Oleg Galkin

Liquid–liquid separation in solutions of normal and sickle cell hemoglobin

We show that in solutions of human hemoglobin (Hb)—oxy- and deoxy-Hb A or S—of near-physiological pH, ionic strength, and Hb concentration, liquid–liquid phase separation occurs reversibly and reproducibly at temperatures between 35 and 40°C. In solutions of deoxy-HbS, we demonstrate that the dense liquid droplets facilitate the nucleation of HbS polymers, whose formation is the primary pathogenic event for sickle cell anemia. In view of recent results that shifts of the liquid–liquid separation phase boundary can be achieved by nontoxic additives at molar concentrations up to 30 times lower than the protein concentrations, these findings open new avenues for the inhibition of the HbS polymerization.

Nucleation of protein crystals: critical nuclei, phase behavior, and control pathways

Abstract We have studied the nucleation of crystals of the model protein lysozyme using a novel technique that allows direct determinations of homogeneous nucleation rates. At constant temperature of 12.6°C we varied the thermodynamic supersaturation by changing the concentrations of protein and precipitant. We found a broken dependence of the homogeneous nucleation rate on supersaturation that is beyond the predictions of the classical nucleation theory. The nucleation theorem allows us to relate this to discrete changes of the size of the crystal nuclei with increasing supersaturation as (10 or 11)→(4 or 5)→(1 or 2). Furthermore, we observe that the existence of a second liquid phase at high protein concentrations strongly affects crystal nucleation kinetics. We show that the rate of homogeneous nucleation of lysozyme crystals passes through a maximum in the vicinity of the liquid–liquid phase boundary hidden below the liquidus (solubility) line in the phase diagram of the protein solution. We found that glycerol and polyethylene glycol (PEG), which do not specifically bind to proteins, shift this phase boundary and significantly suppress or enhance the crystal nucleation rates, although no simple correlation exists between the action of PEG on the phase diagram and the nucleation kinetics. This provides for a control mechanism which does not require changes in the protein concentration, or the acidity and ionicity of the solution. The effects of the two additives on the phase diagram strongly depend on their concentration and this provides opportunities for further tuning of nucleation rates.

Spinodal for the solution-to-crystal phase transformation.

The formation of crystalline nuclei from solution has been shown for many systems to occur in two steps: the formation of quasidroplets of a disordered intermediate, followed by the nucleation of ordered crystalline embryos within these droplets. The rate of each step depends on a respective free-energy barrier and on the growth rate of its near-critical clusters. We address experimentally the relative significance of the free-energy barriers and the kinetic factors for the nucleation of crystals from solution using a model protein system. We show that crystal nucleation is 8-10 orders of magnitude slower than the nucleation of dense liquid droplets, i.e., the second step is rate determining. We show that at supersaturations of three or four k(B)T units, crystal nuclei of five, four, or three molecules transform into single-molecule nuclei, i.e., the significant nucleation barrier vanishes below the thermal energy of the molecules. We show that the main factor, which determines the rate of crystal nucleation, is the slow growth of the near-critical ordered clusters within the quasidroplets of the disordered intermediate. Analogous to the spinodal in supersaturated fluids, we define a solution-to-crystal spinodal from the transition to single-molecule crystalline nuclei. We show that heterogeneous nucleation centers accelerate nucleation not only because of the wettinglike effects that lower the nucleation barrier, as envisioned by classical theory, but by helping the kinetics of growth of the ordered crystalline embryos.

Smooth transition from metastability to instability in phase separating protein solutions

For insight into the structure and dynamics of phases emerging upon crossing the metastability/instability boundary we monitor with optical microscopy, in real time and in real space, the generation of a dense liquid phase in high-concentration solutions of the protein lysozyme after temperature quenches into thermodynamically defined metastable and unstable regions. We show with this system, which is a poor fit to mean-field assumptions, that the evolution of the structure factor during nucleation is similar to that during spinodal decomposition and reveals no singularity predicted upon crossing the metastability boundary. We introduce two kinetic definitions of the metastability/instability boundary that yield values within approximately 1.5 K, i.e., the boundary appears as an area rather than a line, which is near and above the thermodynamic prediction. Delay times for the appearance of the new phase in the unstable regime are significant, i.e., new-phase growth is hindered by kinetic barriers. While our results agree with predictions of the non-mean-field theories of phase transformations, the experimentally observed behavior is richer than the one envisioned by theory.

Free Heme and the Polymerization of Sickle Cell Hemoglobin

In search of novel control parameters for the polymerization of sickle cell hemoglobin (HbS), the primary pathogenic event of sickle cell anemia, we explore the role of free heme, which may be excessively released in sickle erythrocytes. We show that the concentration of free heme in HbS solutions typically used in the laboratory is 0.02-0.04 mole heme/mole HbS. We show that dialysis of small molecules out of HbS solutions arrests HbS polymerization. The addition of 100-260 μM of free heme to dialyzed HbS solutions leads to rates of nucleation and polymer fiber growth faster by two orders of magnitude than before dialysis. Toward an understanding of the mechanism of nucleation enhancement by heme, we show that free heme at a concentration of 66 μM increases by two orders of magnitude the volume of the metastable clusters of dense HbS liquid, the locations where HbS polymer nuclei form. These results suggest that spikes of the free heme concentration in the erythrocytes of sickle cell anemia patients may be a significant factor in the complexity of the clinical manifestations of sickle cell anemia. The prevention of free heme accumulation in the erythrocyte cytosol may be a novel avenue to sickle cell therapy.

