Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M300411200 on April 4, 2003

J. Biol. Chem., Vol. 278, Issue 24, 21860-21868, June 13, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/24/21860    most recent
M300411200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Anderluh, G.
Right arrow Articles by Lakey, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Anderluh, G.
Right arrow Articles by Lakey, J. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Concerted Folding and Binding of a Flexible Colicin Domain to Its Periplasmic Receptor TolA*

Gregor Anderluh {ddagger} § , Qi Hong §, Ruth Boetzel ||, Colin MacDonald ||, Geoffrey R. Moore ||, Richard Virden § and Jeremy H. Lakey § **

From the {ddagger} Department of Biology, Biotechnical Faculty, University of Ljubljana, Vena pot 111, 1000 Ljubljana, Slovenia, § School of Biochemistry and Genetics, University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne, NE2 4HH, United Kingdom, || School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich NR4 7TJ, United Kingdom

Received for publication, January 14, 2003 , and in revised form, April 1, 2003.
   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Compared with folded structures, natively unfolded protein domains are over-represented in protein-protein and protein-DNA interactions. Such domains are common features of all colicins and are required for their translocation across the outer membrane of the target Escherichia coli cell. All of these domains bind to at least one periplasmic protein of the Tol or Ton family. Similar domains are found in Ton-dependent outer membrane transporters, indicating they may interact in a related manner. In this article we have studied binding of the colicin N translocation domain to its periplasmic receptor TolA, by fluorescence resonance energy transfer (FRET) using fluorescent probes attached to engineered cysteine residues and NMR techniques. The domain exhibits a random coil circular dichroism spectrum. However, FRET revealed that guanidinium hydrochloride denaturation caused increases in all measured intramolecular distances showing that, although natively unfolded, the domain is not extended. Furthermore NMR reported a compact hydrodynamic radius of 18 Å. Nevertheless the FRET-derived distances changed upon binding to TolA indicating a significant structural rearrangement. Using 1H-15N NMR we show that, when bound, the peptide switches from a disordered state to an ordered state. The kinetics of binding and the associated structural change were measured by stopped-flow methods, and both events appear to occur simultaneously. The data therefore suggest that this molecular recognition involves the concerted binding and folding of a flexible but collapsed state.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Colicins are produced by many gut bacteria, and each one displays a highly specific antimicrobial activity. They are secreted by immune Escherichia coli strains to kill competing bacteria and may be important in the colonization of new habitats (1). Their interaction with sensitive E. coli proceeds via three steps: binding to one or more outer membrane proteins, translocation across the periplasm with the aid of Tol or Ton protein systems, and subsequent killing that is achieved by enzymatic cleavage of nucleic acids (2) or pore-formation (3). A different structural domain mediates each step. Colicins are assigned to group A or B according to the proteins required for their translocation (4, 5). Group A utilizes the Tol QRAB gene cluster and includes colicins A, N, E1, and K (3). Colicin N is the smallest known pore-forming colicin (42 kDa instead of >60 kDa for the others) (6). Colicin N binds outer membrane receptor OmpF via its receptor binding domain (R-domain). The translocation domain (T-domain)1 then interacts with TolA in the periplasm, and the pore-forming domain (P-domain) is subsequently translocated to the inner membrane. The study of colicins has already given insights into basic biological processes such as protein folding, protein-protein interactions, and ion channel formation, and the simplicity of colicin N makes it a good model for studying mechanisms of translocation across membranes.

TolA spans the periplasm with its central domain (TolA-II), which connects the N-terminal (TolA-I) inner membrane domain to the C-terminal (TolA-III) domain. TolA-II is required for colicin A and N but not E1 activity (7) but the molecular details of this requirement are not yet clear. The TolA-III domain binds to N-terminal domains of both group A colicins and the g3p protein of filamentous phage (8–11), thus intriguingly providing a common entry point for two apparently unrelated invading proteins. The binding of isolated colicin N T-domain to TolA-III can be measured in vitro (12), and this has enabled the definition of the complete colicin N TolA box (13) (the "binding box," Asn42–Gly68) (Fig. 1). The two tryptophans Trp44 and Trp46 are used as intrinsic probes of TolA binding in a fluorescence assay because TolA-III is tryptophan-free. Exposed to the solvent in the soluble form, they are buried after binding to TolA-III causing an increase and blue-shift of their fluorescence (12). Both are needed for binding and contribute equally to the observed fluorescence change (13). Circular dichroism (CD) and fluorescence studies indicate that the T-domain is unstructured (12), in accord with the x-ray crystallographic structure analysis of colicin N (14) that failed to detect electron density for residues 1–90 of the T-domain, indicating that it is disordered in the crystal. This observation was further supported by preliminary 1H NMR spectra of the T-domain, which showed an absence of extensive secondary structure elements.2 Similarly, in the two known group B colicin structures, the N-terminal 83 residues of colicin E3 are invisible (15), whereas the first 23 residues of colicin Ia are not visible, and the next 44 have no defined secondary structure (16). Such an absence of clear electron density could be because of static or dynamic disorder of the protein in the crystal, but the characterization by NMR of intact colicin E9, whose translocation domain is almost identical to that of colicin E3, shows it is in dynamic disorder (17). Inspection of the glycine-rich amino acid sequences of the remaining colicin T-domains indicates that these are likely to be similarly unstructured. Hence the behavior of the colicin N T-domain is representative of the whole colicin family. The unstructured T-domain binds to TolA-III with a Kd = 1–2 µM (from calorimetry and fluorescence) and an association rate of 105 s–1 (stopped-flow fluorescence). The entropy change measured by calorimetry is large and may correlate with the structuring of the T-domain during binding (12, 18). Recent reviews have illustrated the importance of comparable natively unstructured proteins in molecular recognition throughout biology (19–21). Here we have used fluorescence resonance energy transfer (FRET) and NMR to study the solution dynamics and conformational changes of the flexible colicin N T-domain upon TolA-II,III binding (22, 23).



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 1.
Colicin N. Colicin N is composed of three functional domains: translocation domain (checked, T-domain), receptor binding domain (black, R-domain), and pore-forming domain (white, P-domain). The TolA binding box (gray) is located at the C-terminal end of the translocation domain. Two tryptophans are shown in black. Residues that were mutated to cysteines are white. Residues that inhibit binding to TolA when mutated to alanine (non-permissive sites) are shown as black circles; those with no effect as white circles (permissive sites). The line below the sequence marks the length of the binding box peptide used for NMR studies.

