大阪府立大学　学術情報リポジトリ

Study on the reassembly mechanism of

fluorescent protein

著者

磯貝 純美

内容記述

学位授与大学: Osaka Prefecture University(大阪

府立大学), 学位の種類: 博士(生物科学), 学位記

番号: 論理第94号, 学位授与年月日: 2011-03-31,

指導教員: 多田俊治.

大阪府立大学博士

大阪府立大学博士

大阪府立大学博士

大阪府立大学博士（

（

（理学

（

理学

理学

理学）

）

）

）学位論文

学位論文

学位論文

学位論文

Study on the reassembly mechanism of fluorescent protein

(蛍光

蛍光タンパク

蛍光

蛍光

タンパク

タンパク

タンパク質

質

質の

質

の

の再構成機構

の

再構成機構に

再構成機構

再構成機構

に

に

に関

関

関

関する

する

する

する研究

研究

研究)

研究

Masami Isogai

磯

磯

磯

Abbreviations

BiFC bimolecular fluorescence complementation

CD circular dichroism

DEAE diethylaminoethyl

DLS dynamic light scattering

DSC differential scanning calorimetry

E. coli Escherichia coli

Endo. endothermic

EYFP enhanced yellow fluorescent protein

FPLC fast protein liquid chromatography

FRET fluorescence resonance energy transfer

GFP green fluorescent protein

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HPLC high performance liquid chromatography

IPTG isopropyl-1-thio-β-D-galactopyranoside

MALDI matrix assisted laser desorption ionization

mRFP monomeric red fluorescent protein

MS mass spectrometry

NaCl sodium chloride

Ni-NTA nickel-nitrilotriacetic acid

OH group hydroxyl group

PAGE poly acrylamide gel electrophoresis

PCR polymerase chain reaction

PDB protein data bank

PEG polyethylene glycol

r.m.s.d root mean square deviation

SDS sodium dodecyl sulfate

S/N signal-to-noise

Tm temperature max

TOF time of flight

Tris tris-(hydroxymethyl)-aminomethane

Contents

Introduction

CHAPTER I

1. Introduction 6

2. Materials and methods 7

2-1. Expression

2-2. Purification

2-3. Fluorescence spectroscopy

2-4. CD spectroscopy

3. Results and discussion 13

CHAPTER II

1. Introduction 17

2. Materials and methods 19

2-1. MALDI-TOF MS

2-2. DLS measurement

2-3. Crystallization

2-4. Data collection and structure determination

CHAPTER III

1. Introduction 30

2. Materials and methods 32

2-1. Preparation of mutants

2-2. DSC measurement

2-3. Preparation of mCherry

2-4. HPLC analysis of Venus and mCherry

3. Results and discussion 35

Conclusion

References

Introduction

Protein-protein interactions play key roles in mediating signal-transduction

pathways and executing cellular functions. Therefore, defining how each protein

interacts with all possible partners in cells provides new insight into cellular functions

of individual proteins.

Protein-protein interactions have been investigated using many different

approaches. Although many methods have been developed, most methods that enable

direct detection of protein interactions, such as affinity-precipitation or co-purification

require removal of the proteins from their normal environment. The visualization of

protein interactions in living cells provides the potential for direct detection of protein

interactions with minimal perturbation of their normal environment. GFP was

discovered by Shimomura et al. as a companion protein to aequorin, the famous

chemiluminescent protein from jellyfish Aequorea victoria in the early 1990s (Fig. 1).

GFP is a 238 amino acid protein having a β-barrel structure with the chromophore

contained inside the barrel. It has been developed and contributed to visualization of

proteins as a biological marker for tracking gene expression in cells (Heller et al.,

2010). Several assays such as BiFC, FRET, and bioluminescence resonance energy

transfer have been developed and most widely used to visualize protein-protein

interactions in living cells (Shyu et al., 2008). These assays can provide information

regarding when and where protein-protein interactions occur in the cell.

2002; Kerppola, 2006). The basic principle of this technique is the formation of the

fluorescent complex through the association of two non-fluorescent N- and C-terminal

fragments of the fluorescent protein when they are brought together by an interaction

between two proteins fused to the fragments (Figs. 2 & 3). A number of fluorescent

proteins-such as yellow, cyan, green, and red-have been also utilized to study post

translational modifications, protein folding, protein aggregation, protein

conformational change, and protein topology (Hu and Kerppola, 2003; Fan et al.,

2008). In particular, Venus (λex = 515 nm, λem = 528 nm) generated from EYFP has

been recently used in the BiFC assay, because of its fast chromophore maturation rate

and bright yellow fluorescence that is relatively insensitive to changes in pH and ion

concentrations (Nagai et al., 2002; Shyu et al., 2006).

The crystal structure of Venus (whole Venus) was determined by X-ray

crystallography at a resolusion of 2.2 Å (Rekas et al., 2002). The structure of whole

Venus shows an eleven-stranded β-barrel, typical of GFP-derived fluorescent proteins.

Although the Venus-based BiFC assay provides information regarding when and where

protein-protein interactions occur in the cell, background from spontaneous assembly of

the fragments compromises their utility for detecting the interactions. The decrease of

BiFC assay (Kodama and Hu, 2010).

There is no report for the structure and characteristics of reassembled Venus

formed by the complementation between the N- and C-terminal fragments, nevertheless

they should provide important information on the strategies to decrease background

fluorescence in the BiFC assay. Thus, in order to facilitate the application of BiFC to

research of protein-protein interactions, we have attempted to elucidate the structure and

the physicochemical properties of reassembled Venus and its mutants. It was difficult to

obtain the fragment of split fluorescent proteins expressed in E. coli separately due to

intrinsic folding problems and low solubility of these fragments in aqueous solutions.

