- Email: [email protected]
Compactness of the molten globule in comparison to unfolded states as observed by size-exclusion chromatography
BB. ELSEVIER
Biochimica et Biophysica Acta 1209 (1994) 140-143
etB ys a , ta
Radid Report
Compactness of the molten globule in comparison to unfolded states as observed by size-exclusion chromatography Eva Zerovnik a,*, Roman Jerala a, Nata~a Poklar b, Luise Kroon-Zitko a, V. Turk
a
a Department of Biochemistry and Molecular Biology, J. Stefan Institute, P.O. Box 100, 61001 Ljubljana, Slovenia b Department of Chemistry, University ofLjubljana, Murnikova 6, 61000 Ljubljana, Slovenia Received 13 June 1994
Abstract
The volumes of elution of denatured states of four proteins at high urea (8 M) and ethylurea (6 M) concentration were determined. They were found equally unfolded in both solvents. The volumes of elution of the unfolded states were compared to those of the native states and of some molten globule intermediates. It has been shown that the protein proteinase inhibitor stefin B, exhibits 'molten globule'-like properties on acid denaturation. The high salt acidic intermediate (a molten globule) as well as the native state of stefin B eluted as dimers, at 18°C. On thermal denaturation above 42°C, the intermediate dissociated into compact monomers. The more stable stefin A, which is monomeric and does not transform into molten globule intermediates under similar perturbing conditions, was always used for comparison. The states of both, stefin A and B in 50% methanol were found to be monomeric and of native-like compactness. Keywords: Size-exclusion chromatography; Urea and alkylurea denaturation; Unfolded state; Molten globule intermediate; Stefin, A and B
The nature of denatured states of proteins is of considerable interest [1-4]. The belief that the denatured states are completely unfolded and resemble random coils is less widely held. Nuclei of local structure have been detected in several denatured states using NMR, as for example, in the amino-terminal domain of 434-repressor in 7 M urea [5]. The unfolded protein chain is nevertheless characterised by an increased hydrodynamic radius. In contrast to denaturant-unfolded proteins, the so called 'molten globules' [6] are believed to be rather compact [7-9]. A prevailing view is emerging that the molten globule intermediate represents a conformationally different state of proteins with distinct thermodynamics [10-12]. This appears to be consistent with the existence of a 'more structured' molten globule [7] which maintains the overall fold and the secondary structure of the protein. Only the internal, so called 'tertiary' hydrogen bonds would be broken, allowing the blocks of Secondary structure elements to slide along each other. That the structure of molten globule intermediates is less determined is indeed evidenced by the lack of NOE signals in NMR spectra. The lack of persistent tertiary structure of molten globule
* Corresponding author. Fax: + 4 4 742 728697. 0167-4838/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved
SSDI 0 1 6 7 - 4 8 3 8 ( 9 4 ) 0 0 1 4 5 - 6
intermediates is also revealed by a higher rate of H - D exchange and reduced or absent near-UV CD spectra. Anyhow, these intermediate states demonstrate a high content of secondary structure and are compact. (See Ptitsyn [7], Christensen and Pain [8,9] or Haynie [10], for reviews.) Some authors classify the molten globule as 'compact denatured' [1,2], assuming no defined, thermodynamically relevant structure [13,14]. Due to discrepancies in the understanding of the molten globule in the literature, it seems worthwhile to study various aspects of this intermediate state and to collect more experimental data. Our aim with the present contribution is to stress the native-like compactness and association behavior of an acidic intermediate of stefin B which exhibits many characteristics of the 'molten globule' as shown by spectroscopic means (in preparation). Similarly, the compactness of the 'high methanol' state of both stefins A and B was measured. The volumes of elution of unfolded and the native states have also been measured. The hydrodynamic radius can be probed by various techniques as for example by viscosity measurements, X-ray scattering [3] or size-exclusion chromatography (SEC) [15]. By the latter, the Stokes radii of proteins can be determined. A simple inverse relationship between the
E. Zerovnik et aL / Biochimica et Biophysica Acta 1209 (1994) 140-143 Table 1 Ve on the SEC Superdex 75-HR (Pharmacia) column: native vs. denatured states Protein
Aqueous buffer K/ml
6M ethylurea
8M urea
r = (V~U/ V~EU)
Stefin A Stefin B (mon) Cytochrome c b a-ChTgen A b Thyroglobulin GuHC1
13.5+0.1 13.4 + 0.1 13.3 12.5 7.56 20.3
11.6 a 10.0 10.2 8.9 7.48 21.2
11.7 a 9,7 9.9 8.6 7.58 20.0
1.01 0.97 0.97 0.97+0.02 c
a Note that the values for stefin A, denatured in 8 M urea or, in 6 M ethylurea are similar to those for stefin B (native and intermediate states dimer, see Table 2) b marked data in Table 1 were already presented in Poklar et al. (1994) J. Protein Chem., in press. In the last column of the Table, U stands for 8 M urea and EU for 6 M ethylurea. Vc is the volume of elution. c estimated relative error.