 


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Mutagenesis and Protein Production—The T-domain nucleotide sequence, coding for amino acids 1–90 of colicin N, was cloned into a modified pET8c vector, which adds a His6Ser2 tag at the N terminus for easier purification (12). Mutagenesis was carried out using QuikChange (Stratagene) and appropriate oligonucleotides (MWG-Biotech AG). Mutant proteins were expressed in an E. coli BL21(DE3) pLysE strain. 20 ml of overnight culture were used to inoculate 500 ml of M9-LB medium with 100 µg ml–1 of ampicillin and 25 µg ml–1 of chloramphenicol. Protein production was induced at an OD600 of 0.8 by addition of isopropyl-1-thio-{beta}-D-galactopyranoside to a 1 mM final concentration. Cells were grown for an additional 4 h and then spun down, and the pellets were frozen. Cells were thawed into 50 mM NaH2PO4, pH 8.0, 10 mM imidazole, 20 mM {beta}-mercaptoethanol, 1 mg ml–1 lysozyme, 10 µg ml–1 DNase, 20 µg ml–1 RNase, 0.5 mM phenylmethylsulfonyl fluoride, and 1 mM benzamidine (2 ml of lysis buffer per gram dry weight of bacteria). They were incubated on ice for 45 min with occasional vigorous shaking. Bacteria were sonicated on ice for 5 min with a microtip on maximal output (Ultrasonic Generator 7533A, Ultrasonics, Dawe Instruments). Broken cells were centrifuged for 60 min at 4 °C at 45,000 rpm in a 45TI rotor (Beckmann L7 centrifuge). The supernatant was stored at 4 °C, and the pellet was washed with the same buffer (1 ml of buffer/g dry weight of bacteria). Incubation for 30 min on ice was followed by 3-min sonication and centrifugation at the same conditions. The supernatant was merged to the first one and applied to a 3-ml Ni-NTA column (Qiagen, Crawley, UK) equilibrated in 50 mM NaH2PO4, pH 8.0, 10 mM imidazole, and 20 mM {beta}-mercaptoethanol. Unbound protein was eluted with the same buffer, and the bound T-domain was eluted with 300 mM imidazole in the same buffer. Protein was dialyzed three times against 5 liters of 10 mM NaH2PO4, pH 7.4 at 4 °C and aliquoted and stored at –20 °C. A further purification step on a Mono S column (Amersham Biosciences) was performed if impurities were observed on SDS-PAGE gels. In all, wild-type T-domain plus six cysteine mutants were produced: N53C, S58C, S61C, F66C, N70C, and N71C. TolA-II,III (amino acids 43–421 of SwissProt accession no. P19934 [GenBank] ) and TolA-III (amino acids 296–421) were produced in the same expression system. Coding regions for both proteins were amplified from pSKL17 using appropriate oligonucleotide primers (24). They carry XhoI and MluI sites at the N and C terminus, respectively. Products were cloned in a modified pET8c vector and expressed and purified as outlined above. Pure proteins were aliquoted and stored at –20 °C. A 15N-labeled binding box T-domain peptide comprising residues 40–76 with a GSG N-terminal extension (T-domain40–76), was produced as previously described (25) in order to check conformational changes by NMR. In brief, residues 40–76 of the T-domain were fused to the C terminus of TolA-III. Fusion protein was expressed and purified as described above and the peptide cleaved from the TolA-III by enterokinase. The peptide was separated from the uncleaved fusion and TolA-III by ion-exchange chromatography using a Mono S column. The peptide was concentrated with a Speed-Vac (Savant, Albertville, MN) and desalted on PD-10 columns (Amersham Biosciences). Its binding to TolA-II,III resembled the binding of the full-length T-domain since a Kd of 1.3 µM was obtained by the surface plasmon resonance (SPR) measurement. In addition, the tryptophan fluorescence spectrum exhibited the same increase and blue-shift upon TolA-II,III binding.3

Labeling of Cysteine Mutants—Approximately 3 mg of pure cysteine mutant were incubated for at least 3 h at 4 °C in 20 mM NaH2PO4, pH 7.4, 20 mM {beta}-mercaptoethanol to completely reduce thiol groups. This protein solution was added to an equal volume of the same buffer with 300 mM NaCl and applied to a 0.75-ml Ni-NTA-agarose column preequilibrated with the same high salt buffer. The column was well washed with 10 ml of 20 mM NaH2PO4, pH 7.4, 300 mM NaCl (buffer A) in order to remove all {beta}-mercaptoethanol and then washed with a few milliliters of 4 mM IAEDANS (Molecular Probes Inc.) in the same buffer. Flow was stopped after the last milliliter was applied, and the column was incubated in the dark for 4 h at room temperature. The column was then washed with buffer A until all free IAEDANS were eluted (i.e. until A340 dropped below 0.02). Bound labeled protein was eluted from the column with buffer A containing 200 mM imidazole. Fractions with protein were pooled and dialyzed three times for at least 4 h each time against 1 liter of 20 mM NaH2PO4, pH 7.4, at 4 °C. The yield of labeling was calculated from the absorbance at 340 nm for IAEDANS ({epsilon}340 = 6000 M–1 cm–1) and 280 nm for T-domain mutants (calculated from the amino acid composition, {epsilon}280 = 13,000 M–1 cm–1). Labeled samples were analyzed on SDS-PAGE to check whether all probe was covalently attached to the protein. Labeled bands were visualized under a transilluminator. No residual probe was observed. Samples were aliquoted in black plastic 1.5-ml sample tubes (Eppendorf) and stored at –20 °C.

Fluorescence Measurements—Fluorescence spectra were recorded at 25 °C on an SLM 8100 spectrofluorometer operating in ratio mode. Emission bandwidth was 8 nm; excitation bandwidth was 16 and 4 nm for emission and excitation spectra, respectively. Excitation wavelength for emission measurements was 295 nm, and spectra were scanned from 300–400 nm. For excitation spectra, emission was followed at 490 nm, with excitation wavelength scanned from 250 to 400 nm. Excitation spectra required further correction due to polarization anomalies, and this was assisted by placing a Cornu pseudo-depolarizing prism in the excitation beam (26, 27). The correction factor consisted of a normalized ratio mode excitation spectrum of rhodamine (0.3% in ethylene glycol, front face illumination of Helma 111.061QS triangular cuvette). Steady-state polarization anisotropy measurements were carried out using the T-format configuration, and each value was a mean of 10 measurements. Buffer was always 20 mM NaH2PO4, pH 7.4, 300 mM NaCl. A sample of a T-domain mutant, 500 µl, was placed in a 0.5-cm pathlength cuvette (Hellma 101.106). TolA-II,III was added from a stock solution (60 µM), and sample was well mixed before the measurements. The absorbance at the excitation wavelength of the protein samples used was never more than 0.05 to avoid inner-filter effects. The spectra were corrected for dilution and Raman scatter contribution by subtraction of a buffer blank.

Measurements of Resonance Energy Transfer Efficiency—Two methods were used to determine the resonance energy transfer efficiency (E) between the tryptophan residues and IAEDANS. In the first method the tryptophan quantum yields were determined in the absence (QD) and in the presence (QDA) of the acceptor (AEDANS) by comparison to a compound with a known quantum yield, N-acetyl-tryptophanamide (Sigma). The accurate determination of the quantum yields in solution and in the presence of TolA-II,III is mandatory, since the binding of the T-domain to TolA-II,III is accompanied by an increase in tryptophan fluorescence (quantum yield), which makes any direct conclusions about the energy transfer from the changes of fluorescence intensity inaccurate. Quantum yield was calculated using Equation 1,

(Eq. 1)
where F is the protein fluorescence, and FTrp is fluorescence of N-acetyl-tryptophanamide; both integrated between 300 and 400 nm. A and ATrp are the absorbances of protein and N-acetyl-tryptophanamide at 280 nm, respectively. A quantum yield of 0.14 was used for N-acetyl-tryptophanamide. Absorbance was measured in a Cary 4 spectrophotometer under the same conditions as subsequent fluorescent measurements. E was then calculated from Equation 2,

(Eq. 2)
where QDA is the quantum yield of the donor in the presence of acceptor (labeled protein), and QD is the quantum yield of the donor in the absence of acceptor (unlabeled protein). In the second method, the transfer of energy from tryptophan was directly observed as an increase at 290 nm in the AEDANS excitation spectrum normalized at 340 nm (23). E was calculated according to Equation 3,

(Eq. 3)
where F290 and F340 are the corrected intensities of the excitation spectrum and {epsilon}A290 (1200 M–1 cm–1) and {epsilon}A340 (6000 M–1 cm–1) are the AEDANS extinction coefficients at 290 nm (excitation wavelength of tryptophan) and 340 nm (excitation wavelength of AEDANS), respectively. {epsilon}D290 (9200 M–1 cm–1) is the extinction coefficient of the two tryptophan donors at 290 nm.