Recombinant expression of fragment of split fluorescent proteins often results in low

sample yield or most of them express in insolubility form in E. coli cells (Ottmann et al.,

2009). Thus, the author attempted to co-express two fragments of split fluorescent

protein in E. coli and successfully overexpressed reassembled Venus in soluble form.

The author describe the X-ray crystal structure of reassembled Venus at a resolusion of

2.1 Å and its characteristics assessed by chromatography, MALDI-TOF MS, CD

spectra. Thermal stability of reassembled Venus including three mutants, Y143F, Y145F,

H148G, were evaluated by DSC analysis.

The author also attempted to reveal reassembly mechanisms. The folding

mechanism of β-barrel proteins using GFP has been investigated (Kent et al., 2008;

Kent et al., 2009). Proteins composed of β-Strand frequently form oligomers. This

oligomerization is often necessary for protein activity or regulation and its efficiency is

disease, mad cow disease, Parkinson’s disease. This study may provide perspectives

for the folding mechanism of β-strand rich proteins.

Fig. 1 Aequorea victoria

VN155 VC155 Proteins for visualization Reconstitution of fluorescent protein Recovery of fluorescence Interaction of two proteins fused

to each fragments

+

to each fragments VN155 VC155 Proteins for visualization Reconstitution of fluorescent protein Recovery of fluorescence Interaction of two proteins fused

to each fragments

Fig. 3 A topology of fluorescent protein

2

3

11

10

7

8

9

4

5

6

1

X

Split between residue 154 and 155

2

3

11

10

7

8

9

4

5

6

1

X

CHAPTER I

Expression and purification of reassembled Venus

1. Introduction

Monitoring protein–protein interactions and macromolecular complex formation

in living cells is extremely useful for understanding the dynamics and mechanism of

biological processes. In order to facilitate the application of BiFC to research of

protein-protein interactions, the author have attempted to determine the structure of

reassembled Venus complemented between the N-terminal and C-terminal fragments.

Ghosh et al. succeeded to express reassembled GFP in E. coli as a fusion protein

with anti parallel leucine zipper-directed proteins (Ghosh et al., 2000). However, the

fusion protein was expressed in insolubility form in E. coli cells. It was also difficult to

obtain the fragments of split fluorescent proteins expressed in E. coli separately due to

intrinsic folding problems and low solubility of these fragments in aqueous solutions

(Own unpublished results.). Thus, the author developed the coexpression system of

two fragments in E. coli using two plasmids. Fluorescent spectroscopy and CD

analysis were carried out to examine the differences between intact Venus (whole

2. Materials and methods

2-1. Expression

N-terminal fragment (amino acid residues 1-154, VN155) and C-terminal

fragment (amino acid residues 155-238, VC155) of Venus were subcloned into pET-16b

and pET-28a plasmids (Novagen), respectively (Fig. 4). VN155 tagged with (His)10 at

its N-terminus and VC155 tagged with (His)6 at its C-terminus were coexpressed in E.

coli strain BL21(DE3) bacteria (Novagen). Cells were grown using standard culture

medium (LB broth) in a shaker incubator at 37ºC. Expression was induced with 0.5 mM

IPTG at O.D.600nm of 0.5 and cultivation was continued for a further 12 h at 22ºC.

Whole Venus (amino acid residues 1-238) was subcloned into pET-3a plasmid

(Novagen) (Fig. 4) and expressed in E. coli BL21(DE3) bacteria. Cells were cultured

under the same conditions as described above.

2-2. Purification

Cells of reassembled Venus were harvested and resuspended in 50 mM sodium

phosphate buffer, pH 8.0, containing 300 mM NaCl and 10 mM imidazole. After

sonication and centrifugation, the supernatant was loaded onto a Ni-NTA column

(QIAGEN) equilibrated with 20 mM Tris-HCl buffer, pH 8.0, containing 150 mM NaCl

(buffer A) and eluted with 250 mM imidazole. The fractions containing reassembled

Venus were combined and concentrated. The concentrated protein solution was loaded

onto a MonoQ anion-exchange column (GE Healthcare) equilibrated with 20 mM

NaCl (20-1000 mM). The anion-exchange chromatography yielded three peaks showing

yellow fluorescence (Fig. 5). The fraction corresponding to each peak was finally

purified by the gel filtration chromatography using a Superdex 200 column (GE

Healthcare) equilibrated with 20 mM sodium phosphate buffer, pH 7.0 (Fig. 6a-c).

Cells of whole Venus were harvested, resuspended in buffer A, and sonicated. The

lysate was centrifuged to obtain a crude sample. The supernatant was loaded onto a

DEAE sepharose ion-exchange column (GE Healthcare) equilibrated with buffer A and

eluted with a linear gradient of NaCl (100-700 mM). Whole Venus samples were further

purified by anion exchange chromatography using MonoQ column followed by a size

exclusion chromatography using Superdex 200 (Fig. 6d).

All purification steps were carried out at 4ºC and an anion exchange

chromatography using MonoQ and size exclusion chromatography using Superdex 200

were performed on the ÄKTAFPLC system (GE Healthcare).

2-3. Fluorescence spectroscopy

Whole Venus, reassembled Venus were diluted to 0.1 mg/ml with 20 mM sodium

phosphate buffer, pH 7.0 for fluorescence spectral analysis. Fluorescence spectra were

2-4. CD spectroscopy

Samples were diluted to 0.1 mg/mL with 20 mM sodium phosphate buffer, pH

7.0 for CD spectral analysis. CD experiments were performed on a J-820

spectropolarimeter (JASCO) with a Peltier PTC-423L thermo-unit (JASCO). The

far-UV CD spectra (260-190 nm) were recorded using a 0.1 cm path length cell under

constant nitrogen flush with a step size of 0.2 nm, bandwidth of 1 nm, and an

Fig. 4 Constructions of VN155, VC155 and whole Venus.

pET-16b(+)

His-tag

VN155

A

_m

p

Nde I

BamH I

(Mw 20 kDa)

pET-28a(+)

His-tag

VC155

K

an

Nco I

_{Xho I}

(Mw 10 kDa)

pET-3a(+)

Whole Venus

A

m

p

Nde I

BamH I

(Mw 27 kDa)

Fig. 5 Anion exchange chromatogram of reassembled Venus.

