S - 0 * S SAXS study of crotapotin at low pH Jose R. Beltran Abrego, * Aldo F. Craievich,* Yvonne P. Mascarenhas,§ and Carlos J. Laurell *Departamento de Fisica, Instituto de Biociencias, Letras e Ciencias Exatas/UNESP Caixa Postal 136-15055 SAo Jos6 do Rio Preto, SP, Brasil; *Laborat6rio Nacional de Luz Sincrotron/CNPq Campinas, SP, Brasil; and $1nstituto de Fisica/USP SAo Paulo, SP, Brasil; S1nstituto de Fisicia e Qu'imica de S&o Carlos/USP SAo Carlos, SP, Brasil; and I"Departamento de Bioquirmica, Faculdade de Medicina de RibeirAo Preto, USP 14100 Ribeirco Preto, SP, Brasil ABSTRACT The structure of crotapotin, a protein extracted from the venom of the Crotalus durissus terrificus, in solution at pH = 1.5, was studied by SAXS. The experimental results yield structural parameter values of the molecular radius of gyration Rg = 13.6 A, volume v = 16.2 X 103 A3 and maximal dimension Dmax = 46 A. The distance distribution function deduced from the scattering measurements is consistent with an overall molecular shape of an oblate ellipsoid of revolution with assymetry parameter v = 0.45. 1. INTRODUCTION Crotapotin is an acidic subunit of the crotoxin complex extracted from the venom ofthe Crotalus durissus terrif- icus. Such a protein is highly important to enhance the toxicity of crotalus phospholipase A. It is a protein with molecular weight of 9,000 D and isoelectric point ofPKi 3.4. It consists of three polypeptide chains intercon- nected by sulfide bridges. The chains A, B, and C, consist of 40, 34, and 14 aminoacid residues, with molecular weights of 4,300, 3,700 and 1,600 D, respectively (Breithaupt et al., 1974). This paper aims at the study of the structure of crota- potin in acidic solutions, at pH = 1.5, by the small-angle x-ray scattering (SAXS) technique. A low value for the pH was used in order to compare the present results, concerning the crotapotin, with those corresponding to the crotoxin molecule which is not soluble at medium and high pH. 2. MATERIALS AND METHODS 2.1. Samples Purified crotapotin samples were prepared as follows: -400 mg of crude venom ofthe crotalus durissus terrificus were dissolved in 3.0 ml ofammonium formate 0.05 M, pH 3.5, and then filtered through a 4 x 45 cm Sephadex G-75 column, buffered at a flow rate of 0.8 ml/min (Laure, 1975). Fig. 1 shows the chromatographic profile of filtering of the crude venom. The five fractions A, B, C, D, and E are 5, 12, 49, 25, and 3% (wt/wt), respectively. Fraction C, which corresponds to -50% of the whole venom, was identified as being the crotoxin. Three pools of crotoxin were refiltered under the same conditions. These pools correspond to the dashed line in Fig. 1. The horizontal bar indicates the mixed fraction of crotoxin corresponding to peak C. Fig. 2 represents the chromatographic profile obtained by fractioning the crotoxin in crotapotin and phospholipase A in ionic exchange res- ins SP-Sephadex C-25 in the presence of guanidine chloride 1 M, pH 4.3. Fractions I, II, and III correspond to crotapotin, to a protein with phospholipatic activity and to phospholipase A, respectively. These Address correspondence to Jose R. Beltran Abrego. three fractions were eluted using 1.0, 2.0, and 3.0 M, respectively, of guanidine chloride at pH = 3.0, a flux of 30 ml/h and fractions of 5.0 ml/tube in a discontinuous gradient. Fraction I was rechromatized under the same conditions, desalinated, dialyzed, and lyophilized. Elec- trophoresis of the purified crotapotin in a polyacrilamide gel, contain- ing 12% acrilamide, leads to the same polypeptide as that described by Breithaulp et al. ( 1974). 