Article pubs.acs.org/JPCB
The Role of Sulfates on Antifreeze Protein Activity Konrad Meister,† John G. Duman,‡ Yao Xu,†,§ Arthur L. DeVries,∥ David M. Leitner,§ and Martina Havenith*,† †
Lehrstuhl für Physikalische Chemie II, Ruhr Universität, 44801 Bochum, Germany Department of Biological Sciences, University of Notre Dame, Notre Dame, Indiana 46556, United States § Department of Chemistry, University of Nevada, Reno, Nevada 89557, United States ∥ Department of Animal Biology, University of Illinois at Urbana−Champaign, Urbana, Illinois 61801, United States ‡
ABSTRACT: In the present study, we have investigated the effect of sodium sulfate (Na2SO4) buffer on the antifreeze activity of DAFP-1, the primary AFP in the hemolymph of the beetle Dendroides canadensis. In contrast to previous studies, we found evidence that sodium sulfate does not suppress antifreeze activity of DAFP-1. Terahertz absorption spectroscopy (THz) studies were combined with molecular dynamics (MD) simulations to investigate the change in collective hydrogen bond dynamics in the vicinity of the AFP upon addition of sodium sulfate. The MD simulations revealed that the gradient of H-bond dynamics toward the ice-binding site is even more pronounced when adding sodium sulfate: The cosolute dramatically slows the hydrogen bond dynamics on the ice-binding plane of DAFP-1, whereas it has a more modest effect in the vicinity of other parts of the protein. These theoretical predictions are in agreement with the experimentally observed increase in THz absorption for solvated DAFP-1 upon addition of sodium sulfate. These studies support our previously postulated mechanism for AF activity, with a preferred ice binding by threonine on nanoice crystals which is supported by a long-range effect on hydrogen bond dynamics.
■
INTRODUCTION Antifreeze proteins (AFPs) and antifreeze glycoproteins (AFGPs) enable the survival of various organisms in freezing or subfreezing habitats.1,2 AFPs from different species show a great diversity in structures but share a characteristic depression of the freezing point of the solution without affecting the melting point. The difference between the freezing and melting temperatures of this nonequilibrium phenomenon is termed thermal hysteresis (TH) and is widely used as a measure of the antifreeze activity.3 Thermal hysteresis can vary from less than 0.3 °C for certain plant AFPs to over 6 °C in insect AFPs.4 Despite their structural heterogeneity and activity differences, all AFPs are believed to work by an adsorption−inhibition mechanism in which the proteins recognize and quasiirreversibly bind to an ice surface, thereby preventing macroscopic growth of ice crystals.5 Previous studies of our group as well as others have revealed that the mechanism of AFP is more complex than initially believed.6−8 Short-range interactions such as hydroxyl bond formation by threonine is assumed to be responsible for the high affinity to the ice, while long-range interactions such as the inhomogeneous hydration dynamics could provide the necessary selectivity.7 In a recent study, we investigated several mutants of AFP-I of the winter flounder. For this set of mutants, a rigid display of OH groups from the helix is found to determine the activity.9 The fact that the A17L mutant is inactive could be explained now by MD simulations, in which we found a kinking © 2014 American Chemical Society