The elution points of three components of reassembled Venus are shown

along the top.

12.0

15.0

18.0

21.0

24.0

Elutio n volume (ml)

A

b

so

rb

a

n

c

e

a

t

2

8

0

n

m

Fig. 6 Size exclusion chromatograms.

(a) Peak 1 of reassembled Venus. (b) Peak 2 of reassembled Venus.

(c) Peak 3 of reassembled Venus. (d) Whole Venus.

8.00 10.00 12.00 14.00 16.00 18.00 Elution volume (ml) A b so rb a n ce a t 2 8 0 n m 8.00 10.00 12.00 14.00 16.00 18.00 Elution volume (ml) A b so r b a n ce a t 2 8 0 n m 8.00 10.00 12.00 14.00 16.00 18.00 Elution volume (ml) A b so r b a n ce a t 2 8 0 n m 8.00 10.00 12.00 14.00 16.00 18.00 Elution volume (ml) A b so rb a n c e a t 2 8 0 n m (a) (b) (d) (c)

3. Results and discussion

Reassembled Venus was expressed in E. coli as a polyhistidine fusion protein.

The fluorescent fraction isolated by a Ni affinity column chromatography was

separated into three peaks on the anion exchange chromatography (Fig. 5). The profiles

of fluorescent spectra of these three peaks were identical to that of whole Venus (Fig.

7). The N- and C-terminal fragments of split fluorescent protein do not exhibit the

absorption or fluorescence characteristics unless two fragments of split fluorescent

protein completely associate to form a fluorescent complex having a β-barrel

conformation (Sakamoto and Kudo, 2008; Kent et al., 2008). Thus, three peaks should

contain the fluorescent complex formed through the association of two non-fluorescent

fragments at least as a parent compound.

Each peak was further purified by a size exclusion chromatography. The size

exclusion chromatograms of peaks 1 and 2, respectively, showed a single peak (Figs.

6a & 6b). The size exclusion chromatogram of peak 3 yielded two peaks after

incubation for 1 day at 4ºC (Fig. 6c). Both peaks were fluorescent. MALDI-TOF MS

analysis showed that the main components of peak 3 were VN155 and VC155

fragments of Venus as described below. Thus, peak 3a emerged earlier and peak 3b

emerged later were estimated to be an oligomer (octamer to decamer) and a monomer

of reassembled Venus based on their retention volumes, respectively. This view was

supported by CD spectrum of peak 3b which was identical with that of peak 1 (Fig. 8).

After incubation for 3 days at 37ºC, the area of peak 3a further decreased and that of

peak 1, after incubation for 3 days at 37ºC, did not give any other peaks except original

peak. These findings suggest that the oligomer is converted to the monomer in an

irreversible manner or the rate of conversion to oligomer is extremely slow.

Fig. 7 Fluorescence spectra

Fluorescent spectra of whole and reassembled Venus (peak 1) are

indicated solid line and dashed line, respectively.

0

0.2

0.4

0.6

0.8

1

1.2

500

520

540

560

580

600

wavelength (nm)

R

.

F

.

I.

(

a

.

u

.)

Fig. 8 CD spectra

Peak 1 of reassembled Venus and peak 3b of reassembled Venus are

indicated solid line and dashed line, respectively.

-4000

-2000

0

2000

4000

6000

8000

10000

190

210

230

250

Wavelength (nm)

[θ

](

d

eg

.c

m

/d

m

o

l)

Fig. 9 Size exclusion chromatogram of peak 3 of reassembled Venus.

The chromatogram of peak 3 incubated for 1 day at 4ºC is present in gray,

and that of peak3 incubated for 3 days at 37ºC is present in

black. For

comparison, the absorbance is normalized by the maximum absorbance.

The elution points of peak 3a and peak 3b are shown along the top.

8.0

10.0

12.0

14.0

16.0

18.0

Elution volume (ml)

A

b

so

r

b

a

n

ce

a

t

2

8

0

n

m

CHAPTER II

Characterization of reassembled Venus

1. Introduction

GFP has been the subject of continued interest since its gene was first cloned in

1992. Crystal structures of both the monomeric and dimeric forms of GFP have been

solved previously (Ormo et al., 1996; Yang et al., 1996). Chromophore is situated in the

middle of a distorted α-helix that runs through the center of a β-barrel in the folded GFP

structure. Maturation of GFP proceeds in three major steps, beginning as the

238-residue single GFP polypeptide folds into its nearly native conformation. Residues

65–67 of the folded protein then undergo several chemical reactions necessary for

chromophore formation, including cyclization and dehydration. The final rate-limiting

step in the maturation process involves the oxidation of the Cα–Cβ bond of Tyr66 by

aerial oxygen. It has been shown that the protein structure itself plays an essential role

in creating and maintaining a semi-rigid environment around the chromophore where

bulk solvent molecules are excluded and the conformational flexibility of the

chromophore is low. It leads to a state with a high quantum yield.