volume of elution (V~) and the Stokes radius (R s) holds [16], namely: 1/V e = a + bR s
The technique has been used to study various conformational states of proteins [17]'and was applied in the field of protein folding [18-20]. The special type of protein intermediate conformation, the so-called 'molten globule' also has been studied by SEC [21,22]. When the conformational transition is slow in comparison to the flow through the column which is the case with the (late) molten globule intermediate to the native state transition [21], its volume of elution is a direct reflection of the state compactness by the inverse relation mentioned above. To probe the extent of unfolding, several proteins were applied to the SEC Superdex 75-HR column (Pharmacia) equilibrated in 8 M urea and 6 M ethylurea, respectively. The proteins were denatured in high concentration urea or ethylurea solutions, and equilibrated for more than 12 h. In Table 1, the volumes of elution of denaturant-unfolded states are compared to the volumes of elution of the native states. It can be deduced that for three proteins which are completely denatured in 8 M urea, namely, stefin B, cytochrome c, and a-chymotrypsinogen A, the V~ in 6 M ethylurea is a few percent higher (last column in Table 1). This could mean that the degree of unfolding is slightly less in 6 M ethylurea, i.e., that the finally denatured state
141
in ethylurea is slightly less extended and less solvent exposed. Such an observation goes along with some observations by spectroscopic means. For example, a-chymotrypsinogen A, demonstrates a lower intensity change in tryptophan fluorescence emission on denaturation and a blue shift (from 350 to 347 nm) of the A~,ax in 6 M ethylurea as compared to 8 M urea [23]. A high amount of secondary structure at concentrated alkylurea solutions of serum albumin [24], /3-1actoglobulin [25], and ~-chymotrypsinogen A [26] was observed by circular dichroism. In any event, the total number of alkylurea molecules bound to serum albumin seems comparable to urea [27] or, is even higher in 8 M methylurea than in 8 M urea. This might indicate an extended, solvent-exposed denatured state, but of course, has to be confirmed by hydrodynamic measurements, for each case separately. The nature of the denatured state may well depend on the protein composition and sequence. In Fig. 1, the denatured states of stefins A and B are shown, to be later compared to the volumes of elution of the same proteins in their native and molten globule conformations. The recombinant human stefin B is a protein which can adopt molten globule conformations by slight perturbations of pH or GuHC1 concentration [28]. CD measurements on acid denaturation of stefin B (data to be published) reveal that at around pH 3.8 and 0.5 M salt, the protein has a high amount of secondary structure (ellipticity at 208 and 220 nm are even higher than those for the native state). It also exhibits approximately one quarter of the native-state ellipticity in the near UV. The enthalpy change measured by differential scanning calorimetry at pH 3.8 and high salt (to be published) was high enough to be measured, having a melting point at around 42°C. Knowing the above properties of stefin B, it was of interest to study its chromatographic behavior (Table 2). To prevent any electrostatic interaction with the column, a high salt concentration of at least 0.25 molar was always present. The column was thermostated at 18°C. As a control, the volumes of elution of stefin A, which is the more stable protein of the two [29], were measured. The column was first equilibrated under conditions where stefin B behaved as a molten globule, that is 0.02 M acetate (pH 3.8) varying sodium chloride concentration from 0.25 to 0.5 M. The column was then equilibrated in the same buffer composition at high salt and at pH 5.3, where stefin B is native. In Table 2, the volumes of elution are 11.8 and
