Letter pubs.acs.org/JPCL
Stability of a Transient Protein Complex in a Charged Aqueous Droplet with Variable pH Myong In Oh† and Styliani Consta*,†,‡ †
Department of Chemistry, The University of Western Ontario, London, Ontario N6A 5B7, Canada Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, United Kingdom
‡
S Supporting Information *
ABSTRACT: Electrospray ionization mass spectrometry (ESI-MS) has the potential to become a high-throughput robust experimental method for the detection of protein− protein equilibrium constants. Poorly understood processes that affect the stability of weak noncovalent protein complexes in the intervening droplet environment are a significant factor that precludes the advancement of the method. We use molecular dynamics to study the stability of a ubiquitin and ubiquitin-associated domain complex (RCSB PDB code 2MRO) in an aqueous droplet with changing size and charge concentration. We present evidence that a weak protein complex changes conformation and may dissociate in shrinking droplets. Then, the droplets containing these dissociated proteins divide. Our findings suggest that in some cases ESI-MS does not measure the correct association constants. The study intends to stimulate research for systematic development of experimental protocols that stabilize weakly bound protein interfaces in droplets.
T
reactions.16−19 There is accumulated evidence that the rates of many reactions in charged microdrops are enhanced by orders of magnitude relative to those in the bulk solution. The role of the micro (nano)drop environment on the reaction rates is still to be examined.19 When nanosized droplets are sprayed from the bulk solution, each of them may contain at most one protomer or the complex. Therefore, one of the manners in which the equilibrium constant can be modified in the droplet is by the dissociation of the complex. This is the mechanism that is examined here in the latest stage of the droplet lifetime. In the study we use a typical transient complex of ubiquitin (Ub) and ubiquitin-associated domain (UbA) (RCSB PDB20 code 2MRO21). This protein complex is involved in the ubiquitin−proteasome system, which is responsible for proteolytic degradation through which many biochemical pathways are regulated at the cellular level. 2MRO is a good candidate for our purpose for several reasons. The protein complex has 120 amino acids in total, and hence its size allows for computational investigation. 2MRO has been well studied experimentally using multidimensional NMR.21 Ubiquitin is known for its high thermal and acidic resistance, so it has been widely selected for both experimental and computational studies.22 A term that will be used throughout the discussion to characterize the stability of the charged droplet is that of the
he detection of protein−protein and protein−ligand association constants is in the forefront of the electrospray ionization mass spectrometry (ESI-MS) applications.1−11 The reliability of ESI-MS in measuring association constants has been validated by comparisons with other experimental methods, namely, isothermal titration calorimetry.4,6,12−14 The challenges in the equilibrium constant measurements by ESIMS are found in the weakly bound protein complexes. The equilibrium constant of those complexes between the bound and unbound states is in the micromolar range. In the last 20 years there has been considerable improvement in the experimental methodologies that aim at maintaining the weak protein−protein interactions (PPIs) throughout the complex transfer from the bulk solution into the gaseous state.1,3,8,9,15 Despite the improvement, the reliable detection of the association constant of a class of weak noncovalent protein complexes is still debated.1 Zenobi et al.5 have argued that using ESI-MS for analyzing biomolecular complexes is not yet routine, and specific protocols, characteristic of every biomolecular complex, may be required. In particular, transient and weak noncovalent interactions are prone to be ruined during multistep or harsh isolation procedures, as these interactions are often sensitive to many factors such as temperature, acidity, and salt concentration. One of the reasons that an individual analysis protocol is required for every protein complex is that the general principles that determine the stability of complexes in droplets still have not been established. The aim of this research is to establish these principles. From another perspective, the study of the complex dissociation rate in a droplet is in close relation to current research on using electrosprayed droplets as vessels to perform chemical © XXXX American Chemical Society
Received: October 7, 2016 Accepted: December 9, 2016 Published: December 9, 2016 80
DOI: 10.1021/acs.jpclett.6b02319 J. Phys. Chem. Lett. 2017, 8, 80−85
Letter
The Journal of Physical Chemistry Letters Rayleigh limit.23−26 The charge of a droplet at the Rayleigh limit (Zr) is defined as Zr2 = 64π 2ε0γR 03
(1)
where R0 and γ denote the droplet radius and the surface tension, respectively, and ε0 is the permittivity of vacuum. Zr is the maximum charge that a droplet can hold just before it becomes unstable. A droplet is said to be found “below” the Rayleigh limit when the surface tension forces overcome the electrostatic repulsion between the ions. When the strength of the interactions is reversed, the droplet is found “above” the Rayleigh limit. Zr is a constraint in the overall droplet charge, and this in turn determines the charge on the protein, as described in the Supporting Information. R0 is not utilized anywhere in the analysis of the simulation findings. A computationally challenging step in this study is the treatment of the acidity (pH) in the constantly changing environment of the droplet. The computational details in assigning the charge state of a protein27−37 are found in the Supporting Information. To account for the change of the pH during the solvent evaporation of a droplet, we performed equilibrium simulations of the droplet−complex system at various droplet sizes. The equilibrium simulations were performed at certain pH values. The equilibrium simulations allow one to find the mechanism of the complex dissociation. The rate of the complex dissociation is found in the following way: First, we determine the rate of evaporation for the droplet at the temperature at which we perform the simulations. Then, we consider a number of initial relaxed configurations of the system from which we start a number of equilibrium runs. The rate of evaporation provides a good estimate of the time that we will continue those runs. We terminate a run when its time exceeds that that is required for the evaporation of almost all water molecules. If the complex dissociation occurs earlier than the droplet evaporation, then we count this as a complex dissociation event. The study of the complex dissociation has many parameters to examine with respect to the charge state of the protein and the size of the droplet. In our studies we focus on the late stage of the droplet lifetime, where we performed the majority of the simulations in droplets composed of 2000 H2O molecules and a complex charge state +14 e at temperatures in the range of 370−390 K. We simulated 14 systems of the protein complex protonated at +14 e, where +11 e is found in ubiquitin and +3 e is found in UbA. The protein complex is embedded in an aqueous droplet. The charge state +14 e was selected because we expect complexes at the higher charge states to be more susceptible to dissociation than those at the lowest charge states. The protein charge state is assigned using the methodology presented in the Supporting Information. As shown in Figure S3 (Supporting Information) the charge state +14 e is likely to appear in systems with ammonium acetate concentration