Maturation process takes ~3 h at room temperature and its efficiency further

decreases at 37°C. Venus was developed to improve maturation rate at 37ºC. The

crystal structure of Venus (PDB: 1MYW) (whole Venus) was determined by X-ray

crystallography at a resolusion of 2.2 Å (Rekas et al., 2002). The structure of whole

Background from spontaneous assembly of the fragments compromises their

utility for detecting protein-protein interactions. Fluorescent protein fragments have a

finite ability to associate with each other independent of an interaction between

proteins fused to the fragments. To facilitate the application of BiFC, the author have

attempted to determine the X-ray crystal structure of reassembled Venus and its

2. Materials and Methods

2-1. MALDI-TOF MS

The molecular mass of each component was measured on an autoflex II

MALDI-TOF analyzer (Bruker Daltonics) using saturated α-cyano-4-hydroxycinnamic

acid in 50%(v/v) acetonitrile and 0.1%(v/v) trifluoroacetic acid as a matrix.

Reassembled Venus was performed in the 20 mM phosphate buffer (pH 7.0) at

concentrations of 100 µg/mL.

2-2. DLS measurement

Measurements of the hydrodynamic radii of whole and reassembled Venus were

performed at 4ºC in the buffer containing 20 mM phosphate buffer (pH 7.0), 100 mM

NaCl at concentrations of 0.5 mg/mL using a Wyatt DynaPro Nanostar dynamic light

scattering instrument (Wyatt Technology). The data were analyzed using Dynamics 6.0

software.

2-3. Crystallization

Prior to crystallization, the purified monomer of reassembled Venus was dialyzed

against buffer A (20 mM Tris-HCl buffer, pH 8.0, containing 150 mM NaCl) and

concentrated using an Ultafree (Millpore) filter. Protein purity was evaluated by

SDS-PAGE (Fig. 10). Crystallization trials were carried out at 20ºC using the

sitting-drop vapor-diffusion technique by mixing 0.5 µL of 35 mg/mL protein sample

the commercially available sparse-matrix screening kit Index (Hampton Research).

Small but good shaped crystals were obtained (Fig. 11) from the reservoir solution

consisting of 25% (w/v) PEG 3350, 0.2 M Lithium sulfate and 0.1 M HEPES buffer pH

7.5.

2-4. Data collection and structure determination

A diffraction data set was collected to a resolution of 2.1 Å at 100 K on an ADSC

CCD detector using synchrotron radiation of wavelength 1.0 Å at the BL5A station of

Photon Factory, Japan. The data set was processed and scaled with the HKL2000

software package (Otwinowski and Minor, 1997). The crystals belong to the space

group P212121, with unit-cell parameters a = 59.05, b = 116.05, c = 156.85 Å. The

crystal structure was solved by the molecular replacement using the program MOLREP

(Vagin and Teplyakov, 1997) with the structure of Venus (PDB: 1MYW) as a search

model. The solution revealed that the crystal contains a tetramer as the asymmetric unit.

The obtained model was subjected to iterative rounds of model building and refinement

using the programs COOT (Emsley and Cowtan, 2004) and REFMAC5 (Winn et al.,

2003) (Fig. 12). The 2Fo-Fc difference map showed clear electron density

residues were in their most favored regions, 8.6% of the residues were in additionally

allowed regions, 0.3% of the residues were in generously allowed regions and no

residues were in disallowed regions (Fig. 13). Data collection and structure refinement

statistics are summarized in Table 1. All structure figures were prepared with the

program PyMOL (DeLano, 2002, http://www.pymol.org.). Coordinates have been

deposited with the Protein Data Bank with the following accession codes: 3AKO.

Fig. 10 SDS-PAGE. Fig. 11 A crystal of reassembled Venus.

Fig. 12 The 2Fo-Fc electron density map of the chromophore

The 2Fo-Fc electron density map contoured at 1.5 σ showing the refined

structure of the chromophore of reassembled Venus.

Table 1 Summary of data collection and refinement statistics.

Data Set reassembled Venus Data collection

X-ray source BL5A, PF

Wavelength (Å) 1.00 Space group P2₁2₁2₁ Unit cell (Å) a = 59.05 b = 116.05 c = 156.85 Resolution (Å) 50.0 – 2.10 (2.18-2.10) Unique reflections 63893 (6263) Completeness (%) 99.9 (99.9) Rmerge (%)b 10.2 (32.9) I/

σ

Values in parentheses are for the highest resolution shell.

b

Rmerge=ΣhΣj|Ihj-<Ih>|/ΣhΣj|Ihj|, where h represents a unique reflection and j represents symmetry-equivalent indices. I is the observed intensity and <I> is the mean value of I.

c

3. Results and discussion

Based on its retention volume in size exclusion chromatography (Fig. 6) and

MALDI-TOF MS data (Table 2), peak 1 was elucidated to be a monomer of

reassembled Venus. The structure of the species corresponding to peak 1 was

confirmed by X-ray structure analysis. The retention volume indicated that peak 2 was

also a monomer of reassembled Venus. However, the molecular weight of VN155

found in MS is approximately 180 Da larger than the calculated value (Table 2). These

results suggested that a part of reassembled Venus might be modified during incubating

in E. coli.

DLS data (Table 3) indicated that peak 3 was a molecular aggregate which have

a molecular weight corresponding to an octamer of reassembled Venus. However, as

previously indicated in CHAPTER I, this octamer is converted to a monomer in an

irreversible manner or the rate of conversion to oligomer is extremely slow.

The crystal structure of monomeric reassembled Venus was determined at 2.1 Å

resolution. There are four reassembled Venus molecules formed through the

association of two VN155 and VC155 fragments in an asymmetric unit (Fig. 14). Each

molecule has an eleven-stranded β-barrel fold including a chromophore in the middle,

concentration of protein and in high-salt concentration. The dimer structure may be

stabilized in crystals. The locations of the residues surrounding the chromophore of

reassembled Venus are quite similar to those surrounding the chromophore of whole

Venus (Fig. 16). Fluorescent spectrum of reassembled Venus was identical to whole

Venus (Fig. 7). There is no difference between the environment surrounding

chromophore of reassembled Venus and that of whole Venus.

