Protein Engineering, Vol. 12, No. 6, 485-490,
June 1999
© 1999 Oxford University Press
Direct energy transfer to study the 3D structure of non-native proteins: AGH complex in molten globule state of apomyoglobin
1 Laboratory of Experimental and Computational Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892-5677, USA, 2 Institute of Macromolecular Compounds, Russian Academy of Sciences, 199004 St Petersburg and 4 Institute of Protein Research, Russian Academy of Sciences, 142292 Pushchino, Moscow Region, Russia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionReferences |
|---|
The direct energy transfer technique was modified and applied to probe the relative localization of apomyoglobin A-, G- and H-helixes, which are partly protected from deuterium exchange in the equilibrium molten globule state and in the molten globule-like kinetic intermediate. The non-radiative transfer of tryptophan electronic energy to 3-nitrotyrosine was studied in different conformational states of apomyoglobin (native, molten globule, unfolded) and interpreted in terms of average distances between groups of the protein chain. The experimental data show that the distance between the middle of A-helix and the N-terminus of G-helix as well as the distance between the middle of the A-helix and the C-terminus of the H-helix in the molten globule state are close to those in the native state. This is a strong argument in favor of similarity of the overall architecture of the molten globule and native states.
Keywords: apomyoglobin/direct energy transfer/fluorescence/molten globule
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
When the concept of the molten globule state was proposed for the first time (Ptitsyn, 1973
), it was suggested that this state is a productive one, i.e. can be transformed into the native state without a global rearrangement. This means that a productive intermediate must have a native-like overall architecture (folding pattern), otherwise this state would be no more than an impasse from which the folding molecule has to return before it folds correctly into the native 3D structure.
The revealing of equilibrium and kinetic intermediates in protein folding (see Ptitsyn, 1995, for references) has confirmed that the equilibrium intermediate shares a number of structural features with the kinetic intermediate (Baldwin, 1993) and, thus, can be considered as its static model. The study of the equilibrium intermediate is of fundamental interest itself since the molten globule might exist in a living cell and participate in important physiological processes (Bychkova and Ptitsyn, 1993). Moreover, it was predicted (Bychkova and Ptitsyn, 1995) and confirmed experimentally (Zhang and Peng, 1996) that a number of genetic diseases can be caused by mutations that stop protein folding at the molten globule stage.
The first detailed analysis of the structural similarity of the molten and the native globules was achieved by a combination of hydrogen exchange with 2D NMR techniques. It was established that some 
-helixes in cytochrome c (Roder et al., 1988; Jeng et al., 1990), -lactalbumin (Baum et al., 1989), apomyoglobin (Hughson et al., 1990; Jennings and Wright, 1990), lysozyme (Redford et al., 1992) and other proteins form at earlier stages of the protein folding and are relatively stable in the equilibrium molten globule state. Since these helixes form clusters in the native state of the corresponding proteins, it was suggested that these clusters might be formed also in equilibrium and kinetic intermediates. However, this assumption never has been supported by direct experimental data and nothing is known about the structure of these possible complexes.
Very little progress has been made so far in the experimental investigation of the 3D structural similarity of native and molten globule states. The molten globule cannot be crystallized and its NMR spectrum shows very broad resonance and a small dispersion of chemical shifts. Ewbank and Creighton (1991) proposed testing the similarity of these structures by the restoration of S–S bonds in a reduced molten globule state. They found that S–S bonds are formed randomly in the molten globule state of human -lactalbumin and concluded that the molten globule state in -lactalbumin is much closer to an unfolded, but collapsed, form than to an expanding native conformation. Peng and Kim (1994), however, showed that only native S–S bonds are formed by four SH groups of the single -helical subdomain of this protein in the molten globule state. On the other hand, the same group found a random formation of S–S bonds in the molten globule state of intact -lactalbumin containing both -helical and ß-structural subdomains (Wu et al., 1995). Finally, these authors concluded that in the molten globule state of -lactalbumin its helical subdomain has a native-like folding pattern whereas its ß-sheet domain is relatively unstructured. Apparently, these results are related to a fundamental difficulty in studying the conformation of large semi-flexible molecules by a chemical reaction. Clearly, the chemical reaction shows that two reacting groups can touch each other, which does not necessarily mean that their average distance is small.