2.2. SAXS measurements Protein solutions of several concentrations were placed in Lindemann glass capillary tubes of 1.0 mm diameter (Marck-Korchen, Berlin) for SAXS measurements. The solvent was a solution of30% formic acid in water, at pH 1.5. First, a 60 mg/ml stock solution of 3.0 mg of protein in 50 X 10-6 ml solvent was prepared. The remaining solutions of different concentrations (50, 40, 30, and 20 mg/ml) were obtained by increasing dilution steps. The SAXS experiments were performed using a Rigaku small-angle goniometer and a classic X-ray generator with a Cu tube at 40 Kv and 30 mA. The scattered x-ray intensity was measured by means of a Tennelec PSD-1000 position-sensitive detector located at 500 mm from the sample and a Tracor Northern TN- 17 10 multichannel ana- lyzer in a 256 channel mode. The x-ray radiation from the Cu tube was monochromatized using a Ni filter. Three vertical collimation slits were employed. Two of them (0.5 x 10 mm and 0.3 x 10 mm) define the incoming beam having a linear cross-section. The third set was located close to the sample, reducing the parasitic scattering from the collimating slits. A vacuum beam path was placed between the sample and the detector in order to reduce scattering by the air. A linear 2.8 mm wide beam-stopper was inserted close to the detector inside the vacuum path tube. Typical counting times of 5 h were used for recording the scattering curves corresponding to each concentration. The maximum total num- ber of photons was 3 x 104/channel at the lower angles for the most concentrated solution. Lower photon numbers were counted at higher angles and lower concentrations. The experimental SAXS intensity for each concentration was deter- mined as a function of the modulus of the scattering vector, h, defined as h = 4-r sin 0/ X, where ( is half the scattering angle and X the x-ray wavelength used in the experiments (X = 1.54 A). The scattering inten- sity was measured in the h-range between 2.0 x 10'2 A-' and 30 x 10-2 A-. The parasitic scattering produced by the solvent, capillary tube, air, and slits was subtracted from the total scattering curves corresponding to all the concentrations. The resulting scattering curves, were plotted in Fig. 3a, were corrected to normalize them equivalent sample absorp- tion, concentration and thickness. Appropriate extrapolations for each scattering angle, as described by I. Piltz ( 1982), on the set offour SAXS 560 0006-3495/93/02/560/05 $2.00 Biophys. J. © Biophysical Biophys. J. FBiophysical Society Volume 64 February 1993 560-564 560 0006-3495/93/02/560/05 $2.00 a _4 20 40 60 Number of tubes 80 TABLE 1 Structural parameters of crotapotin Rg (A) V(103 x A3) Dm. (A) 13.6±0.4 16.2±0.4 46±2 3. RESULTS AND DISCUSSION 3.1. Scattering curves and molecular parameters The desmeared and smoothed scattering curve corre- sponding to zero concentration is plotted in Fig. 4 a. The radius of gyration, Rg, of the protein was determined using Guinier's law from the slope y ofthe linear region of the log I(h) versus h' curve at low h (Guinier and Fournet, 1955): FIGURE 1 Chromatographic peaks obtained from the crude venom of the Brazilian snake (crotalus durissus terrificus) when filtered through a 4 x 45 cm, Sephadex G-75 column. Peak C corresponds to the cro- toxin fraction and the dashed peak to refiltered crotoxin pools. experimental curves corresponding to different concentrations, yielded the IN(h) curves at infinite dilution (or zero concentration), which is shown in Fig. 3 b. This extrapolation procedure allows the SAXS inten- sity function to be obtained free from intermolecular interference ef- fects. The scattering curves, corresponding to the different crotapotin con- centrations, were analyzed after a mathematical desmearing process to eliminate the geometric effects due to the finite linear cross-section using the ITP program (Glatter, 1982). The described theory was ap- plied to the desmeared and smoothed scattering curves for the