Download Word file (51.0 KB ) - Nature

erectboboΛογισμικό & κατασκευή λογ/κού

14 Δεκ 2013 (πριν από 3 χρόνια και 7 μήνες)

81 εμφανίσεις



1

Supplementary data


Materials and methods


Expression and purification of wild
-
type and mutant Kai proteins

Full
-
length KaiA and KaiC proteins from
S. elongatus

were produced as described previously

(Mori
et al
, 2002; Pattanayek
et al
, 2004). Over
-
expressi
on constructs for the
Th
KaiA and
Th
KaiC full
-
length proteins with fused (His)
6

tags at their N
-
termini were generated by
amplifying the
ThkaiA

and
ThkaiC

open reading frames with PCR (
PfuUltra

high
-
fidelity DNA
polymerase, Stratagene), using
T. elongatus

B
P
-
1 genomic DNA as a template (GenBank
accession number: BA000039; gift from Dr. Satoshi Tabata, Kazusa DNA Research Institute).
Amplified DNA fragments were subcloned into the pRSF
-
2 Ek/LIC vector (EMD Biosciences
Novagen) by the ligation
-
independent clon
ing method. The constructs were confirmed by DNA
sequencing and introduced into
E. coli

BL21(DE3) cells for over
-
expression.

Both proteins were purified using affinity (Talon/Co
2+
, Clontech) and gel filtration
chromatography performed using a Superdex200

HR 10/30 column (Amersham). They were
eluted as a single peak in the gel filtration column. Purity was assessed with SDS
-
PAGE.
Th
KaiA and
Th
KaiC were purified as a dimer and monomer, respectively. AMPPNP (1mM) and
MgCl
2

(5 mM) were added to the
Th
KaiC mo
nomer (the protein buffer contained 20 mM Tris
-
HCl (pH 7.8), 150 mM NaCl and 1mM DTT) and hexamer formation was achieved by
incubation overnight at 4°C

(Hayashi
et al
, 2004a). Various amounts of KaiA were added to the
KaiC hexamer to form the KaiA
-
KaiC com
plex that was then verified by a native gel
-
electrophoretic assay (Figure 4).

To produce

25C
Th
KaiC, the last 25 amino acids of
Th
KaiC were removed by a
KpnI/PacI restriction digest of the pRSF
-
2 Ek/LIC
-
Th
KaiC plasmid. The linear plasmid was
purified wit
h the QIAquick (Qiagen) PCR purification kit, and the single
-
stranded ends were
filled in with Klenow DNA polymerase I (Invitrogen) treatment. Following another round of
purification with the QIAquick purification kit, recircularization was performed by bl
unt
-
end
ligation using T4 DNA ligase (Promega). After confirming the DNA sequence the plasmid was
transformed into BL21 (DE3)
E. coli

cells for over
-
expression and purification was carried out as
described above for wild
-
type
Th
KaiC.



2


X
-
ray crystallograph
y

Diffraction data used for determination of the crystal structure of
S. elongatus

KaiC (PDB
-
ID
1TF7

(Pattanayek
et al
, 2004)) in combination with the coordinates of
S. elongatus

P
-
KaiC
(phosphate groups at residues S431 and T432, PDB
-
ID 1U9I

(Xu
et al
, 20
04)) were used to
compute Fourier sum (2F
o
-
F
c
) and difference (F
o
-
F
c
) electron density maps. Beginning with
residue T498, C
-
terminal peptide chains of each monomer
-
subunit were extended by one amino
acid at a time. The programs TURBO
-
FRODO

(Cambillau and

Roussel, 1997) or O

(Jones,
2004) were used for map display and model building. Conjugate

gradient energy refinement
was run with CNS

(Brünger
et al
, 1998), using data between 30 and 2.8 Å resolution, followed
by repeated cycles of constraint positional
and B
-
factor refinement. Omit maps were calculated
to monitor the correctness of chain tracing. Numerous alternating cycles of building and
refinement allowed construction of complete C
-
termini for the A and F subunits and partial
models for the B, C, D, a
nd E chains. Attempts at extending the model were stopped when the
electron density became too poorly defined. Selected crystal data and refinement statistics are
summarized in Table S1.


