1. Trang chủ
  2. » Luận Văn - Báo Cáo

Tài liệu Báo cáo khoa học: Steady-state and time-resolved fluorescence studies of conformational changes induced by cyclic AMP and DNA binding to cyclic AMP receptor protein from Escherichia coli ppt

11 499 0

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 11
Dung lượng 348,61 KB

Các công cụ chuyển đổi và chỉnh sửa cho tài liệu này

Nội dung

Measurement of fluorescence energy transfer FRET between labeled Cys178 and Trp85 showed that the binding of cAMP in the CRP–cAMP2complex caused a substan-tial increase in FRET efficiency..

Trang 1

Steady-state and time-resolved fluorescence studies of conformational changes induced by cyclic AMP and DNA binding to cyclic AMP

Agnieszka Polit, Urszula Błaszczyk and Zygmunt Wasylewski

Department of Physical Biochemistry, Faculty of Biotechnology, Jagiellonian University, Krako´w, Poland

cAMP receptor protein (CRP), allosterically activated by

cAMP, regulates the expression of several genes in

Escheri-chia coli As binding of cAMP leads to undefined

conform-ational changes in CRP, we performed a steady-state and

time-resolved fluorescence study to show how the binding of

the ligand influences the structure and dynamics of the

protein We used CRP mutants containing a single

trypto-phan residue at position 85 or 13, and fluorescently labeled

with 1,5-I-AEDANS attached to Cys178 Binding of cAMP

in the CRP–(cAMP)2complex leads to changes in the Trp13

microenvironment, whereas its binding in the CRP–

(cAMP)4complex alters the surroundings of Trp85

Time-resolved anisotropy measurements indicated that cAMP

binding in the CRP–(cAMP)2complex led to a substantial

increase in the rotational mobility of the Trp13 residue

Measurement of fluorescence energy transfer (FRET)

between labeled Cys178 and Trp85 showed that the binding

of cAMP in the CRP–(cAMP)2complex caused a substan-tial increase in FRET efficiency This indicates a decrease in the distance between the two domains of the protein from 26.6 A˚ in apo-CRP to 18.7 A˚ in the CRP–(cAMP)2 com-plex The binding of cAMP in the CRP–(cAMP)4complex resulted in only a very small increase in FRET efficiency The average distance between the two domains in CRP–DNA complexes, possessing lac, gal or ICAP sequences, shows an increase, as evidenced by the increase in the average distance between Cys178 and Trp85 to 20 A˚ The spectral changes observed provide new structural information about the cAMP-induced allosteric activation of the protein

Keywords: allosteric regulation; cAMP receptor protein; emission anisotropy; Escherichia coli; fluorescence

cAMP receptor protein (CRP), which is allosterically

activated by cAMP, regulates transcription of over 100

genes in Escherichia coli [1,2] Upon binding the cyclic

nucleotide, CRP undergoes an allosteric conformational

change that allows it to bind specific DNA sequences with

increased affinity [3] CRP is a dimeric protein, composed of

two identical 209-amino-acid subunits Each subunit of

CRP has a molecular mass of 23.6 kDa, as deduced from

the amino-acid sequence Individual subunits fold into two

domains [4] The larger N-terminal domain (residues 1–133)

is responsible for dimerization of CRP and for interaction

with the allosteric effector, cAMP The smaller C-terminal

domain (residues 139–209) is responsible for interaction

with DNA through a helix–turn–helix motif CRP

recog-nizes a 22-bp, symmetric DNA site [5] Amino-acid residues

134–138 form a flexible hinge which covalently couples two domains Recent studies of the crystal structure of the CRP– DNA complex showed that each protein subunit binds two cAMP molecules with different affinities [6] Higher-affinity sites, where the nucleotide binds in the anti conformation, are buried within the N-terminal domains, whereas lower-affinity binding sites (where the bound cAMP has a syn conformation) are located at the interface formed by the two C-terminal domains of the CRP subunits, interacting with a helix–turn–helix motif and, indirectly, with the DNA Crystallographic observations have been supported by recent NMR [7] and isothermal titration calorimetry studies [8] Therefore, it has been suggested that CRP exists in three conformational states: free CRP, CRP with two cAMP molecules bound to N-terminal domains [CRP–(cAMP)2], and CRP with four cAMP molecules bound to both N-terminal and C-terminal domains [CRP–(cAMP)4] An earlier hypothesis suggested [9] that the three conforma-tional states of CRP consisted of the following species: free CRP, CRP–(cAMP)1and CRP–(cAMP)2, which has been reinterpreted by Passner & Steitz [6] It is important to note that the behavior of CRP at different concentrations of cAMP is essentially biphasic, so two different conformers exist at lower and higher concentrations of cAMP In the presence of 100 lMcAMP, CRP becomes activated and is able to recognize and bind specific DNA sequences and stimulate transcription [10], whereas at millimolar concen-trations of cAMP, there is a loss of affinity and sequence specificity for DNA binding and, consequently, loss of transcription stimulation [11] In the crystal phase, the CRP

Correspondence to Z Wasylewski, Department of Physical

Biochemistry, Faculty of Biotechnology, Jagiellonian University,

ul Gronostajowa 7, 30-387 Krako´w, Poland.