Molecular mechanisms of microheterogeneity‐induced defect formation in ferritin crystallization

We apply in situ atomic force microscopy to the crystallization of ferritins from solutions containing ≈5% (w/w) of their inherent molecular dimers. Molecular resolution imaging shows that the dimers consist of two bound monomers. The constituent monomers are likely partially denatured, resulting in increased hydrophobicity of the dimer surface. Correspondingly, the dimers strongly adsorb on the crystal surface. The adsorbed dimers hinder step growth and on incorporation by the crystal initiate stacks of up to 10 triple and single vacancies in the subsequent crystal layers. The molecules around the vacancies are shifted by ≈0.1 molecular dimensions from their crystallographic positions. The shifts strain the lattice and, as a consequence, at crystal sizes > 200 μm, the accumulated strain is resolved by a plastic deformation whereupon the crystal breaks into mosaic blocks 20–50 μm in size. The critical size for the onset of mosaicity is similar for ferritin and apoferritin and close to the value for a third protein, lysozyme; it also agrees with theoretical predictions. Trapped microcrystals in ferritin and apoferritin induce strain with a characteristic length scale equal to that of a single point defect, and, as a consequence, trapping does not contribute to the mosaicity. The sequence of undesired phenomena that include heterogeneity generation, adsorption, incorporation, and the resulting lattice strain and mosaicity in this and other proteins systems, could be avoided by improved methods to separate similar proteins species (microheterogeneity) or by increasing the biochemical stability of the macromolecules against oligomerization. Proteins 2001;43:343–352.

On the methods of determination of homogeneous nucleation rates of protein crystals

A recently published analysis of an experimental method to directly determine the homogeneous rates of nucleation of protein crystals or other solid phases of proteins inferred that the method underestimates the homogeneous nucleation rates by several orders of magnitude. Here, we consider the presented arguments and find that the analysis contains certain inconsistencies. We conclude that the homogeneous nucleation rate data produced using the direct determination technique are, most likely, correct.

Differential pathways in oxy and deoxy HbC aggregation/crystallization

CC individuals, homozygous for the expression of βC‐globin, and SC individuals expressing both βS and βC‐globins, are known to form intraerythrocytic oxy hemoglobin tetragonal crystals with pathophysiologies specific to the phenotype. To date, the question remains as to why HbC forms in vivo crystals in the oxy state and not in the deoxy state. Our first approach is to study HbC crystallization in vitro, under non‐physiological conditions. We present here a comparison of deoxy and oxy HbC crystal formation induced under conditions of concentrated phosphate buffer (2g% Hb, 1.8M potassium phosphate buffer) and viewed by differential interference contrast microscopy. Oxy HbC formed isotropic amorphous aggregates with subsequent tetragonal crystal formation. Also observed, but less numerous, were twisted, macro‐ribbons that appeared to evolve into crystals. Deoxy HbC also formed aggregates and twisted macro‐ribbon forms similar to those seen in the oxy liganded state. However, in contrast to oxy HbC, deoxy HbC favored the formation of a greater morphologic variety of aggregates including polymeric unbranched fibers in radial arrays with dense centers, with infrequent crystal formation in close spatial relation to both the radial arrays and macroribbons. Unlike the oxy (R‐state) tetragonal crystal, deoxy HbC formed flat, hexagonal crystals. These results suggest: (1) the Lys substitution at β6 evokes a crystallization process dependent upon ligand state conformation [i.e., the R (oxy) or T (deoxy) allosteric conformation]; and (2) the oxy ligand state is thermodynamically driven to a limited number of aggregation pathways with a high propensity to form the tetragonal crystal structure. This is in contrast to the deoxy form of HbC that energetically equally favors multiple pathways of aggregation, not all of which might culminate in crystal formation. Proteins 2001;42:99–107.

Interactions of hemin with model erythrocyte membranes.

To address the interactions of hemin with phospholipid bilayers, we introduce hemin to a solution of dimyristoylphosphatidylcholine (DMPC), a long chain phospholipid, and 3-(cholamidopropyl)(dimethylammonio)-2-hydroxy-1-propanesulfonate (CHAPSO), a detergent, in which vesicles form at around 37 degrees C. We show that vesicles composed of DMPC/CHAPSO form and grow, following a mechanism that does not trap solution and excludes larger solutes, such as hemin, from the vesicle interior. The existence of a limited number of patches of likely 2D crystalline hemin embedded in the phospholipid bilayer suggests that this layer is saturated with hemin molecules. We show that despite this saturation, even after prolonged contact with hemin-containing solution outside the vesicles, hemin is not released on the other side of the membrane; i.e., the phospholipid bilayer is impermeable to hemin. Comparison of the properties of the model membrane to those of the erythrocyte membrane suggests that the latter might also be impermeable to hemin and, given the absence of pores suitable to hemin in the erythrocyte membranes, that hemin might accumulate in erythrocytes after its release due to hemoglobin instability.

Are protein crystallization mechanisms relevant to understanding and control of polymerization of deoxyhemoglobin S

Abstract We investigated the homogeneous nucleation and subsequent evolution of polymers of sickle cell hemoglobin (HbS) using differential interference contrast (DIC) microscopy. The same technique was employed to determine the liquid–liquid separation boundaries for a variety of conditions in solution of sickle cell and normal human hemoglobin. The HbS polymers were also imaged using atomic force microscopy. We found that the location of Liquid–Liquid phase boundary under conditions that mimic those in the erythrocytes is consistent with previous determinations of the spinodal for this phase transition. Increasing the ionic strength shifts this phase boundary to significantly lower temperatures and Hb concentrations. We also found that the nucleation of individual HbS fibers indicates that the process is random and follows statistics similar to those established for nucleation of crystals or liquid droplets from vapors.