Calculation of Donor-Acceptor Separation—The measured transfer efficiencies were used to calculate the donor-acceptor separation (R) by Equation 4,

(Eq. 4)
where R0 is the Förster critical distance at which E is 50%. R0 was calculated in the usual manner in Equation 5,

(Eq. 5)
where n is the refractive index of solution (taken as 1.4), QD is the donor quantum yield, determined in each case exactly as described above, {kappa}2 is the orientation factor and was taken as two-thirds. J is the overlap integral (in cm3 M–1) as given by Equation 6,

(Eq. 6)
where F({lambda})is the corrected fluorescence of the protein sample excited at 295 nm, and {epsilon} is the extinction coefficient of attached AEDANS expressed in M–1cm–1. J was numerically integrated at 1-nm intervals.

Stopped-flow Measurements—An Applied Photophysics model SX-17MV fitted with a Hamamatsu Photonics 150W Hg-Xe lamp (model L2482) was used to follow the enhancement of tryptophan and AEDANS fluorescence in the presence of TolA-II,III. Excitation wavelength was 295 nm with a bandwidth of 2.35 nm, while all emission above 320 or 420 nm was collected via a WG320 Schott high-pass filters for unlabeled tryptophan fluorescence or labeled AEDANS fluorescence, respectively. The cell pathlength was 2 mm (excitation) and 1 mm (emission), and the dead time of mixing was ~1.5 ms. Buffer used in all measurements was 20 mM NaH2PO4, pH 7.4, 300 mM NaCl, filtered through a 0.2-µm filter prior use. Protein samples were allowed to equilibrate at 25 °C prior to use and centrifuged to pellet particulate material if necessary. The fluorescence versus time data were analyzed using SX.17MV software version 4.22 exactly as described (13). Concentration of the T-domain mutants was kept at 0.1 µM, while TolA-II,III was at least at 10-fold higher concentrations. Under these conditions the reaction should follow pseudo first-order kinetics. The plot of kobs against [TolA-II,III] allowed estimation of k1 from the slope and k–1 from the ordinate intercept. As the ratio of k1/k–1 represents the equilibrium constant Ka, the affinity constant of T-domain-TolA-II,III interaction can be determined.

SPR Measurements—SPR measurements of T-domain binding to TolA were performed on a BIAcore system as described previously (13). TolA-II,III was immobilized on a CM5 chip. The binding kinetics were measured by passing a concentration series of T-domain mutants or binding box peptide over the surface of the chip. Rate constants, ka and kd, were recovered from the fits of the binding curves using BIAevaluation software. The equilibrium dissociation constant, Kd, was then calculated from the ratio kd/ka.

NMR Measurements—NMR measurements were carried out using Varian Unity Inova 500 or 600 spectrometers, operating at 1H frequencies of 499.865 MHz and 599.162 MHz and 15N frequencies of 50.66 MHz and 60.72 MHz, respectively and with pulse sequences incorporated into the Varian Protein Pack suite of experiments. NMR data were processed using Varian VNMR or NMRPipe (28). Prior to Fourier transformation, a sine bell-shaped window function with a 90° shift was applied to each dimension for apodization. The indirect dimensions were linear-predicted once, and then the spectra were zero-filled in all dimensions to double the number of data points. Proton chemical shifts were referenced to external DSS, and nitrogen chemical shifts indirectly to DSS as described by Wishart et al. (29)

1H-15N HSQC and 15N NOESY-HSQC spectra were acquired using a ~300 µM solution of T-domain40–76 in 50 mM acetate buffer. The pH was adjusted to ~5 for long term stability. Subsequently, aliquots of unlabeled TolA-III were added and further spectra acquired at molar ratios ranging from 1:0.25 to 1:4. A 15N NOESY-HSQC spectrum (mixing time of 75 ms) was obtained at the 1:2 ratio.

Diffusion experiments for extraction of the hydrodynamic radius Rh were performed with unlabeled T-domain using the longitudinal encode-decode gradient-echo sequence (30, 31) through acquisition of spectra with varying gradient strengths and plots of variations in signal intensity as a function of gradient strength, according to Equation 7,

(Eq. 7)
where A refers to the peak intensity for a given gradient strength G; A0 to the peak intensity in the absence of gradients; {Delta} denotes the separation between leading edges of the gradients; {delta} is the duration of the gradient pulse of strength G, and {gamma} the magnetogyric ratio of 1H. A non-invasive compound, dioxane (with ), was added to the protein solutions (32), and was obtained from the slopes, Sprot and Sref, of the semilogarithmic fits of Equation 7 for resonances of the protein and reference compounds, respectively, according to Equation 8.

(Eq. 8)

The diffusion experiments were performed on the T-domain peptide in 100% D2O containing 20 µl of 1% dioxane in D2O, with gradient strengths between 1.7 and 21.5 G/cm at 298 K. The gradient strength was calibrated using the diffusion coefficient of water at 298 K (2.3 x 10–5 cm2/s). The delays {Delta} and {delta} were held constant for all experiments, at 113 and 7.8 ms, respectively. Peak intensities for several peaks across the spectrum in the aromatic and aliphatic regions were measured using VNMR. All data fitting were performed in Origin 4.1 (Microcal Software Inc.).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Properties of Cysteine Mutants—In order to study conformational changes of the T-domain upon binding to its periplasmic receptor TolA-III, six cysteine mutants were produced within the binding box of the colicin N T-domain (Fig. 1). Two mutants (S58C and S61C) were chosen at sites in the middle of the binding box known not to affect the binding affinity (13). One mutant (N53C) was set just before the residues critical for the binding, and two mutants (N70C and N71C) were placed just after the binding box. These two consecutive residues were mutated to have a reliable double control for the calculation of the final distances between the tryptophans and AEDANS. Another mutant (F66C) was placed at a non-permissive site and was used as a negative control, i.e. in the absence of the binding no changes in either tryptophan or AEDANS fluorescence should be observed. Isolated mutants were pure as viewed by SDS-PAGE under reducing conditions. The binding to TolA-II,III was measured by tryptophan fluorescence and SPR (Fig. 2 and Table I). S58C, S61C, N70C, and N71C exhibit binding comparable to the wild-type T-domain. N53C showed weaker binding. As expected F66C did not bind, but S58C showed slightly tighter binding than the wild-type T-domain. All mutants were fully labeled. This was expected since the thiol side chains of the T-domain should be largely exposed. Labeling did not affect the binding to TolA-II,III, as the determined Kd values were approximately the same for labeled or unlabeled samples.