Only a few physical methods provide direct information on intramolecular distances and fluorescence energy transfer is one of them, suitable for the investigation of macromolecules in solutions. The first attempt in this direction was made two decades ago and was based on the covalent binding of extrinsic fluorophores to protein functional groups. This approach, however, has not been widely used because of possible labeling heterogeneity, structural alteration resulting from the labeling per se and the effect of labeling on the protein folding (Lakowicz, 1983; Cheung, 1991). Fortunately, towards the end of the 1960s it was found that the reaction of tyrosine (Tyr) with tetranitromethane results in the formation of 3-nitrotyrosine (Lundblad, 1991). As a consequence, nitrated tyrosine, Tyr(NO2), becomes an acceptor of tryptophan (Trp) electronic excitation energy (Borkman and Phillips, 1985). This means that the decrease in Trp fluorescence in the presence of Tyr(NO2) depends on the average distance RDA between Trp and Tyr(NO2) residues. The strong dependence of an energy transfer on a donor–acceptor distance allows us to measure RDA with high accuracy. If one can nitrate the Tyr residues of the protein chain and measure the decrease in Trp fluorescence following the modification then one can measure the average distance between two definite points of the protein chain. We applied this method to compare the 3D structure of horse apomyoglobin in the native and molten globule states.
The combination of deuterium exchange and 2D-NMR revealed that the native A-, G-, and H-helixes of apomyoglobin become relatively stable at an early stage of protein folding (Jennings and Wright, 1990) and are relatively stable in the equilibrium molten globule (Hughson et al., 1990). Multi-dimensional NMR spectroscopy provides (Eliezer et al., 1998) the backbone conformation of residues in the molten globule state, which suggests the localization and probability of these helixes. However, the relative positions of the most stable helixes A, G and H in the molten globule are still questionable. We attempted to obtain this information by measuring the intramolecular energy transfer from Trp residues to Tyr(NO2). Fortunately, horse myoglobin contains Trp7 and Trp14 in the middle of the A-helix, Tyr103 at the beginning of the G-helix and Tyr146 at the end of the H-helix (Figure 1). The comparison of Trp fluorescence in non-modified and selectively and entirely modified samples permits one to evaluate the average distances from Trp to Tyr residues, i.e. from the middle of the A-helix to the beginning of the G-helix and to the end of the H-helix.
|
Recently, two distances between A- and H-helixes were obtained from time-resolved fluorescence experiments with selectively nitrated Tyr146 (Rischel et al., 1996
). The results suggest that the distances between Tyr146 and Trp7 or Trp14 are similar (20 Å) in the native state, but different (16 and 23 Å, respectively) in the molten globule state. The conclusion was drawn that `the average helix–helix orientation in the molten globule state may be non-native' (Rischel et al., 1996). However, energy transfer data based on the analysis of time-resolved Trp fluorescence should be interpreted with caution owing to the problems inherent in using multiple Trp donors. First, the Trp fluorescence decay shows non-exponential behavior even in apomyoglobins with a single Trp residue (Hochstrasser and Johnson, 1993). The nature of this behavior is poorly understood. The suggested explanations include the existence of Trp rotamers with different local environments, the mixing of 1La and 1Lb transitions either in absorption or emission and partial internal 1Lb 