Fax: + 48 12 25 26 902, Tel.: + 48 12 25 26 122,

E-mail: wasylewski@mol.uj.edu.pl

Abbreviations: 1,5-I-AEDANS,

N-iodoacetylaminoethyl-1-naphthyl-amine-5-sulfonate; AEDANS-CRP, CRP covalently labeled with

1,5-I-AEDANS attached to Cys178; apo-CRP, unligated CRP;

CRP, cAMP receptor protein; FRET, fluorescence resonance

energy transfer.

(Received 31 October 2002, revised 19 December 2002,

accepted 3 February 2003)

Trang 2

dimer with two molecules of cAMP bound to anti-cAMP

binding sites is asymmetric, i.e one monomer is in an open

form, in which the a-helices are swung out away from the

N-terminal domain, and the other monomer is in a closed

form, in which the a-helices are swung in close to the

N-terminal domain [4] X-ray crystal structure studies

revealed that the cAMP-ligated CRP dimer complexed to

a 30-bp DNA sequence is exclusively in the closed form [12]

Each subunit of CRP contains two tryptophan residues at

positions 13 and 85 (Fig 1) Both residues are located in the

N-terminal domain, Trp85 near the cAMP-binding pocket

and Trp13 on the surface of the protein Trp13 is much more

accessible to solvent than Trp85 19F-NMR studies have

revealed that binding of cAMP induces not only changes in

the immediate environment of the cAMP-binding site but

also far-reaching conformational changes (perturbation of

chemical shift of Trp13) [13] Fluorescence studies have

shown that Trp13 is responsible for 80% of the tryptophan

fluorescence in CRP with 20% of the signal originating

from Trp85 [14] Acrylamide-mediated and iodide-mediated

fluorescence quenching studies indicate that Trp13 is

solvent-exposed and accessible to the quenching agents On the other

hand, Trp85 is inaccessible to quenching agents Heyduk &

Lee [9] have shown that, at micromolar cAMP

concentra-tions, there is no detectable change in fluorescence intensity

of protein tryptophan residues The increase was only

detected in the millimolar range of cAMP concentration

Biochemical and biophysical studies have demonstrated

that binding of cAMP allosterically induces CRP to assume a

conformation that binds to DNA and interacts with RNA

polymerase However, the details of the mechanism by which

cAMP mediates the allosteric activation of CRP remain obscure, because the crystal structure of apo-CRP has not yet been elucidated Therefore, we decided to study the changes

in the apo-CRP structure induced by binding of cAMP and DNA by measuring fluorescence resonance energy transfer (FRET) and fluorescence anisotropy decay Two tryptophan residues, Trp13 and Trp85, were used as intrinsic donors for FRET analysis The fluorescence anisotropy decay was used

to investigate the dynamics of CRP

Experimental procedures

Materials Acrylamide, KCl, EDTA, phenylmethanesulfonyl fluoride, Tris and N-iodoacetylaminoethyl-1-naphthylamine-5-sulfo-nate (1,5-I-AEDANS) were purchased from Sigma cAMP and dithiothreitol were from Fluka The Fractogel EMD

SO3650 (M) was from Merck, and Q-Sepharose Fast Flow, Sephacryl S-200 HR and Sephadex G-25 were from Amersham Pharmacia Biotech DNA sequence containing the CRP-binding sites was from TIB MOLBIOL (Poznan´, Poland) The nutrients for bacterial growth were from Life Technologies All other chemicals were analytical-grade products of POCh-Gliwice (Gliwice, Poland) All measure-ments were performed in buffers prepared in water purified with the Millipore system

The sequences of duplexes used in this study were as follows:

lac(26 bp), 5¢-ATTAATGTGAGTTAGCTCACTCATT A-3¢ and 3¢-TAATTACACTCAATCGAGTGAGTAAT-5¢; gal (26 bp), 5¢-AAAAGTGTGACATGGAATAAATT AGT-3¢ and 3¢-TTTTCACACTGTACCTTATTTAATCA-5¢; ICAP (28 bp), 5¢-AATTAATGTGACATATGTCACAT TAATT-3¢ and 3¢-TTAATTACACTGTATACAGTGTAAT TAA-5¢

The recognition half-sites are shown in bold An equimolar amount of complementary strand was added, and the mixture was heated for 1 min at 96C and slowly cooled to room temperature The double-stranded DNA was stored at)20 C in experimental buffer

Protein purification The tryptophans at positions 13 and 85 of CRP were replaced with phenylalanine and alanine, respectively The mutagen-esis was performed using the overlap extension method with PwoDNA polymerase pHA7 plasmid encoding mutant crp genes was introduced into E coli strain M182Dcrp, kindly provided by Dr S Busby (The University of Birmingham, UK) The bacteria were grown on Luria–Bertani medium at

37C overnight in a Biostat B fermentor from B Braun Biotech International (Melsungen, Germany) Proteins were purified at 4C, essentially as described previously [15] but with one modification After ion-exchange chromatography

on Q-Sepharose, the proteins were additionally purified by gel filtration on Sephacryl S-200 HR After this procedure, the proteins were highly pure (> 97%), as judged by SDS/ PAGE and Coomassie Brilliant Blue staining

For spectrophotometric determination of concentrations, the following absorption coefficients were used: 14 650

)1Æcm)1at 259 nm for cAMP [16] and 6000 )1Æcm)1at

Fig 1 Structure of the CRP dimer The locations of tryptophan

resi-dues are marked in red, and the location of Cys178 residue is indicated

in yellow The figure was generated with WEBLAB VIEWERPRO (version

3.7) using atomic coordinates for the CRP–cAMP complex [29] The

coordinates were obtained from the Brookhaven Protein Data Bank

(accession code 1G6N).