of this mutant. Thus, we confirmed that an exact positioning of the threonines to form strong hydroxyl bonds with the OH of the water molecules as a precondition for antifreeze activity in AFP-I.9 Whereas some mutants showed a helical structure, we observed a second necessary condition for AF activity of AFP-I: a long-range interaction via protein-induced changes in the water dynamics extending several hydration shells from the protein surface. When comparing the results for different AFPs, we could propose that structurally different AF(G)Ps optimize either one or both of these mechanisms to a different extent.7,9,10 The Davies group proposed that antifreeze proteins that lack an ordered array of threonine residues in their ice-binding site (IBS) could instead function by an array of anchored clathrate water molecules.8,11,12 Enhancement of antifreeze activity with cosolutes of low molecular mass is a well-known but not fully understood phenomenon. Various groups of molecules have been identified to enhance activity, including various salts.13−16 The magnitude of enhancement varies significantly between different structural AFPs, ranging from 2-fold (type-I fish AFP) to 6-fold (DAFP) increase, with sodium citrate being the most efficient one.7 Special Issue: James L. Skinner Festschrift Received: January 20, 2014 Revised: March 30, 2014 Published: May 12, 2014 7920
dx.doi.org/10.1021/jp5006742 | J. Phys. Chem. B 2014, 118, 7920−7924
The Journal of Physical Chemistry B
Article
difference Δα between the integrated THz absorbance of the AF(G)P sample solution (αsample) and of the aqueous reference (αreference) in the frequency range between 2.4 and 2.7 THz: Δα = αsample − αreference. THz absorbance measurements were performed at low humidity ( 15 mg/mL, and lower in sulfate solution at a 7921
dx.doi.org/10.1021/jp5006742 | J. Phys. Chem. B 2014, 118, 7920−7924
The Journal of Physical Chemistry B
Article
we found an increase of Δα = ∼15 cm−1 compared to buffer. When we add 0.5 M sodium sulfate to the solvated protein, we observe an increase (or THz excess) of Δα = 9 cm−1. In the case of a suppression of the antifreeze activity of DAFP-1 upon addition of sodium sulfate, we would expect a decrease in THz absorption as seen for other inactive AF(G)Ps.7,9,10 Instead, we find an increase in THz absorption which is in agreement with the postulation of antifreeze activity. Accompanying MD simulations were carried out in order to reveal a molecular understanding: Figure 2B shows the hydrogen bond lifetime correlation functions CHB(t) for water molecules around the binding and nonbinding plane of DAFP1 in aqueous and sodium sulfate solution. A remarkable difference between the water dynamics in the vicinity of the binding and nonbinding planes of DAFP-1 is observed. CHB(t) reaches a value of 0.2 for hydrogen bonds between water and the atoms of the ice-binding site after 72 ps compared to 45 ps for bonds between water and the atoms at the nonbinding plane. These values can be compared with the value of 23 ps for hydrogen bonds in bulk water. In the presence of sodium sulfate, the simulations predict an increase in hydrogen bond lifetimes especially around the ice-binding plane. A similar behavior was observed previously for DAFP-1 in the presence of sodium citrate, a well-known enhancer of its antifreeze activity.7 In contrast, an inactive mutant of DAFP-1 exhibited decreased hydrogen bond lifetimes and the difference between the hydrogen bond dynamics over the binding and nonbinding planes is reduced.7 In all of these studies, the same water model and force field was used. Importantly, the effect on hydrogen bond and water dynamics is not merely local. It extends beyond the distinct hydrogen bond dynamics observed on the icebinding and non-ice-binding regions of the protein. As detailed in our previous study of DAFP-1,7 the vibrational density of states of water in the hydration layer shifts to higher frequency when cosolute is present, consistent with a more retarded hydrogen binding dynamics in the dynamical hydration shell. The concentration dependence of the detected THz absorbance gives information on the size of this dynamical hydration shell which includes all water molecules which are affected in their hydrogen bond dynamics in this special
Figure 1. Freezing point hysteresis of DAFP-1 in water (black ■) and in 0.5 M Na2SO4 (pink ■) determined using the capillary freezing melting/point method. Symbols are connected for easier visualization.
