Kinetics of hemoglobin S polymerization and gelation under shear: I. Shape of the viscosity progress curve and dependence of delay time and reaction rate on shear rate and temperature

Polymerization and gelation of deoxyhemoglobin S makes red blood cells (RBCs) rigid and is the immediate basis of pathogenesis in sickle cell disease. Hence, characterization of hemoglobin S viscosity and its time- dependent development as RBCs pass through the microvasculature is important in understanding pathogenesis. Because RBCs and the intraerythrocytic milieu in vivo are subject to shear, the shear dependence of polymerization kinetics is also important. In steady- state cone-plate viscometry we find: (1) gelation under shear progresses exponentially with time; (2) shear markedly increases exponential rate and (3) shortens delay time independent of when in the delay time it is applied; (4) shear greatly decreases the temperature dependence of the exponential rate and delay time; (5) simultaneous with its acceleratory effect on polymerization, shear breaks down gel structure. We conclude that shear acts to accelerate gelation by breaking fibers and creating new growing ends, a process that occurs in addition to the homogeneous and heterogeneous nucleation of new fibers that occurs in the absence of shear. Fibers that break are part of a gel network rather than in free solution. The shear dependence of gelation rates means that the critical clinical issue, whether the delay time is long enough and gelation slow enough to permit deoxygenated cells to pass through the microvasculature before they rigidify, depends on in vivo shear rates as well as on degree of unsaturation and hemoglobin concentration.     

Hemoglobin S polymerization and gelation under shear II. The joint concentration and shear dependence of kinetics

The kinetics of hemoglobin S gelation are critical in sickle disease because microvascular obstruction can be avoided if red blood cells pass these vessels during the delay time, before polymerization and gelation occur in sufficient degree to rigidify the cells. Kinetics, including the delay time and the closely related exponential progress rate, are highly sensitive to hemoglobin concentration and degree of deoxygenation. Kinetics are also greatly accelerated by shear, an effect that may contribute to pathogenesis, since red blood cells deform and can undergo shear in vivo. Here we examine the joint dependence of kinetics on shear and hemoglobin concentration. As shear rate increases, the concentration dependence of the exponential progress rate decreases. The large decrease in concentration dependence supports the conclusion that acceleration of gelation by shear is due to breakage and not to enhancement of heterogeneous nucleation. Under shear, new fibers are created by breakage of existing ones, as well as by heterogeneous nucleation. At high shear, the rate of new fiber creation by breakage is very great and dominates that by heterogeneous nucleation. Therefore, if breakage depended only on shear rate and were independent of the concentration of hemoglobin in solution, the concentration dependence of kinetics should vanish. Although it decreases, it does not disappear. The concentration dependence that remains at high shear arises from (1) the direct contribution of fiber growth rate to the exponential progress rate, (2) the dependence of breakage rate on fiber growth rate, and (3) the dependence of solution viscosity on hemoglobin concentration.     

Kinetics and Mechanism of Deoxyhemoglobin S Gelation: A New Approach to Understanding Sickle Cell Disease

We report the results of a kinetic investigation on the gelation of purified deoxyhemoglobin S. Gelation was induced by raising the temperature and was monitored by measuring both the heat absorbed, with a microcalorimeter, and the appearance of linear birefringence, with a microspectrophotometer. The kinetics are unusual. Prior to the onset of gelation there is a delay period, followed by a sigmoidal progress curve. The delay time is formally dependent on approximately the 30th power of the deoxyhemoglobin S concentration; a decrease in concentration from 23 to 22 g/dl increases the delay time by a factor of four. It is also extremely temperature dependent; a 1 degrees C temperature rise in the range 20-30 degrees C almost halves the delay time. From these results we conclude that the initial rate is controlled by the nucleation of individual fibers. We present a kinetic model that accounts for the concentration, temperature, and time dependence of the initial phase of the gelation reaction. Extrapolation of our data to physiological conditions predicts that changes in intracellular hemoglobin concentration and oxygen saturation, realizable in vivo, produce enormous changes in the delay time. The range of delay times spans both the mean capillary transit and total circulation times. This result points to the delay time as an extremely important variable in determining the course of sickle cell disease, and suggests a new approach to therapy.     