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 2.
Properties of cysteine mutants. Fluorescence was measured at 25 °C. Excitation wavelength was 295 nm, emission was scanned from 300 to 400 nm. The concentration of the mutants was 2 µM in 20 mM NaH2PO4, pH 7.4, 300 mM NaCl. A, fluorescence of N71C before (bottom trace) and after addition of 0.96-pmol aliquots of TolA-II,III (other traces that show increase in fluorescence and blue-shift). B, tryptophan fluorescence at 337 nm from curves in A was fitted to a single binding site model (solid line), which allows the determination of Kd. C, tryptophan fluorescence in solution and in the presence of 11 µM TolA-II,III is shown for each mutant. D, polarization anisotropy of AEDANS fluorescence is shown for each mutant in the presence of different concentrations of TolA-II,III (bottom right). The values represent the mean of ten measurements ± S.D.

 

View this table:
[in this window]
[in a new window]
  		TABLE I
Properties of T-domain cysteine mutants

 

Tryptophan anisotropy values were similar for all mutants in solution (Table I) and increased to values between 0.062 and 0.067 for mutants showing complete binding. These similar increases indicate that in all cases tryptophans were placed in the same environment and that the binding was complete at 11 µM TolA-II,III. As expected, mutants S53C and F66C showed small or non-existent increases in tryptophan anisotropy. Also as expected, all mutants in solution showed low AEDANS polarization anisotropy values (Table I and Fig. 2D), which increased in the presence of 11 µM TolA-II,III and showed site-specific values for each mutant. When bound to the larger TolA-II,III, local mobility becomes important, and differences in anisotropy can be interpreted in terms of local steric effects, i.e. AEDANS is the least mobile in S61C, which is next to a residue (Tyr62) whose mutation completely abolishes binding.

Emission and Excitation Spectra—The emission and excitation spectra of S58C, F66C, and N71C are shown in Fig. 3. The spectrum of S61C was qualitatively similar to S58C, and those of N70C were similar to N71C (data not shown). The decrease in the tryptophan fluorescence in labeled mutants is the consequence of energy transfer to AEDANS because: (i) enhanced AEDANS fluorescence was measured in all cases and (ii) a tryptophan contribution between 270 and 300 nm is present in the AEDANS excitation spectra, (Fig. 3, right). Labeled S58C, S61C, N70C, and N71C mutants exhibited the expected increase and blue-shift of tryptophan fluorescence upon addition of TolA-II,III. TolA-II,III does not contain any tryptophans, and its fluorescence caused by tyrosines is negligible when excited at 295 nm. The blue-shift of tryptophan fluorescence was similar for all mutants, shifting from 348–350 nm in solution to 333–334 nm in the presence of TolA-II,III. By contrast, the tryptophan fluorescence yield was much more variable, indicating either differences in quenching or in the extent of energy transfer due to changes in intramolecular distances. When energy transfer occurs, the donor quenching should be accompanied by acceptor enhancement values that lead to similar values of E. The large variations in AEDANS fluorescence changes upon TolA-II,III binding do match this expectation and indicate the occurrence of distance changes. Despite a large increase in tryptophan fluorescence, the emission from S58C AEDANS decreased in the bound state by ~10% in comparison to solution. On the other hand, despite a small increase in tryptophan fluorescence, AEDANS fluorescence increased by ~80% in N71C. The energy transfer was further confirmed in excitation spectra, where the tryptophan component increased for N70C and N71C and decreased for S58C and S61C (Fig. 3, right). The emission maximum of AEDANS remained the same, 489–490 nm, upon TolA-II,III binding in all four mutants. Thus in the bound state, the probe remains in a similar environment to that in solution, i.e. it is largely exposed to the aqueous phase. The labeled mutant F66C did not bind to the receptor since there was no change in the tryptophan fluorescence in the presence of TolA-II,III. AEDANS fluorescence remained the same for this mutant in both conditions. There were also no changes observed in its excitation spectrum. These results indicate that changes in AEDANS fluorescence result from the receptor-bound state and that binding and structural changes are linked.



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 3.
Emission and excitation spectra. The corrected emission (left) and excitation (right) spectra of 2 µM labeled mutants in 20 mM NaH2PO4, pH 7.4, 300 mM NaCl at 25 °C are shown. The excitation wavelength was 295 nm for the emission spectrum. Bandwidths were 4 nm in the excitation beam and 8 nm in the emission beam. Emission wavelength was 490 nm for excitation spectrum. Bandwidths were 16 nm for excitation beam and 8 nm for emission beam. The fluorescence of TolA-II,III in the buffer was subtracted from the spectrum to correct for the contribution of tyrosines present in TolA-II,III. Thick solid line, unlabeled mutant in solution; continuous line, labeled mutant in solution; broken line, labeled mutant in the presence of 11 µM TolA-II,III. The spectrum of mutant S61C was qualitatively similar to S58C, N70C to N71C, and N53C to F66C and are not included.

 

In 5 M GdmHCl all labeled mutants showed increases in tryptophan yield compared with the solution state. This suggested that the distance between the tryptophans and the probe increased. This was confirmed by measuring AEDANS excitation spectra in 5 M GdmHCl. The observed FRET clearly correlated with the number of residues between tryptophans and AEDANS (Fig. 4). Mutants closer to the probe (i.e. N53C) showed larger FRET than those located farther away (i.e. N71C). In all mutants FRET was less than in solution (i.e. compare the excitation spectrum of N71C in solution (Fig. 3) with the spectrum in GdmHCl (Fig. 4)).



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 4.
Spectroscopic ruler. Excitation spectra of 2 µM labeled mutants in 20 mM NaH2PO4, pH 7.4, 300 mM NaCl, 5 M GdmHCl at 25 °C are shown together with the spectrum of a {beta}-mercaptoethanol-AEDANS adduct. The experimental conditions are the same as described in the legend to Fig. 3.

 

Donor-Acceptor Separation—For all functional mutants the tryptophan quantum yields altered in the presence of TolA-II,III. Since J, the overlap integral between the combined tryptophan emission and the AEDANS absorption, and consequently R0, are dependent on tryptophan quantum yield, it was necessary to determine its value for each case. The results are collected in Table II. The values of J for all mutants in solution were between 7.11 and 7.15 x 10–15 M–1 cm3. Consistent with the fact that tryptophans are exposed in the T-domain and fully accessible to the solvent, similar values were obtained when mutants were denatured with GdmHCl. Values dropped to between 6.95 and 7.04 x 10–15 M–1 cm3 when bound to TolA-II,III. R0 was then calculated for each condition. Values of R0 for mutants in solution were between 20.0 and 20.3 Å, for mutants in GdmHCl between 19.9 and 20.3 Å and for TolA-II,III-bound mutants between 23.4 and 23.9 Å. The highest uncertainty in R0 determination is in the value of {kappa}2, since it is difficult to correctly estimate it. However in our case, the donor orientation is highly anisotropic due to its low measured polarization, which may result from our use of a pair of neighboring tryptophans. It was therefore concluded that {kappa}2 = 2/3 is a reasonable value in distance calculations. Values obtained for tryptophan quantum yield, overlap integral, and R0 agree well with the values from the literature for both states, in solution or receptor-bound form when tryptophans are buried (33).