In the β-barrel fold between reassembled Venus and whole Venus, a slight but

significant difference was found. R.m.s.d. values of the mainchain calculated using

LSQKAB with monomer of reassemble and whole Venus are 0.59 Å for N-terminal

fragment (amino acid residues 1-154) and 1.66 Å for C-terminal fragment (amino acid

residues 155-230), respectively. As compared with whole Venus, the seventh β-strand

(β7) of reassembled Venus is shortened (Figs. 17 & 18). Unlike whole Venus, two

amino acid residues, Asn146 and Ser147, didn’t contribute to form β7. The loop

(amino acid residues from 138 to 147) of reassembled Venus was expanded and more

flexible than that of whole Venus by lacking of hydrogen bonds between β7 and β10.

This means that the formation of the β-barrel structure of reassembled Venus is

Table 2 TOF-MS of reassembled Venus.

Comparison between the found and calculated masses of proteins.

Table 3 Hydrodynamic Radii, polydispersity values (Pd%) and

molecular weight of reassembled Venus, as determined by DLS

analysis.

peak1

Found

Calculated

VN155

VC155

VN155

VC155

19925

19920

10558

10552

20104

10553

19917

10552

peak2

peak3

Radius (nm) %Mass Mw (kDa)

2.5 6.3 30 245 100.0 99.8 peak3 %Pd 13.2 19.1 Polymerization monomer octamer

Fig. 14 Crystal structure of reassembled Venus.

Ribbon diagram of a tetramer of reassembled Venus, which consists of

VN155 (Chain A, C, E, G) and VC155 (Chain B, D, F, H).

Chain A

Chain B

Chain E

Chain F

Chain C

_{Chain D}

Chain G

Chain H

Chain A

Chain B

Chain E

Chain F

Chain C

_{Chain D}

Chain G

Chain H

Chain A

Chain B

Chain E

Chain F

Chain C

_{Chain D}

Chain G

Chain H

Fig. 15 Superposition of the main chain of the dimer structure of

reassembled (yellow) and whole Venus (cyan).

chromophore

Tyr203

His148

Asn146

Ser205

Glu222

Thr62

Arg96

Gln94

Fig. 17 Comparison of reassembled Venus and whole Venus.

Whole Venus is represented in cyan. VN155 and VC155 of reassembled

Venus are represented in orange and yellow, respectively.

Fig. 18 β

β

β

β7 of reassembled Venus is shorter than that of whole Venus.

Hydrogen bond (dashed line) between β7, β10 and β11of reassembled

(yellow) and whole Venus (cyan).

β ββ β7 β ββ β8 β β β β9 β ββ β10 β β β β11 β ββ β7 β β β β8 β β β β9 β ββ β10 β ββ β11 β β β β4 ββββ4 C C N N C N β β β β7 β β β β10 β ββ β11 β ββ β7 β β β β10 β ββ β11

CHAPTER III

Thermal stabilities and reassembly mechanism

1. Introduction

Since the identification of the GFP from jellyfish Aequorea victoria in the early

1990s, a large number of fluorescent proteins have been isolated from natural sources,

primarily from marine animals and corals (Hsu et al., 2010). mFruits are

second-generation mRFPs that improved brightness and photostability compared to the

first-generation mRFP1 (Shu et al., 2006). mCherry (λex = 587 nm, λem = 610 nm)

derived from Discosoma sp. (Figs. 19 & 20), has been developed for its fast

chromophore maturation rate as same as yellow fluorescent protein, Venus (Shaner et

al., 2004). The brilliant redness, short maturation time, and the long excitation and

emission wavelengths of mCherry make the new BiFC system for analyzing protein–

protein interactions in living cells and for studying multiple protein – protein

interactions when coupled with other BiFC systems. This new red BiFC system was

developed by splitting mCherry, into two fragments between amino acids 159-160 as

well as splitting Venus.

Fig. 19 Discosoma sp.

2. Materials and Methods

2-1. Preparation of mutants

Three mutants of VN155 (Y143F, Y145F, H148G) were constructed by

site-directed mutagenesis of plasmids carrying VN155 (pET-16b) by PCR using Pfu

Turbo (Stratagene) and primers (Table 4). The sequence of mutants were verified by

DNA sequencing with a dye terminator cycle sequencing kit (Beckman Coulter) and a

CEQ2000 fragment analysis system (Beckman Coulter). Reassembled Venus

consisting of VN155 mutants and VC155 (denoted as rV-Y143F, rV-Y145F,

rV-H148G) were coexpressed and purified by the same procedures as those used for

reassembled Venus.

2-2. DSC measurement

Calorimetric experiments were carried out with a nanoDSC (TA instruments).

Samples were prepared in concentrations of 0.5 and 1.0 mg/mL. The buffer used for the

sample was 20 mM sodium phosphate, pH 7.0. Experiments were performed over a

temperature range of 25-95ºC at a scan rate of 1 ºC/min and excess pressure of 2.8 atm.

in E. coli strain BL21(DE3) bacteria (Novagen). Whole mCherry (amino acid residues

1-238) was subcloned into pRSET-B plasmid and expressed in E. coli strain

BL21(DE3)pLysS bacteria. Cells were grown using LB broth in a shaker incubator at

37ºC. Expression was induced with 0.5 mM IPTG at O.D.600nm of 0.5 and cultivation

was continued for a further 12 h at 22ºC.

Cells of reassembled mCherry were harvested and resuspended in 50 mM

sodium phosphate buffer,pH 8.0, containing 300 mM NaCl and 10 mM imidazole.

After sonication and centrifugation, the supernatant was loaded onto a Ni-NTA column

equilibrated with 20 mM Tris-HCl buffer, pH 8.0, containing 150 mM NaCl (buffer A)

and eluted with 250 mM imidazole. The fractions containing reassembled Venus were

combined and concentrated. Reassembled mCherry was finally purified by the gel

filtration chromatography using a Superdex 200 column (GE Healthcare) equilibrated

with 20 mM sodium phosphate buffer, pH 7.0.