Electron microscopy

The
T. elongatus
BP
-
1 and
S. elongatus

KaiA and

KaiC proteins were imaged by negative stain
electron microscopy in parallel with native gel analysis of complex formation. Proteins were
stored in Tris•HCl buffer, pH 7.8, 100 mM NaCl, 1mM AMP
-
PNP, 5 mM MgCl
2
, 1 mM
dithiothreitol at 4°C and diluted from 1

mg/ml to 50

g/ml in 20 mM HEPES pH 7.6, 100 mM
KCl, 1 mM AMP
-
PNP. Protein solutions were adsorbed to charged carbon support films on EM
grids for 2 minutes and then negatively stained with 0.75% uranyl formate. Digital electron
micrographs were collected

on an FEI Tecnai
-
12 transmission electron microscope (120 kV,
LaB
6

filament) equipped with a Gatan UltraScan 1000 CCD camera (2048 x 2048 pixels). A
range of underfocus levels (0.5 to 1.0

m) and a nominal magnification of 67,000X were used
for data acquisition. Calibration with a negatively stained TMV grid indicated a pixel size of
1.55 Å on the molecular scale. The digital micrographs were binned by a factor of two to
produce a pixel siz
e of 3.1Å prior to image processing.



3

Individual particle images were semi
-
automatically selected from micrographs using the
EMAN program Boxer

(Ludtke
et al
, 1999). The datasets for
Th
KaiC,
Se
KaiC,
Th
KaiA+KaiC,
and
Se
KaiA+KaiC include 3,869, 3,978, 3,994
and 3,816 particle images respectively. The
dataset for

25C
Th
KaiC mixed with
Th
KaiA has 2,638 particle images. The particle images
were normalized, low
-
pass filtered to 20 Å resolution, and translationally aligned to the
rotationally averaged sum of the p
article images using the IMAGIC software package

(van Heel
et al
, 1996). An initial reference
-
free classification step was performed by IMAGIC’s
multivariate statistical analysis procedure, followed by a multi
-
reference alignment (MRA) step
to improve the
translational and rotational alignment of the particle images. Subsequently a
second classification step was performed using the MRA
-
aligned particle images and the initial
class sum images as references. The angular reconstitution approach in IMAGIC was u
sed to
produce initial three
-
dimensional reconstructions for
Th
KaiC and
Se
KaiC with imposed C6
symmetry. One round of refinement was performed with IMAGIC, and the resulting
Th
KaiC and
Se
KaiC reconstructions were used as starting models for six rounds of r
efinement in EMAN with
the Refine program. The resolution of the final
Th
KaiC and
Se
KaiC reconstructions was
determined to be 24 Å for both datasets using the EMAN program Eotest and the Fourier shell
correlation (FSC) 0.5 threshold criterion.

For the
Th
K
aiA
-
KaiC and
Se
KaiA
-
KaiC datasets the angular reconstitution approach in
IMAGIC was used to produce initial three
-
dimensional reconstructions without imposed
symmetry. The
Th
KaiA
-
KaiC reconstruction revealed weak density for KaiA protruding from
the KaiC h
exameric barrel, while the
Se
KaiA
-
KaiC reconstruction revealed no KaiA density
presumably because of the low percentage (<5%) of particle images and class sum images with
discernible KaiA density. Further refinement in IMAGIC and EMAN of the
Se
KaiA
-
KaiC
da
taset produced reconstructions of the KaiC hexameric barrel but without density for KaiA. One
round of refinement was performed for
Th
KaiA
-
KaiC with IMAGIC, followed by six rounds of
refinement in EMAN, all without imposed symmetry. In addition, the final
Th
KaiC
reconstruction was used as an alternate starting model for the
Th
KaiA
-
KaiC dataset in EMAN.
Both refinement schemes produced similar reconstructions of
Th
KaiA
-
KaiC containing weak
and diffuse KaiA density protruding from one end of the KaiC hexamer
ic barrel structure. The
two final
Th
KaiA
-
KaiC reconstructions have resolutions of 24 and 25 Å by the FSC 0.5


