| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Biochemistry and Molecular Biology, The Pennsylvania State University, College of Medicine, Hershey, Pennsylvania 17033, USA
Reprint requests to: Ira J. Ropson, Department of Biochemistry and Molecular Biology, The Pennsylvania State University, College of Medicine, Hershey, PA 17033, USA; e-mail: iropson{at}psu.edu; fax: (717) 531-7072.
(RECEIVED August 15, 2003; FINAL REVISION September 30, 2003; ACCEPTED September 30, 2003)
 |
Abstract |
|---|
 TOP Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Keywords: proline isomerization; protein folding; folding intermediates; 
-sheet proteins; structural homology; stopped-flow kinetics, iLBP
2 Present address: Provid Pharmaceuticals, 10 Knightsbridge Road, Piscataway, NJ 08854, USA. 
Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.03317804.
|
Introduction |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
-sheet proteins function in the intracellular trafficking of small amphipathic ligands (Banaszak et al. 1994; Zimmerman and Veerkamp 2002). The structures of these proteins are very similar, consisting of 10 antiparallel -strands and two short 
-helices (Banaszak et al. 1994). The topology of the protein is a simple -meander. The first two strands are connected by a helix–turn–helix motif, and the remaining strands are connected by reverse turns (Fig. 1
). The -strands are organized into two nearly orthogonal -sheets surrounding a large internal ligand-binding cavity, which is filled with solvent in the absence of ligand. Only one of these proteins has a disulfide bond and all of the prolines are in the trans configuration in the native structures (Banaszak et al. 1994).
|
One member of the family, cellular retinoic acid binding protein I (CRABP-I) both folds and unfolds significantly slower than other family members, resulting in a similar stability in the absence of urea. The refolding of this protein shows four distinct kinetic phases, a burst phase, followed by three additional observed phases (Burns et al. 1998; Clarke et al. 1998; Eyles and Gierasch 2000; Burns and Ropson 2001). The slowest refolding phase has a small amplitude, minimal dependence of the rate on the final denaturant concentration, and is slow enough to be caused by a proline isomerization event in the unfolded state (Nall 1994). There are four prolines in CRABP-I, all of which are in the trans configuration in the native state. Conventional double-jump mixing experiments suggest that proline isomerization in the unfolded state is not an important factor in the folding of this protein (data not shown). However, these experiments are inconclusive for this protein because the rate at which unfolding occurs is similar to that of proline isomerization in the unfolded state (Nall 1994; Eyles and Gierasch 2000). The addition of peptidyl prolyl isomerase during refolding did not alter the folding kinetics (Eyles and Gierasch 2000; data not shown). Replacement of Pro 85 with alanine abolished this slow phase, supporting a role for cistrans isomerization of this residue in the unfolded state as responsible for the slowest refolding phase (Eyles and Gierasch 2000). The two most closely related proteins in this family, cellular retinoic acid binding protein II and cellular retinol binding protein I (CRABP II and CRBP II, respectively), have valines at this sequence location instead of proline and lack this slow refolding phase. As such, the replacement of Pro 85 with the sequence-conserved valine was expected to confirm the role of proline isomerization in this slow phase. However, as described below, this mutant protein does have a slow folding phase that is very similar in rate and amplitude to that of wild-type protein, making the role of proline isomerization in the folding of this protein uncertain.
|
Results |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
-sheet minimum at 216 nm and a small minimum at 232 nm. The minima at 232 nm were shown to be due to favorable interactions between the 
electrons of the aromatic ring of Trp 109 and the positively charged side chain of Arg 111, providing a unique spectral signature for this region of the protein (Fig. 1
; Clark et al. 1996). The presence of these minima in the P85V-CRABP I CD spectra indicated that this specific tertiary interaction was intact in the variant protein. The CD spectra of the denatured proteins were identical, suggesting that the two proteins were equally unfolded in 8 M urea. Similarly, the fluorescence emission spectra of denatured WT-CRABP I and P85V-CRABP I were coincident with a maximum of 354 nm. As such, the three tryptophans of both proteins were equally exposed in the unfolded state (Creighton 1993). However, the environment of the tryptophans in the native state for the wild-type and the P85V mutant were not equivalent. Native P85V-CRABP I was more fluorescent and the emission maximum was redshifted approximately 10 nm in comparison to the wild-type protein. There are three tryptophans in CRABP I (Trp 7, Trp 87, and Trp 109). Cys 95 is a significant quencher of Trp 109 (Clark et al. 1998) and is also within 4.4 Å of Pro 85 in the wild-type protein (Kleywegt et al. 1994; Thompson et al. 1995). Given the enhanced fluorescence of the P85V-CRABP I mutant, it appears that the mutation perturbs the structure near Cys 95, resulting in less efficient quenching of Trp 109. Trp 109 is in an unusually polar environment for the interior of a protein. Polar residues with side-chain atoms within 5 Å of the Trp 109 side chain include Glu 73, Ser 83, Cys 95, Gln 97, and Arg 111. Several water molecules are also within 5 Å of the Trp 109 side chain in both the apo- and holo-structures (Kleywegt et al. 1994; Thompson et al. 1995). The emission spectrum of this tryptophan is redshifted compared to the other tryptophans of the protein (Clark et al. 1998). Reduced quenching of the emission of this tryptophan would cause a redshift in the emission maxima of the entire protein. However, the overall orientation of the aromatic ring of Trp 109 with respect to Arg 111 must remain similar to that of the wild-type protein because the inflection at 232 nm in the CD spectrum was present.