Cells of whole mCherry were harvested, resuspended in buffer A, and sonicated.

The lysate was centrifuged to obtain a crude sample. The supernatant was loaded onto a

DEAE sepharose ion-exchange column (GE Healthcare) equilibrated with buffer A and

eluted with a linear gradient of NaCl (100-700 mM). Whole mCherry samples were

eluted in the wash buffer. Whole mCherry samples were further purified by a size

2-4. HPLC analysis of Venus and mCherry

Whole Venus, reassembled Venus, whole mCherry and reassembled Venus were examined by HPLC analysis at room temperature. HPLC analysis was performed using L-6200 Intelligent Pump (HITACHI), ELITE LaChrom L-2400 UV Detector (HITACHI), D-2500 Chromato-Integrator (HITACHI). Columns were TSKgel SuperSW2000 4.5*300 (TOSOH) and TSK guardcolumn SuperSW 4.6*3.5 (TOSOH). mCherry and Venus were detected at 587 nm and 515 nm, respectively.

Table 4 List of primers.

Y143F forward 5’- CACAAGCTGGAGTTCAACTACAACAGC -3’

Y143F reverse 5’- GCTGTTGTAGTTGAACTCCAGCTTGTG -3’

Y145F forward 5’- GGAGTACAACTTCAACAGCCACAAC -3’

Y145F reverse 5’- GTTGTGGCTGTTGAAGTTGTACTCC -3’

H148G forward 5’- CAACTACAACAGCGGCAACGTCTATATC -3’

H148G reverse 5’- GATATAGACGTTGCCGCTGTTGTAGTTG -3’

3. Results and discussion

Based on this structural features, mutations in β7 of VN155, Y143F, Y145F and

H148G (Figs. 21 & 22), were introduced to see if any change in the thermal stability of

reassembled fluorescent complex is observed.

DSC studies were carried out on whole Venus, reassembled Venus and mutants

(Fig. 23, Table 5). The profiles were analyzed mainly in terms of the peak temperatures,

because all of the samples showed irreversible transition. The thermogram of whole

Venus shows a sharp single peak at 89.0ºC (Tm). On the other hand, the thermogram of

monomeric reassembled Venus shows a sharp single peak at a lower temperature of

77.5ºC. The Tm value of the monomeric reassembled Venus did not fall at a lower

concentration of 0.1 mg/mL. It can therefore be presumed that the dissociation and the

thermal denaturation of monomeric reassembled Venus occur simultaneously at around

Tm value.

The thermograms of three mutants of reassembled Venus, rV-Y143F, rV-Y145F

and rV-H148G, showed Tm values of 77.1ºC, 83.8ºC and 72.1ºC, respectively (Table 5,

Fig. 23). The OH group of Tyr143 forms a hydrogen bond with the carbonyl O atom of

Ser208 (Fig. 22a). Despite the hydrogen bond is lacking, Tm value of the rV-Y143F

mutant was the almost same as that of reassembled Venus. Surprisingly, the

substitution of Tyr145 with phenylalanine enhanced the thermal stability compared

with monomeric reassembled Venus. Tyr145 and His169 are linked by the hydrogen

bond network through two water molecules (Fig. 22b). The substitution seems to cause

hydrophobic amino acids, Val61 and Ile167, or cause the rearrangement of those amino

acids into alternative dense-packing structure. In contrast, the significant decrease in

thermal stability was observed when His148 was substituted with glycine. This mutant

seems to be useful for BiFC assay, because the lacking of the hydrogen bonds (Fig.

22c) may weaken the binding force between N- and C-fragments of Venus.

DSC measurement for the oligomer of reassembled Venus showed two peaks;

one peak is identical to that for the monomer of reassembled Venus and has a

maximum at 77.6ºC, and the other peak spreads through the range from 40ºC to 65ºC

(Fig. 24, cyan). The size exclusion chromatography indicated that the oligomer was

readily converted to the monomer. Thus, it can be considered that the conversion from

oligomer to monomer occurs while temperature rises to ~65ºC.

HPLC analysis was carried out on whole Venus, reassembled Venus, whole

mCherry and reassembled mCherry to confirm whether reassembled mCherry forms an

oligomer as well as reassembled Venus. Reassembled mCherry is a split red fluorescent

protein at amino acid residue 159/160. The size exclusion chromatography of whole

mCherry yielded one peak as same as whole Venus (Fig. 25). On the other hand, the

chromatography of reassembled mCherry yielded two peaks as same as reassembled

of mCherry has a great thermal stability compared to those of Venus.

BiFC assay utilizes the formation of the fluorescent complex through the

association of two non-fluorescent N- and C-terminal fragments of the fluorescent

protein when they are brought together by an interaction between two target proteins

fused to the fragments. Thus, oligomerization will not be preferable for BiFC assay,

because it will prevent the interaction between two target proteins and increase

background. In the β-strand swapping, it has been also reported that proline residues

play an important role (Bergdoll et al., 1997). The cis-trans proline isomerization, a

very low energy barrier, is a relatively slow process that can affect the protein folding

pathway. Interestingly, reassembled Venus has three proline residues, Pro187, Pro192

and Pro196, and mCherry has two proline residues, Pro186, Pro190, on the loop region

between β9 and β10 (Fig. 27) and they seem to play a critical role in the domain,

including β10- and β11-strands or β8- and β9-strands, swapping. It is suggested that

replacing a proline with another amino acid should depress the formation of oligomers.