4

threshold criterion. The 24 Å resolution reconstruction, which resulted from using the IMAGIC
Th
KaiA
-
KaiC reconstruction as the EMAN starting mode
l, is presented in Figure 5.

The two final
Th
KaiA
-
KaiC reconstructions were used as starting models for refinement
of the

25C
Th
KaiC+
Th
KaiA dataset in EMAN without imposed symmetry. After two rounds of
refinement the KaiA density, which was present in bot
h starting models, was gone from the
Th
KaiA+

25C
Th
KaiC reconstructions. Following six rounds of refinement both image
processing schemes produced reconstructions resembling the KaiC hexameric barrel structure
but without any discernible KaiA density. The r
esolutions of the final
Th
KaiA+

25C
Th
KaiC
reconstructions are 24 and 25 Å by the FSC 0.5 threshold criterion. The 24 Å resolution
reconstruction is shown in Figure 6.

The crystal structures of KaiA and KaiC were converted to 25 Å filtered density
represen
tations for comparison with the EM data using the EMAN program Pdb2mrc. The
isosurface thresholds for the
Th
KaiC and
Se
KaiC reconstructions, as well as the filtered crystal
structures of KaiA and KaiC, were chosen using the EMAN program Volume and the in
-
h
ouse
Fortran program Density
-
mult to enclose 120% of the predicted volume assuming a protein
density of 1.35 g/cm
3
. The figures were produced using EMAN, IMAGIC, the AVS5
visualization software package (Advanced Visual Systems, Inc.), and Chimera

(Pettersen
et al
,
2004)
.


S. elongatus strains, KaiC mutagenesis, immunoblotting assay, and in vivo rhythm
measurements

Growth and transformation of cyanobacteria were described previously

(Xu
et al
, 2003). The
host strain for expression of wild
-
type or mutant KaiCs has an in
-
frame deletion of the
endogenous
kaiC

gene (
kaiC
-
null;

Ditty
et al
, 2005) and harbors a
kaiBC
p::luxAB reporter in
neutral site I

(Xu
et al
, 2003). The R488A mutation in

Se
KaiC (
Se
KaiC
R488A
) was generated by
conversion of the
1462
CGG
1464

codon to
GC
G using site
-
directed mutagenesis

(Xu
et al
, 2004).
Deletion of the C
-
terminal 30 residues of
Se
KaiC (

30C
Se
KaiC; deleting residues 490
-
519) was
made by introducing a translational stop codon TAG immediately downstream of
1465
ATT
1467

and removing the 3’
-
end DNA sequence from 1468
-
1560 using a PCR method based on
PfuUltra

high
-
fidelity DNA polymerase (Strat
agene). Mutagenized
Se
KaiCs were confirmed by
both DNA sequencing and immunoblotting. The endogenous clock promoter
kaiBC
p can be


5

functionally replaced by an IPTG
-
derepressible heterologous promoter (
trc
p

(Xu
et al
, 2003));
therefore, the constructs for ex
pression of wild
-
type
Se
KaiC or mutant
Se
KaiCs were driven by
trc
p from neutral site II. The expression levels of different
Se
KaiCs were regulated with or
without application of IPTG at different concentrations. Native
-
gel
-
based immunoblotting assays
and
k
aiBC
p::luxAB luminescence rhythm recordings of populations of cyanobacteria were
performed as described previously

(Xu
et al
, 2004).