concentration of 10−15 mg/mL. Below 5 mg/mL, the thermal hysteresis of both solutions is about the same. In order to investigate the effect of addition of Na2SO4 on the hydrogen bond network dynamics, we have carried out terahertz absorption measurements along with accompanying molecular dynamics simulations. Terahertz absorption spectroscopy has proven to be a valuable tool to investigate collective protein−water network motions. In previous studies, we could show that all active antifreeze proteins showed an experimentally observed THz excess.7,9,10 As our previous studies showed, an increase of THz absorption in the frequency range between 2.4 and 2.7 THz is associated with a retardation in the hydrogen bond dynamics in the dynamical hydration shell of a protein.30,31 Here we compare the THz absorption of solvated DAFP-1 to DAFP-1 in a sodium sulfate buffer. The results are shown in Figure 2A, where we display the change of the integrated THz absorption coefficient of the sample, using buffer as a reference, as a function of protein concentration. For solvated DAFP-1,
Figure 2. (A) Concentration-dependent terahertz absorption of DAFP-1 in water (black ■) and in 0.5 M Na2SO4 (pink ◆). Measurements were carried out at 293.15 (20 °C) ± 0.5 K. The corresponding buffer absorption was subtracted to yield Δα = αprotein+buffer − αbuffer at each protein concentration. The absorption was integrated in the frequency range from 2.4 to 2.7 THz. (B) Hydrogen bond lifetime correlation functions CHB(t) for water molecules around the binding (black, solid line) and nonbinding plane (black, dashed line) of DAFP-1 at 300 K. CHB(t) is also plotted for hydrogen bonds between water molecules at the binding plane (magenta, solid line) and nonbinding plane (magenta, dashed line) in sodium sulfate solution. 7922
dx.doi.org/10.1021/jp5006742 | J. Phys. Chem. B 2014, 118, 7920−7924
The Journal of Physical Chemistry B
Article
Figure 3. Antifreeze activities of AFP-1, AF(G)P1−5, AF(G)P7−8, and DAFP-1 in water (black ■) and in the presence of 1 M MgSO4 (red ▼) and 1 M (NH4)2SO4 (blue ▲) determined with a Clifton Nanoliter Osmometer. Symbols are connected for easier visualization.
frequency window. The protein concentration at which the maximum is found (cmax) upon addition of sodium sulfate (see Figure 2 for DAFP-1 in sodium sulfate solution) is within the same protein concentration interval within the experimental uncertainty. On the basis of the fact that we find similar values for the concentration at which the maximum is reached, we concluded that the size of this dynamical hydration shell is not affected by the presence of the Na2SO4. In order to gain more insight into the overall influence of cosolutes on the antifreeze activity, we have investigated the effect of ammonium and magnesium sulfate on various AF(G)Ps. The cations ammonium and magnesium were chosen according to the Hofmeister series, since they can serve as examples for a weakly and strongly hydrated ion, respectively. Figure 3 illustrates the experimentally determined relationship between thermal hysteresis of different AF(G)Ps in the presence of sodium, ammonium, and magnesium sulfate using a Clifton Nanoliter Osmometer. Unfortunately, for sodium sulfate, we were not able to determine an experimental freezing point hysteresis due to solubility problems after having frozen the sample to −30 °C, which made it impossible to discriminate between sodium sulfate and ice crystals at the melting and hysteresis freezing point temperatures. All AF(G)Ps show a significantly enhanced antifreeze activity in the presence of magnesium and ammonium sulfate, independent of the position in the Hofmeister series. The maximal amount of enhancement varies for the different structural AF(G)Ps, ranging from a 100% increase for DAFP up to 230% for AF(G)P. The differences between the enhancement in the presence of either magnesium or ammonium sulfate are marginal, which is in agreement with a previous
study which found no significant influence upon exchange of the cation.20 In summary, our results indicate that upon addition of sodium sulfate the antifreeze activity of DAFP-1 is unaffected. This is in contrast to a previous study13 which postulated a suppression but which might have faced similar experimental problems as we did. We attribute the misinterpretation of Li et al.13 most likely to an artifact by the temperature-dependent solubility of sodium sulfate, which makes it extremely challenging to distinguish sodium sulfate from ice crystals. The molecular dynamics simulations of the hydrogen bond dynamics which are presented here and in a previous study7 with citrate solution provide the following picture: Different regions of DAFP-1 recognize, interact with, and affect water distinctly. The hydrogen bond dynamics in the vicinity of the ice-binding plane is significantly retarded compared to the hydrogen bond dynamics in the vicinity of other parts of the protein. This inhomogeneity becomes even more pronounced when a cosolute is added. We see in Figure 2 that addition of sodium sulfate considerably slows the hydrogen bond dynamics on the ice-binding plane of DAFP-1 (the typical time increases from 72 to 90 ps). In contrast, in the vicinity of the nonbinding parts of the protein, the change in hydrogen bond dynamics is much smaller. This supports our previous results for DAFP-1 in which we proposed that the gradient of the long-range retardation of the H-bond dynamics toward the ice-binding site plays a crucial role in antifreeze activity. Here we find now a clear indication that a more pronounced H-bond gradient caused by addition of cosolutes results in more efficient antifreeze activity without changing the structure or the positioning of the threonines. Addition of cosolutes shifts the vibrational density of states of water in the hydration layer 7923
dx.doi.org/10.1021/jp5006742 | J. Phys. Chem. B 2014, 118, 7920−7924
The Journal of Physical Chemistry B
Article
to higher frequency,7 consistent with the assumption that AFP leads to a retardation of the water in the hydration layer in the presence of a cosolute.