Table 2 Ve on the SEC Superdex 75-HR (Pharmacia) column: native vs. molten globule states Protein
0.02 M acetate pH 5.3 0.5-0.25 M NaCI
0.02 M acetate pH 3.8 0.5-0.25 M NaCI
pH 2.6, 50% methanol 0.25 M NaC1
T/°C
Stefin A / m o n Stefin B / m o r t Stefin B / d i m Stefin A / m o n Stefin B / m o n
13.2 13.4 (15%) 11.8 (85%) -
13.5-13.7 12.0-12.3 14.3 14.4
13.2 13.2 -
18 18 18 45 45
E. Zerovnik et aL / Biochimica et Biophysica Acta 1209 (1994) 140-143
142
12.0 ml, for the native-state and the molten-globule intermediate at 0.5 M salt, respectively. It is obvious (see Fig. 2) that the native state of recombinant stefin B is dimeric and that only around 15% of the protein elutes as monomer at a V~ of 13.4 ml. (The tendency to form dimers has already been detected for the tissue-isolated human stefin B [30]. With the recombinant protein, where the Cys [3] has been substituted for a Ser [31], the covalent dimer formation was prevented.) The V~ of 13.2 ml for stefin A native state, the homologous protein which does not associate or denature to molten globule intermediates under any of the conditions studied, is close to that of stefln B monomer. By CD and microcalorimetric measurements (to be published), it could be shown that the high salt, dimeric intermediate of stefin B at pH 3.8 undergoes denaturation above 42°C. Therefore, the Superdex 75-HR column was thermostated at 45°C and the volumes of elution of stefins A and B were measured. The V~ of 14.4 ml of stefin B observed at this temperature compares favourably to the V~ of stefin A which elutes at 14.3 ml (Table 2). It has been shown that stefin A remains stable at pH 5 up to 90°C [29], therefore, the somewhat higher retention of the native-state monomer of stefin A must be due to the influence of the higher temperature on the properties of the column. It can be concluded that the acidic intermediate of stefin B at high salt (which was dimeric at around room temperature) dissociated on its thermal denaturation.
Den states 2.5 MGuHC[
J
I
6MEtU
I
L..._
i
J st B
8 M Urea
j
I
t
stA
L.,
10
f
I
A
~J
pH = 5.3
st B
MO states pH = 3.8
st B
50% MeOH, pH = 2.6
st B st A
I---~ 10
~ "-15
mt Fig. 2. Volumesof elution of stefinsA and B at severalpH values; native vs. 'molten globule' and 'high methanol' states.
st B
I
,
pH=5.3
I 5
st B stA
5
Nat states
15 Ve/ml
Fig. 1. Volumes of elution of stefins A and B unfolded states in denaturing concentrations of GuHC1, ethylurea and urea (from top to bottom).
The conformation of proteins at very high methanol concentrations (more than 50%) have been studied by microcalorimetry and NMR [32,33]. An interesting conclusion was reached that the folded state of the protein can be maintained, even at 60% methanol, where no solvent ordering effect is present. The volumes of elution of steflns A and B in 50% methanol (pH 2.6 and 0.25 salt) were measured. They are given in Table 2 and in Fig. 2. Near-UV CD measurements demonstrate that the 'high methanol' states of both stefins exhibit around 60% of the native-state ellipticity value and that the far-UV CD remains at the native-state level (R.J., unpublished results). Being aware that the limits of the definition of molten globules are somewhat arbitrary, we find the 'high methanol' state cannot be termed as 'molten globule', since it has still many tertiary structure interactions (revealed by the near-UV CD). In conclusion, the acidic intermediate state of stefin B at high salt with reduced tertiary structure interactions (reflected in the substantial decrease of the near-UV CD) which is shown to be dimeric at 18°C and monomeric after