1La conversion before emission. Basically, it suggests that even in a system with a single Trp residue the Trp fluorescence decay testifies to the existence of at least two potential donors (with short and long lifetimes). Second, to obtain the distance distribution for Trp7 and Trp14 one needs to know the relative contribution of each donor (Trp7 and Trp14) to the total decay. The discrimination between Trp residues in apomyoglobins still presents difficulties. For instance, our experiments with apomyoglobin mutants (with only Trp7 or only Trp14) show the similar fluorescence behavior of Trp residues in both mutants. Data obtained by benign modification of Trp7 in sperm whale apomyoglobin (Postnikova et al., 1991) suggest that the fluorescence of Trp14 is 50% stronger than that of Trp7 in the native state, whereas in the molten globule state the situation is the reverse. Clearly, an understanding of the parameters determining the Trp fluorescence is required to take full advantage of this high potential yield of information. Attempts to use model parameters to obtain the distance distribution may lead to an incorrect conclusion, which in any case needs to be confirmed. It appears that time-resolved fluorescence data with the Trp residue as a donor are capable nowadays of providing information about the average donor–acceptor separation. Apparently this information can be obtained by the simpler steady-state fluorescence method. The last concern is about the single helix–helix distance used as the only parameter of the 3D structure of the protein. To reach a conclusion about the 3D structure of the molten globule one has to obtain a set of data characterizing the mutual positions and orientations of different structural elements. This paper deals with this problem for the AGH complex in apomyoglobins. |
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Materials
Horse heart myoglobin was purchased from Sigma. The heme was removed by the method of Teale (1959) followed by gel chromatography with Pharmacia PD-10 columns containing Sephadex G-25. After heme removal, the proteins were dialyzed against water at 4°C and then lyophilized for storage. The contamination of the apoprotein by myoglobin was assessed by spectroscopic means. No significant absorption was observed in the Soret region. The homogeneity of the apomyoglobin preparations was tested by SDS–PAGE (Laemmli, 1970). A single band was always observed for the proteins investigated. The protein concentration was estimated by measuring the absorption at 280 nm with a Hewlett-Packard Model 8452A diode-array spectrophotometer. The molar extinction at 280 nm was calculated from the Trp and Tyr content by using extinction coefficients (
) of 5600 and 1250 M–1 cm–1, respectively.
The reaction to convert selectively Tyr146 into Tyr146(NO2) was similar to that previously described in detail (Rischel and Poulsen, 1995). To nitrate both Tyr residues the protein was dissolved at 6 µM in 0.1 M Tris buffer (pH 7.7) containing 6 M GuHCl. A 150 µl aliquot of a 1 wt% solution of tetranitromethane in ethanol was added to 1700 µl of protein solution. The mixture was left with stirring in the dark for 5 min and thereafter applied to the gel filtration columns equilibrated with water. The proteins were dialyzed against water at 4°C and lyophilized for storage.
The extent of Tyr modification was assessed by spectroscopic means. We used the fact (Riordan et al., 1967) that in an alkane medium 3-nitrotyrosine has an absorption maximum at 428 nm, whereas in an acidic medium the absorption maximum shifts to 360 nm with an isosbestic point at 381 nm ( = 2200 M–1 cm–1). Figure 2 illustrates the variation of the protein absorption following the modification. The absorption peak of Tyr(NO2) at 360 nm is clearly visible. The calculation of Tyr(NO2) concentration with 381 nm = 2200 M–1 cm–1 shows that the molar contents of Tyr(NO2) per protein have a value of 1 and 2 for selectively and entirely modified samples, respectively. Further, the efficiency of modification was estimated by mass spectrometry (data are not shown). We found that the mass spectrum of horse apomyoglobin corresponds to the theoretical mass of 16 951. Meanwhile, the mass spectra of selectively and entirely modified samples showed shifts of 45 and 90 units, respectively. The fraction of modified molecules was found to be about 90%. The selectivity of the modification was verified by proteolytic cleavage of the `selectively' modified protein (with only nitrated Tyr146) and mass spectrometry of the resulting fragments essentially as described (Rischel and Poulsen, 1995). Only the fragments with Tyr146 showed a mass deviation of 45 units in accord with the mass of the NO2 group.
