A distinct intermediate of RNase A is induced by sodium dodecyl sulfate at its pKa

Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

A distinct intermediate of RNase A is induced by sodium dodecyl sulfate at its pKa A.A. Moosavi-Movahedi a,∗ , M. Gharanfoli a , K. Nazari b , M. Shamsipur c , J. Chamani a , B. Hemmateenejad d , M. Alavi b , A. Shokrollahi c , M. Habibi-Rezaei e , C. Sorenson f , N. Sheibani g a

Institute of Biochemistry and Biophysics, University of Tehran, Tehran, Iran Inhibitors Department, Research Institute of Petroleum Industry, N.I.O.C., Tehran, Iran c Department of Chemistry, Razi University, Kermanshah, Iran d Department of Chemistry, Shiraz University, Shiraz, Iran e Department of Biology, Faculty of Science, University of Tehran, Tehran, Iran f Department of Pediatrics, University of Wisconsin, Madison, WI, USA Departments of Ophthalmology and Visual Sciences and Pharmacology, University of Wisconsin, Madison, WI, USA b

g

Received 5 October 2004; accepted 4 April 2005 Available online 8 June 2005

Abstract The chemical denaturation of RNase A was found to be mediated by sodium dodecyl sulfate (SDS) at various pH. The characterization of the unfolding pathway was investigated by spectrophotometry and differential scanning calorimetry (DSC), and was analyzed by multivariate curve resolution (MCR) as a chemometric method. The spectrophotometric titration curve of RNase A upon interaction with SDS indicated a distinct complex intermediate in glycine buffer at pH 3.3. This was accompanied with the catalytic activation of the enzyme and was concurrent with maximum population of the intermediate, determined by MCR. This was confirmed by the DSC profile of RNase A in the presence of SDS, indicated by two transitions in thermal unfolding. The kinetic data on the unfolding process of RNase A upon addition of SDS showed a twophase pathway under the same conditions. The intermediate appeared at low pH especially at the pKa of SDS (pH 3.3). These results provide strong evidence of the influence of low pH (around the pKa of SDS) on the existence of an intermediate upon interaction of RNase A with SDS. © 2005 Elsevier B.V. All rights reserved. Keywords: RNase A; Sodium dodecyl sulfate; Chemometry; Intermediate; Denaturation; Differential scanning calorimetry

1. Introduction Bovine pancreatic ribonuclease A (RNase A) has played a crucial role as a model system in the studies of protein structure, folding and unfolding pathways and enzyme catalysis [1,2]. In most of the reported studies, the unfolded state of the protein was achieved using changes in temperature, pH, urea and guanidine hydrochloride as well as the use of chemical Abbreviations: DSC, differential scanning calorimetry; SDS, sodium dodecyl sulfate; cCMP, cytidine 2 ,3 -cyclic phosphate; MCR, multivariate curve resolution; FA, factor analysis ∗ Corresponding author. Tel.: +98 21 640 3957; fax: +98 21 640 4680. E-mail address: [email protected] (A.A. Moosavi-Movahedi). 0927-7765/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2005.04.008

denaturants such as detergents [3–7]. A cooperative, twostate reversible unfolding transition has been observed by thermal, urea and guanidine hydrochloride induced unfolding [8]. In contrast, other investigators demonstrate a stable intermediate during the thermal denaturation [9–12]. Later studies suggested that RNase A folds and unfolds through multiple pathways determined by transition intermediates [13,14]. Anionic detergents, such as sodium dodecyl sulfate (SDS), can denature proteins at low concentrations of the order of millimolar. SDS binds to most proteins with a high affinity via interactions between the sulfate head group and the positively charged amino acid chains of the protein, on the one hand, and between the surfactant alkyl chain and the protein’s hydrophobic side chains, on the other [15]. Recently,