E. Zerovnik et al. /Biochimica et Biophysica Acta 1209 (1994) 140-143
its thermal denaturation at 45°C, as well as the high methanol states of both stefin A and B, are highly compact. Therefore, these non-native states cannot be treated equally as the unfolded states produced by denaturants. Minor increase in the volumes of elution in 6 M ethylurea is not comparable to the increase in the volumes of elution (i.e., decrease in Stokes' radius) observed with the molten globule and high methanol states, at least not for stefins. Our thanks go to Professor Savo Lapanje (Department of Chemistry, University of Ljubljana) for useful discussions. We wish to thank Dr. K. Lohner (Institute for Biophysics and X-ray Structure Research, Graz, Austria) and the Austrian Academy of Sciences for their help by purchasing the Superdex 75-HR column from Pharmacia. The support of the Ministry for Science and Technology of the Republic of Slovenia is greatly appreciated. References [1] Dill, K.A. and Shortle, D. (1991) Annu. Rev. Biochem. 60, 795-825. [2] Alonso, D.O.V., Dill, K.A. and Stigter, D. (1991) Biopolymers 31, 1631-1649. [3] Goldenberg, D.P. and Zhang, J.X. (1993) Proteins 15, 322-329. [4] Flanagan, J.M., Kataoka, M., Fujisawa, T. and Engelman, D.M. (1993) Biochemistry 32, 10359-10370. [5] Ned, D., Billeter, M., Wider', G. and Wiitrich, K. (1992) Science 257, 1559-1563. [6] Kuwajima, K. (1989) Proteins 6, 87-103. [7] Ptitsyn, O.B. (1992), in Protein Folding (Creighton, T., ed.), pp. 243-300, Freeman, New York. [8] Christensen, H. and Pain, R.H. (1994) in Mechanisms of Protein Folding (Pain, R.H., ed.), Oxford Univ. Series, Oxford University Press, Oxford. [9] Christensen, H. and Pain, R.H. (1991) Eur. Biophys. J. 19, 221-229. [10] Haynie, D.T. and Freire, E., (1993) Proteins 16, 115-140. [11] Xie, D., Bhakuni, V. and Freire, E., (1993) J. Mol. Biol. 232, 5-8.
143
[12] Xie, D., Bhakuni, V. and Freire, E., (1991) Biochemistry 30, 10673-10678. [13] Yutani, K., Ogasahara, K. and Kuwajima, K. (1992) J. Mol. Biol. 228, 347-350. [14] Griko, Y.V. and Privalov, P.L. (1994) J. Mol. Biol. 235, 1318-1325. [15] Ackers, G.K. (1967) J. Biol. Chem. 242, 3237-3238. [16] Davis, L.C. (1983) J. Chromatogr. Sci. 21, 214-217. [17] Martenson, R.E. (1978) J. Biol. Chem. 253, 8887-8893. [18] Shalongo, W., Ledger, R., Jagannadham, M.V. and Stellwagen, E. (1987) Biochemistry 26, 3135-3141. [19] Shalongo, W., Jagannadham, M. and Steilwagen, E. (1993) Biopolymers 33, 135-145. [20] Wada, A., Saito, Y. and Ottogushi, M. (1983) Biopolymers 22, 93-99. [21] Ptitsyn, O.B., Pain, R.H., Semisotnov G.V., Zerovnik, E. and Razgulyaev, O.I. (1990) FEBS Lett. 262, 20-24. [22] Uversky, V. (1993) Biochemistry 32, 13288-13298. [23] Poklar, N., Vesnaver, G. and Lapanje, S. (1994), J. Protein Chem. 13, 323-331. [24] Pavli~, A. and Lapanje, S. (1981) Biochim. Biophys. Acta 669, 60-64. [25] Kranjc, Z. and Lapanje, S. (1993) Int. J. Pep. Protein Res. 42, 320-325. [26] Lapanje, S. and Kranjc, Z. (1982) Biochim, Biophys. Acta 705, 1i11-1116. [27] Lapanje, S., Simi~, M. and Pavli~, A. (1981) Croat. Chem. Acta 54, 481-488. [28] ~erovnik, E., Jerala, R., Kroon-Zitko, L., Pain, R.H. and Turk, V. (1992) J. Biol. Chem. 267, 9041-9046. [29] 7~erovnik, E., Lohner, K., Jerala, R., Laggner, P. and Turk, V. (1992) Eur. J. Biochem. 210, 217-221. [30] Lenar~i~, B., Ritonja, A., Sali, A., Kotnik, M. and Turk, V. (1986) in Cysteine Proteinases and Their Inhibitors, pp. 473-487, Walter de Gruyter, Berlin. [31] Jerala, R., Trstenjak, M., Lenar~i~, B. and Turk, V. (1988) FEBS Lett. 239, 41-44. [32] Woolfson, D.N., Cooper, A., Harding, M.M., Williams, D.H. and Evans, P.A. (1993) J. Mol. Biol. 229, 502-511. [33] Harding, M.M., Williams, D.H. and Woolfson, D.N. (1991) Biochemistry 30, 3120-3128.