|
The experiments were carried out in a 10 mM sodium acetate–10 mM sodium phosphate buffer mixture containing 30 mM NaCl. The native and the molten globule states were obtained by adjustment of the pH of the protein solution to 6.5 and 4.2, respectively. The unfolded conformation was obtained in 10 mM HCl (pH 2). The experiments were carried out with an ionic strength of 0.05 to stabilize the molten globule state (Goto and Fink, 1990). The protein concentration was adjusted to be less than 10 µM. All reagents were of reagent grade or better. The temperature was kept at 20 ± 0.05°C with an RTE-111 Neslab water-bath. Refractive index measurements were performed with an AO Abbé refractometer. The unmodified apomyoglobin was employed as a system characterizing the fluorescence properties of the donor (Trp) in the absence of the acceptor. Two other samples were used as samples with different amounts of the acceptor Tyr(NO2) per protein chain.
Steady-state fluorescence energy transfer experiments
Fluorescence spectra were measured with SPEX Fluorolog-2 spectrofluorimeter supplied with DM-3000 software. Excitation was set at 280 nm. Fluorescence was measured in the ratio mode and background fluorescence from a solvent blank was subtracted. Spectra were corrected for the wavelength dependence of the instrument response. Fluorescence quantum yields 
of the protein in solution were measured by comparison with twice-recrystallized N-acetyltryptophanamide (standard) as follows:
|
|
st of the standard was reported to be 0.14 in aqueous solution (Eisinger, 1969). However, the Trp absorbance in the modified samples may be evaluated by the steady-state method with only limited accuracy because of the contribution of Tyr(NO2) to the total spectrum (Figure 2
). To reduce this uncertainty we made similar estimates by fluorescence lifetime measurements, which have the advantage of being independent of concentration. The scattering of the quantum yields measured by both the steady-state and the kinetic methods was found to be <10% of the absolute value.
The efficiency of energy transfer E was calculated as the relative loss of the donor fluorescence due to the interaction with the acceptor (Förster, 1948):
|
|
D and D,A are the fluorescence quantum yields of the donor in the absence and presence of the acceptor, respectively. The energy transfer efficiency is directly related to the donor–acceptor distance RDA (Förster, 1948):
|
|
|
|
D is as defined above, n is the refractive index of the medium, N is Avogadro's number, 
is the wavelength, FD() is the fluorescence spectrum of the donor with the total intensity normalized to unity and A() is the molar extinction coefficient of the acceptor. The overlap integral <J> expresses the degree of spectral overlap between the donor emission FD() and the acceptor absorption A(). Figure 2 shows that the overlap of Trp emission (300–450 nm) with the near-UV region of Tyr(NO2) absorption is evident. Meanwhile, the spectrum of Trp fluorescence in unfolded molecules is shifted to the longer wavelengths (max = 350 nm) in comparison with that in the molten globule (max = 338 nm) and native (max = 333 nm) states. As a consequence, the spectral overlap with the Tyr(NO2) absorption (max = 360 nm) is higher for unfolded molecules, and thereby the R0 value increases (Figure 3). The overlap <J> integral was computed for each state of apomyoglobin and used for further calculation of R0.