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

it has been demonstrated that the binding of SDS results in a conformational change in the protein to form a partially denatured state, before the occurrence of the major unfolding transition [16] or the induction of a molten globule state at very low concentration of SDS [17]. Differential scanning calorimetry (DSC) has been extensively used to study the protein thermal denaturation [8]. Analysis of DSC transitions can provide a direct measurement of small structural transitions and lead to the characterization of their thermodynamic aspects [18–20]. On the other hand, spectroscopic methods are in general simple, highly sensitive and very suitable for the study of chemical reactions in solutions. When the components involved in the chemical reaction have distinct spectral responses, their concentration can be monitored directly. However, in many cases, the spectral responses of two and sometimes even more components overlap considerably and the analysis is no longer straightforward. Nowadays, by using chemometric methods, one can analyze whole spectra, thereby utilizing all spectral information [21]. This enables chemists to monitor the complex chemical reactions by spectrophotometric techniques even if the component spectra are highly overlapped. Spectral curve deconvolution or multivariate curve resolution methods are chemometrics techniques for extraction of the pure spectra of components involved and their corresponding concentration profiles from evolutionary processes [22]. Self-modeling methods extract the concentration profiles without having any information about the shape of the spectra. Several self-modeling approaches have been developed since the pioneering work by Lawton and Sylvestre in 1971 [23]. Among these are the factor analysis-based methods such as automated spectral isolation (ASI), iterative target transformation factor analysis (ITTFA), evolving factor analysis; (EFA), iterative key set factor analysis (IKS-FA) [22], windows factor analysis; (WFA) and alternative least squares (ALS) [21–24]. Here, we provide further information regarding the intermediate state produced upon interaction of RNase A with SDS utilizing spectroscopic titration via chemometrics and calorimetric indications. This was accomplished through multivariate curve resolution as well as a kinetic model in order to know more about the mechanisms for intermediate pathway during the unfolding state.

151

the reverse rate constants. The concentration of a particular specie at each time (ci(t) ), could be followed as a function of time on the basis of relaxation time method [25] as below:  
−t ci(t) = ai exp (2) τi where i is the number of kinetic phases and a, t, and τ are amplitude, time and relaxation time, respectively. Amplitudes included the initial concentrations of the denaturant and the enzyme and the microscopic rate constants at final conditions. A physical property that has a linear dependency on concentration of specie i of protein (e.g., absorbance) could be used for monitoring the process [26] as:  
−t At = A∞ + ai exp (3) τi where At is absorbance of phase i at time t and A∞ is absorbance at t = ∞ corresponding to the end of the reaction. In the exact treatment of the three-state model, Eq. (4) for two kinetic phases can be used:     −t1 −t2 At − A∞ = a1 exp + a2 exp (4) τ1 τ2 where terms 1 and 2 denote the kinetic phases 1 and 2, respectively. Details of the theory of kinetics of protein denaturation and the expression of τ 1 and τ 2 have been previously reported [26]. In summary, the correlation between observed relaxation times and microscopic rate constants depends on the unfolding and refolding mechanisms. In a three-state model including a single intermediate, the expressions for τ 1 and τ 2 and microscopic rate constants according to Eq. (1) are given as follows [25,27]:   1 k2 (5) = k1 + k−1 τ1 k2 + k−2 1 = k2 + k−2 τ2 or

  1 KID = (k1 + k−1 ) τ1 KID + 1

(6)

(7)

where k2 [D] = k−2 [I]

2. Theories

KID =

2.1. Kinetics of denaturation

under limiting conditions, in a complete unfolding experiment, since k2  k−2 and k1  k−1 therefore, 1/τ 1 = k1 and 1/τ 2 = k2 .

The theoretical analysis of the linear three-state model for protein unfolding is as follows: k1

k2

N
I
D k−1 k−2

(8)

2.2. MCR analysis (1)

where N, I, D are native, intermediate and denatured states, k1 and k2 are the forward rate constants, k−1 and k−2 are

In this method, the spectral data, recorded at each reaction step, were collected in a data matrix (D) with m × n dimension, m being the number of spectra collected and n being the

152

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

number of data points per spectrum. If there are p absorbing components in the reaction system, the recorded absorbance at each wavelength is assumed to be the sum of contributions of all components: dj (λ) =

k


si (λ)cij

(j = 1, n)

(9)

i=1

where dj (λ) is the absorbance of sample j, cij is the concentration of component i in sample j, and n is the number of samples. The above equation can be written in matrix notation as: D = SC