determi- nation of the structural parameters of the studied molecule. E 40 iM 2M 3M .000- c44CNI30 00 -O2.000- co I(h) = I(O)e-1/3Rh2 and Rg(A) = 2.628 y (A2). (1) In order to determine other structural parameters, the protein solution was assumed to be a two-electronic- density system composed ofmonodisperse particles, i.e., protein molecules of space-constant electron density, and of equal size, immersed in a homogeneous solvent. Under this assumption, the SAXS intensity was ex- pected to have an assymptotic behavior described by Porod's law (Guinier and Fournet, 1955): I(h) = a/h4, (2) where a is a constant. For real systems, Eq. 2 is not gener- ally obeyed because of a usually weak contribution to SAXS from short range electronic density fluctuations in the protein molecule. This provides a constant contribu- tion to the SAXS intensity. Under this condition, the expected assymptotic behavior of the scattered intensity by two-phase systems, including electronic short range density fluctuations, can be written as follows: IN(h)h' = a + ph4(h --c4). (3) This behavior was observed experimentally as shown in Fig. 4 b. In order to determine the intensity function I(h) associated with crotapotin and its solvent, described as a two-density system, the constant contribution from the density fluctuations was subtracted: I(h) = IN(h) - (4) 10 20 Number of tubes being deduced from the plot shown in Fig. 4 b. The volume of the molecule was determined by using the following equation (Guinier and Fournet, 1955): V- 27r2I(0) Q (5) SAXS Study of Crotapotin at Low pH 561 3.000- E 0 c- 0 o .0 0 Cl) 2.000 - 1.000 - I I'| I I I I I I I I I I I D I I I I I I I I AI I I I~~~~ FIGURE 2 Chromatographic peaks obtained from crotoxin when fil- tered through a SP-Sephadex C-25 (2.5 x 40 cm) column. Peaks I and III correspond to crotapotin and phospholipase A, respectively. Abrego et al. SAXS Study of Crotapotin at Low pH 561 1000 800 1-c 600 400 200 0 200 -c3 150 100 50 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 FIGURE 3 (a) Experimental SAXS intensities before normalization and correction for collimation effects. Protein concentrations are (A) 20, (m) 30, (0) 40, (0) 50, and (LOI) 60 mg/ ml. (b) SAXS intensities extrapolated to zero concentration. The intensities corresponding to fixed h values were extrapolated after normalization of each scattering curve to equivalent concentration and absorption. where I(O) is the value of the extrapolated scattered in- tensity at h = 0 and Q is the "invariant" parameter given by: Q = I(h)h2 dh. (6) The value ofQ was calculated numerically from a mini- mum value of h, hm, up to the maximum value, hM, for which the SAXS intensity was measured. For h < hm and h > hM, Guinier and Porod behaviors of the scattering function (Eqs. 1 and 2, respectively) were assumed. The distance distribution function of the molecule, P(r), was determined from the experimental SAXS in- tensity as follows (Glatter and Kratky, 1982): P(r) = 22 I(h)hr sin (hr) dh. 27r2 (7) P(r) depends on the molecular shape and defines an additional structure parameter, the maximal molecular dimension Dm,,. The experimental distance distribution function, was derived from the scattering intensity IN(h) by applying Eq. 7 and using the ITP8 1 program (Glatter, 1982). The analysis of the SAXS experimental results led to the structural parameters for the crotapotin mole- cule listed in Table 1. Rg and V were determined from Eqs. 1 and 5, respectively, and Dmax from the P(r) func- tion (Eq. 7). The following step was to establish the mo- lecular shape consistent with the parameter values shown in Table 1. We have disregarded the simplest as- sumption of a spherical shape because the volume of a 562 Biophysical Journal Volume 64 FebruaryVolume 64 February 1993562 Biophysical Joumal 500 400 6 300 -e 200 100 0 FIGURE 4 (a) SAXS curve after desmearing, smoothing and extrapolation down to h = 0 by using the ITP program (Glatter, 1982). (b) Porod plot of the curve of Fig. 4 a. sphere calculated from the gyration radius Rg = 13.6 A, Vsphere = 22.6 x i03 A, is higher than the equal to the experimental volume, Vexp = 16.2 x 1O3 A3. The subsequent assumed molecular shape was that represented by an ellipsoid of revolution, with semi-axes a, b = a, and c, and assymetry parameter v = c/a. In Fig. 5 the volume associated to an ellipsoid ofrevolution with a radius of gyration of 12.6 A versus the ratio of the semi-axis v was plotted. The intercepts ofthis curve with the horizontal line corresponding to the experimental molecular volume, Vexp = 16.2 x 1O3 A3 yield the semi- axes ratio Po = 0.45 and vP = 1.95, which are related to oblate and prolate ellipsoids, respectively. Both shapes are then consistent with the experimental Rg and V values. In order to decide between the prolate and oblate 2J5 OSLATE PROLATE OS \ -fV shapes for crotapotin, a comparison of the experimental and theoretical distance distribution function P(r) was tried. The experimental PE( r) distance distribution func- tion calculated using Eq. 7 is plotted in Fig. 6. The theo- retical distance distribution functions PT(r), which were calculated using the Multibody program (Glatter, 1982) for ellipsoidal molecules assuming Rg = 13.6 A, and v = 0.45 and 1.95, for oblate and prolate ellipsoids, respec- tively, are also plotted in Fig. 6. It is apparent that the theoretical function associated with an oblate ellipsoid with semi-axes a = b = 22 A and c = 10 A agrees with the experimental function better than that ofa prolate ellip- soid. This implies that the presented experimental SAXS results are consistent with a molecular shape ofan oblate ellipsoid of revolution. r (AI FIGURE 6 Comparison between the experimental (0) and theoretical distance distribution functions P( r) for two structural models ofcrota- potin. The continuous and dashed lines correspond to oblate and pro- late ellipsoids, respectively. SAXS Study of Crotapotin at Low pH 563 FIGURE 5 Volume associated with an ellipsoid ofrevolution with R. = 13.6 A as a function of the assymetry factor. The horizontal line indi- cates the volume VO = 16.2 x 10 3 .3. hrx-,l .. [k410-3] SAXS Study of Crotapotin at Low pHAbrego et a]. 563 CONCLUSION The presented experimental SAXS results indicate that the overall shape of crotapotin in solution at pH = 1.5 corresponds to an oblate ellipsoid of revolution with semi-axes a = b = 22 A and c = 10 A. Received for publication on 30 April and in finalform 28 September 1992. REFERENCES Breithaupt, H., K. Ribsamen, and E. Habermann. 1974. Biochemistry and pharmacology of the crotoxin complex. Eur. J. Biochem. 49:333-45. Glatter, 0. 1980. Computation of distance distribution functions and scattering functions of models for small-angle scattering experi- ments. Acta Phys. Austr. 52:243-56. Glatter, 0. 1982. Practical aspects to the use of indirect Fourier trans- formations methods. Makromol. Chem. 183:465-79. Guinier, A., and G. Fournet 1955. Small-Angle Scattering of X-rays. John Wiley and Sons, New York. 5-82. Laure, C. J. 1975. Die Primostruktur des Crotamins. Hoppe-Seyter's. Z. Physiol. Chem. 356:213-5. Pilz, I. 1982. Proteins. Small-Angle X-ray Scattering. 0. Kratky, and 0. Glatter, editors. Academic Press, London. 239-293. Pilz, I., 0. Glatter, and 0. Kratky. 1979. Small-Angle X-ray Scattering. In Methods in Enzymology. C. H. Hin, and S. N. Timasheff, editors. Academic Press, London. 148-249. Porod, G. 1982. General theory. In Small-Angle X-ray Scattering. 0. Kratky, and 0. Glatter, editors. Academic Press, London. 17-51. 564 Biophysical Journal Volume 64 February 1993