|
The parameter <k2> represents the effect of the relative orientations of the donor and acceptor transition dipoles on the energy transfer efficiency (Förster, 1948
). For each donor–acceptor couple this parameter can be calculated as follows:
|
|
is the angle between transition moments of the donor and acceptor and ß and 
are the angles between the donor and acceptor transition moments and the donor–acceptor vector. Equation (5) shows that k2 can vary between 0 to 4. This means that the uncertainty in k2 can cause the greatest error in distance determination by energy transfer experiments and needs to be clearly estimated (see below). Nevertheless, if one can measure the energy transfer efficiency (Equation 2) and estimate the model parameters (Equation 4), then one can calculate the donor–acceptor distance by Equation 3. |
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
In order to follow the overall geometry of the apomyoglobin AGH cluster during folding, one can nitrate the tyrosine residues residing in the G- and H-helixes and measure the quenching of the fluorescence of the tryptophan residues of the A-helix by the certain acceptor Tyr(NO2). The selective nitration of Tyr146 in the apomyoglobin molecules has recently been achieved (Rischel and Poulsen, 1995
) and has been reproduced in the present work. New information about the geometry of the AGH cluster in the molten globule state can be obtained by monitoring the Trp energy transfer to acceptor Tyr(NO2) situated in the G-helix, i.e. to the nitrated Tyr103. We found that it is possible to use the distinctly different reaction pattern to produce an entirely modified sample. In fact, whereas the reaction with the native apomyoglobin results in specific modification of Tyr146, the reaction with the unfolded protein results in the modification of both Tyr103 and Tyr146. This approach allows us to obtain selectively and entirely modified samples (Figure 1) and, therefore, to measure the average distances between Trp and Tyr(NO2) located at certain positions in the protein chain or, in other words, to measure the `AG' and `AH' average separation. Direct energy transfer in apomyoglobin
The fluorescence spectra of unmodified apomyoglobin and selectively and entirely modified samples in the native, molten globule and unfolded states are shown in Figure 4. One can see that the Trp fluorescence follows the protein folding in a non-monotonic way. It is substantially stronger in the molten globule state than in the native or unfolded state. This phenomenon is usually observed for apomyoglobins (Irace et al., 1981; Postnikiva et al., 1991). The increase in Trp fluorescence at the `unfolded–molten globule' transition reflects Trp residues being hidden from the water environment in the interior of the protein molecule. The further decrease in fluorescence at the `molten globule–native' transition suggests that in the native state Trp residues located in the A-helix may be tightly packed with a quencher.
|
Figure 4
demonstrates that in unfolded apomyoglobin molecules the Trp fluorescence is almost unaffected by the presence of Tyr(NO2), either in selectively or in entirely modified samples. Meanwhile, the modification of Tyr results in a sharp decrease in the Trp fluorescence in both the native and molten globule states. We find for compact apomyoglobin molecules that the higher the Tyr(NO2) content per protein molecule, the shorter is the lifetime of Trp fluorescence (data not shown). This means that the quenching observed by the steady-state method results from the interaction of the excited Trp donor with the unexcited Tyr(NO2) acceptor and, therefore, may be interpreted in terms of non-radiative transfer of the electronic excitation energy of the Trp to the Tyr(NO2). To calculate the donor–acceptor separation we have to calculate the Trp Tyr(NO2) energy transfer efficiency for each acceptor.
We measure the quenching induced by the presence of either only Tyr146(NO2) or both Tyr103(NO2) and Tyr146(NO2). Considering the additive character of the energy transfer, the efficiency per acceptor [Tyr146(NO2) or Tyr103(NO2)] can be calculated in the following way:
|
|
|
|
.
|
|
One can see from the data in Table II
that in unfolded apomyoglobin the efficiency is very low and does not exceed 8% for the entirely modified sample. This being the case, the donor and acceptor dipoles sample a large number of random orientations and <k2> becomes 2/3, a statistical average (Förster, 1948; Blumen, 1980). With a <k2> value of 2/3, the average distances between the (Trp7–Trp14) segment and Tyr146 and/or Tyr103 in unfolded apomyoglobin were found to be more than 50 Å.