(10)

where S is a (p × m) matrix of the molar absorbance and C is an (n × p) matrix containing concentration profiles. The number of components or chemical species (p) present in the system is estimated by factor analysis [23]. Briefly, factor analysis (FA) is a methodology to analyze large data sets by reducing the data to their lowest dimensionality. This is achieved through abstract matrices, which have no physical meanings. FA is normally applied to the determination of the number of principal factors of a data matrix without using supplementary chemical information. The multivariate curve resolution–alternative least squares (MCR–ALS) algorithm proposed by Tauler and coworkers [28] was used to resolve the components’ pure spectra and their corresponding concentration profiles. In this method, through an iterative procedure, C and S are calculated so that the CS product constructs the original data matrix D with the optimal fit (i.e., a minimal residual error, E) based on the two following matrix equations: S = C+ D

(11)

C = DS+

(12)

The superscript (+) denotes the pseudo-inverse of a matrix. Initial guesses of concentration profile or pure spectra are needed to start the ALS optimization. Here, evolving factor analysis (EFA) was used to obtain the first estimate of the concentration profiles of the components. In each iterative cycle of optimization, some constrains were applied as follows; non-negativity, unimodality and closure constrains were used for concentration profiles and non-negativity and unimodality were applied to the spectral profiles.

3.2. Methods 3.2.1. Spectrophotometry The RNase A samples were prepared from a stock solution by dissolving 1 mg of enzyme into 1 ml of buffer. The enzyme concentration was determined spectrophotometrically at 277 nm using a molar extinction coefficient of 9630 M−1 cm−1 [29]. Unfolding measurements were performed with appropriate concentrations of SDS solution at various pH by a Shimadzu UV–vis 3100 spectrophotometer using a quartz cell of 1 cm path length in a thermostatically controlled cell compartment and maintained at 25 ◦ C using a Haake D8 water bath. The instrumental reading was adjusted to zero with RNase A solutions in both cuvettes, and then the spectra were obtained by adding portions of surfactant solutions to one cuvette after 3 min. SDS denaturation curves were made at pH 3.3, 7.5 and 10 with the appropriate buffers. 3.2.2. Kinetics of denaturation The kinetics of denaturation was determined at 277 nm and 25 ◦ C. After instantaneous mixing of the RNase A and SDS solutions, the course of reaction was monitored by following the change in UV absorbance during 400 s. A proper computer program was prepared using the MATLAB software which enabled us to fit the experimental data (the time profile) into the two exponential terms Eq. (4) at a high level of confidence. 3.2.3. Differential scanning calorimetry Differential scanning calorimetry (DSC) was carried out on a Scal-1 microcalorimeter (Russia); the heating rate was fixed at 1 K min−1 . An additional pressure of 1.5 atm was applied during all DSC runs, in order to prevent any possible degassing of the solutions during heating. All experiments were repeated three times, and the enzyme solutions were prepared by dissolving 0.4 mg of the RNase A in 1 ml of the buffers at various pHs.

3.1. Materials

3.2.4. MCR analysis A spectral curve deconvolution program (MCR–ALS) was written in MATLAB (version 5.1, the Math Work Inc.). This program contained all the necessary calculation methods as described elsewhere [21]. The UV–vis spectra of solutions (220–450 nm) were recorded in each reaction step. Sixteen digitized absorbance spectra were recorded in 0.5 nm intervals and the data were collected in a (460 × 16) data matrix. This data matrix was subjected to the multivariate curve resolution analysis and the concentration profiles and pure spectra of the components were obtained.

Ribonuclease A (EC 3.1.27.5) was obtained from Sigma and sodium dodecyl sulfate was from Merck. Other reagents were of analytical grade. Buffers used were: glycine–HCl (50 mM, pH 2–6, I = 0.1 M), Tris (50 mM, pH 7.5, I = 0.1), and glycine–NaOH (50 mM, pH 10, I = 0.1).