Trang 3

340 nm for AEDANS [18] Absorption coefficients of CRP

mutants were determined using the method described

elsewhere [19] as 29 700M )1Æcm)1and 33 100M )1Æcm)1at

278 nm for the W13F and W85A dimers, respectively

Measurements were performed in 50 mM Tris/HCl

buffer, pH 8.0, containing 100 mM KCl and 1 mM

EDTA (buffer A), and 50 mM Tris/HCl buffer, pH 7.8,

supplemented with 100 mM KCl and 1 mM EDTA

(buffer B)

Fluorescence labeling of CRP

Covalent modification of Trp mutants with 1,5-I-AEDANS

was carried out as described elsewhere [20] with several

modifications The protein and label were mixed at a molar

ratio of 1 : 10 and incubated at room temperature for 2 h,

and then at 4C overnight in the dark The labeled CRP

was purified on a Sephadex G-25 column equilibrated with

buffer A Fractions displaying a high absorbance at both

280 and 340 nm were combined and dialyzed extensively

against buffer A

Mapping of modified residues

Mapping of labeled residues was performed as described

previously [21] with several modifications Peptides were

separated by an HPLC system consisting of (a) a Shimadzu

LC-9A pump equipped with FCV-9AL low-pressure

pro-portioning valve, (b) a Knauer A0263 manual injector

equipped with a 100-lL loop, (c) a Supelcosil LC-318

HPLC (5 lm) cartridge column (250· 4.6 mm) with

20· 2.1 mm Supelguard LC-318 precolumn, (d) a

Merck-Hitachi L-4000A detector, (e) a Shimadzu RF-535

fluores-cence monitor, and (f) a Shimadzu Class-VP 1-2 hardware/

software system for data acquisition and analysis Solvent A

was 0.1% trifluoroacetic acid in water, and solvent B was

0.08% trifluoroacetic acid in 80% acetonitrile A linear

gradient of 10–70% solvent B over 40 min was applied at a

flow rate of 1 mLÆmin)1, with spectrophotometric detection

at 215 nm and fluorescence detection at an excitation

wavelength of 336 nm and an emission wavelength of

490 nm

Steady-state fluorescence measurements

Steady-state fluorescence was measured with an Hitachi

F-4500 spectrofluorimeter All studies were carried out at

room temperature and excitation at 295 nm The

experi-ments were conducted in buffer A or buffer B The protein

solution had an initial absorbance at the excitation

wave-length lower than 0.1

The effect of cAMP on tryptophan fluorescence was

monitored by a fluorescence titration of CRPW13F and

CRPW85A Tryptophan emission was scanned from 310 to

480 nm When energy transfer was measured, the emission

spectra were recorded in the range 310–570 nm The

fluorescence quantum yield of the donor in the absence of

the acceptor (QD) was calculated from the equation:

QD¼ QRF

SDARF

SRFAD

ð1Þ

where SD and SRF are the respective areas under the emission spectra of the donor and a reference compound, and ARF and AD are the respective absorbances of the reference compound and donor at the excitation wave-length QRFis the quantum yield of the reference compound

L-tryptophan and was taken to be 0.14 in water at 25C [22]

All spectra were corrected for sample dilution and the inner filter effect, introduced by cAMP and DNA at the excitation wavelength, according to the following formula [23]

Fcor¼ F  10ðPþDAÞ=2 ð2Þ where F and Fcorare fluorescence intensity before and after the correction, and P and DA denote the initial sample absorbance at the excitation wavelength and the change in absorbance introduced by the ligand, respectively

Time-resolved fluorescence measurements Fluorescence decays were measured using a homemade time-correlated single-photon counting system based on Ortec electronics (Oak Ridge, USA) It consisted of (a) a Philips 2020Q photomultiplier with a 1.5-ns response time, (b) a 1-GHz preamplifier, (c) a quad constant fraction discriminator model 935, and (d) a time-to-amplitude converter (TAC) model 457 A nanosecond flash lamp nF

900 from Edinburgh Instruments was used as a light source (e) In the case of anisotropy measurements, (f) Glan– Thompson prism polarizers were also used

All measurements were performed at 20 ± 0.2C Before measurements, all samples were filtered through a microporous filter (0.45 lm; Millipore) to remove insoluble impurities

FRET measurements Energy transfer was observed between the tryptophan residues and the 1,5-I-AEDANS moiety covalently attached to Cys178 The tryptophans were excited at 297 nm Fluorescence decays were observed

at wavelengths between 320 and 400 nm using two cut-off filters Measurements were performed in buffer A Fluor-escence decays were recorded at a resolution of 23 ps per channel, resulting in a total time window of 100 ns Intensity decay data were analyzed using the following multiexponential decay law:

It¼X

i

aiexpðt=siÞ ð3Þ

where aiand siare the pre-exponential factor and decay time of component i, respectively The fractional fluores-cence intensity of each component is defined as ƒi¼ aisi/