Supersaturation in sickle cell hemoglobin solutions.

The kinetic inhibition of the gelation of hemoglobin S is compared to the change in hemoglobin S soulbility, when the solubility is altered by carbon monoxide, pH, or urea. By means of a new technique, the delay time and the extent of gelation are measured on the same sample. They delay time, td, is found to be proportional to a high power (30-40) of the hemoglobin S solubility. Togehter with the previously reported concentration dependence, this result demonstrates that the rate is proportional to a high power of the supersaturation, S, defined as the ratio of the total hemoglobin S concentration to the equilibrium solubility. The results obey the supersaturation equation td-1 = gammaSn, where gamma is an empirical constant (about 10(-7) sec-1) and n is about 35. The supersaturation equation can successfully account for observations on the kinetics of cell sickling and is therefore used to estimate the increase in the delay time for sickling necessary to produce significant clinical benefit to patients with sickle cell disease.     

Hemoglobin Variants in Cuba

Non-ideal behavior of sickle cell hemoglobin (Hb S) and varying experimental conditions have made gelling kinetic studies difficult. A new gelation kinetic procedure is developed in this study. This technique uses low Hb S concentrations (20-fold less than in conventional methods) without the need for a temperature jump. Gelation progress, monitored with a newly developed turbidimetric procedure at 815 nm, was fitted to a mono-exponential and a sigmoid-E_max models. These models allowed precise definitions of gelation delay periods, rate of rapid Hb S gelation and a parameter, T_1/2 which combines information from the delay and rapid gelation stages. All these parameters vary linearly with Hb S concentration. The merits of using T_1/2 are discussed.     

Pairings and polarities of the 14 strands in sickle cell hemoglobin fibers.

Sickle cell anemia results from the formation of hemoglobin S fibers in erythrocytes, and a greater understanding of the structure of these fibers should provide insights into the basis of the disease and aid in the development of effective antisickling agents. Improved reconstructions from electron micrographs of negatively stained single hemoglobin S fibers or embedded fiber bundles reveal that the 14 strands of the fiber are organized into pairs. The strands in each of the seven pairs are half-staggered, and from longitudinal views the polarity of each pair can be determined. The positions of the pairs and their polarities (three in one orientation; four in the opposite orientation) suggest a close relationship with the crystals of deoxyhemoglobin S composed of antiparallel pairs of half-staggered strands.     

Studies on abnormal hemoglobins. VIII. The gelling phenomenon of sickle cell hemoglobin: its biologic and diagnostic significance

1. When sufficiently concentrated sickle cell hemoglobin containing solutionsare exposed to a constant stream of CO₂ gas, the hemolysates gel. This gellingphenomenon is indicative of the presence of S hemoglobin and cannot be obtainedwith any other type of human hemoglobin in the absence of S pigment. Thelowest S hemoglobin concentration (Gm. per cent) of a hemolysate at which thegelling phenomenon can still be elicited is designated as its lowest gelling point.

2. A simple apparatus was developed to analyze the gelling phenomenon understandardized conditions. It could be shown that the lowest gelling points ofhemolysates prepared from erythrocytes of the sickle cell trait (containing A +S hemoglobins), of the "C variant" (containing C + S hemoglobins), and fromsickle cell anemia cells (containing S + F hemoglobins) differ distinctly. Furtherexperiments suggest that the presence of A hemoglobin decreases the minimalamount of S pigment required for gel formation, and that type C hemoglobinreduces this amount even further. F hemoglobin seems to exert no significantinfluence on the gelling phenomenon. Serum albumin is also capable of decreasingthe amount of S hemoglobin required for gelation.

3. A sickled erythrocyte is visualized as an S hemoglobin tactoid or gel, specifically influenced by the companion pigment which interacts with the S compound.Thus, in the sickle cell trait, a positive sickling test is not only caused by thepresence of S hemoglobin, but also by its interaction with A hemoglobin. Onlyin the sickle cell anemia cells does sickling seem to depend solely upon the interaction of the S hemoglobin molecules.