View this table:
[in this window]
[in a new window]
  		TABLE II
Data for FRET calculations

 

The distance between tryptophans and AEDANS was estimated from the experimental data for all three conditions: in solution, in the presence of TolA-III, and when denatured by GdmHCl (Table III). There are two clear effects when comparing soluble and receptor-bound states. In mutants S58C and S61C tryptophans move away from the probe upon TolA-II,III binding, while in mutants N70C and N71C tryptophans move closer to the probe. The difference between the two states is larger in the first case, where the distance increases by 3–4 or 5–6 Å for S58C and S61C, respectively. In mutants N70C and N71C the difference between the two states is in the range 1–3 Å. Although small, for each pair the same changes are observed in both mutants in both emission and excitation spectra.


View this table:
[in this window]
[in a new window]
  		TABLE III
Calculated resonance transfer efficiencies (E) and distances between the tryptophans and probe are reported for two different approaches.

 

Stopped-flow Fluorescence Measurements—Fluorescent enhancement of both tryptophan and AEDANS versus time after mixing TolA-II,III and the T-domain was measured as described previously. The increase in tryptophan fluorescence was measured with unlabeled N71C, and the increase in AEDANS fluorescence was measured with labeled N71C. This mutant was chosen because it showed the largest change in AEDANS fluorescence upon TolA-II,III binding. The increase in fluorescence was measured also for S58C. In this case the amplitude of the AEDANS signal change was very low and consequently data (not shown) could not be analyzed properly. The data obtained for N71C were fitted by non-linear least squares assuming a pseudo first-order reaction (Fig. 5, upper panel). T-domain concentration was kept low (0.1 µM) in order to obtain as large an excess of TolA-II,III as possible. Concentrations of TolA-II,III higher than 5 µM, which was used as the highest concentration, gave significant background fluorescence. These were avoided by performing all measurements at 10–50-fold excess of TolA-II,III (1–5 µM TolA-II,III). A linear relationship between the observed rates and [TolA-II,III] was observed, confirming that a pseudo first-order kinetic model was appropriate. The slope of the line gave the second-order association rate constant (k1) and intercept gave k–1. The k1 values were identical, 1.1 ± 0.13 x 105 and 1.1 ± 0.05 x 105 M–1 s–1 for unlabeled and labeled sample, but the k–1 values were somewhat different. For labeled sample k–1 was five times higher (0.051 ± 0.037 s–1) than for the unlabeled one (0.011 ± 0.012 s–1). This gives Kd values of 0.5 µM 1 for N71C and 0.1 µM for N71C-AEDANS. The k1 values are similar to those obtained by the 90-residue wild-type T-domain (1.6 x 105 M–1 s–1) in the same system (13), while the k–1 value for the wild-type was higher (0.37 s–1). Thus these mutants show a higher affinity compared with the wild-type T-domain (2.3 µM), which qualitatively agrees with a higher affinity shown by SPR (N71C = 0.66 µM). However SPR (see earlier) does not show the above difference between N71C and N71C-AEDANS.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 5.
Kinetics of binding. Kinetics of binding as measured for the N71C mutant are shown. Experimental (squares) and fitted data (lines) are shown on the upper panel. Protein concentrations were 0.1 µM for N71C and 2 µM for TolA-II,III. An unlabeled mutant was used for the measurements of kinetics of tryptophan fluorescence. Excitation was set to 295 nm, and all emission above 320 nm was followed (open squares). A labeled mutant was used for the kinetics of AEDANS fluorescence change. Excitation was set to 295 nm, and all emission above 420 nm was followed (solid squares). kobs versus [TolA-II,III] is shown on the bottom panel. The points are means and S.D. of eight or more injections.

 

NMR Measurements—Following our initial one-dimensional 1H NMR evidence for the absence of secondary structure in the T-domain we determined the hydrodynamic radius of the T-domain peptide to investigate whether it adopted a compact three-dimensional structure; for an extended strand the hydrodynamic radius will be much larger than for a compact structure. The value observed was 19.3 ± 0.2 Å, compared with the expected radii for a fully folded and a completely unfolded protein of the same length of 18 and 30 Å, respectively (34). This demonstrates that the T-domain does not exist as an extended strand in solution, but rather adopts a compact average conformation without forming long stretches of secondary structure.

The initial 1H-15N HSQC spectrum of the T-domain40–76 peptide in solution showed at least 35 peaks, excluding side chain signals, suggesting that all the backbone NH resonances were detected between 7.7 and 8.7 ppm (1H). NOESY data collected on the peptide showed very few cross-peaks, presumably owing to its flexibility. The 1H-15N HSQC peaks of the T-domain40–76 are clearly not clustered in the narrow region (~0.45 ppm 1H) to be expected for a random coil, but are more dispersed, covering a range of 1 ppm (1H), which indicates that the peptide has at least some regions of non-random structure. However, if the peptide were rigidly structured the NOESY spectrum would be expected to show considerably more cross-peaks than were observed.

Adding unlabeled TolA-III to the labeled T-domain peptide immediately caused new peaks to appear with some of the original peaks reducing in intensity. The number and intensity of the new signals continued to increase with further additions of TolA-III and the intensity of the original peaks reducing further. This trend continued until a 1:2 molar ratio was obtained. Increasing the ratio to 1:4 gave no further changes, indicating that the proteins were already fully bound. The appearance and position of the new peaks showed the peptide to be bound to TolA-III with almost all the peptide resonances being shifted and broadened (Fig. 6). The lowest field signals of the unbound peptide arise from the tryptophan side chains at 9.9 and 10.1 ppm (1H), and simultaneous with the decrease in intensity of these peaks, a new peak appeared at even lower field, 11.5 ppm 1H(inset in Fig. 6). If this signal also arises from one or both tryptophans, this represents a large shift and would be in agreement with tryptophans 44 and 46 being in the binding site as suggested from fluorescence measurements. Thus, the NMR data clearly demonstrate that the colicin N T-domain40–76 peptide and TolA-III formed a complex and that the peptide adopts an organized structure in the complex.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 6.
1H,15N HSQC spectra of binding box peptide in solution and in the presence of TolA-III. The 1H-15N HSQC spectrum of 15N-labeled binding box peptide in black is overlaid with that of the peptide in the presence of excess (1:2 molar ratio) TolA-III, in red. Inset is the low field signal at 11.5 ppm. Spectra were recorded at 15 °C and in 50 mM acetate buffer at a pH of 5.

 


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The Properties of the Binding Box—In this article we have studied the changing dynamics of the flexible colicin N T-domain as it binds to TolA-III. Single cysteine mutants of the isolated 90-residue T-domain, created for site-specific FRET studies, provided additional data about the T-domain-TolA-III interaction. By comparison with the binding constants of the wild-type T-domain (Table I) a new mutation, N53C, was shown to inhibit binding; while F66C confirmed (13) the need for the benzyl side chain at this site. The other four mutants behaved similarly to the wild type. The normal binding for S58C and S61C was expected, as previous mutations to alanine at these sites did not reduce binding. Most probably these two polar groups are normally oriented toward solvent when in the complex. Similarly, the side chains of Asn70 and Asn71 are beyond the defined binding box. The binding of T-domain of colicin N to periplasmic receptor is unusual in its absolute dependence upon a large number of side chains. So far 14 of 27 residues in the minimal box have been shown to be required for the T-domain-TolA-III interaction. This is a much larger proportion than is generally found in scanning mutagenesis studies of protein-protein interactions (35) and may indicate the special requirements of an interaction where residues must maintain both the bound and the folded state.