Saisi The data were analyzed with the software from Edinburgh Instruments Best-fit parameters were obtained

by minimization of the reduced v2value

The average efficiency of energy transferÆEæ was calcu-lated from the average donor lifetime in the presenceÆsDAæ and absence of acceptorÆsDæ

< E >¼ 1 <sDA>

<sD> ð4Þ

Trang 4

The average lifetime was obtained from the equation:

<s >¼

P

i

ais2 i

P

i

AsÆsDAæ and ÆsDæ are obtained without the need to know the

absolute protein concentrations, uncertainties associated

with protein concentration determination are eliminated in

the time-resolved fluorescence measurements

The average distance between the donor–acceptor pair

ÆRæ was calculated from the equation:

< R >¼ R0hE1 11=6i

½ ˚AA ð6Þ where R0 is the Fo¨rster critical distance (the distance at

which 50% energy transfer occurs) R0is given by:

R0¼ 9:78  103j2n4QDJð Þk1=6

½ ˚AA ð7Þ where n is the refractive index of the medium, QDis the

quantum yield of the donor, J(k) is a spectral overlap

integral of the donor fluorescence and acceptor absorption,

and j2 is the orientation factor and accounts for relative

orientation of the donor emission and acceptor absorption

transition dipole Generally, j2is assumed to be equal to

2/3, which is the value for donors and acceptors that

randomized by rotational diffusion before energy transfer

Fluorescence anisotropy decay measurements The W13F

and W85A CRP mutants were used to measure the

rotational correlation time of the protein The excitation

wavelength for tryptophan residues was 297 nm

Fluores-cence anisotropy decays were observed using a cut-off

filter > 320 nm Experiments were performed at several

concentrations of the proteins (1.0–8.5 lM) for each species

The sample was excited with vertically polarized light

Fluorescence anisotropy decays with vertical and horizontal

emission polarization were alternatively recorded All

measurements were repeated at least twice for each sample

Fluorescence anisotropy decays were recorded at a

resolu-tion of 46 ps per channel, resulting in a total time window

of 200 ns

Anisotropy decay data were analyzed according to the

impulse reconvolution model decay law:

R tð Þ ¼ R1þXn

i¼1

Aiexpðt=hiÞ ð8Þ

where Ai are the amplitudes of the components with

rotational correlation time hI, and Rlis limiting anisotropy

The time-zero anisotropy r(0) was obtained from the

equation:

rð Þ ¼ R0 1þXn

i¼1

In each case, the best-fit parameters were obtained by

minimization of the reduced v2 test value The v2 and

residuals distribution were utilized to judge the goodness of

the fit The software used for analysis was from Edinburgh

Instruments

Results

Fluorescence labeling of CRP Each subunit of CRP possesses three cysteine residues, two

in the N-terminal domain (Cys19 and Cys92) and one in the C-terminal domain (Cys178) Only Cys178 can be chemi-cally modified under native conditions; Cys19 and Cys92 seem to be buried [24,25] To confirm the selectivity of the labeling, CRP mutants modified with 1,5-IAEDANS were denaturated and completely digested with trypsin and chymotrypsin The peptides liberated were examined by HPLC As expected, only one peptide fragment had been modified with thiol-reactive probe Thus, we conclude that CRP was uniformly labeled with 1,5-IAEDANS at the SH group of Cys178

The stoichiometry of the labeling was determined from the absorption spectrum of the labeled CRP When CRP was incubated with the fluorescence reagent, 1,5-I-AEDANS, at pH 8.0, a mean of 2 mol was bound per mol protein dimer The effect of the label on the secondary structure of CRP was investigated using CD spectroscopy No differences were observed between the modified and unmodified variants of CRP (data not shown) The insertion of the fluorescent probes also did not significantly alter the biological activity of CRP [20]

Steady-state fluorescence data The effect of cAMP on tryptophan fluorescence was monitored Changes in CRP tryptophan fluorescence were monitored by titrating the CRP solution with

1–2-lL aliquots of concentrated cAMP solution Measure-ments were performed in the cAMP concentration range

50 lM to 1 mM The fluorescence emission spectra of Trp13 and Trp85 in the presence and absence of cAMP are given in Fig 2A and Fig 3A, respectively When CRP was titrated with cAMP, the fluorescence intensity

of the Trp13 residue decreased with increasing ligand concentration However, this decrease could only by detected in the micromolar range of cAMP concentra-tions In addition, the emission maximum of Trp13 shifted from  342.5 nm to  340 nm The reduction in fluorescence intensity and the blue shift indicate a conformational transition of CRPW85A on binding of cAMP The effect of different concentrations of cAMP

on the fluorescence intensity of Trp13 is shown in Fig 2B When 200 lMcAMP was added to the solution

of CRPW85A, a 13% decrease in fluorescence intensity was observed

In contrast with the observation made with CRPW85A, the addition of cAMP to CRPW13F caused an increase in the fluorescence intensity of Trp85 with no change in the emission maximum The maximum wavelength of emission for CRPW13F was  339 nm The effect of different concentrations of cAMP on the fluorescence intensity of Trp85 is shown in Fig 3B In the case of Trp85, the addition

of 200 lM cAMP caused only a very small change in the fluorescence intensity, increasing it by  3.4% A pro-nounced increase in the Trp85 fluorescence intensity was

Trang 5

detected only at high concentrations of cAMP (> 2 mM)

(data not shown)