■
(15) Amornwittawat, N.; Wang, S.; Duman, J. G.; Wen, X. Polycarboxylates enhance beetle antifreeze protein activity. Biochim. Biophys. Acta 2008, 1784, 1942−1948. (16) Wu, D. W.; Duman, J. G.; Xu, L. Enhancement of Insect Antifreeze-Activity by Antibodies. Biochim. Biophys. Acta 1991, 1076, 416−420. (17) Ahmed, A. I.; Yeh, Y.; Osuga, Y. Y.; Feeney, R. E. Antifreeze glycoproteins from Antarctic fish. Inactivation by borate. J. Biol. Chem. 1976, 251, 3033−3036. (18) Zittle, C. A. Reaction of borate with substances of biological interest. Adv. Enzymol. Relat. Subj. Biochem. 1951, 12, 493−527. (19) Kristiansen, E.; Pedersen, S. A.; Zachariassen, K. E. Salt-induced enhancement of antifreeze protein activity: A salting-out effect. Cryobiology 2008, 57, 122−129. (20) Evans, R. P.; Hobbs, R. S.; Goddard, S. V.; Fletcher, G. L. The importance of dissolved salts to the in vivo efficacy of antifreeze proteins. Comp. Biochem. Physiol. A 2007, 148, 556−561. (21) Wang, S.; Amornwittawat, N.; Banatlao, J.; Chung, M.; Kao, Y.; Wen, X. Hofmeister Effects of Common Monovalent Salts on the Beetle Antifreeze Protein Activity. J. Phys. Chem. B 2009, 113, 13891− 13894. (22) Ramløv, H.; DeVries, A. L.; Wilson, P. W. Antifreeze glycoproteins from the antarctic fish Dissostichus mawsoni studied by differential scanning calorimetry (DSC) in combination with nanolitre osmometry. Cryo-Lett. 2005, 26, 73−84. (23) DeVries, A. L. Antifreeze glycopeptides and peptides: interactions with ice and water. Methods Enzymol. 1986, 127, 293− 303. (24) Zachariassen, K. E.; Husby, J. A. Antifreeze effect of thermal hysteresis agents protects highly supercooled insects. Nature 1982, 298, 865−867. (25) Ebbinghaus, S.; Kim, S. J.; Heyden, M.; Yu, X.; Heugen, U.; Gruebele, M.; Leitner, D. M.; Havenith, M. An extended dynamical hydration shell around proteins. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 20749−20752. (26) Berendsen, H. J. C.; Vanderspoel, D.; Vandrunen, R. Gromacs- a message-passing parallel molecular-dynamics implementation. Comput. Phys. Commun. 1995, 91, 43−56. (27) Duan, Y.; Wu, C.; Chowdhury, S.; Lee, M. C.; Xiong, G. M.; Zhang, W.; Yang, R.; Cieplak, P.; Luo, R.; Lee, T.; Caldwell, J.; Wang, J. M.; Kollman, P. A point-charge force field for molecular mechanics simulations of proteins based on condensed-phase quantum mechanical calculations. J. Comput. Chem. 2003, 24, 1999−2012. (28) Leitner, D. M.; Havenith, M.; Gruebele, M. Biomolecule largeamplitude motion and solvation dynamics: modelling and probes from THz to X-rays. Int. Rev. Phys. Chem. 2006, 25, 553−582. (29) Bagchi, B. Water Dynamics in the Hydration Layer around Proteins and Micelles. Chem. Rev. 2005, 105, 3197−3219. (30) Born, B.; Kim, S. J.; Ebbinghaus, S.; Gruebele, M.; Havenith, M. The terahertz dance of water with the proteins: the effect of protein flexibility on the dynamical hydration shell of ubiquitin. Faraday Discuss. 2009, 141, 161−173. (31) Heyden, M.; Havenith, M. Combining THz spectroscopy and MD simulations to study protein-hydration coupling. Methods 2010, 52, 74−83.