Model building

The

Th
KaiA
-
KaiC
reconstruction at a low isosurface value (Figure 5C, bottom
row
) was used as
an envelope i
nto which the X
-
ray structures of
Se
KaiC (
PDB
-
ID

1TF7, Pattanayek
et al
, 2004
)
and
Se
KaiA (
PDB
-
ID

1R8J,
Ye
et al
, 2004
) were fit (Figure 8).
Se
KaiC was translated and
rotated by hand into the symmetrical hexamer
-
shaped

portion of the density. It was then r
efined
iteratively using the Chimera FitMap tool

(Petterson
et al
, 2004). Optimization of
Se
KaiA in the

engaged
’ and ‘
tethered
’ orientations was
accomplished

using similar procedures. The presence
of the second plume of KaiA density in the
Th
KaiA
-
KaiC
rec
onstruction (Figure 8A, left plume)
implies that KaiA can occupy a variety of positions relative to
KaiC
. We chose to build only one
tethered model, because one model is sufficient to make the point that KaiA can be positioned
~35 Å away from the hexameric

barrel of KaiC if a region of KaiC (aa 485
-
500) is extended. We
positioned KaiA into the right plume of density, since this plume has a bi
-
lobed shape similar to
the KaiA crystal structure. Given the weak and diffuse nature of the reconstructed KaiA densi
ty,
it is not surprising that the KaiA crystal structure protrudes somewhat from the isosurface of the
EM reconstruction.

The NMR structure of the complex between the C
-
terminal domain of
Th
KaiA and the C
-
terminal peptide of
Th
KaiC (PDB
-
ID 1SUY, Vakonakis

and LiWang, 2004

) was aligned with
the tethered and engaged positions of the
Se
KaiA dimer

(r.m.s.d. 1.5 Å). To complete the
tethered KaiA

KaiC model (Figure 8A) we: (i) mutated the
Th
KaiC peptide residues in the NMR
structure of the complex into the corr
esponding
Se
KaiC residues (Figure 3), (ii) retained the
conformation of the E501
-
E518 region of the
Th
KaiC C
-
terminal peptide as it is in the NMR
structure, and (iii) extended residues N485
-
D500 of the NMR KaiC peptide by changing the
backbone torsion angl
es of residue V499 so that an extended linker would be created between
the portion of KaiC bound to KaiA and residue R484 in the
Se
KaiC crystal structure. To finish


6

the engaged KaiA

KaiC model (Figure 8B) we: (i) mutated the
Th
KaiC peptide residues in the
NMR structure of the complex into the corresponding
Se
KaiC residues as above, (ii) preserved
the conformation of the
Th
KaiC peptide I490
-
E518 as in the NMR structure of the complex, and
(iii) changed the backbone torsion angles of residue I489 to form a sh
ort compact linker region
(N485
-
I489, Figure 3, gray box) connecting to residue R484 in the
Se
KaiC crystal structure.



References


Brünger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse
-
Kunstleve RW, Jiang JS,
Kuszewski J, Nilges M, Pannu NS, Read RJ,

Rice LM, Simonson T, Warren GL (1998)
Crystallography and NMR System: a new software suite for macromolecular structure
determination.
Acta Cryst D

54:

905
-
921.

Cambillau C, Roussel A (1997) Turbo Frodo, Version OpenGL.1, Université Aix
-
Marseille II,
Mars
eille, France.

Ditty JL, Canales SR, Anderson BE, Williams SB, Golden SS (2005) Stability of the
Synechococcus elongatus

PCC 7942 circadian clock under directed anti
-
phase expression
of the
kai

genes.
Microbiology

151
: 2605
-
2613.

Jones TA (2004) Interact
ive electron
-
density map interpretation: from INTER to O.
Acta Cryst D

60:

2115
-
2125.

Hayashi F, Ito H, Fujita M, Iwase R, Uzumaki T, Ishiura M (2004a) Stoichiometric interactions
between cyanobacterial clock proteins KaiA and KaiC.
Biochem Biophys Res Com
m

316:

195
-
202.