3.2.5. pH–metric titration The base used for the potentiometric pH titration of the protonated form of SDS was carbonate free potassium hydroxide, which was standardized against the primary standard oven-dried potassium hydrogen phthalate. A CO2 -free

3. Materials and methods

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

atmosphere for the base was ensured throughout. The potentiometric apparatus use consisted of a 50 ml glass jacketed cell, a constant temperature bath (MLW thermostat, 25.0 ± 0.1 ◦ C), a combined glass electrode and a 10 ml capacity piston burette, the tip of which was sealed in the cap of the titration cell with a clamp and O-rings. Atmospheric CO2 was excluded from the titration cell with a stream of purified nitrogen gas. The electrodes were calibrated in the thermostat cell with standard acid–base to read pH directly. In all experiments, a 1.0 × 10−4 M solution of the protonated SDS was used. The ionic strength was adjusted to 0.05 M with KNO3 . The surfactant protonation constant and the corresponding distribution curves were evaluated using the program BEST method [30]. In all experiments, the utilized concentration of SDS was less than the critical micelle concentration (cmc) (equal to 8.3 mM), and the molecular weight of RNase A was 13,700 [31].

4. Results Fig. 1 shows the influence of pH on the extinction coefficient (ε277 ) of RNase A upon interaction with sodium ndodecyl sulfate at pH 3.3, 7.5, and 10. The curve at pH 3.3 indicates a two-stage interaction; the intermediate inflection point was observed at a concentration of 0.09 mM SDS. The concentration of RNase A solution was 10 ␮M and the molar ratio of [SDS]/[RNase A] = 9. As seen in the Figure there is no change at pH 10. Fig. 2 shows typical kinetic progress curves for denaturation of RNase A upon addition of SDS at a molar ratio of [SDS]/[RNase A] = 9 and pH 3.3 and 7.5 at 25 ◦ C (the concentrations of RNase A and SDS are 30 ␮M and 0.27 mM, respectively). Knowing the number of exponential terms related to the experimental data according to Eq. (3), one can recognize the number of phases present in the denatu-

Fig. 1. Extinction coefficient (ε277) vs. SDS concentration: SDS-induced unfolding curves for RNase A at 25 ◦ C in glycine buffer pH 3.3 (); in Tris–EDTA buffer pH 7.5 (䊉); and in glycine buffer pH 10 (). The concentration of RNase A was 10 ␮M.

153

Fig. 2. Kinetic progress curves 1 and 2 (solid curves) for unfolding of RNase A in the presence of SDS with molar ratio of [SDS]/[RNase A] = 9 at pH 3.3 and pH 7.5, respectively. The dash curves are the fit of data points using a three-state model and two-state model for pH 3.3 and 7.5, respectively. The kinetic progress curves were constructed by measuring the change in UV absorbance at 277 nm over time.

ration path. In order to reach satisfactory results during the experimental and model data comparison, a proper computer program was prepared using the MATLAB software. This enabled us to fit the experimental data to Eq. (3) with a high level of confidence. The program, using the non-linear least squares and the Newton method, examines different numbers of exponential terms based on Eq. (3) in order to minimize the function in relation to the corresponding parameters (ai and τ i ). All obtained parameters from the fitting of kinetic progress curves at pH 3.3 and pH 7.5 are summarized in Table 1. The results revealed that the denaturation of RNase A by SDS at pH 7.5 follows a two-state mechanism, whereas the result of cited process at pH 3.3 obeys a two-term exponential equation or two-phase kinetic diagram (as a three-state mechanism) based on Eq. (4). The corresponding mechanistic pathway at pH 3.3 is given in Eq. (1). A multivariate curve resolution analysis was also conducted to specify the denaturation species of RNase A induced by SDS. To do so, the titration of the RNase A by SDS was carried out at various pH. The UV–vis absorbance spectra of the solutions were recorded at different titration steps. Factor analysis was applied to the absorbance data to determine the number of coexisting components in the reaction mixtures. The variations of the logarithm of eigen-values as a function of the number of factors are shown in Fig. 3 for pH 3.3 and 7.5. As observed, there is a large separation between eigen-values number 3 and 4 for pH 3.3, which indicate the presence of three factors at this pH; while at pH 7.5 a large separation is observed between the eigen-values number 2 and 3 indicating the presence of two factors at this pH. The mathematical factors obtained at pH 3.3 and 7.5 can be attributed to the respective three and two different forms of the RNase A in the presence of different amounts of the added SDS. Multivariate curve resolution analysis was then applied to find the population of the different RNase A species at different concentrations of SDS and their pure absorbance spectra.