FRET Determination of Intramolecular Distances—In order to investigate the complex binding of this flexible peptide we wished to compare dimensions of the fully denatured, free, and bound forms in solution using FRET. FRET can accurately provide absolute intramolecular distances comparable to known three-dimensional structures (23) or models (36). The changes in conformation in different conditions can then be evaluated by reference to the calibrated distances. In the case of the T-domain the calculated distances of Trp-AEDANS separation have no independent three-dimensional information available to check the accuracy of the distance determination. Moreover, the estimated distances could be liable to error because of the presence of two tryptophans present in the T-domain. However, both tryptophans (Trp44 and Trp46) can be considered as a single donor of resonance energy due to their proximity. The use of multiple donors and their proximity combine to decrease the polarization anisotropy of the fluorescence (37) increasing the likelihood that a value of {kappa}2 = 2/3 is appropriate. In support of this low anisotropy, values (r <0.07) were recorded for all mutants even when bound to TolA (Table I). This compares with a minimum value of 0.087 in the P-domain of colicin A used by us for previous FRET studies (38). {kappa}2 is frequently listed as the biggest limiting factor in FRET studies. However in reports where it has been investigated it has been shown that the isotropic orientation of absorption and emission probes is apparently satisfied (39). Additionally, we measured each distance using two separate mutants, which were AEDANS-labeled on neighboring cysteines. Agreement between these, especially in the distance changes, provides a way to confirm the lack of order in {kappa}2. A final control in our case was to measure the labeled mutants denatured with GdmHCl, and in this case the expected separation according to the distance of AEDANS from tryptophan was observed (Fig. 4). The assumption that each tryptophan is an equal donor of energy could introduce additional uncertainty but Gokce et al. (13) showed that their quantum yields are similar. Specific FRET distance measurements are only valid if the donor-acceptor distance is unchanged during the lifetime of the excited state. Our assumption is that T-domain bound to TolAIII has a defined conformation whereas the free and denatured forms do not. Thus in these two forms we have to assume that we are measuring the average E value for ensembles of states that are largely unstructured. This view is supported by our NMR investigations of the unbound peptide. The dependence of E upon the distance between the probe and the donor is sigmoidal meaning that the most accurate estimation is when E is between 30 and 70%, where the dependence is quasi-linear. This corresponds to distances of 17.4–23.0 Å, which accounts for most of our data. Mutants at the C terminus of the binding box in the GdmHCl-denatured state are the only exception. Here E was below 14% for F66C, N70C, and N71C, and the calculated R may be underestimated in these cases.

Bearing in mind these limits to accuracy, the following clear conclusions can be drawn from our study. (i) The dimensions of the T-domain binding box are different in solution and when bound to TolA-III. While the distance to the end of the binding box (mutants 70 and 71) changes only slightly, the change in the distance to the middle part (mutants 58 and 61) is considerable. (ii) For all the mutants studied the distances get considerably longer when the T-domain is denatured by GdmHCl. (iii) The distribution of the distances for the free state does not follow the pattern expected for a statistical coil (Fig. 7), which would be expected if the T-domain is completely unstructured in solution. (iv) According to stopped-flow data the binding and structural change occur simultaneously.



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 7.
Comparison of calculated distances for T-domain in solution to statistic coil model. The distances between the tryptophan and AEDANS determined from excitation spectra for native (open circles) and denatured (5 M GdmHCl (open squares)) are shown in comparison to statistical coil model. The data for statistical coil (solid line) were taken from Ref. 48.

 

The Unbound T-domain Peptide Ensemble Has a Compact Average Conformation—Our FRET (see above) and NMR (Fig. 6) data lead to a conclusion that the T-domain is not an elongated random coil in solution but shows evidence of intramolecular non-covalent associations. Incubation in 5 M GdmHCl increases the end-to-end distances by up to 7 Å, and since the T-domain is monomeric in solution (by gel filtration chromatography) the apparent shortened distances of the native peptide are likely to be due to a partial collapse. This is not a molten globule state since the far UV-CD is close to random coil, the tryptophan residues show almost maximum water exposure (12), and there is no secondary structure detectable by NMR (this work). However, our 1H NMR measurements of the diffusion coefficient give a value of 8.57 x 10–7 cm2/s, which predicts a hydrodynamic radius of 19.3 ± 0.2 Å. Using the empirical formulas from Ref. 34 for a protein of 99 residues, Rh is expected to be 18 Å for a folded globular protein and 30 Å for a fully unfolded protein of the same size. This supports the conclusion that the unbound T-domain is disordered with one or more fluctuating clusters of side chain interactions holding it together. No electron density is observed for the T-domain in crystals of full-length colicin N, and thus the disordered state is not a consequence of its isolation from the rest of the protein. Hence we should regard the T-domain in intact colicin N as a restricted flexible structure rather than a fully extended fishing line for Tol receptors.

T-domain Folds into a New Ordered Structure When Binding to TolA-III—Although only a few residues need to be directly involved in the binding interface, the combined folding and binding cause almost all the T-domain peptide resonances to be shifted and broadened (Fig. 6). This concerted folding and binding must have a significant additional cost in conformational entropy compared with the association of two prefolded partners, but our extensive ITC data is of little use for estimating this entropic term since the measured entropy change contains significant contributions from other sources, notably solvent and differences in heat capacity ({Delta}Cp) upon binding (18, 40).

However, it is clear that upon binding, the T-domain peptide has an altered, and more defined, structure that involves its entire backbone. The FRET data indicate that the central region (58–61) is stretched away from the N terminus by 3–6 Å, whereas the overall length of the binding box does not alter significantly. The simplest model that describes such a result is one in which the globular collapsed state forms an elongated structure or U-shape with the central region at the apex. From previous data the tryptophans at positions 44 and 46 form part of the interface, while we assume that the other side chains important for binding are also facing TolA-III. Importantly, the stopped-flow data, which measured FRET and tryptophan intensity changes, shows no evidence for separate binding and folding steps within the resolution of our measurements, suggesting that a single association event is taking place in which cooperative folding and binding takes place with a second-order association rate constant of 105 M–1 s–1.

Biological Implications—Although the broad definition of the various T-domains may include regions of secondary structure, in all three cases of known structure the first 65–90 residues are either undefined or in an extended conformation (14–17). Residues 1–90 and 1–83 are unresolved in the x-ray structures of colicin N and E3, respectively (14, 15). In colicin Ia the first 22 amino acids are not determined, and residues 23–67 are in an extended conformation held between the T- and P-domains (16). The relationship between the colicin N T-domain and the known structure is unclear but its affinity for TolA-III increases when it is expressed as a separate domain (12). As extensive unfolding occurs during translocation (41) it is reasonable to assume that in all cases the T-domain becomes a freely exposed flexible peptide. The Ton B-interacting porins have a parallel in the TonB box peptides of FhuA and BtuB, which change from a helical to extended conformation upon the binding of ligand to the receptor (42–44). This indicates that, in general, interaction with Ton or, by analogy (45) Tol, domains requires a flexible peptide. Since the FhuA peptide does traverse the outer membrane, flexibility is not solely due to the need for a translocation step. One possibility is that the transporter (e.g. FhuA), or the colicin bound to its outer membrane receptor, needs a flexible interaction to recruit the periplasmic partner with which it may not have a fixed spatial relationship. The T-domains are longer than the FhuA peptide and it has been proposed that an unstructured ColN T-domain might pass through the OmpF porin channel (13). The flexibility might indeed be required for this or similar processes but the interaction of the folded g3p phage protein with TolA-III under similar conditions suggests that again flexibility is not a prerequisite for translocation. Although colicin A and E require both TolA and TolB, colicin N and the B group colicins show that flexibility is not the consequence of the need to accommodate two periplasmic binding sites.