Typical fluorescence spectra of CRPW13F, unmodified

and modified with 1,5-I-AEDANS, are shown in Fig 4

When excited at 295 nm, tryptophan residues in the

unlabeled protein had a fluorescence emission maximum

near 339 nm In the presence of 1,5-I-AEDANS,

trypto-phan fluorescence intensity was significantly reduced

com-pared with an approximately equal concentration of an

unmodified protein The maximum wavelength of

trypto-phan emission in the labeled mutant W13F was shifted to

 327 nm The addition of cAMP and DNA increased

energy transfer from Trp85 residue to the AEDANS

moiety

The quantum yields of tryptophan fluorescence at

25C in buffer A were determined to be 0.09 for mutant

CRPW13F alone and 0.094 for mutant CRPW13F in the

presence of 200 lM cAMP The quantum yield of the donor increased upon protein–DNA complex formation The change in the observed value of the quantum yield was  20% A similar result was obtained for CRP– (cAMP)4

Time-resolved fluorescence data FRET measurements Lifetime measurements on the mutant CRPW13F labeled with 1,5-I-AEDANS indicated

a decreased lifetime of the tryptophan fluorescence, as expected when energy transfer occurs Figure 5 shows the time-dependent donor decays for the proteins bearing donor alone and those with donor and acceptor In the case of mutant CRPW85A, no energy transfer was observed Lifetime measurements were repeated several times for each species The fluorescence decays were

Fig 3 Fluorescence emission spectra of

CRPW13F in the absence (—) and presence of

cAMP at 50 l M (ÆÆÆÆ) and 1 m M cAMP (- - -).

Excitation was at 295 nm All spectra were

recorded in buffer B, pH 7.8 The inset in the

plot shows fluorescence intensity change in

Trp85 as a function of cAMP concentration.

F and F 0 are the fluorescence intensities of the

protein in the presence and absence of the

ligand, respectively The range of cAMP

con-centrations used was from 50 l M to 1 m M

The line was drawn only to indicate the trend

of the data.

Fig 2 Fluorescence emission spectra of

CRPW85A in the absence (—) and presence of

cAMP at 50 l M (ÆÆÆÆ) and 1 m M cAMP (- - -).

Excitation was at 295 nm All spectra were

recorded in buffer B, pH 7.8 The inset shows

fluorescence intensity change in Trp13 as a

function of cAMP concentration F and F 0 are

the fluorescence intensities of the protein in the

presence and absence of the ligand,

respect-ively The range of cAMP concentrations used

was from 50 l M to 1 m M The line was drawn

only to indicate the trend of the data.

Trang 6

analyzed as a multiexponential decay, and for each case the

double exponential decay was characterized by lower values

of reduced v2 In some cases the triple exponential decays

were recorded; however, the pre-exponential factor a3for

these fits was close to zero, and therefore this component of

the decay is not included in the calculated average values of

fluorescence lifetimesÆsDæ and ÆsDAæ The values for average

lifetimes of Trp85 in the absence and presence of acceptor

and efficiency of energy transfer are presented in Table 1

The transfer efficiency varied between apoprotein and after

binding cAMP at micromolar or millimolar concentrations

In the absence of a specific ligand, CRPW13F showed an

efficiency of transfer of 24.2 ± 8.1% The addition of

cAMP had a significant effect on energy transfer The value

for CRPW13F with two cAMP molecules bound to

anti-cAMP-binding sites was considerably higher

(72.3 ± 2.5%) than for apoprotein The result determined

for the mutant W13F in the presence of 2 mMcAMP was

similar to the above value, averaging 74.3 ± 2.4% Energy

transfer in CRPW13F bound to the specific fragments of

DNA was also measured and displayed similar values of

efficiency for each complex examined ( 62%)

The energy-transfer efficiency values were used to

calcu-late the average distance between the donor and the

acceptor To approximate the distance between the

Trp85–Cys178 pair, we assumed that the Fo¨rster distance

for the tryptophan–IAEDANS pair was 22 A˚ [26,27] The

estimated distances are shown in Table 1 There was a

significant difference between the calculated distance in

apo-CRPW13F and CRPW13F complexed with cAMP

and DNA

Time-domain anisotropy data The fluorescence

aniso-tropy decays of Trp13 and Trp85 were measured to

determine the changes in the rotational diffusion of CRP

after cAMP binding The anisotropy decay of Trp13 in the

CRP complexed with two cAMP molecules is shown in

Fig 6A Analysis of the anisotropy decays was carried out

according to Eqn (8) with an increasing number of

exponents, until the fit no longer improved In all cases,

the anisotropy decays of Trp13 and Trp85 could be described by one exponent with single rotational correlation time Addition of the second exponent did not significantly alter the goodness of fit The v2 value obtained with the single-exponential and double-exponential analysis indi-cates that the two-component analysis is not significantly better than the one-component analysis Also the distribu-tion of the residuals did not improve on the addidistribu-tion of the second component (Fig 6B,C) However, the data obtained for the single exponential analysis suggest the presence of an additional segmental mobility This is evident from appar-ent time-zero anisotropy r(0), which is lower than the fundamental anisotropy r0of tryptophan at this excitation wavelength [28] For both tryptophan residues, the initial anisotropy was in the range 0.22–0.23 (Table 2) It indicates that anisotropy decay contains a fast component which cannot be resolved with our device