Biochimica et Biophysica Acta 1209 (1994) 140-143
etB ys a , ta
Radid Report
Compactness of the molten globule in comparison to unfolded states as observed by size-exclusion chromatography Eva Zerovnik a,*, Roman Jerala a, Nata~a Poklar b, Luise Kroon-Zitko a, V. Turk
a
a Department of Biochemistry and Molecular Biology, J. Stefan Institute, P.O. Box 100, 61001 Ljubljana, Slovenia b Department of Chemistry, University ofLjubljana, Murnikova 6, 61000 Ljubljana, Slovenia Received 13 June 1994
Abstract
The volumes of elution of denatured states of four proteins at high urea (8 M) and ethylurea (6 M) concentration were determined. They were found equally unfolded in both solvents. The volumes of elution of the unfolded states were compared to those of the native states and of some molten globule intermediates. It has been shown that the protein proteinase inhibitor stefin B, exhibits 'molten globule'-like properties on acid denaturation. The high salt acidic intermediate (a molten globule) as well as the native state of stefin B eluted as dimers, at 18°C. On thermal denaturation above 42°C, the intermediate dissociated into compact monomers. The more stable stefin A, which is monomeric and does not transform into molten globule intermediates under similar perturbing conditions, was always used for comparison. The states of both, stefin A and B in 50% methanol were found to be monomeric and of native-like compactness. Keywords: Size-exclusion chromatography; Urea and alkylurea denaturation; Unfolded state; Molten globule intermediate; Stefin, A and B
The nature of denatured states of proteins is of considerable interest [1-4]. The belief that the denatured states are completely unfolded and resemble random coils is less widely held. Nuclei of local structure have been detected in several denatured states using NMR, as for example, in the amino-terminal domain of 434-repressor in 7 M urea [5]. The unfolded protein chain is nevertheless characterised by an increased hydrodynamic radius. In contrast to denaturant-unfolded proteins, the so called 'molten globules' [6] are believed to be rather compact [7-9]. A prevailing view is emerging that the molten globule intermediate represents a conformationally different state of proteins with distinct thermodynamics [10-12]. This appears to be consistent with the existence of a 'more structured' molten globule [7] which maintains the overall fold and the secondary structure of the protein. Only the internal, so called 'tertiary' hydrogen bonds would be broken, allowing the blocks of Secondary structure elements to slide along each other. That the structure of molten globule intermediates is less determined is indeed evidenced by the lack of NOE signals in NMR spectra. The lack of persistent tertiary structure of molten globule
* Corresponding author. Fax: + 4 4 742 728697. 0167-4838/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved
SSDI 0 1 6 7 - 4 8 3 8 ( 9 4 ) 0 0 1 4 5 - 6
intermediates is also revealed by a higher rate of H - D exchange and reduced or absent near-UV CD spectra. Anyhow, these intermediate states demonstrate a high content of secondary structure and are compact. (See Ptitsyn [7], Christensen and Pain [8,9] or Haynie [10], for reviews.) Some authors classify the molten globule as 'compact denatured' [1,2], assuming no defined, thermodynamically relevant structure [13,14]. Due to discrepancies in the understanding of the molten globule in the literature, it seems worthwhile to study various aspects of this intermediate state and to collect more experimental data. Our aim with the present contribution is to stress the native-like compactness and association behavior of an acidic intermediate of stefin B which exhibits many characteristics of the 'molten globule' as shown by spectroscopic means (in preparation). Similarly, the compactness of the 'high methanol' state of both stefins A and B was measured. The volumes of elution of unfolded and the native states have also been measured. The hydrodynamic radius can be probed by various techniques as for example by viscosity measurements, X-ray scattering [3] or size-exclusion chromatography (SEC) [15]. By the latter, the Stokes radii of proteins can be determined. A simple inverse relationship between the
E. Zerovnik et aL / Biochimica et Biophysica Acta 1209 (1994) 140-143 Table 1 Ve on the SEC Superdex 75-HR (Pharmacia) column: native vs. denatured states Protein
Aqueous buffer K/ml
6M ethylurea
8M urea
r = (V~U/ V~EU)
Stefin A Stefin B (mon) Cytochrome c b a-ChTgen A b Thyroglobulin GuHC1
13.5+0.1 13.4 + 0.1 13.3 12.5 7.56 20.3
11.6 a 10.0 10.2 8.9 7.48 21.2
11.7 a 9,7 9.9 8.6 7.58 20.0
1.01 0.97 0.97 0.97+0.02 c
a Note that the values for stefin A, denatured in 8 M urea or, in 6 M ethylurea are similar to those for stefin B (native and intermediate states dimer, see Table 2) b marked data in Table 1 were already presented in Poklar et al. (1994) J. Protein Chem., in press. In the last column of the Table, U stands for 8 M urea and EU for 6 M ethylurea. Vc is the volume of elution. c estimated relative error.