154

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

Table 1 The parameters deduced from kinetic data fitting at SDS with molar ratio of [SDS]/[RNase A] = 9 and different pHs pH

A∞

A1

A2

τ 1 (s)

τ 2 (s)

3.3 7.5

1.15 ± 0.22 0.23 ± 0.05

−0.06 ± 0.01 −0.05 ± 0.01

−0.3 ± 0.02 –

25.44 ± 1.4 151 ± 1.6

208 ± 0.8 –

The results obtained at pH 3.3 are shown in Fig. 4. Fig. 4a shows the population fraction of different components during unfolding of RNase A at the SDS concentration range of 0–0.12 mM. In this figure, FN , FI , and FD indicate the population fractions of native, intermediate and denatured states, respectively. As is obvious, FN is about 1 at low concentration of SDS (0–0.03 mM) and FI is the predominant component in the concentration range of 0.04–0.09 mM SDS, and FD starts at SDS concentration greater than 0.06 mM. The pure absorbance spectra of the native, denatured and intermediate species are represented in Fig. 4b. As observed, the three species have relatively the same absorbance spectra and no significant wavelength shift is observed. The major difference is due to the molar absorptivity of the native, denatured and intermediate species. It is important to note that the chemometric analysis of the UV-spectra for cited interactions at pH 7.5 and 10 were also carried out, but the populated fraction of FI was not observed in these cases. Fig. 5 shows DSC profiles for RNase A in the absence and presence of SDS with molar ratio of [SDS]/[RNase A] = 9 (the concentration of RNase A and SDS were 30 ␮M, 0.27 mM respectively). Fig. 5a indicates a single DSC thermal profile of RNase A-SDS complex at pH 7.5 with a Tm of 337 K. The inset shows that the thermal transition of RNase A at pH 7.5 in the absence of SDS possesses only one peak with Tm of (336 K). However, the thermal unfolding of RNase A in the presence of SDS at pH 3.3 shows two distinct transitions with different melting temperatures, Tm1 = 341 K and Tm2 = 312 K (Fig. 5b). The inset in Fig. 5b shows only one thermal transition of RNase A in the absence of SDS at pH 3.3. Variation of

Cp in the unfolding process of RNase A- SDS complex at pH 2 and10 was not observed (data not shown). Fig. 5c shows a comparison of thermal profiles for RNase A in the presence of 0.3 mM SDS at different pH (pH 2.9, 3.3, 4). Table 2 indicates the values of H for intermediate transitions at the cited interactions at various pH. Also the value of H for the major transitions were 309.1 and 257 kJ/mol for pH 7.5 and 3.3, respectively (see Table 2). The specific activities of RNase A in the absence and presence of SDS at different pH are shown in Tables 3 and 4. The enzyme activity was also measured at 325 K (between the two transitions in the DSC curve, see Fig. 5b) and was found to be negligible. The species distribution diagram against pH at 25 ◦ C with a pKa value of about 3.3 is shown at Fig. 6.

Fig. 3. Plots of the logarithm of the eigen-value of absorbance spectra in the titration of RNase A with SDS at pH 3.3 and 7.5 as a function of the number of factors.

Fig. 4. Concentration profiles (a) and pure spectra (b) of the native (N), intermediate (I) and denatured (D) species of RNase A at pH 3.3 obtained by multivariate curve resolution analysis of the absorbance data.

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

155

Fig. 5. (a) DSC thermal scan of RNase A at pH 7.5, in the presence of SDS with molar ratio of [SDS]/[RNase A] = 9 (inset: in the absence of SDS); (b) DSC thermal scan of RNase A at pH 3.3, in the presence of SDS with molar ratio of [SDS]/[RNase A] = 9 (inset: in the absence of SDS); (c) DSC thermal scans of RNase A at various pH (2.9, 3.3, 4) in the presence of 0.3 mM SDS. Table 2 The calorimetric enthalpy (H) for major and minor intermediate transitions of RNase A in the presence of 0.3 mM SDS at various pH pH H (kJ/mol) for minor intermediate transition H (kJ/mol) for major transition