4. The readily demonstrable differences of the lowest gelling points of hemolysates prepared from the various types of sickling red cells form the basis of thediagnostic gelling test which distinguishes sharply between sickle cell anemia andsickle cell trait erythrocytes. By this procedure atypical cases of sickle cell disease,for example, those whose erythrocytes contain C hemoglobin, may also bedetected.

Submitted on April 21, 1953 Accepted on May 25, 1953     

Structure of Fibers of Sickle Cell Hemoglobin

Electron microscope studies have been conducted on individual fibers of human deoxyhemoglobin S (sickle cell hemoglobin). The fibers are obtained by injection of gelled samples into a large excess of glutaraldehyde, which quickly stabilizes the fibers by cross-linking. The fibers are negatively stained with phosphotungstic acid or shadowed with platinum-carbon. The fibers are approximately 200 A in diameter, and display long and short helical striations with an opposite handedness. The long striations occur at an angle of about 15 degrees from the fiber axis, and complete one turn around the helix at a distance of about 2 x 10(3) A along the fiber axis. The short striations occur at an angle of about 80 degrees from the fiber axis, with a spacing of about 65 A, and complete one turn around the helix at a distance along the fiber axis of about 130 A. The structure of the fiber appears to be a sextuple helix in terms of the long striations, and a double helix in terms of the short striations. The shadowed samples are consistent with a left-handed screw sense for the short striations, thus implying a right-handed sense for the long striations. A structural model incorporating these features is compatible with the atomic structure of hemoglobin, with individual molecules oriented with their dyad axis of symmetry perpendicular to the fiber axis and their alpha(1)-beta(1) pseudo-dyad axis roughly parallel to the fiber axis. This orientation places the two beta-6 regions of each molecule (sites of the sickle cell mutation) in contact with the beta-6 regions of the molecules above and below along the long striations. Both the long and short striations are accounted for by individual hemoglobin molecules arranged in double helical arrays with 6.4 molecules per turn in each array.     

Electron microscopy of fibers and discs of hemoglobin S having sixfold symmetry

Aggregated forms of deoxyhemoglobin S were examined with a field emission transmission electron microscope. Images of isolated helical fibers were obtained from sickled cell lysates stained directly on the electron microscope grid. Optical and digital analyses of the electron micrographs showed that the fibers are similar to those characterized by J. T. Finch, M. F. Perutz, J. F. Bertles, and J. Döbler [(1973) Proc. Natl. Acad. Sci. USA 70, 718-722] in that they consist of stacked discs each composed of six hemoglobin molecules. The fibers exhibit an outer diameter of 160-170 Å and an inner diameter of about 60 Å with an axial spacing of 58 Å per disc. The fiber can be described as a helix consisting of 56 discs per helical turn. We observed discs of six hemoglobin molecules, which may be stable substructural components of the fibers. They were observed in preparations of hemoglobin fibers and exhibited 6-fold symmetry by power spectrum analysis. A reconstructed image of a disc digitally filtered for 6-fold symmetry has a maximum external diameter of ∼170 Å and a central hole of 60 Å diameter and is similar to the axial projection of a single disc from a low-resolution, three-dimensional reconstructed model of a fiber.     