AUTHOR INFORMATION
Corresponding Author
*Phone: +49 234 32-28249. Fax: +49 234 32-14183. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was funded by the Volkswagen Stiftung (M.H. and D.M.L.). Additional funding was provided by Ruhr-Universität Bochum (K.M. and M.H.) and NSF OPP (A.L.D.) and NSF CHE-0910669 (D.M.L.). The authors thank E. Bründermann for the initial setup of the p-Ge spectrometer. This work is supported by the Cluster of Excellence RESOLV (EXC 1069) funded by the Deutsche Forschungsgemeinschaft.
■
REFERENCES
(1) Duman, J. G. Antifreeze and Ice Nucleaotor Proteins in Terrestrial Arthropods. Annu. Rev. Physiol. 2001, 63, 327−357. (2) DeVries, A. L.; Wohlschlag, D. E. Freezing Resistance in Some Antarctic Fishes. Science 1969, 163, 1073−1075. (3) DeVries, A. L. Glycoproteins as Biological Antifreeze Agents in Antarctic Fishes. Science 1973, 172, 1152−1155. (4) Middleton, A. J.; Marshall, C. B.; Faucher, F.; Bar-Dolev, M.; Braslavsky, I.; Campbell, R. L.; Walker, V. K.; Davies, P. L. Antifreeze Protein from Freeze-Tolerant Grass Has a Beta-Roll Fold with an Irregularly Structured Ice-Binding Site. J. Mol. Biol. 2012, 416, 713− 724. (5) Yeh, Y.; Feeney, R. E. Antifreeze Proteins: Structures and Mechanisms of Function. Chem. Rev. 1996, 96, 601−618. (6) Sharp, K. A. A peek at ice binding by antifreeze proteins. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 7281−7282. (7) Meister, K.; Ebbinghaus, S.; Xu, Y.; Duman, J. G.; DeVries, A.; Gruebele, M.; Leitner, D. M.; Havenith, M. Long-range protein−water dynamics in hyperactive insect antifreeze proteins. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 1617−1622. (8) Garnham, C. P.; Campbell, R. L.; Davies, P. L. Anchored clathrate waters bind antifreeze proteins to ice. Proc. Natl. Acad. Sci. U.S.A. 2011, 108 (18), 7363−7367. (9) Ebbinghaus, S.; Meister, K.; Prigozhin, M. B.; DeVries, A. L.; Havenith, M.; Dzubiella, J.; Gruebele, M. Functional Importance of Short-Range Binding and Long-Range Solvent Interactions in Helical Antifreeze Peptides. Biophys. J. 2012, 103, 20−22. (10) Ebbinghaus, S.; Meister, K.; Born, B.; DeVries, A. L.; Gruebele, M.; Havenith, M. Antifreeze Glycoprotein Activity Correlates with Long-Range Protein−Water Dynamics. J. Am. Chem. Soc. 2010, 132, 12210−12211. (11) Sun, T.; Lin, F.-H.; Campbell, R. L.; Allingham, J. S.; Davies, P. L. An Antifreeze Protein Folds with an Interior Network of More Than 400 Semi-Clathrate Waters. Science 2014, 343, 795−798. (12) Kondo, H.; Hanada, Y.; Sugimoto, H.; Hoshino, T.; Garnham, C. P.; Davies, P. L.; Tsuda, S. Ice-binding site of snow mold fungus antifreeze protein deviates from structural regularity and high conservation. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 9360−9365. (13) Li, N.; Andorfer, C. A.; Duman, J. G. Enhancement of insect antifreeze protein activity by solutes of low molecular mass. J. Exp. Biol. 1998, 201, 2243−2251. (14) Amornwittawat, N.; Wang, S.; Banatlao, J.; Chung, M.; Velasco, E.; Duman, J. G.; Wen, X. Effects of polyhydroxy compounds on beetle antifreeze protein activity. Biochim. Biophys. Acta 2009, 1794, 341−346. 7924
dx.doi.org/10.1021/jp5006742 | J. Phys. Chem. B 2014, 118, 7920−7924