Ligand binding studies
CRABP I has a nanomolar affinity for retinoic acid and binds this ligand with a 1:1 stoichiometry (Norris et al. 1994). The binding of retinoic acid by P85V-CRABP I and WT-CRABP I is shown in Figure 2. The binding isotherms of the two proteins were nearly coincident (Kd of P85V-CRABP I, 16 ± 3 nM; Kd of WT-CRABP I, 16 ± 1 nM). The binding stoichiometry of the variant protein was 1:1 (inset of Fig. 2), identical to that of WT-CRABP I. As such, replacement of Pro 85 with valine did not affect the function of the mutant protein.
|
. The data for WT-CRABP I have been reported previously (Burns et al. 1998). All folding transitions for the variant protein were completely reversible and overlaid well. The data collected by both CD and fluorescence were best fit by a two-state model, suggesting that no significant population of intermediates was present.
|
GH2O value of 0.77 kcal/mole at 3.76 M urea. Because replacement of proline residues usually destabilizes the native state, the increase in stability of this variant was unexpected. The imino ring of prolines decreases the rotational freedom of this side chain, restricting the degree of randomness in the unfolded state. Proline residues can stabilize the native state by narrowing the entropic energy barrier between the native and unfolded states.
Refolding kinetics
Similar to WT-CRABP I, the folding transitions of P85V-CRABP I by fluorescence were best fit by a tri-exponential equation (Fig. 4). As such, at least two intermediates were present on the folding path. The three observed folding rates differed by up to three orders of magnitude from each other and displayed a logarithmic dependence on the final urea concentration (Fig. 5). A significant burst phase was also present. This burst phase intermediate (IB) had a 4.5-fold greater intensity than that expected for the denatured state in 1.1 M urea (Fig. 6). In comparison, the wild-type burst phase intermediate was twofold greater in intensity than that expected for the denatured state under similar conditions (Burns et al. 1998). Because both the equilibrium spectra of the denatured states and the slope of the denatured state baselines for the mutant and wild-type protein were virtually identical, differences in the extrapolated denatured state fluorescence did not account for the fluorescent intensity of IB.
|
|
|
 |
where B is the burst rate and k1R, k2R, and k3R are the observed rates for the folding process (Fig. 6). The amplitudes associated with k1R, k2R, and k3R when folding in 1.1 M urea were 76%, 13%, and 11% of the total amplitude change, respectively (data not shown). Therefore, most of the detectable tertiary changes (76%) occurred during the formation of I2. The proportion of the amplitude change associated with each rate were comparable to those observed for wild type.
The folding rates were independent of protein concentration (5–80 µM), suggesting that the fluorescence changes were not the result of protein association (data not shown). Stopped-flow CD measurements showed a burst phase and a single observed phase corresponding to the fastest observed fluorescence phase (Fig. 5).
Unfolding kinetics
The rates of unfolding as a function of urea concentration by both fluorescence and CD are shown in Figure 5. Data collected by both methods were best fit by a mono-exponential equation at all urea concentrations. The entire expected amplitude associated with the rate was present at each final concentration of denaturant (Fig. 6). Thus, the kinetic model that best fit the unfolding data of P85V-CRABP I was
 |
where N and U stand for the native and unfolded states, respectively. This behavior was different from that of WT-CRABP I (Burns et al. 1998). The unfolding transition of WT-CRABP I was biphasic by fluorescence but monophasic by CD. The kinetic model that best fit the data is
 |
where I represents the intermediate state. The rate detected by CD during the unfolding of WT-CRABP I coincided with the slower rate observed by fluorescence. Thus, the intermediate observed during unfolding of WT-CRABP I has nativelike secondary structure associated with it.