Here, the author proposes the model of reassembly mechanism of fluorescent

protein (Fig. 28). After coexpression in E. coli, most of VN155 and VC155 form a

metastable oligomer by β-strand swapping as described above. Then, this oligomer is

converted irreversibly to a more stable monomer. However, ∆H of oligomer (16.3

KJ/mol) was lower than that of monomer (22.9 KJ/mol). It is suggested that a part of

Fig. 21 Three mutants of VN155 of reassembled Venus.

Fig. 22 Close-up views of three amino acids around β

β

β7.

β

Hydrogen bonds are indicated by dashed lines. Water molecules are drawn

with pink spheres. (a) Tyr143. (b) Tyr145. (c) His148.

Tyr143 Ser208 His169 Tyr145 His148 chromophore Arg168 (a) (b) (c)

Fig. 23 DSC of whole and reassembled Venus and its mutants.

Heat capacity curves for whole Venus (black), peak 1 of reassembled

Venus (red), Y143F (green), Y145F (pink), H148G (blue).

Table 5 Comparison of T

value (ºC).

20 30 40 50 60 70 80 90 100 E x ce ss H ea t C ap ac it y Temperature / oC 20 kJ K-1 mol-1

reassembled Venus

77.5

Mutant rV-Y143F

77.1

rV-Y145F

83.8

Fig. 24 DSC of reassembled Venus.

Heat capacity curves for peak 1 of reassembled Venus (red) and peak 3 of

reassembled Venus (cyan).

20 30 40 50 60 70 80 90 100 E x ce ss H ea t C ap ac it y Temperature / oC 20 kJ K-1 mol-1 peak 1 peak 3 ∆H = 22.9 KJ/mol ∆H = 16.3 KJ/mol

Fig. 25 Chromatogram of whole (top) and reassembled mCherry

whole mCherry

reassembled mCherry

0

20 min

Fig. 26 DSC of Venus and mCherry.

Heat capacity curves for peak 1 of reassembled Venus (green), whole

mCherry (pink), reassembled mCherry (blue), respectively.

30 40 50 60 70 80 90 100 110 60 80 100 120 140 160 k J K -1 m o l -1 Temperature / oC VNVC1 1.0 mg/ml cherry 1.0 mg ml cherryR 1.0 mg/ml

30 40 50 60 70 80 90 100

110

Temperature (ºC)

160

140

120

100

80

60

E

n

d

o

.

E

x

ce

ss

H

ea

t

C

ap

ac

it

y

(k

J

K

m

o

l

)

Fig. 27 Position of prolines in reassembled Venus and mCherry.

Prolines are depicted as spheres.

reassembled Venus

mCherry

2

3

11

10

7

8

9

4

5

6

1

Pro Pro Pro

2

3

11

10

7

8

9

4

5

6

1

2

3

11

10

7

8

9

4

5

6

1

Pro Pro Pro

Pro187

Pro192

Pro196

Pro186

Fig. 28 Model of reassembly mechanism of fluorescent protein.

oligomer

misfold

Conclusion

Bimolecular fluorescence complementation (BiFC) assay has been used

widely to visualize protein-protein interactions in cells. However, there is a problem

that fluorescent protein fragments have an ability to associate with each other

independent of an interaction between proteins fused to the fragments. To facilitate the

BiFC assay, the author attempted to determine the structure and characteristics of

reassembled fluorescent protein, Venus.

The author succeeded to coexpress of N- and C- terminal fragments of

fluorescent protein, Venus. In the coexpression using E. coli, there were two distinct

thermodynamically stable and metastable forms, monomer and oligomer of

reassembled Venus. The author also succeeded to identify these distinct components by

use of MALDI-TOF MS, DLS measurements, CD spectroscopy, DSC analysis and

X-ray crystallography. Crystal structure of reassembled Venus was determined at 2.1 Å

resolution. It had an eleven-stranded β-barrel fold, typical of GFP-derived fluorescent

proteins.

Despite the overall structure of monomeric reassembled Venus was quite

aggregation including oligomerization is said to occur by specific intramolecular

associations involving the recognition of a sequence partner in another molecule rather

than in the same molecule during the folding process. The author proposed that

cis-trans proline isomerization caused β-strand swapping during folding process. Thus,

it can be expected that replacing a proline with another amino acid depresses the

formation of oligomers. The interface plays an important role in the complementary

association of two non-fluorescent N- and C-terminal fragments. Based on the

structural features, the author mutated amino acids adjacent β7 and measured Tm

values. The results have clearly showed that the mutation was susceptible to thermal

stability of reassembled fluorescent complex. Thus, the substitution of amino acids in

four β-strands (β8, β9, β10 and β11) of the C-fragment and three β-strands (β3, β4 and

β7) of N-fragment is also expected to facilitate the application of BiFC to research of

protein-protein interactions. The crystal structure of the monomer of reassembled

Venus including water molecules provides new insights into decreasing background

References

M. Bergdoll, M.H. Remy, C. Cagnon, J.M. Masson, P. Dumas,

Proline-dependent oligomerization with arm exchange. Structure. 5 (1997) 391-401.

W. DeLano, The PyMOL Molecular Graphics System. DeLano Scientific, San

Carlos, CA. http://www.pymol.org., (2002).

P. Emsley, K. Cowtan, Coot: model-buildings tools for molecular graohics.

Acta Cryst. D. 60 (2004) 2126-2132.

J.Y. Fan, Z.Q. Cui, H.P. Wei, Z.P. Zhang, Y.F. Zhou, Y.P. Wang, X.E. Zhang,

Split mCherry as a new red bimolecular fluorescence complementation system for

visualizing protein-protein interactions in living cells. Biochem. Biophys. Res.

Commun. 367 (2008) 47-53.

I. Ghosh, A.D. Hamilton, L. Regan, Antiparallel leucine zipper-directed

protein reassembly: Application to the green fluorescent protein. J. Am. Chem. Soc.