The anisotropy decay parameters for various CRP species are reported in Table 2 The mean ± SD value

of rotational correlation times determined from the study

of fluorescence anisotropy decays of Trp85 in the absence

of specific ligand was 20.5 ± 2.4 ns The value for Trp13 was considerably lower, averaging 15.3 ± 1.8 ns The addition of 100 lM cAMP probably did not affect the rotational correlation time of CRPW13F The uncer-tainty of this value was too large to ascertain any changes in the CRP dynamic on cAMP binding Trp13 exhibited different behavior The rotational correlation time determined from the study of fluorescence aniso-tropy decays of Trp13 in the presence of cAMP decreased to 10.45 ± 3.0 ns

Discussion

cAMP binding to CRP has been studied using a variety of methods, which have shown that the ligand binding mediates changes in the protein conformation It is believed that these changes allow the protein molecule to switch from the low-affinity and nonspecific DNA-binding state to the state characterized by high affinity and sequence specificity

Fig 4 Fluorescence emission spectra of unmodified and modified CRPW13F The excitation wavelength was 295 nm and emission was scanned from 310 to 570 nm Measurements were performed at 25 C in buffer A, pH 8.0 (— Æ Æ —) Unmodified CRPW13F; (—) modified CRPW13F; ( -) modified CRPW13F in the presence of 200 l M

cAMP; (ÆÆÆÆ) modified CRPW13F bound to DNA in the presence of cAMP.

Trang 7

for the DNA promoter [2] As the X-ray crystal structure of

apo-CRP has not yet been resolved, it is believed that the

binding of cAMP, which leads to a switch to active protein

conformation, involves subunit realignment and hinge

reorientation between the protein domains [2,29] For a

long time, it has been a paradigm that CRP undergoes a

cAMP concentration-dependent transition between three

conformations: apo-CRP, CRP–(cAMP)1 and CRP–

(cAMP), and each conformer possesses a unique structure

and activity [2] The reinterpretation of this paradigm was proposed by Passner & Steitz [6] on the basis of the crystal structure of the CRP–cAMP complex, and it has recently been supported by NMR [7] and isothermal titration calorimetry [8] studies in solution NMR experiments have shown that CRP possesses two anti-cAMP-binding sites in each monomer, and the next two syn-cAMP sites are formed by an allosteric conformational change in the protein on biding of two anti-AMP at the N-terminal

Fig 5 Trp85 fluorescence intensity decays for mutant CRPW13F without and with 1,5-I-AEDANS covalently attached to Cys178 The dark grey dotted curve shows the intensity decay of the donor alone (D), and the grey dotted curve shows the intensity decay of the donor in the presence of the acceptor (DA) The black solid lines and weighted residuals (lower panels) are for the best triple exponential fits Experiments were performed in buffer A at 20 C.

Trang 8

domain [7] The isothermal titration calorimetry

measure-ments demonstrated that, at low cAMP concentration,

there are two identical interactive high-affinity sites for

cAMP and at least one low-affinity cAMP-binding site at

high concentration of the ligand [8]

The idea of the four cAMP-binding sites in CRP, for its

anticonformation (at low concentration of the ligand) and

the next two binding it in syn conformation (at high cAMP

concentration) has been used to describe the results of fast

kinetic studies [15] and investigations with dynamic light

scattering and time-resolved fluorescence anisotropy

meas-urements [30] In these studies, we have shown that the

binding of cAMP in anti conformation in the N-terminal

domain of CRP leads to the conformational changes in the

helix–turn–helix motif of the C-terminal domain,

respon-sible for the interaction with DNA, as well as to the changes

in the global hydrodynamic structure of CRP The

satura-tion of the low-affinity sites with cAMP in syn conformasatura-tion

of cAMP results in the changes in the microenvironment of

the Trp85 residue, localized in the N-terminal domain of the

protein, without further substantial changes in the global

hydrodynamic structure [30]

The results presented in this report provide further

evidence for conformational changes induced by cAMP

binding to the anti-cAMP-binding sites of CRP, which in

turn trigger specific pathways of signal transmission from

the cyclic nucleotide-binding domain to the DNA-binding

domain of the protein We have used single

tryptophan-containing mutants of CRP The mutations were localized

in the N-terminal domain at position 85 or 13 in order to

follow conformational changes in their microenvironment

on binding of cAMP, both in anti-and syn-conformation

We have also used AEDANS for fluorescent labeling of

Cys178, located at the turn of the helix–turn–helix motif, in

order to detect FRET between Trp85 and the label We

have shown that binding of cAMP at a concentration of

200 lM to anti-cAMP-binding sites results in  13%

decrease in the Trp13 fluorescence intensity along with the

blue shift in its maximum of emission by 2.5 nm Probable

candidates for quenching residues in CRP are Thr10,

Asn109 and His17, which are located within a distance up to

5 A˚, as has been determined from the X-ray crystal

structure of the CRP–cAMP complex (PDB code 1G6N)