|
Discussion |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
However, the replacement of Pro 85 with valine did not remove this phase. An alternative explanation is that the alanine replacement relaxes a steric constraint on the native structure involving the correct positioning of the imino side chain of proline into the native structure. As such, a slow folding reaction may be caused by a barrier associated with the correct positioning of the proline ring in the final folded structure rather than proline isomerization. The replacement of proline with the -branched amino acid valine may mimic the steric problems associated with the correct positioning of the proline side chain at this structural location, resulting in the appearance of the slowest phase in folding.
Proline 85 is located in a region of the protein that has been postulated to be an initiating site for folding of these proteins (Ropson and Frieden 1992; Dalessio and Ropson 2000) and is within the strand that spans both the top and bottom -sheet (Fig. 1). As such, the constraints on the folding of this region of the protein may be more stringent.
An alternative explanation postulates that the valine substitution may have created a new barrier not present in the wild-type protein that causes a similar folding rate. In this case, the slowest phase in the folding of the WT-CRABP I was indeed due to proline isomerization. It is also possible that the replacement of proline with the smaller alanine residue reduced the observed amplitude of the slowest phase to the point where it could no longer be detected, even though it was still there.
It has long been known that processes other than proline isomerization can cause slow folding phases. In the cases of human lysozyme (Herning et al. 1991) and trp aporepressor (Mann et al. 1995; Gloss et al. 2001) removal of all of the proline residues failed to remove the slowest folding phase. The data shown here indicate that the loss of a particular folding phase in a mutant protein where proline is replaced with alanine is not definitive proof that proline isomerization is responsible for that phase.
|
Materials and methods |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Reagents
Denaturant stock (10 M urea) solutions were prepared and stored at –20°C until the day of the experiment as previously described (Burns et al. 1998; Dalessio and Ropson 1998). Working solutions (8 M urea) were prepared by adding buffer components (final concentrations: 25 mM NaPO4, 75 mM NaCl, 0.1 mM EDTA, 0.1 mM DTT) to the denaturant stock and adjusting the pH to 8.0. Final denaturant concentrations were determined by refractive index measurements (Pace 1986). All buffers were filtered using a 0.2-mm Whatman nylon membrane. All chemicals were reagent grade.
All-trans retinoic acid (Sigma) was reconstituted with 100% ethanol, aliquoted into microfuge tubes in the dark, and stored at –20°C. The concentration of retinoic acid was estimated using an extinction coefficient of 45,000 M–1cm–1 (Horwitz and Heller 1973).
Equilibrium studies
Equilibrium unfolding and refolding transitions as a function of denaturant concentration were monitored using far-UV CD and intrinsic tryptophan fluorescence as previously described (Burns et al. 1998; Dalessio and Ropson 1998). All measurements were made at 25°C after at least 1 h equilibration. Reversibility of unfolding was demonstrated by coincident equilibrium curves for native protein and protein that was previously denatured. The data were corrected for the background signal of the buffer and urea solutions (Burns et al. 1998; Dalessio and Ropson 1998).
Fluorescence kinetic studies
Unfolding and refolding kinetics were monitored using an Applied Photophysics DX-17MV stopped-flow spectrophotometer as previously described (Burns et al. 1998; Dalessio and Ropson 1998). Excitation was at 290 nm (2.3 nm bandpass) using a 0.2-cm path-length. Emission intensity was monitored above 305 nm at 90° through a WG345 Schott glass filter (Oriel) at 25°C. Five parts of denaturant solution were mixed with one part of protein solution (11 µM, final protein concentration). Three to seven kinetic traces were averaged at each final denaturant concentration. The dead time of the instrument was determined to be between 5 and 10 msec (Ropson and Dalessio 1997).
CD kinetic studies
The kinetics of unfolding and refolding were monitored at 218 nm (2 nm bandpass) in a 0.1-cm cell with a Jasco J-710 spectropolarimeter (CD) in conjunction with a RX1000 stopped-flow apparatus (Applied Photophysics) as previously described (Ropson and Dalessio 1997). For all experiments five parts of denaturant were mixed with one part of protein (32 µM protein, final concentration). Five to seven transitions were averaged at each denaturant concentration. The dead time of the instrument was 15–30 msec (Ropson and Dalessio 1997).