122 (2000) 5658-5659.

W.T. Heller, H.M. O’Neill, Q. Zhang, G.A. Baker, Characterization of the

influence of the ionic liquid 1-butyl-3-methylimidazolium chloride on the structure and

C.D. Hu, Y. Chinenov, T.K. Kerppola, Visualization of interactions among

bZIP and Rel family proteins in living cells using bimolecular fluorescence

complementation. Mol. Cell. 9 (2002) 789-798.

C.D. Hu, T.K. Kerppola, Simultaneous visualization of multiple protein

interactions in living cells using multicolor fluorescence complementation analysis.

Nat. Biotechnol. 21 (2003) 539-545.

K.P. Kent, W. Childs, S.G. Boxer, Deconstructing green fluorescent protein. J.

Am. Chen. Soc. 130 (2008) 9664-9665.

T.K. Kerppola, Design and implementation of bimolecular fluorescence

complementation (BiFC) assays for the visualization of protein interactions in living

cells. Nat. Protoc. 1 (2006) 1278-1286.

Y. Kodama, C.D. Hu, An improved bimolecular fluorescence

complementation assay with a high signal-to-noise ratio. Biotechniques. 49 (2010)

793-805.

T. Nagai, K. Ibata, E.S. Park, M. Kubota, K. Mikoshiba, A. Miyawaki, A

variant of yellow fluorescent protein with fast and efficient maturation for

cell-biological applications. Nat. Biotechnol. 20 (2002) 87-90.

M. Ormö, A.B. Cubitt, K. Kallio, L.A. Gross, R.Y. Tsien, S.J. Remington,

Crystal structure of the Aequorea victoria green fluorescent protein. Science. 273

C. Ottmann, M. Weyand, A. Wolf, J. Kuhlmann, C. Ottmann, Applicability of

superfolder YFP bimolecular fluorescence complementation in vitro. Biol. Chem. 390

(2009) 81-90.

Z. Otwinowski, W. Minor, Processing of X-ray diffraction data collected in

oscillation mode. Methods Enzymol. 276 (1997) 307-326.

A. Rekas, J.R. Alattia, T. Nagai, A. Miyawaki, M. Ikura, Crystal structure of

Venus, a yellow fluorescent protein with improved maturation and reduced

environmental sensitivity. J. Biol. Chem. 277 (2002) 50573-50578

S. Sakamoto, K. Kudo, Supramolecular control of split-GFP reassembly by

conjugation of β-cyclodextrin and coumarin units. J. Am. Chem. Soc. 130 (2008)

9574-9582.

N.C. Shaner, R.E. Campbell, P.A. Steinbach, B.N.G. Giepmans, A.E. Palmer,

R.Y. Tsien, Improved monomeric red, orange and yellow fluorescent proteins derived

from Discosoma sp. red fluorescent protein. Nat. Biotechnol. 22 (2004) 1567-1572.

X. Shu, N.C. Shaner, C. A. Yarbrough, R.Y. Tsien, S.J. Remington, Novel

chromophores and buried charges control color in mFruits. Biochemistry. 45 (2006)

9639-9647.

Y.J. Shyu, C.D. Suarez, C.D. Hu, Visualization of AP-1-NF-κB ternary

complexes in living cells by using a BiFC-based FRET. Proc. Natl. Acad. Sci. U S A.

105 (2008) 151-156.

E. Takács, O. Barabás, M.V. Petoukhov, D.I. Svergun, B.G. Vértessy,

Molecular shape and prominent role of β-strand swapping in organization of dUTPase

oligomers. FEBS Lett. 583 (2009) 865-871.

A. Vagin, A. Teplyakov, MOLREP: an automated program for molecular

replacement. J. Appl. Cryst. 30 (1997) 1022-1025.

M. Winn, G.N. Murshudov, M.Z. Papiz, Macromolecular TLS refinement in

REFMAC at moderate resolutions. Methods Enzymol. 374 (2003) 300-321.

F. Yang, L.G. Moss, G.N. Phillips, The molecular structure of green

List of publication

Masami Isogai, Yoshihiro Kawamoto, Kazuto Inahata, Harumi Fukada, Kenji

Sugimoto and Toshiji Tada, Structure and characteristics of reassembled fluorescent

protein, a new insight into the reassembly mechanism, Bioorg. Med. Chem. Lett., In

Acknowledgements

The author expresses gratitude to Dr. Toshiji Tada, Professor of Graduate school

of Science, Osaka Prefecture University, for his kind guidance, invaluable advice and

stimulating discussion on this work.

The author wishes to thank Dr. Satoru Tokutomi and Dr. Ikuo Fujii, Professors of

Graduate school of Science, Osaka Prefecture University, for their helpful advice and

critical reading of this thesis.

The author also wishes to thank Dr. Kenji Sugimoto, Professor of Graduate

school of Life and Environmental Sciences, Osaka Prefecture University, for providing

materials and helpful suggestions.

Thanks are due to Dr. Takayoshi Kinoshita, Associate professor of Graduate

school of Science, Osaka Prefecture University, and Dr. Maki Onda, Assistant

professor of Graduate school of Science, Osaka Prefecture University, for their helpful

suggestions and observations.

The author is grateful to Dr. Harumi Fukada, Associate professor of Graduate

school of Life and Environmental Sciences, Osaka Prefecture University, Osaka

Prefecture University, for her kind cooperation.

Thanks are also due to Dr. Shigenori Nishimura, Assistant professor of Graduate

school of Life and Environmental Sciences, Osaka Prefecture University, for his kind

The author would like to thank the beamline staff at KEK-PF BL5A for

assistance in the synchrotron experiments.

Finally, special thanks are due to all members of Laboratory of Structural

Biology, Department of Biological Science, Graduate school of Science, Osaka