[29] The observed changes in the microenvironment of

Trp13 on filling of the high-affinity sites are also supported

by the time-resolved anisotropy measurements Binding of

the ligand to anti-cAMP-binding sites results in a decrease in

rotational correlation time by  5 ns, from the value of

15.3 ns, detected for apo-CRP, to the value of 10.4 ns for

CRP–(cAMP) complex The decrease in rotational time

indicates an increase in the mobility of helix A of the protein As Trp13 is located in the vicinity of the activation region AR2 of the protein [1], which is responsible for the activation of the second class of E coli promoters such as galP1, one can speculate that this conformational change may play an important role in a signal transmission in the protein molecule, which in turn may allow CRP to adopt a conformation appropriate for the interaction with the aNTD domain of RNA polymerase in the transcription complex On the other hand, Trp13 of CRP directly interacts with another gene regulatory protein, CytR [31], and the observed conformational changes in Trp13 micro-environments on cAMP binding to anti-cAMP sites may play a significant role in the CRP–CytR–DNA complex

In contrast with Trp13 of CRP, binding of cAMP to anti-cAMP-binding sites does not lead to a significant change in fluorescence intensity of Trp85, and only a 3.4% increase

in the intensity has been observed at 200 lM cAMP However, cAMP binding to the syn-cAMP-binding sites at concentration of the ligand of 1 mMcauses a 6% increase

in its fluorescence intensity, which indicates that this residue

is sensitive to cAMP binding to low-affinity sites The rotational correlation time of Trp85 in apo-CRP of 20.5 ns indicates that this residue is immobilized within the N-terminal domain of the protein and exhibits motion characteristic of the whole protein (within experimental error), while the respective value for the AEDANS-labeled apo-CRP has been estimated at 23.3 ns [30] Binding of cAMP to anti-cAMP-binding sites increases the rotational correlation time to 22 ns for the CRP–(cAMP)2complex, which is much lower than the correlation time of 30 ns determined for this complex using CRP labeled at Cys178 with the AEDANS fluorescent probe [30] These discre-pancies can probably be explained by the fact that the average fluorescence lifetime of Trp85, 5.6 ns in the case

of the CRP–(cAMP)2 complex, is too short to allow observations of the longer rotational correlation times The allosteric activation of CRP involves conformational changes in the N-terminal domain of the protein and leads

to changes in the CRP molecule, enabling it to recognize the specific DNA sequence [1,2] As the crystal structure of apo-CRP has not yet been established, it was suggested that cAMP binding may cause reorientation of the coiled-coil C helices, consequently altering the relative position of the protein dimer subunits [29] These authors also suggested that, in the absence of cAMP in apo-CRP, some b strands

of the N-terminal domain of the protein may collapse into the cAMP-binding pocket, causing reorientation of the smaller domain in relation to the larger one, and bringing these domains closer together This suggestion has been

Table 1 Summary of energy transfer measurements In the CRPW13F–(cAMP) 2 complex, the concentration of cAMP was 200 l M , whereas in the case of CRPW13F–(cAMP) 4 the concentration of cAMP was 2 m M The molar ratio CRP to DNA in the protein–DNA complex was 1 : 1.

Trang 9

supported recently by NMR studies [7]; these authors argue that binding in solution of two cAMP molecules to high-affinity anti-cAMP-binding sites at the N-terminal domain causes the C-terminal domain to shift further to the N-terminal domain of CRP To confirm this suggestion,

we used FRET to detect the distance between the C-terminal and N-terminal domains of CRP on binding of cAMP in anti

as well as syn conformation to the protein For this purpose,

we used time-resolved fluorescence lifetime measurements

Table 2 Parameters of Trp13 and Trp85 anisotropy decays in the

presence and absence of cAMP.

CRPW13F–(cAMP) 2 22.1 ± 6.9 0.23 ± 0.03 1.036

CRPW85A–(cAMP) 2 10.45 ± 3.0 0.23 ± 0.03 1.146

Fig 6 Time-domain fluorescence anisotropy decay of Trp13 in the presence of 100 l M cAMP The solid line corresponds to the best single exponential fit of the data (dotted curve) according to Eqn (8) The grey cross-haired curve represents the lamp profile The plots of the residuals for the best single exponential fit (B) and the double exponential fit (C) are also shown Measurements were performed at 20 C in buffer B, pH 8.0, with a CRPW85A concentration of 1.1 l M Excitation was at 297 nm.