Ligand binding studies
The binding of retinoic acid to CRABP I and P85V-CRABP I was monitored by fluorescence using an Aminco-Bowman Series 2 luminescence spectrometer. Inner filter corrections were unnecessary due to the low absorbance of retinoic acid over this concentration and wavelength range (Birdsall et al. 1983). Samples were prepared by mixing retinoic acid (0–1.5 µM, concentration range) and protein (0.7 µM, final concentration). Samples were incubated for 1 h at 25°C in the dark before analysis. The data were corrected for the background signal of the buffer and free retinoic acid and were done in duplicate. The stoichiometry of ligand binding was determined using the continuous variation method (Huang 1982). Samples contained 0–1 mole fraction of ligand and 0–1 mole fraction of protein. The total concentration of protein plus ligand for any given sample was 1 µM. Samples were analyzed by fluorescence as described above. Each sample was corrected for the signal of protein in the absence of ligand. The stoichiometry of ligand binding was determined from two normalized data sets.
Computer fitting of data
Ligand binding isotherms were normalized and fit to the following equations by nonlinear least squares regression with KaleidaGraph (Synergy Software):
 |
where 
is the fractional saturation, F is the normalized and corrected fluorescence, F0 is the fluorescence of the protein in the absence of ligand, F
is the fluorescence of the protein at saturation, [P]total and [L]total are the total concentration of protein and ligand, respectively, and A =[P]total + [L]total + Kd.
Ligand binding stoichiometry was determined by the value where the two lines that best fit the data intersected in the continuous variation plot (Huang 1982). The lines were fit by linear regression with KaleidaGraph.
Nonlinear least-square fits of the equilibrium data were determined with KaleidaGraph as previously described (Burns et al. 1998). Midpoints of the transition for the equilibrium data were determined by substituting G H2O/midpoint for mG into the six-term equation as previously described (Dalessio and Ropson 1998).
The program supplied by Applied Photophysics was used to determine the nonlinear least-square fits of the kinetic data to mono-, bi-, and tri-exponential equations using Kaleidagraph (Burns et al. 1998).
The extent of the burst phase reaction occurring in the dead time of the stopped-flow CD and fluorescence experiments was determined by constructing an equilibrium plot using the normalized initial and final intensities obtained from both folding and unfolding transitions over a range of urea concentrations. The intensities were normalized using minimum and maximum values. The expected intensities of the native and denatured states at high and low denaturant concentrations were determined from the native and denatured state baselines in this plot. The extent of the reaction occurring in the dead time was determined by comparing the calculated intensity change with the observed intensity change.
Standard criteria were used to assess the quality of the fit to the various models (Mannervik 1982; Motulsky and Ransnas 1987). When the standard error of a parameter exceeded the value of the fitted parameter, the parameter was eliminated from the equation, and the fit repeated. This process continued until all of the remaining terms were significant.
|
Acknowledgments |
|---|
The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
|
References |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Birdsall, B., King, R.W., Wheeler, M.R., Lewis, C.A.J., Goode, S.R., Dunlap, R.B., and Roberts, G.C. 1983. Correction for light absorption in fluorescence studies of protein-ligand interactions. Anal. Biochem. 132: 353–361.[CrossRef][Medline]
Burns, L.L. and Ropson, I.J. 2001. Folding of intracellular retinol and retinoic acid binding proteins. Proteins 43: 292–302.[CrossRef][Medline]
Burns, L.L., Dalessio, P.M., and Ropson, I.J. 1998. Folding mechanism of three structurally similar -sheet proteins. Proteins 33: 107–118.[CrossRef][Medline]
Clark, P.L., Liu, Z.P., Zhang, J., and Gierasch, L.M. 1996. Intrinsic tryptophans of CRABP I as probes of structure and folding. Protein Sci. 5: 1108–1117.[Abstract]
Clark, P.L., Weston, B.F., and Gierasch, L.M. 1998. Probing the folding pathway of a -clam protein with single tryptophan constructs. Fold. Des. 3: 401–412.[CrossRef][Medline]
Creighton, T.E. 1993. Proteins: Structures and molecular properties. W.H. Freeman and Company, New York.