2 0 –

2.9 8.9 –

5. Discussion The denaturation of RNase A by SDS was investigated at pH 3.3, 7.5 and 10, and found to be markedly influenced by the solution pH. The DSC experiments revealed that the Table 3 The activity percent of RNase A in the absence and presence of SDS with molar ratio of [SDS]/[RNase A] = 9 upon reaction with cCMP [SDS] (mM)

pH

a Activity

0

3.3 7.5 10.0

35 100 5.3

3.3 7.5 10.0

186 103 1.7

0.09

3.3 77.25 257

4 21.42 –

6 0 –

7.5 0 309.1

unfolding of RNase A in the absence and presence of SDS at pH 7.5 is a reversible two-state process (Fig. 5a). Furthermore, the extinction coefficient curve (Fig. 1) was sigmoidal at pH 7.5, cited in literature previously as showing a two-state mechanism for the denaturation of RNase A by SDS at this pH [32,33]. However, the results at pH 3.3 were quite different from those obtained at pH 7.5 and 10. The extinction coefficient curve at this low pH indicated two distinct stages for RNase A upon interaction with SDS. The correspond-

(%)

Initial velocities were determined by spectrophotometric assay at a series of pH values and 25 ◦ C according to Crook method [38]. a These data are relative to the case of native form (pH 7.5 and [SDS] = 0).

Table 4 The activity percent of RNase A in the presence of [SDS] upon reaction with cCMP at pH 3.3 [SDS] (mM)

Activity (%)

0 0.006 0.09 0.15 0.2

23 74 100 48 0

156

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

Fig. 6. pH-metric plot for species distribution of SDS at various pH, at 25 ◦ C and ionic strength 0.05 M. pKa is 3.3.

ing interaction shows an alteration at 0.09 mM SDS and a steep increase above 0.09 mM SDS. The formation of RNase A–SDS complexes is related to the first stage which induced the activation of the enzyme and the second stage is related to the denatured state. These results indicate the existence of a stable intermediate in the interaction pathway of RNase A during its interaction with SDS as indicated in Eq. (1). It should be mentioned that no intermediate was seen at pH 4.5 and 5.0 and no change of absorptivity was seen at pH 2.2. This is because the positive surface charge on the protein increases at low pH, thereby inducing the SDS–RNase A complexes, with an optimum pH of 3.3. The conformation of RNase A is changed at pH less than 2.9 and this condition is not suitable for complex formation with SDS. At pH greater than 4 the positive surface charge on the protein decreases, diminishing the interaction with anionic surfactant such as SDS. The fitting of kinetic progress curves (Fig. 2) indicates three-state and two-state denaturation mechanisms for RNase A upon addition of SDS at pH 3.3 and 7.5, respectively. The kinetics of denaturation at pH 7.5 shows one kinetic phase, whereas at pH 3.3 the kinetic map can be represented by the sum of two exponential terms, each term describing an independent kinetic phase. All kinetic parameters including τ 1 , τ 2 have been deduced from the exponential fit of the data to a kinetic mechanism for the pathway intermediate in which all three species (N, I, D) are populated (see Table 1). Our observation supports the fact that the native RNase A (N), upon interaction with SDS at pH 3.3, was rapidly converted to an intermediate state (I), with slow formation of the final unfolded state (D). This biphasic mechanism is quite consistent with the literature reports for RNase A denaturation by other denaturants [13,14]. It should be mentioned that data fitting was done at all of the investigated conditions of SDS concentration and pH. Only the data at pH 7.5 and 3.3 are shown since it was under these conditions that the distinct behavior indicating a two-state model (pH 7.5) and a biphasic mechanism (pH 3.3) accompanied by the formation of intermediate complexes was seen. Since this mechanism was deduced from the fitting of the kinetic data, the next step