Trang 10

using single tryptophan-containing mutants of CRP

Fluo-rescence energy transfer could be detected between Trp85,

localized close to the cAMP-binding pocket of the

N-terminal domain, and Cys178, fluorescently labeled by

AEDANS, localized in the helix–turn–helix motif of the

C-terminal domain of the protein The lifetimes obtained for

the fluorescence donor, Trp85, indicate that binding of

cAMP to anti-cAMP-binding sites leads to a dramatic

increase in FRET efficiency This observation clearly shows a

decrease in the average distance between the two domains of

CRP on cAMP binding If one assumes the Fo¨rster distance,

R0, for the pair donor–acceptor such as tryptophan–

AEDANS to be 22 A˚ [26,27], the distance between Trp85

and Cys178-AEDANS in the apo-CRP can be calculated to

be 26.6 A˚ This distance decreases by about 8 A˚ to 18.7 A˚ on

binding of cAMP to the anti-cAMP-binding sites of the

protein The distance between the sulfur atom of Cys178 and

the C9–C10 bond of the indole ring of Trp85, derived from

the crystal structure of CRP–(cAMP)2(PDB code 1G6N) is

18.9 A˚ and 21.9 A˚ for the subunit present in the closed and

open conformation, respectively [29] The structural

asym-metry of the CRP–(cAMP)2 complex resulting from

con-formational differences between subunits has been

questioned [32], and from molecular dynamics simulation,

it has been predicted that, in solution, both subunits of CRP

adopt a closed conformation If this is so, the distance (equal

to 18.7 A˚), determined in this work by FRET, is in good

agreement with the value of 18.9 A˚ predicted for the closed

conformation This supports experimentally the dynamic

simulation studies [32] and indicates that, in solution, both of

the protein subunits exist in closed conformation in the

CRP–(cAMP)2complex

Because a variety of spectroscopic effects, at least in

theory, could influence energy transfer efficiency, one can

argue that the good agreement determined for the distance

between Cys178 and Trp85 residues in CRP–(cAMP)2may

also result from the assumed value of Fo¨rster distance R0

However, as binding of cAMP at a concentration of 200 lM

to CRP leads only to a 4.3% increase in the fluorescence

quantum yields from the value of 0.09 for apo-CRP to the

value of 0.094 for the CRP–(cAMP)2 complex, and no

substantial changes in the shape of the emission spectra of

the donor have been observed, this justifies the lack of the

alteration of the Fo¨rster distance between apo and holo

forms of the protein As the AEDANS label attached to the

Cys178 enjoys local freedom of movement, in both

apo-CRP and the apo-CRP–(cAMP)2 complex [30], the distance

obtained from the crystal structure between the sulfur atom

of Cys178 and the indole ring of Trp85 seems to be realistic

We have also tried to measure fluorescence energy transfer

between Trp13 and Cys178-AEDANS; however, we have

not detected any energy transfer in either the apo-or

holo-form This could be because of the distance between the two

residues, which is about 45 A˚, as can be calculated from the

crystal structure of CRP–(cAMP)2[29]

Binding of cAMP in syn conformation to the

low-affinity binding sites in the CRP–(cAMP)4 complex leads

to only a small increase in the efficiency of energy

transfer, which, with an assumed R0 value of 22 A˚,

corresponds to the small decrease in average distance

between the N-terminal and C-terminal domains of CRP,

estimated at 18.4 A˚ However, the fluorescence quantum

yield of the Trp85 donor increases by  20% at a concentration of cAMP of 2 mM from the value characteristic of apo-CRP, which in turn may be responsible for this very small change We have also measured the distance between the two domains of CRP

in the complexes with DNA containing various

sequenc-es, such as lac and gal promoters and with the symmetric sequence ICAP For each CRP–DNA complex, the increase in the distance between the two CRP domains has been observed with the average distance of 20.2 A˚ This value is in good agreement with the value of 20.7 A˚, calculated from the crystal structure of the CRP– DNA complex [33]

The present results show that the binding of anti-cAMP

in the CRP–(cAMP)2complex results in the movement of the C-terminal domain of CRP by  8 A˚ towards the N-terminal domain, which in consequence leads to rearrangement of DNA-binding domains and cAMP-binding domains of the protein This finding clarifies the suggestion derived from the NMR measurements [7] that the C-terminus is closer to the N-terminal domain in apo-CRP than in cAMP-bound apo-CRP Binding of cAMP to anti-cAMP-binding sites leads to an increase in the structural dynamic motion around Trp13, which is close

to the activation region AR2, responsible for the interac-tion of CRP with the a subunit of RNA polymerase The changes in the CRP dynamics on cAMP binding have recently been observed by the hydrogen exchange method [34] In that paper, it was shown that binding of the ligand to the protein causes the C-terminal domain of CRP to become more flexible, in contrast with the N-terminal domain which is shifted to a less dynamic conformation Our results extend this observation and suggest that the binding of cAMP to anti-cAMP-binding sites of CRP leads to the increase in the structural dynamic motion of at least Trp13, which is located in the N-terminal domain of the protein

Acknowledgements

We are grateful to Dr S Garges for supplying us with the plasmid for production of CRP This work was supported by grant no.

6 P04A 031 16 from the State Committee for Scientific Research.

References

1 Busby, S & Ebright, R (1999) Transcription activation by Cata-bolite Activator Protein (CAP) J Mol Biol 293, 199–213.

2 Harman, J.G (2001) Allosteric regulation of the cAMP receptor protein Biochim Biophys Acta 2, 1–17.

3 de Crombrugghe, B., Busby, S & Buc, H (1984) Cyclic AMP receptor protein: role in transcription activation Science 224, 831–838.

4 Weber, I.T & Steitz, T.A (1987) Structure of a complex of cata-bolite gene activator protein and cyclic AMP refined at 2.5 A˚ resolution J Mol Biol 198, 311–326.

5 Parkinson, G., Wilson, C., Gunasekera, A., Ebright, Y.W., Ebright, R.E & Berman, H (1996) Structure of the CAP–DNA complex at 2.5 A˚ resolution: a complete picture of the protein– DNA interface J Mol Biol 260, 395–408.

6 Passner, J.M & Steitz, T.A (1997) The structure of a CAP–DNA complex having two cAMP molecules bound to each monomer Proc Natl Acad Sci USA 94, 2843–2847.

Ngày đăng: 20/02/2014, 23:20

TỪ KHÓA LIÊN QUAN

TÀI LIỆU CÙNG NGƯỜI DÙNG

TÀI LIỆU LIÊN QUAN

🧩 Sản phẩm bạn có thể quan tâm