Dalessio, P.M. and Ropson, I.J. 1998. pH dependence of the folding of intestinal fatty acid binding protein. Arch. Biochem. Biophys. 359: 199–208.[CrossRef][Medline]
———. 2000. -sheet proteins with nearly identical structures have different folding intermediates. Biochemistry 39: 860–871.[CrossRef][Medline]
Eyles, S.J. and Gierasch, L.M. 2000. Multiple roles of prolyl residues in structure and folding. J. Mol. Biol. 301: 737–747.[CrossRef][Medline]
Gloss, L.M., Simler, B.R., and Matthews, C.R. 2001. Rough energy landscapes in protein folding: Dimeric E. coli Trp repressor folds through three parallel channels. J. Mol. Biol. 312: 1121–1134.[CrossRef][Medline]
Herning, T., Yutani, K., Taniyama, Y., and Kikuchi, M. 1991. Effects of proline mutations on the unfolding and refolding of human lysozyme: The slow refolding kinetic phase does not result from proline cistrans isomerization. Biochemistry 30: 9882–9891.[CrossRef][Medline]
Hodsdon, M.E. and Frieden, C. 2001. Intestinal fatty acid binding protein: The folding mechanism as determined by NMR studies. Biochemistry 40: 732–742.[CrossRef][Medline]
Horwitz, J. and Heller, J. 1973. Interactions of all-trans, 9-, 11-, and 13-cis-retinal, all-trans-retinyl acetate, and retinoic acid with human retinol-binding protein and prealbumin. J. Biol. Chem. 248: 6317–6324.
Huang, C.Y. 1982. Determination of binding stoichiometry by the continuous variation method: The Job plot. Methods Enzymol. 87: 509–525.[Medline]
Kiefhaber, T. 1995. Protein folding kinetics. In Protein stability and folding (ed. B.A. Shirley), Vol. 40, pp. 313–341. Humana Press, Totowa, NJ.
Kleywegt, G.J., Bergfors, T., Senn, H., Le Motte, P., Gsell, B., Shudo, K., and Jones, T.A. 1994. Crystal structures of cellular retinoic acid binding proteins I and II in complex with all-trans-retinoic acid and a synthetic retinoid. Structure 2: 1241–1258.[Medline]
Mann, C.J., Shao, X., and Matthews, C.R. 1995. Characterization of the slow folding reactions of trp aporepressor from Escherichia coli by mutational analysis of prolines and catalysis by a peptidyl-prolyl isomerase. Biochemistry 34: 14573–14580.[CrossRef][Medline]
Mannervik, B. 1982. Regression analysis, experimental error, and statistical criteria in the design and analysis of experiments for discrimination between rival kinetic models. Methods Enzymol. 87: 370–390.[Medline]
Motulsky, H.J. and Ransnas, L.A. 1987. Fitting curves to data using nonlinear regression: A practical and nonmathematical review. FASEB J. 1: 365–374.[Abstract]
Nall, B.T. 1994. Proline isomerization as a rate-limiting step. In Mechanisms of protein folding (ed. R.H. Pain), pp. 80–99. Oxford University Press, New York.
Norris, A.W, Cheng, L., Gigue're, V., Rosenberger, M., and Li, E. 1994. Measurement of subnanomolar retinoic acid binding affinities for cellular retinoic acid binding proteins by fluorometric titration. Biochim. Biophys. Acta 1209: 10–18.[CrossRef][Medline]
Pace, C.N. 1986. Determination and analysis of urea and guanidine hydrochloride denaturation curves. Methods Enzymol. 131: 266–280.[Medline]
Pace, C.N., Vajdos, F., Fee, L., Grimsley, G., and Gray, T. 1995. How to measure and predict the molar absorption coefficient of a protein. Protein Sci. 4: 2411–2423.[Abstract]
Ropson, I.J. and Dalessio, P.M. 1997. Fluorescence spectral changes during the folding of intestinal fatty acid binding protein. Biochemistry 36: 8594–8601.[CrossRef][Medline]
Ropson, I.J. and Frieden, C. 1992. Dynamic NMR spectral analysis and protein folding: Identification of a highly populated folding intermediate of rat intestinal fatty acid-binding protein by 19F NMR. Proc. Natl. Acad. Sci. 89: 7222–7226.
Thompson, J.R., Bratt, J.M., and Banaszak, L.J. 1995. Crystal structure of cellular retinoic acid binding protein 1 shows increased access to the binding cavity due to formation of an intermolecular -sheet. J. Mol. Biol. 252: 433–446.[CrossRef][Medline]
Yeh, S.R., Ropson, I.J., and Rousseau, D.L. 2001. Hierarchical folding of intestinal fatty acid binding protein. Biochemistry 40: 4205–4210.[CrossRef][Medline]
Zimmerman, A.W. and Veerkamp, J.H. 2002. New insights into the structure and function of fatty-acid binding proteins. Cell. Mol. Life Sci. 59: 1096–1116.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