volume of elution (V~) and the Stokes radius (R s) holds [16], namely: 1/V e = a + bR s
The technique has been used to study various conformational states of proteins [17]'and was applied in the field of protein folding [18-20]. The special type of protein intermediate conformation, the so-called 'molten globule' also has been studied by SEC [21,22]. When the conformational transition is slow in comparison to the flow through the column which is the case with the (late) molten globule intermediate to the native state transition [21], its volume of elution is a direct reflection of the state compactness by the inverse relation mentioned above. To probe the extent of unfolding, several proteins were applied to the SEC Superdex 75-HR column (Pharmacia) equilibrated in 8 M urea and 6 M ethylurea, respectively. The proteins were denatured in high concentration urea or ethylurea solutions, and equilibrated for more than 12 h. In Table 1, the volumes of elution of denaturant-unfolded states are compared to the volumes of elution of the native states. It can be deduced that for three proteins which are completely denatured in 8 M urea, namely, stefin B, cytochrome c, and a-chymotrypsinogen A, the V~ in 6 M ethylurea is a few percent higher (last column in Table 1). This could mean that the degree of unfolding is slightly less in 6 M ethylurea, i.e., that the finally denatured state
141
in ethylurea is slightly less extended and less solvent exposed. Such an observation goes along with some observations by spectroscopic means. For example, a-chymotrypsinogen A, demonstrates a lower intensity change in tryptophan fluorescence emission on denaturation and a blue shift (from 350 to 347 nm) of the A~,ax in 6 M ethylurea as compared to 8 M urea [23]. A high amount of secondary structure at concentrated alkylurea solutions of serum albumin [24], /3-1actoglobulin [25], and ~-chymotrypsinogen A [26] was observed by circular dichroism. In any event, the total number of alkylurea molecules bound to serum albumin seems comparable to urea [27] or, is even higher in 8 M methylurea than in 8 M urea. This might indicate an extended, solvent-exposed denatured state, but of course, has to be confirmed by hydrodynamic measurements, for each case separately. The nature of the denatured state may well depend on the protein composition and sequence. In Fig. 1, the denatured states of stefins A and B are shown, to be later compared to the volumes of elution of the same proteins in their native and molten globule conformations. The recombinant human stefin B is a protein which can adopt molten globule conformations by slight perturbations of pH or GuHC1 concentration [28]. CD measurements on acid denaturation of stefin B (data to be published) reveal that at around pH 3.8 and 0.5 M salt, the protein has a high amount of secondary structure (ellipticity at 208 and 220 nm are even higher than those for the native state). It also exhibits approximately one quarter of the native-state ellipticity in the near UV. The enthalpy change measured by differential scanning calorimetry at pH 3.8 and high salt (to be published) was high enough to be measured, having a melting point at around 42°C. Knowing the above properties of stefin B, it was of interest to study its chromatographic behavior (Table 2). To prevent any electrostatic interaction with the column, a high salt concentration of at least 0.25 molar was always present. The column was thermostated at 18°C. As a control, the volumes of elution of stefin A, which is the more stable protein of the two [29], were measured. The column was first equilibrated under conditions where stefin B behaved as a molten globule, that is 0.02 M acetate (pH 3.8) varying sodium chloride concentration from 0.25 to 0.5 M. The column was then equilibrated in the same buffer composition at high salt and at pH 5.3, where stefin B is native. In Table 2, the volumes of elution are 11.8 and