was to confirm and characterize the species that appeared based on the chemometrics, DSC and specific activity data. To obtain the required information, we needed to find conditions that stabilize the intermediate and maximize its concentration [I]. The chemometric analysis indicated that SDS in a concentration range of 0.06–0.09 mM should maximize the concentration of intermediate I. In this work, the MCR method was used as a chemometric technique to study the denaturation kinetics of RNase A upon interaction with SDS. The chemometric analysis of the UV–vis spectra of RNase A solutions during the SDSinduced unfolding showed that three chemical components co-exist in the process. The first of them is the native molecule (N). This component predominates up to a SDS concentration of about 0.03 mM and then decreases with increasing SDS concentration. The second specie is an intermediate I, the concentration of which is maximal between 0.06–0.09 mM SDS. The third component is an unfolded molecule which begins to from at 0.06 mM SDS and increases with increasing SDS concentration. It is important to note that the population profile of the unfolding process (Fig. 4) supports the existence of the intermediate specie (I) during the unfolding process. The first product of the interaction of RNase A with SDS is intermediate I, which then converts to the unfolded specie due to denaturation of the intermediate molecule. These results confirm our previous data shown in Figs. 1 and 2. Differential scanning calorimetry (DSC) was also used to obtain direct measurements of the fine structural transitions showing the thermal denaturation of RNase A in the presence of SDS at various pH (2–7.5). The RNase A undergoes a biphasic denaturation in the presence of SDS at pH 3.3, and two distinct peaks were observed in the corresponding DSC profile with Tm1 , 312 K and Tm2 , 342 K. At pH 7.5, on the other hand, the DSC profile showed only a single peak with a Tm at 336 K. The DSC measurements thus demonstrated the distinct RNase A–SDS complex as an intermediate at pH 3.3 and confirm the proposed kinetic mechanism with the pathway intermediate. To confirm the effect of pH on the dissociation of SDS in the thermal unfolding process of RNase A, we compared the DSC profiles of the RNase A–SDS mixture at various pH (2.9, 3.3, 4) (see Fig. 5c). The formation of the intermediate increases from pH 2.9, maximizes at pH 3.3 (the pKa of SDS) and then diminishes at pH 4. The intermediate was not seen at pH 6 and 7.5. We calculated the enthalpy (H, the area under the thermal profile) as an extensive parameter indicating the amount of intermediate which is maximized at pH 3.3 (pKa of SDS). The total enthalpies including major and minor intermediate transitions for RNase A–SDS complexes at pH 3.3 (total enthalpy 334.25 kJ per mol) indicate a higher value compared to pH 7.5 (total enthalpy 309.1 kJ per mol). This suggests that the more stable intermediate formed at pH 3.3 (see Table 2). To investigate the characteristics of the intermediate complexes, we carried out an enzymatic assay with RNase A and RNase A–SDS complexes at various pH. The enzyme activity of RNase A as measured by the hydrolysis of cyti-

A.A. Moosavi-Movahedi et al. / Colloids and Surfaces B: Biointerfaces 43 (2005) 150–157

dine 2 ,3 -cyclic phosphate (cCMP) is highly pH dependent [34]. It should be mentioned that the enzymatic activity of RNase–SDS complexes at pH 3.3 and 325 K was negligible, indicating a disordered structure of the protein at this temperature. Our results showed that the activity of the RNase–SDS complexes at pH 3.3 and molar ratio of SDS]/[RNase A] = 9 was higher than the activity of RNase A alone. It is important to note that SDS activates some enzymes at low concentration [35,36] since under these conditions electrostatic interactions with the protein predominate over hydrophobic interactions [17]. As shown above, the unfolding process of RNase A by SDS goes through a pathway involving an intermediate, indicated by spectrophotometric, DSC and chemometric analysis. The reason for the increased intermediate population and activity of the intermediate at pH 3.3 relative to pH 7.5 is because of the presence of DSH (positive net charge) and DS− (negative net charge) in equilibrium (see Fig. 6). Due to the positive net charge at low pH, the contribution of electrostatic interactions between SDS at low concentration and protein is predominant (i.e., the electrostatic interactions tend to occur predominantly between the positive charge on the protein surface and the head group of SDS at low concentration). Due to the negative net charge on the protein at high pH, the contribution of electrostatic interactions is decreased and perhaps the hydrophobic moiety is increased [37]. Here the data revealed that the intermediate is induced at low pH between 2.9 and 4, especially at the pKa of SDS (pH 3.3) (see Fig. 6). This could be an important factor for the manifestation of RNase A intermediate in the presence of SDS at pKa accompanied by activation of the enzyme (see Table 2). 6. Conclusion These results indicate that the presence of RNase A intermediate is induced in the presence of SDS at low pH (2.9–4). The main factor for induction of intermediate is a strong electrostatic contribution through RNase A–SDS complexes, especially at pH 3.3, the pKa of SDS. Therefore, equilibrium between DSH and DS− in SDS dissociation may play a major role in the stabilization of RNase A intermediate at this pH. Acknowledgment The financial support of Research Council of the University of Tehran and Iran National Science Foundation (INSF) are gratefully acknowledged.