A fundamental reason for the reliance upon flexible peptides may be one of parsimony. Many side chains in the 27 amino acid box contribute strongly to binding and move from a solvated to dehydrated environment. This could lead to a large change in {Delta}Cp per residue as seen in binding of flexible peptides to DNA (46). On the other hand most structured binding domains interact via a small proportion of their residues (35). If the T-domains are simply an anchor to stabilize one end of the translocated state then it is highly efficient (or extremely generous) of the host bacterium to provide a rigid protein template for the compact linear toxin sequence of interacting residues to bind to. This may explain why unstructured regions of proteins show more protein-protein interactions than folded domains (46). The remaining flexible non-binding regions may then provide linkers, which impose little restriction on the interactions of the defined binding box. Shoemaker et al. (47) have proposed the fly-casting role for flexible domains where receptor densities are low. Receptors captured over a wide radius then bind tightly because of a concerted folding of the previously flexible polypeptide. Although OmpF is present at 100,000 copies per cell, the TolA protein numbers less than 500, indicating that fly-casting Tol-recognition regions of colicins could be a requirement for efficient translocation.

In this study we have not considered the effects of colicin binding on the TolA-III protein. This has been studied by NMR using the interaction with both g3p and the larger (172 residue) T-domain of colicin A (11). While the g3p interaction appears to leave the TolA-III structure unchanged, the colicin domain appears to destabilize it. The colicin A domain shows no sequence homology to the colicin N peptide used here, and it will be interesting to compare the ways in which these two apparently similar domains behave.


   FOOTNOTES
 
* This work was supported by Grants 056232, 040422, 055979, 066850, and 049735 from the Wellcome Trust and the Biotechnology and Biosciences Research Council (BBSRC). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Recipient of a long-term Federation of European Biochemical Societies (FEBS) fellowship. Back

** A BBSRC Research Development Fellow. To whom correspondence should be addressed. Tel.: 191-222-8865; Fax: 191-222-7424; E-mail: j.h.l{at}ncl.ac.uk.

1 The abbreviations used are: T-domain, translocation domain; AEDANS, amino-ethyl-amino naphthalene-1-sulfonic acid; DSS, 2,2-di-methyl-2-silapentane-5-sufonic acid; FRET, fluorescence resonance energy transfer; GdmHCl, guanidinium hydrochloride; IAEDANS, 5-((((2-iodoacetyl)amino)ethyl)amino)naphthalene-1-sulfonic acid; NTA, nitrilotriacetic acid; SPR, surface plasmon resonance; P-domain, poreforming domain. Back