Table 2 Ve on the SEC Superdex 75-HR (Pharmacia) column: native vs. molten globule states Protein
0.02 M acetate pH 5.3 0.5-0.25 M NaCI
0.02 M acetate pH 3.8 0.5-0.25 M NaCI
pH 2.6, 50% methanol 0.25 M NaC1
T/°C
Stefin A / m o n Stefin B / m o r t Stefin B / d i m Stefin A / m o n Stefin B / m o n
13.2 13.4 (15%) 11.8 (85%) -
13.5-13.7 12.0-12.3 14.3 14.4
13.2 13.2 -
18 18 18 45 45
E. Zerovnik et aL / Biochimica et Biophysica Acta 1209 (1994) 140-143
142
12.0 ml, for the native-state and the molten-globule intermediate at 0.5 M salt, respectively. It is obvious (see Fig. 2) that the native state of recombinant stefin B is dimeric and that only around 15% of the protein elutes as monomer at a V~ of 13.4 ml. (The tendency to form dimers has already been detected for the tissue-isolated human stefin B [30]. With the recombinant protein, where the Cys [3] has been substituted for a Ser [31], the covalent dimer formation was prevented.) The V~ of 13.2 ml for stefin A native state, the homologous protein which does not associate or denature to molten globule intermediates under any of the conditions studied, is close to that of stefln B monomer. By CD and microcalorimetric measurements (to be published), it could be shown that the high salt, dimeric intermediate of stefin B at pH 3.8 undergoes denaturation above 42°C. Therefore, the Superdex 75-HR column was thermostated at 45°C and the volumes of elution of stefins A and B were measured. The V~ of 14.4 ml of stefin B observed at this temperature compares favourably to the V~ of stefin A which elutes at 14.3 ml (Table 2). It has been shown that stefin A remains stable at pH 5 up to 90°C [29], therefore, the somewhat higher retention of the native-state monomer of stefin A must be due to the influence of the higher temperature on the properties of the column. It can be concluded that the acidic intermediate of stefin B at high salt (which was dimeric at around room temperature) dissociated on its thermal denaturation.
Den states 2.5 MGuHC[
J
I
6MEtU
I
L..._
i
J st B
8 M Urea
j
I
t
stA
L.,
10
f
I
A
~J
pH = 5.3
st B
MO states pH = 3.8
st B
50% MeOH, pH = 2.6
st B st A
I---~ 10
~ "-15
mt Fig. 2. Volumesof elution of stefinsA and B at severalpH values; native vs. 'molten globule' and 'high methanol' states.
st B
I
,
pH=5.3
I 5
st B stA
5
Nat states
15 Ve/ml
Fig. 1. Volumes of elution of stefins A and B unfolded states in denaturing concentrations of GuHC1, ethylurea and urea (from top to bottom).
The conformation of proteins at very high methanol concentrations (more than 50%) have been studied by microcalorimetry and NMR [32,33]. An interesting conclusion was reached that the folded state of the protein can be maintained, even at 60% methanol, where no solvent ordering effect is present. The volumes of elution of steflns A and B in 50% methanol (pH 2.6 and 0.25 salt) were measured. They are given in Table 2 and in Fig. 2. Near-UV CD measurements demonstrate that the 'high methanol' states of both stefins exhibit around 60% of the native-state ellipticity value and that the far-UV CD remains at the native-state level (R.J., unpublished results). Being aware that the limits of the definition of molten globules are somewhat arbitrary, we find the 'high methanol' state cannot be termed as 'molten globule', since it has still many tertiary structure interactions (revealed by the near-UV CD). In conclusion, the acidic intermediate state of stefin B at high salt with reduced tertiary structure interactions (reflected in the substantial decrease of the near-UV CD) which is shown to be dimeric at 18°C and monomeric after