In compact states of apomyoglobin, the energy transfer efficiency was measured to be 60 and 30% for Tyr146(NO2) and Tyr103(NO2), respectively. To obtain the donor–acceptor distances one has to assume or directly measure the orientation factor <k2> for the system of interest. For proteins with known 3D structure, the <k2> value might be obtained from data from X-ray and energy transfer experiments. Here we describe one version of this approach. First, for native molecules we measure energy transfer efficiencies E and calculate RDA/R0 ratios by Equation 3. Second, the average distances <RDA> between Trp7 and/or Trp14 and each acceptor are calculated from the crystallographic data reported for horse myoglobin (Evans and Brayer, 1988, 1990; PDB 1ymb), since it is believed that native apomyoglobin resembles native holomyoglobin (Hughson et al., 1990). Moreover, it has been established that the difference in the linear dimensions of native apo- and holomyoglobin molecules does not exceed 12% (Kataoka et al., 1995). We find that the average distances from (Trp7–Trp14) to Tyr146 and Tyr103 in native molecules are 22 ± 2 and 24 ± 3 Å, respectively. The scattering shows the difference in donor–acceptor distances calculated for each Trp–Tyr couple. Third, with the RDA/R0 and <RDA> values experimentally obtained we calculate the characteristic radius <R0> for native apomyoglobin and then calculate <k2>eff to be 0.18 (see Equation 4). Another assumption to be made is that the orientation factors have similar values in native and molten globules of apomyoglobin. Fortunately, R0 depends on <k2>1/6, i.e. is a very slow function. Therefore, the uncertainty induced by this assumption should be negligible for <k2> varying from 0.18 (native) to 0.67 (unfolded). Finally, with a <k2>eff value of 0.18 and the energy transfer efficiency measured in the molten globule state we estimate the average distances from the (Trp7–Trp14) segment to Tyr146 and Tyr103 to be 22 ± 2 and 27 ± 3 Å, respectively. However, the distance of 27 ± 3 Å between Trp residues and Tyr103 might be overestimated. In fact, the energy transfer efficiency observed in the selectively modified sample is about 60%. This means that the energy transfer in the entirely modified sample may lead to complete quenching of the Trp fluorescence. The fluorescence observed in this sample (Figure 4) might come from the unmodified fraction of protein molecules.
The most valuable observation in this work is that the direct energy transfer efficiency is almost the same in the molten globule and in the native states of apomyoglobin in selectively and entirely modified samples. The major conclusion to be drawn is that the distances between the middle of the A-helix and the N-terminus of the G-helix and the distance between the middle of the A-helix and the C-terminus of the H-helix in the molten globule state of apomyoglobin are close to those in its native state. This suggests that the molten globule state of horse apomyoglobin has an AGH complex whose overall structure (in terms of `AG' and `AH' separations) is similar to that of the native state. These measurements lend strong support to the hypothesis (Ptitsyn, 1973) of the similarity of the overall architectures of the molten globule and native states. However, to complete this conclusion for apomyoglobins one needs supplementary information about the location of other structural elements. Energy transfer experiments to probe the average distance between G- and H-helixes in the molten globule state of apomyoglobin are in progress.
In summary, this paper demonstrates the feasibility of using steady-state fluorescence data to study the structure of non-native proteins. Owing to the relative simplicity of generating such data, the steady-state energy transfer method may be of widespread usefulness in studies of non-native protein structures in solution, including the structure of folding intermediates.
|
Acknowledgments |
|---|
The authors are grateful to W.A.Eaton for enormous support of this research, J.R.Knutson for help with the kinetic fluorescence experiments and L.K.Pannell for help with the mass spectrometric measurements. This study was partly supported by an INTAS grant (93-3472ext).
|
Notes |
|---|
3 To whom correspondence should be addressed, at the first address. E-mail: otcherka{at}lmmb.nci.nih.gov
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Baldwin,R.L. (1993) Curr. Opin. Struct. Biol., 3, 84–91.
Baum,J., Dobson,C.M., Evans,P.A. and Hanly,C. (1989) Biochemistry, 28, 7–13.[Medline]
Blumen,A. (1980) J. Chem. Phys., 74, 6926–6933.
Borkman,R.A. and Phillips,S.R. (1985) Exp. Eye Res., 40, 819–826.[Web of Science][Medline]
Bychkova,V.E. and Ptitsyn,O.B. (1993) Chemtracts Biochem. Mol. Biol., 4, 133–163.
Bychkova,V.E. and Ptitsyn,O.B. (1995) FEBS Lett., 359, 6–8.[Web of Science][Medline]
Cheung,H.C. (1991) In Lakowicz,J.R. (ed.), Topics in Fluorescence Spectroscopy, Vol. 2. Plenum Press, New York, pp. 127–176, and references cited therein.
Eisinger,M. (1969) J. Photochem. Photobiol., 9, 247–258.