157

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38]

J.L. Neira, M. Rico, Fold Des. 2 (1997) 1. R.T. Rains, Chem. Rev. 98 (1998) 1045. C. Tanford, Adv. Protein Chem. 23 (1968) 121. C. Tanford, Adv. Protein Chem. 24 (1970) 1. F. Ahmad, C.C. Bigelow, J. Mol. Biol. 131 (1979) 607. C.N. Pace, D.V. Laurents, J.A. Thamson, Biochemistry 29 (1990) 2564. A.A. Moosavi-Movahedi, G.A. Naderi, B. Farzami, Thermochim. Acta 239 (1994) 61. P.L. Privalov, Annu. Rev. Biophys. Biophys. Chem. 18 (1989) 47. O.W. Howarth, L.Y. Lian, Biochemistry 23 (1984) 3515. J. Zhang, X.D. Peng, A. Jones, J. Jones, Biochemistry 34 (1995) 8631. Y. Tamura, K. Gekko, Biochemistry 34 (1995) 1878. S.D. Stelea, P. Pancoska, A.S. Benight, T.A. Keiderling, Protein Sci. 10 (2001) 970. Y.J. Li, D.M. Rothwarf, H.A. Schreraga, Natl. Struct. Biol. 2 (1995) 489. Y. Yong-Bin, J. Bo, Z. RI-Qing, Z. Hai-Meng, Protein Sci. 10 (2001) 321. A.A. Moosavi-Movahedi, Encyclopedia of Surface and Colloid Science, Marcel Dekker Inc., New York, 2002, p. 5344. D.E. Otzen, Biophys. J. 83 (2002) 2219. A.A. Moosavi-Movahedi, J. Chamani, Y. Goto, G.H. Hakimelahi, J. Biochem. 133 (2003) 93. M.L. Galisteo, P.L. Mateo, J.M. Sanchez-Ruiz, Biochemistry 30 (1990) 2061. J.F. Brandts, L. Lung-Nan, Biochemistry 29 (1990) 6927. M. Gharanfoli, A.A. Moosavi-Movahedi, S. Safarian, J. Chamani, A.A. Saboury, Thermochim. Acta 411 (2004) 37. M. Shamsipure, B. Hemmateenejad, M. Akhond, K. Javidnia, R. Miri, J. Pharm. Biomed. Anal. 31 (2003) 1013. B.K. Lavine, Anal. Chem. 72 (2000) 91. W.H. Lawton, E.A. Sylvestre, Technometric 13 (1971) 617. M. Kubista, R. Sjoback, B. Albinson, Anal. Chem. 65 (1993) 994. P.J. Hagerman, R.L. Baldwin, Biochemistry 15 (1976) 1462. H. Utiyama, R.L. Baldwin, Methods Enzymol. 131 (1986) 51. B.T. Nall, J.R. Garel, R.L. Baldwin, J. Mol. Biol. 118 (1978) 317. R. Tauler, B.R. Kowalski, S. Flemming, Anal. Chem. 65 (1993) 2040. G. Taborsky, J. Biol. Chem. 234 (1959) 2915. A.E. Martell, R.J. Motekaitis, Determination and Use of Stability Constants, second ed., VCH, New York, 1992. C.H. Hirs, S. Moore, W.H. Stein, J. Biol. Chem. 219 (1956) 623. M. Yao, D.W. Bolen, Biochemistry 34 (1995) 3771. P.L. Privalov, N.N. Khechinashvili, J. Mol. Biol. 86 (1974) 665. M.R. Eftink, R.L. Biltonen, Biochemistry 22 (1983) 5123. M.N. Jones, A. Finn, A.A. Moosavi-Movahedi, Biochim. Biophys. Acta 913 (1987) 395. A.A. Moosavi-Movahedi, IUPAC 66 (1994) 71. K. Nazari, A.A. Saboury, A.A. Moosavi-Movahedi, Thermochim. Acta 302 (1997) 131. E.M. Crook, A.P. Mathias, B.R. Rabin, Biochem. J. 74 (1960) 234.