S
8

T
o
investigate

t
he biological significance of dimerization of
p62
UBA
in vivo
, the Tet
-
On
/Tet
-
Off

system

was used
as
we
reported previ
o
usly
(
Ichimura, Y., et al.
,

J Biol Chem
,
283
, 22847

(2008)
)
.

A

regulator gene cassette, CAG
-

rtTA and TRE
-
GFP
-
fused wild type and a
p62
mutan
t
lack
ing

Tyr 435 and its
C
-
terminal residues
(
Δ
C
8
)

were
introduc
ed into immortalized
p62

knock
-
out and
Atg7/p62

double knock
-
out mouse embryonic fibroblasts (MEFs) to abolish the effect of
endogenous wild
-
type p62. The presence of
doxycycline (
Dox
)

in the culture medium
should induce the expression of GFP
-
p62 and the mutants, whereas Dox removal should
allow suppression of the expression. To examine the degradation of p62, the cells were
cultured in media containing Dox for 24 h, and then the level
of GFP
-
p62 was chased for

16

36 h by culturing Dox
-
free media.
Both wild
-

type and mutant GFP
-
p62 were efficiently
expressed in each cells.

Autophagy
-
dependent degradation of p62 UBA derivatives was
not significantly affected in these cells
, although basal le
vel of
the deletion mutant

was
less than
that of
wild
-
type

(A)
. However, immunoprecipitation experiments

using
anti
-
GFP

antibody
with proteasome inhibitors
(lactacystin)
indicated that impeding p62
UBA dimerization promoted proteasomal degradation of
GFP
-
p
62

(
B
), showing

that the
UBA dimerization plays some role in regulating p62 lifetime.

The degradation was not
affected by lysosomal inhibitors (
E64d
and
pepstatin A
).
In (B), t
he cells were cultured
in the absence or presence of 10 µM lactacystin
or a
cocktail

of lysosomal inhibitors
(
E64d
and
pepstatin A
, 10
µ
g/mL)
for
12

hr. The cell lysates were immunoprecipitated
with anti
-
GFP

antibody, then subjected to SDS
-
PAGE and analyzed by immunoblotting
with anti
-
p62

antibod
y
. The bands corresponding to
GFP
-
p6
2 is

shown.






17

Supplementary
Table
S
1
. Molecular weight estimation of p62 UBA and mutants in solution


Analytical gel
-
filtration

Sample


Retention Vol. (ml)


Mw Estimate (kDa)

Ubiquitin

13.9



9.5

UBA

WT


13.1



13

W414A


14.0



9.0

W414K


14.0



8.9

W414F


14.1



8.5

L418A


14.2



8.4

L418V


14.4



7.6

I433A


14.5



7.4

I433V


13.0



14


Analytical ultracentrifugation

Sample


Sedimentation velocity (kDa)


Sedimentation equilibrium (kDa)


WT


11




10.8


W414F


5.8




6.7


I433A


6.0




6.2