2 E. Raggett, J. H. Lakey, R. Boetzel, and G. R. Moore, unpublished observations. Back

3 G. Anderluh, Q. Hong, J. H. Lakey, unpublished observations. Back


   ACKNOWLEDGMENTS
 
We thank and gratefully acknowledge the Higher Education Funding Council for England for a 500 MHz NMR spectrometer. We also thank W. Webster for the TolA strains. The excellent technical assistance of Pauline Heslop is gratefully acknowledged.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Kerr, B., Riley, M. A., Feldman, M. W., and Bohannan, B. J. M. (2002) Nature 418, 171–174[CrossRef][Medline] [Order article via Infotrieve]
  2. James, R., Kleanthous, C., and Moore, G. R. (1996) Microbiology (UK) 142, 1569–1580[Free Full Text]
  3. Lakey, J. H., and Slatin, S. L. (2001) in Pore-forming Toxins (Van Der Goot, F. G., ed) Vol. 257, pp. 131–161, Springer Verlag, Heidelberg
  4. Davies, J. K., and Reeves, P. (1975) J. Bacteriol. 123, 102–117[Abstract/Free Full Text]
  5. Davies, J. K., and Reeves, P. (1975) J. Bacteriol. 123, 96–101[Abstract/Free Full Text]
  6. Pugsley, A. P. (1987) Mol. Microbiol. 1, 317–325[CrossRef][Medline] [Order article via Infotrieve]
  7. Schendel, S. L., Click, E. M., Webster, R. E., and Cramer, W. A. (1997) J. Bacteriol. 179, 3683–3690[Abstract/Free Full Text]
  8. Click, E. M., and Webster, R. E. (1998) J. Bacteriol. 180, 1723–1728[Abstract/Free Full Text]
  9. Holliger, P., Riechmann, L., and Williams, R. L. (1999) J. Mol. Biol. 288, 649–657[CrossRef][Medline] [Order article via Infotrieve]
  10. Lubkowski, J., Hennecke, F., Pluckthun, A., and Wlodawer, A. (1999) Struct. Fold. Des. 7, 711–722[Medline] [Order article via Infotrieve]
  11. Deprez, D., Blanchard, L., Guerlesquin, F., Gavioli, M., Simorre, J.-P., Lazdunski, C., Marion, D., and Lloubes, R. (2002) Biochemistry 41, 2589–2598[CrossRef][Medline] [Order article via Infotrieve]
  12. Raggett, E. M., Bainbridge, G., Evans, L. J. E., Cooper, A., and Lakey, J. H. (1998) Mol. Microbiol. 28, 1335–1344[CrossRef][Medline] [Order article via Infotrieve]
  13. Gokce, I., Raggett, E. M., Hong, Q., Virden, R., Cooper, A., and Lakey, J. H. (2000) J. Mol. Biol. 304, 621–632[CrossRef][Medline] [Order article via Infotrieve]
  14. Vetter, I. R., Parker, M. W., Tucker, A. D., Lakey, J. H., Pattus, F., and Tsernoglou, D. (1998) Structure 6, 863–874[Medline] [Order article via Infotrieve]
  15. Soelaiman, S., Jakes, K., Wu, N., Li, C. M., and Shoham, M. (2001) Mol. Cell 8, 1053–1062[CrossRef][Medline] [Order article via Infotrieve]
  16. Wiener, M., Freymann, D., Ghosh, P., and Stroud, R. M. (1997) Nature 385, 461–464[CrossRef][Medline] [Order article via Infotrieve]
  17. Collins, E. S., Whittaker, S. B. M., Tozawa, K., MacDonald, C., Boetzel, R., Penfold, C. N., Reilly, A., Clayden, N., Osborne, M. J., Kleanthous, C., James, R., and Moore, G. R. (2002) J. Mol. Biol. 122, 145–162
  18. Cooper, A., Johnson, C. M., Lakey, J. H., and Nollmann, M. (2001) Biophys. Chem. 93, 215–230[CrossRef][Medline] [Order article via Infotrieve]
  19. Dyson, H. J., and Wright, P. E. (2002) Curr. Opin. Struct. Biol. 12, 54–60[CrossRef][Medline] [Order article via Infotrieve]
  20. Dunker, A. K., Brown, C. J., Lawson, J. D., Iakoucheva, L. M., and Obradovic, Z. (2002) Biochemistry 41, 6573–6582[CrossRef][Medline] [Order article via Infotrieve]
  21. Uversky, V. N. (2002) Protein Sci. 11, 739–756[CrossRef][Medline] [Order article via Infotrieve]
  22. Stryer, L. (1978) Annu. Rev. Biochem. 47, 819–846[CrossRef][Medline] [Order article via Infotrieve]
  23. Lakey, J. H., Baty, D., and Pattus, F. (1991) J. Mol. Biol. 218, 639–653[CrossRef][Medline] [Order article via Infotrieve]
  24. Levengood-Freyermuth, S. K., Click, E. M., and Webster, R. E. (1993) J. Bacteriol. 175, 222–228[Abstract/Free Full Text]
  25. Anderluh, G., Gokce, I., and Lakey, J. H. (2002) Protein Expression Purif. 28, 173–181
  26. Marriott, G., Kirk, W. R., Johnsson, N., and Weber, K. (1990) Biochemistry 29, 7004–7011[CrossRef][Medline] [Order article via Infotrieve]
  27. Lakey, J. H., Duché, D., González-Mañas, J.-M., Baty, D., and Pattus, F. (1993) J. Mol. Biol. 230, 1055–1067[CrossRef][Medline] [Order article via Infotrieve]
  28. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) J. Biomol. NMR 6, 277–293[Medline] [Order article via Infotrieve]
  29. Wishart, D. S., Bigam, C. G., Yao, J., Abildgaard, F., Dyson, H. J., Oldfield, E., Markley, J. L., and Sykes, B. D. (1995) J. Biomol. NMR 6, 135–140[Medline] [Order article via Infotrieve]
  30. Stejskal, E. O., and Tanner, J. E. (1965) J. Chem. Phys. 42, 288–292[CrossRef]
  31. Gibbs, S. J., and Johnson, C. S. (1991) J. Magn. Reson. 93, 395–402
  32. Jones, J. A., Wilkins, D. K., Smith, L. J., and Dobson, C. M. (1997) J. Biomol. NMR 10, 199–203[CrossRef]
  33. Van Der Meer, B. W., Coker, G., and Chen, S. (1991) Resonance Energy Transfer, Theory and Data, Wiley-VCH, New York
  34. Wilkins, D. K., Grimshaw, S. B., Receveur, V., Dobson, C. M., Jones, J. A., and Smith, L. J. (1999) Biochemistry 38, 16424–16431[CrossRef][Medline] [Order article via Infotrieve]
  35. Stites, W. E. (1997) Chem. Reviews 97, 1233–1250[CrossRef][Medline] [Order article via Infotrieve]
  36. Eisenhawer, M., Cattarinussi, S., Kuhn, A., and Vogel, H. (2001) Biochemistry 40, 12321–12328[CrossRef][Medline] [Order article via Infotrieve]
  37. Wahl, P. (1975) in Biochemical Fluorescence: Concepts (Chen, R. F., and Edelhoch, H., eds) Vol. 1, pp. 1–42, Marcel Dekker, New York
  38. Lakey, J. H., Massotte, D., Heitz, F., Dasseux, J. L., Faucon, J. F., Parker, M. W., and Pattus, F. (1991) Eur. J. Biochem. 196, 599–607[Medline] [Order article via Infotrieve]
  39. Chen, R. F. (1990) in Practical Fluorescence (Guilbault, G. G., ed), Marcel Dekker, New York
  40. Raggett, E. M., Bainbridge, G., Evans, L. J. A., Cooper, A., and Lakey, J. H. (1997) Protein Sci. 6, 63
  41. Benedetti, H., Lloubes, R., Lazdunski, C., and Letellier, L. (1992) EMBO J. 11, 441–447[Medline] [Order article via Infotrieve]
  42. Locher, K. P., Rees, B., Koebnik, R., Mitschler, A., Moulinier, L., Rosenbusch, J. P., and Moras, D. (1998) Cell 95, 771–778[CrossRef][Medline] [Order article via Infotrieve]
  43. Vakharia, H. L., and Postle, K. (2002) J. Bacteriol. 184, 5508–5512[Abstract/Free Full Text]
  44. Merianos, H. J., Cadieux, N., Lin, C. H., Kadner, R. J., and Cafiso, D. S. (2000) Nat. Struct. Biol. 7, 205–209[CrossRef][Medline] [Order article via Infotrieve]
  45. Witty, M., Sanz, C., Shah, A., Grossmann, J. G., Mizuguchi, K., Perham, R. N., and Luisi, B. (2002) EMBO J. 21, 4207–4218[CrossRef][Medline] [Order article via Infotrieve]
  46. Lundback, T., Chang, J.-T., Phillips, K., Luisi, B., and Ladbury, J. E. (2000) Biochemistry 39, 7570–7579[CrossRef][Medline] [Order article via Infotrieve]
  47. Shoemaker, B. A., Portman, J. J., and Wolynes, P. G. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 8868–8873[Abstract/Free Full Text]
  48. Navon, A., Ittah, V., Landesman, P., Scheraga, H. A., and Haas, E. (2001) Biochemistry 40, 105–118[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Microbiol. Mol. Biol. Rev.Home page
E. Cascales, S. K. Buchanan, D. Duche, C. Kleanthous, R. Lloubes, K. Postle, M. Riley, S. Slatin, and D. Cavard
Colicin Biology
Microbiol. Mol. Biol. Rev., March 1, 2007; 71(1): 158 - 229.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
W. L. Vos, R. B. M. Koehorst, R. B. Spruijt, and M. A. Hemminga
Membrane-bound Conformation of M13 Major Coat Protein: A STRUCTURE VALIDATION THROUGH FRET-DERIVED CONSTRAINTS
J. Biol. Chem., November 18, 2005; 280(46): 38522 - 38527.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
S. Pommier, M. Gavioli, E. Cascales, and R. Lloubes
Tol-Dependent Macromolecule Import through the Escherichia coli Cell Envelope Requires the Presence of an Exposed TolA Binding Motif
J. Bacteriol., November 1, 2005; 187(21): 7526 - 7534.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
N. G. Housden, S. R. Loftus, G. R. Moore, R. James, and C. Kleanthous
Cell entry mechanism of enzymatic bacterial colicins: Porin recruitment and the thermodynamics of receptor binding
PNAS, September 27, 2005; 102(39): 13849 - 13854.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
G. Anderluh, I. Gokce, and J. H. Lakey
A Natively Unfolded Toxin Domain Uses Its Receptor as a Folding Template
J. Biol. Chem., May 21, 2004; 279(21): 22002 - 22009.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
G. E. Fanucci, N. Cadieux, R. J. Kadner, and D. S. Cafiso
Competing ligands stabilize alternate conformations of the energy coupling motif of a TonB-dependent outer membrane transporter
PNAS, September 30, 2003; 100(20): 11382 - 11387.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/24/21860    most recent
M300411200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Anderluh, G.
Right arrow Articles by Lakey, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Anderluh, G.
Right arrow Articles by Lakey, J. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2003 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement