Modulation of the intermolecular interaction of myoglobin by removal of the heme
aGraduate School of Advanced Integration Science, Chiba University, 1-33 Yayoi, Inage, Chiba 263-8522, Japan, bDepartment of Chemistry, Okayama University, 3-1-1 Tsushima-Naka, Kita-ku, Okayama 700-8530, Japan, cDepartment of Biotechnology, Toyama Prefectural University, 5180 Kurokawa, Imizu, Toyama 939-0398, Japan, and dDepartment of Pharmacy, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan
*Correspondence e-mail: firstname.lastname@example.org
Toward understanding intermolecular interactions governing self-association of proteins, the present study investigated a model protein, myoglobin, using a small-angle X-ray scattering technique. It has been known that removal of the heme makes myoglobin aggregation-prone. The interparticle interferences of the holomyoglobin and the apomyoglobin were compared in terms of the structure factor. Analysis of the structure factor using a model potential of Derjaguin–Laudau–Verwey–Overbeek (DLVO) suggests that the intermolecular interaction potential of apomyoglobin is more attractive than that of holomyoglobin at short range from the protein molecule.
Precipitation or aggregation of proteins is often problematic when handling protein solution for research and commercial use. Developing methods to inhibit undesired aggregation of proteins is one of the critical issues in the biopharmaceutical field (Shire et al., 2004; Yang et al., 2010). To our knowledge, in protein design studies, we often come across water-insoluble artificial proteins (Isogai et al., 2005; Imamura et al., 2012). In contrast, a mechanism which prevents an uncontrolled aggregation is likely to exist in naturally occurring proteins (Isogai, 2006; Monsellier & Chiti, 2007). Understanding physicochemical mechanisms governing self-association of proteins will be useful for future protein engineering. In the present study, we used a small-angle X-ray scattering (SAXS) technique to analyze protein's interparticle interference, which yields an intermolecular interaction of a protein in solution through use of the theory for liquids (Tardieu et al., 1999). We employed myoglobin as a model protein. Removal of heme is regarded as a kind of simple protein engineering. Interestingly, it was reported that the apomyoglobin, in which the heme is removed, is prone to aggregate, while holomyoglobin, in which the heme is binding, is highly water-soluble (Fändrich et al., 2001, 2003). Crystallization of proteins would be controlled by protein's self-interaction. To date, a three-dimensional structure of apomyoglobin by X-ray crystal analysis has not been uploaded to the Protein Data Bank, while holomyoglobin was the first success in the X-ray crystal analysis of the protein structures (Kendrew et al., 1958). Unravelling how removal of heme changes the protein–protein interaction of myoglobin will be a step toward understanding key factors for the aggregation and the solubility.
Holomyoglobin from equine skeletal muscle was purchased from Sigma. Apomyoglobin was prepared according to the following methods (Teale, 1959; Hapner et al., 1968). 10 mg ml−1 aqueous solution of myoglobin was adjusted to pH ∼1.5 with HCl at 297 K. An equal volume of ice-cooled 2-butanone was mixed with the solution, and then the organic phase containing the extracted heme was discarded. This procedure was repeated three times. The aqueous phase was dialyzed against sodium bicarbonate (0.05 g l−1) and then ultrapure water (Milli-Q, Millipore). The apomyoglobin solution was finally lyophilized. The solutions of holomyoglobin and apomyoglobin for the measurements were prepared by dissolving the powders into the ultrapure water. The pH of the holomyoglobin and apomyoglobin solutions were 7.5 and 6.7, respectively. The concentrations of both the proteins were determined by using an ∊410nm of 1.6 × 105 for holomyoglobin and ∊280nm of 1.43 × 104 for apomyoglobin (Goto et al., 1990).
SAXS was measured using a SAXSess camera (Anton Paar; Graz, Austria) with a sealed-tube X-ray source (PANalytical, The Netherlands), operating at 40 kV and 50 mA, and line collimation. X-ray intensities were recorded using an imaging plate (100 µm × 100 µm pixel size; Fuji Film, Japan), which was read using a Cyclone scanner (PerkinElmer, USA). The X-ray wavelength, λ, was 1.542 Å (Cu Kα); the camera length was 264.5 mm. The image data were integrated into one-dimensional scattering intensities using the program ImageJ (developed by Wayne Rasband, National Institute of Health, Bethesda, MD, USA) with the macros in Utah SAXS Tools written by Professor David P. Goldenberg (University of Utah, USA). The scattering parameter q is defined as q = 4πsinθ/λ, where 2θ is the scattering angle of X-rays. The width of the integration area (detector slit width) was 10 mm. Protein solution and water were measured at 293 K in 1 mm-diameter glass capillaries for 1 h.
2.3. Calculation of a structure factor from an intermolecular potential
The SAXS intensity of a protein, I(q), is described by
where c is the protein concentration, k is a constant, P(q) is a form factor (protein's self-scattering) and S(q) is a structure factor (protein's interparticle interference). The SAXS profiles of the dilute protein solutions (0.5 wt%), at the concentration of which the interparticle interferences are negligibly small, yields kP(q) by an indirect Fourier transform program GNOM (Svergun, 1992) involving correction of the slit smeared effect. S(q) can be connected to the intermolecular potential of a protein molecule, V(r), by solving the Ornstein–Zernike equation with an appropriate closure. In the present analysis, we used the closure obtained from a random phase approximation, under which S(q) is described by
(Hansen & McDonald, 1976; Narayanan & Liu, 2003), where S0(q) is the structure factor of the reference system, n is the protein's number density, kB is the Boltzmann constant and V(q) is the Fourier transform of V(r). S0(q) has been evaluated using the empty core model (Croxton, 1975; Kelkar et al., 1992). The present study employed the Derjaguin–Laudau–Verwey–Overbeek (DLVO) model potential (Verwey & Overbeek, 1948), in which V(r) is expressed as
where r is the distance between the protein molecules. The term VHS(r) is the hard sphere potential given by
where σ is the diameter of the protein. The terms VC(r) and VAY(r) are the screened repulsive Coulomb potential and the attractive potential, respectively [here, VAY(r) is assumed to be a Yukawa-type potential],
where Z is the net charge on the protein, e is the elementary charge, ∊ is the dielectric constant of the medium, κ is the reciprocal Debye–Hückel screening length, J is the depth of attractive potential at r = σ, and d is the range of the attractive potential. S(q) can be simulated, according to the equations described above and the given values for the parameters (Z, σ, J and d). In the present study, we used Z = 1 and σ = 41.2 Å, the values of which were from the literature (Longeville et al., 2003). The procedure for the present analysis was performed by using IGOR Pro version 6.22A (Wavemetrics, USA).
Fig. 1 shows the observed SAXS profiles of holomyoglobin and apomyoglobin at 0.5 and 6.3 wt%, where the intensities are normalized by the protein concentrations. For both the 6.3 wt% solutions, the intensities at low scattering vectors (q < ∼0.1 Å−1) were depressed compared with that for 0.5 wt% solutions, which are due to the interparticle interference, i.e. the structure factor. We analysed the structure factor component in the SAXS intensities based on the DLVO model interaction potential and the equations above. A slit-smeared SAXS intensity, Is(q), was simulated from I(q), which is from the determined kP(q) and the theoretically derived S(q) with equation (1), and the instrumental beam profile used as a weighting function (Glatter & Kratky, 1982). To find the values of the parameters (J and d), which reproduce the experimental Is(q), we compared the experimental Is(q) with the calculated Is(q), varying the values of J (0.01–9 kBT) and d (1–15 Å). The maps of the residual sum of squares are shown in Figs. 2(a) and 2(b). It is found that smaller J and larger d tend to reduce the residuals for holomyoglobin, while larger J and smaller d tend to reduce the residuals for apomyoglobin. The protein–protein interaction potentials, V(r), calculated with the optimal values of J and d, which give the best fits, are depicted in Fig. 2(c). At small r (r < ∼45 Å), the potential of apomyoglobin is more attractive than that of holomyoglobin, due to the larger J value. This could explain the irreversible aggregation-prone property of apomyoglobin (Fändrich et al., 2003). In addition, the attractive potential of holomyoglobin is relatively small negative at small r (r < ∼45 Å), but still remains at larger r (r < ∼55 Å) due to the larger d value. The attractive potential allows the proteins to be close to each other, but to be separated because the magnitude is comparable with thermal energy (∼kBT) at short range. This seems to underlie the highly water-soluble and less irreversible aggregation-prone properties of holomyoglobin.
In the present study we investigated the role for the heme in the self-interaction of myoglobin molecules by SAXS. From the fitting of the experimental SAXS data with the simulated SAXS intensities based on the DLVO model and the equations in the liquid theory, it is suggested that change in the SAXS profile by removal of the heme could be explained by an increase in the intermolecular attractive interaction potential at short range from the protein molecule. We remark that the potential we can obtain depends on the selected model. Therefore, we progress a study developing a method to determine the intermolecular potential of proteins from experimental structure factors without assuming any model potential by applying an integral equation theory.
This work was supported in part by the Global Center-of-Excellence Program `Advanced School for Organic Electronics' supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan (to HI).
Croxton, C. A. (1975). Introduction to Liquid State Physics. London: Wiley.
Fändrich, M., Fletcher, M. A. & Dobson, C. M. (2001). Nature (London), 410, 165–166. Web of Science PubMed
Fändrich, M., Forge, V., Buder, K., Kittler, M., Dobson, C. M. & Diekmann, S. (2003). Proc. Natl Acad. Sci. USA, 100, 15463–15468. Web of Science CrossRef PubMed
Glatter, O. & Kratky, O. (1982). Small-Angle X-ray Scattering. New York: Academic Press.
Goto, Y., Calciano, L. J. & Fink, A. L. (1990). Proc. Natl Acad. Sci. USA, 87, 573–577. CrossRef CAS PubMed Web of Science
Hansen, J. P. & McDonald, I. R. (1976). Theory of Simple Liquids. New York: Academic Press.
Hapner, K. D., Bradshaw, R. A., Hartzell, C. R. & Gurd, F. R. (1968). J. Biol. Chem. 243, 683–689. CAS PubMed Web of Science
Imamura, H., Isogai, Y. & Kato, M. (2012). Biochemistry, 51, 3539–3546. Web of Science CrossRef CAS PubMed
Isogai, Y. (2006). Biochemistry, 45, 2488–2492. Web of Science CrossRef PubMed CAS
Isogai, Y., Ito, Y., Ikeya, T., Shiro, Y. & Ota, M. (2005). J. Mol. Biol. 354, 801–814. Web of Science CrossRef PubMed CAS
Kelkar, V. K., Narayanan, J. & Manohar, C. (1992). Langmuir, 8, 2210–2214. CrossRef CAS Web of Science
Kendrew, J. C., Bodo, G., Dintzis, H. M., Parrish, R. G., Wyckoff, H. & Phillips, D. C. (1958). Nature (London), 181, 662–666. CrossRef PubMed CAS Web of Science
Longeville, S., Doster, W. & Kali, G. (2003). Chem. Phys. 292, 413–424. Web of Science CrossRef CAS
Monsellier, E. & Chiti, F. (2007). EMBO Rep. 8, 737–742. Web of Science CrossRef PubMed CAS
Narayanan, J. & Liu, X. Y. (2003). Biophys. J. 84, 523–532. Web of Science CrossRef PubMed CAS
Shire, S. J., Shahrokh, Z. & Liu, J. (2004). J. Pharm. Sci. 93, 1390–1402. Web of Science CrossRef PubMed CAS
Svergun, D. I. (1992). J. Appl. Cryst. 25, 495–503. CrossRef Web of Science IUCr Journals
Tardieu, A., Le Verge, A., Malfois, M., Bonneté, F., Finet, S., Riès-Kautt, M. & Belloni, L. (1999). J. Cryst. Growth, 196, 193–203. Web of Science CrossRef CAS
Teale, F. W. (1959). Biochim. Biophys. Acta, 35, 543. CrossRef PubMed Web of Science
Verwey, E. J. W. & Overbeek, J. T. G. (1948). Theory of the Stability of Lyophobic Colloids. Amsterdam: Elsevier.
Yang, Y., Petkova, A., Huang, K., Xu, B., Hua, Q. X., Ye, I. J., Chu, Y. C., Hu, S. Q., Phillips, N. B., Whittaker, J., Ismail-Beigi, F., Mackin, R. B., Katsoyannis, P. G., Tycko, R. & Weiss, M. A. (2010). J. Biol. Chem. 285, 10806–10821. Web of Science CrossRef CAS PubMed
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.