E. Zerovnik et al. /Biochimica et Biophysica Acta 1209 (1994) 140-143
its thermal denaturation at 45°C, as well as the high methanol states of both stefin A and B, are highly compact. Therefore, these non-native states cannot be treated equally as the unfolded states produced by denaturants. Minor increase in the volumes of elution in 6 M ethylurea is not comparable to the increase in the volumes of elution (i.e., decrease in Stokes' radius) observed with the molten globule and high methanol states, at least not for stefins. Our thanks go to Professor Savo Lapanje (Department of Chemistry, University of Ljubljana) for useful discussions. We wish to thank Dr. K. Lohner (Institute for Biophysics and X-ray Structure Research, Graz, Austria) and the Austrian Academy of Sciences for their help by purchasing the Superdex 75-HR column from Pharmacia. The support of the Ministry for Science and Technology of the Republic of Slovenia is greatly appreciated. References [1] Dill, K.A. and Shortle, D. (1991) Annu. Rev. Biochem. 60, 795-825. [2] Alonso, D.O.V., Dill, K.A. and Stigter, D. (1991) Biopolymers 31, 1631-1649. [3] Goldenberg, D.P. and Zhang, J.X. (1993) Proteins 15, 322-329. [4] Flanagan, J.M., Kataoka, M., Fujisawa, T. and Engelman, D.M. (1993) Biochemistry 32, 10359-10370. [5] Ned, D., Billeter, M., Wider', G. and Wiitrich, K. (1992) Science 257, 1559-1563. [6] Kuwajima, K. (1989) Proteins 6, 87-103. [7] Ptitsyn, O.B. (1992), in Protein Folding (Creighton, T., ed.), pp. 243-300, Freeman, New York. [8] Christensen, H. and Pain, R.H. (1994) in Mechanisms of Protein Folding (Pain, R.H., ed.), Oxford Univ. Series, Oxford University Press, Oxford. [9] Christensen, H. and Pain, R.H. (1991) Eur. Biophys. J. 19, 221-229. [10] Haynie, D.T. and Freire, E., (1993) Proteins 16, 115-140. [11] Xie, D., Bhakuni, V. and Freire, E., (1993) J. Mol. Biol. 232, 5-8.
143
[12] Xie, D., Bhakuni, V. and Freire, E., (1991) Biochemistry 30, 10673-10678. [13] Yutani, K., Ogasahara, K. and Kuwajima, K. (1992) J. Mol. Biol. 228, 347-350. [14] Griko, Y.V. and Privalov, P.L. (1994) J. Mol. Biol. 235, 1318-1325. [15] Ackers, G.K. (1967) J. Biol. Chem. 242, 3237-3238. [16] Davis, L.C. (1983) J. Chromatogr. Sci. 21, 214-217. [17] Martenson, R.E. (1978) J. Biol. Chem. 253, 8887-8893. [18] Shalongo, W., Ledger, R., Jagannadham, M.V. and Stellwagen, E. (1987) Biochemistry 26, 3135-3141. [19] Shalongo, W., Jagannadham, M. and Steilwagen, E. (1993) Biopolymers 33, 135-145. [20] Wada, A., Saito, Y. and Ottogushi, M. (1983) Biopolymers 22, 93-99. [21] Ptitsyn, O.B., Pain, R.H., Semisotnov G.V., Zerovnik, E. and Razgulyaev, O.I. (1990) FEBS Lett. 262, 20-24. [22] Uversky, V. (1993) Biochemistry 32, 13288-13298. [23] Poklar, N., Vesnaver, G. and Lapanje, S. (1994), J. Protein Chem. 13, 323-331. [24] Pavli~, A. and Lapanje, S. (1981) Biochim. Biophys. Acta 669, 60-64. [25] Kranjc, Z. and Lapanje, S. (1993) Int. J. Pep. Protein Res. 42, 320-325. [26] Lapanje, S. and Kranjc, Z. (1982) Biochim, Biophys. Acta 705, 1i11-1116. [27] Lapanje, S., Simi~, M. and Pavli~, A. (1981) Croat. Chem. Acta 54, 481-488. [28] ~erovnik, E., Jerala, R., Kroon-Zitko, L., Pain, R.H. and Turk, V. (1992) J. Biol. Chem. 267, 9041-9046. [29] 7~erovnik, E., Lohner, K., Jerala, R., Laggner, P. and Turk, V. (1992) Eur. J. Biochem. 210, 217-221. [30] Lenar~i~, B., Ritonja, A., Sali, A., Kotnik, M. and Turk, V. (1986) in Cysteine Proteinases and Their Inhibitors, pp. 473-487, Walter de Gruyter, Berlin. [31] Jerala, R., Trstenjak, M., Lenar~i~, B. and Turk, V. (1988) FEBS Lett. 239, 41-44. [32] Woolfson, D.N., Cooper, A., Harding, M.M., Williams, D.H. and Evans, P.A. (1993) J. Mol. Biol. 229, 502-511. [33] Harding, M.M., Williams, D.H. and Woolfson, D.N. (1991) Biochemistry 30, 3120-3128.