Ludtke SJ, Baldwin PR, Chiu W (1999) EMAN: Semiautomated software for high
-
resolution
single
-
particle reconstructions.
J Struct Biol

128:

82
-
97.

Mori T, Saveliev SV, Xu Y, Stafford WF, Cox MM, Inman RB, Johnson CH (2002) Circadian
clock pro
tein KaiC forms ATP
-
dependent hexameric rings and binds DNA.
Proc Natl
Acad Sci USA

99:

17203
-
17208.

Pattanayek R, Wang J, Mori T, Xu Y, Johnson CH, Egli M (2004) Visualizing a circadian clock
protein: crystal structure of KaiC and functional insights.
Mol

Cell

15:

375
-
388.



7

Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE (2004)
UCSF Chimera
-

A Visualization System for Exploratory Research and Analysis.
J Comp
Chem

25:

1605
-
1612.

Vakonakis I, LiWang AC (2004) Structure of the

C
-
terminal domain of the clock protein KaiA in
complex with a KaiC
-
derived peptide: implications for KaiC regulation.
Proc Natl Acad
Sci USA

101:

10925
-
10930.

Van Heel M, Harauz G, Orlova EV, Schmidt R, Schatz M (1996) A new generation of the
IMAGIC image

processing system.
J Struct Biol

116:

17
-
24.

Xu Y, Mori T, Johnson CH (2003) Cyanobacterial circadian clockwork: roles of KaiA, KaiB,
and the kaiBC promoter in regulating KaiC.
EMBO J

22:

2117
-
2126.

Xu Y, Mori T, Pattanayek R, Pattanayek S, Egli M, Johnso
n CH (2004) Identification of key
phosphorylation sites in the circadian clock protein KaiC by crystallographic and
mutagenetic analyses.
Proc Natl Acad Sci USA

101:

13933
-
13938.




Table S1

Crystal data and refinement statistics

Crystal Data


Space gro
up

P
2
1
2
1
2
1

Unit cell parameters (Å)
a

=

132.87

b

=

135.58

c

=

204.95

Refinement statistics


Number of:


Protein atoms

23,432

ATP atoms

372

Water molecules

67

R
-
factor (%)

0.23 (0.45)
1

R
-
free (%)

0.29 (0.55)

RMS deviations from ideal:


Bond leng
ths (Å)

0.025

Bond angles (°)

2.15

Avg. B
-
factor (residues 498


㔱㤬⃅
2
)

91

Avg. B
-
factor (residues 14


㐹㜬⃅
2
)

74

Avg. B
-
factor (complete model, Å
2
)

74

1

Numbers in parentheses refer to the last resolution shell





8

Figure S2

Complete sets of EM cl
ass sum images of
Th
KaiC,
Th
KaiA
-
KaiC, and
Th
KaiA +

25C
Th
KaiC. Two hundred class sum images were generated for each sample based on particle
images from negative
-
stain electron micrographs using the IMAGIC multivariate statistical
analysis approach. These

class sum images were generated after the second round of
classification using multiple
-
reference aligned particle images. (
A
)
Th
KaiC class sum images
occasionally show a second KaiC hexamer nearby or small regions of randomly positioned
density next to K
aiC. (
B
)
Th
KaiA
-
KaiC class sum images show a plume of KaiA density
protruding from one end of the KaiC hexameric barrel in ~30% of the images. (
C
)
Th
KaiA +

25C
Th
KaiC class sum images have occasional additional density near the KaiC hexamer, but it
is rare
ly connected to the hexamer or reproducibly positioned relative to KaiC. These results
indicate that
Th
KaiA binds to full
-
length
Th
KaiC but not to the truncation mutant

25C
Th
KaiC,
which lacks the C
-
terminal 25 residues of
Th
KaiC.