Eliezer,D., Yao,J., Dyson,H.J. and Wright,P.E. (1998) Nature Struct. Biol., 5, 148–155.[Web of Science][Medline]
Evans,S.V. and Brayer,G.D. (1988) J. Biol. Chem., 263, 4263–4268.
Evans,S.V. and Brayer,G.D. (1990) J. Mol. Biol., 213, 885–897.[Web of Science][Medline]
Ewbank,J.J. and Creighton,T.E. (1991) Nature, 350, 518–520.[Medline]
Förster,T. (1948) Ann. Phys., 2, 55–75.
Goto, Yu. and Fink,A.L. (1990) J. Mol. Biol., 214, 803–805.[Web of Science][Medline]
Hochstrasser,R.M. and Johnson,C.K. (1993) In Kaiser,W. (ed.), Topics in Applied Physics, 2nd edn, Vol. 60. Springer, Berlin, pp. 368–372.
Hughson,F.M., Wright,P.E. and Baldwin,R.L. (1990) Science 249, 1544–1548.
Irace,G., Balestrieri,C., Parlato,G., Servillo,L. and Colonna,G. (1981) Biochemistry, 20, 792–799.[Medline]
Jeng,M.-F., Englander,S.W., Elöve,G.A., Wand,A.J. and Roder,H. (1990) Biochemistry, 29, 10433–10437.[Medline]
Jennings,P.A. and Wright,P.E. (1990) Science, 262, 892–896.
Kataoka,M., Nishii,I., Fujisawa,T., Ueki,T., Tokunaga,F. and Goto, Yu. (1995) J. Mol. Biol., 249, 215–228.[Web of Science][Medline]
Laemmli,U.K. (1970) Nature, 227, 680–685.[Medline]
Lakowicz,J.R. (1983) Principles of Fluorescence Spectroscopy. Plenum Press, New York, pp. 303–339, and references cited therein.
Lundblad,R.L. (1991) Chemical Reagents for Protein Modification. CRC Press, Boca Raton, FL, pp. 239–266, and references cited therein.
Peng,Z.-Y. and Kim,P.S. (1994) Biochemistry, 33, 2136–2141.[Medline]
Postnikova,G.V., Komarov, Yu. E. and Yumakova,E.M. (1991) Eur. J. Biochem., 198, 223–232.[Web of Science][Medline]
Ptitsyn,O.B. (1973) Dokl. Akad. Nauk SSSR, 210, 1213–1215.[Medline]
Ptitsyn,O.B. (1995) Adv. Protein Chem., 47, 83–229.[Web of Science][Medline]
Redford,S.E., Dobson,C.M. and Evans,P.A. (1992) Nature, 358, 302–307.[Medline]
Riordan,J.F., Sokolovsky,J.F. and Vallee,B.L. (1967) Biochemistry, 6, 358–361.[Medline]
Rischel,C. and Poulsen,F.M. (1995) FEBS Lett., 374, 105–109.[Web of Science][Medline]
Rischel,C., Thyberg,P., Rigler,R. and Poulsen,F.M. (1996) J. Mol. Biol., 257, 877–885.[Web of Science][Medline]
Roder,H., Elöve,G.A. and Englander,S.W. (1988) Nature, 335, 700–704.[Medline]
Teale,F.W.J. (1959) Biochim. Biophys. Acta, 35, 543.[Medline]
Wu,L.C., Peng,Z.-Y. and Kim,P.S. (1995) Nature Struct. Biol., 2, 281–285.[Web of Science][Medline]
Zhang,B. and Peng,Z.-Y. (1996) J. Biomol. Chem., 271, 28734–28737.
Received January 7, 1999; revised March 30, 1999; accepted April 16, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  O. Tcherkasskaya, V. E. Bychkova, V. N. Uversky, and A. M. Gronenborn Multisite Fluorescence in Proteins with Multiple Tryptophan Residues. APOMYOGLOBIN NATURAL VARIANTS AND SITE-DIRECTED MUTANTS J. Biol. Chem., November 10, 2000; 275(46): 36285 - 36294. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||