Iron in sickle cell disease: a review why less is better

Sickle cell anemia (SCA) is an inherited disorder of hemoglobin synthesis that is characterized by life-long severe hemolytic anemia, attacks of pain crisis, and chronic organ system damage. A third of the hemolysis in SCA is intravascular and the resulting urinary losses of iron may lead to iron deficiency. There is no evidence of iron overload in SCA and iron deficiency may be more common than suspected, especially in men. Absence of bone marrow iron remains a gold standard for the diagnosis of iron deficiency in these patients. Although low serum ferritin is highly specific for the diagnosis of iron deficiency, its sensitivity is quite low in SCA because of non-specific elevation due to increased red cell turnover. The kinetics of sickling is strongly concentration dependent such that small decreases in the mean corpuscular deoxyhemoglobin-S concentration (MCHC-S) cause a substantial delay in sickle hemoglobin polymerization. Prolongation of the "delay time of gelation" in excess of the capillary transit time may allow the erythrocyte to traverse the capillary bed to escape to the arterial side before there is rheologic impairment of the erythrocyte from polymerization of sickle hemoglobin. Overt iron deficiency lowers the MCHC-S and thereby decreases the sickling tendency and the severity of hemolysis. The clinical improvement in SCA following the induction of iron deficient erythropoiesis by repeated phlebotomies or by erythrocytapheresis has been reported. Prospective controlled studies are needed to evaluate further, the therapeutic strategy of inducing controlled iron-deficient erythropoiesis in selected patients with SCA.     

Some properties of hemoglobin mobile (alpha 2 beta 2 73 Asp----Val)

A hemoglobin variant was identified as hemoglobin Mobile in which valine replaces the normal aspartic acid at beta 73. Studies of its oxygen equilibria and of its interactions in gelation when mixed with hemoglobin S were carried out. Hemoglobin Mobile had an oxygen affinity lower than that of hemoglobin A, as observed by others. However, in mixtures with hemoglobin S, hemoglobin Mobile appeared to impair gelation or increase solubility to a slightly greater extent than did hemoglobin A. Beta 73 is a known site of intermolecular interactions in polymers of hemoglobin S. Our studies suggest that the impairment of hemoglobin S polymer formation by altered intermolecular interactions is significantly less in Hb Mobile than in Hb Korle-Bu in which beta 73 is asparagine.     

Cross-sectional views of hemoglobin S fibers by electron microscopy and computer modeling.

Fibers of deoxyHb S have been investigated by thin-section electron microscopy, utilizing a tannic acid embedding procedure. On the basis of numerous measurements of cross-sectional center-to-center distances for adjacent fibers in pairs or arrays, fiber diameters (mean +/- SD) of 205 +/- 5 A in embedded cells and 212 +/- 8 A in embedded hemolysates were obtained. This is an agreement with values obtained by conventional embedding procedures [Crepeau, R. H., Dykes, G., Garrell, R. L. & Edelstein, S. J. (1978) Nature (London) 274, 616--617]. The use of tannic acid has resulted in improved resolution of fiber cross sections, revealing individual strands of Hb S molecules. Because the section thickness corresponds to approximately one-fifth of the fiber helical repeat distance, the strands in projection superimpose to form characteristic image patterns. Additional superposition patterns arise in sections taken at small deviations from perpendicularity to the longitudinal fiber axis. These patterns are consistent with the 14-strand structure for hemoglobin S fibers [Dykes, G., Crepeau, R. H. & Edelstein, S. J. (1978) Nature (London) 272, 506--510], as indicated by computer models of cross-sectional patterns for various thicknesses and angular deviations of sections.     

An increased Bohr effect in sickle cell anemia.

Recent findings that hemoglobin S gelation and sickling are pH-dependent and also influence oxygen affinity suggested that the red cells containing this hemoglobin variant might show an abnormal Bohr effect. We therefore studied the effects of pH variation on the in vitro oxygen affinity of whole blood from persons with sickle cell anemia (SS) and normal donors (at 37 degrees C and constant carbon dioxide tension of 40 mm Hg). The Bohr effect in SS blood was greatly increased only between blood pH 7.4 and 7.2 (cell pH 7.2 and 7.0, a shift that strongly affects gelation), with delta log p50/deltapH= - 0.92 to -0.99 (normal = -0.42 to -0.46). Thus a drop in SS blood pH below 7.4 in tissue capillaries yields twice the normal decrease in oxygen affinity and a large release of oxygen from red cells, whose risk of sickling is high. Even mild transient acidosis would seem hazardous for patients with sickling disorders.     

Viscometric and spectrophotometric measurements of hemoglobin S polymerization kinetics

The rates of polymerization and depolymerization of identical concentrated deoxygenated hemoglobin S (HbS) solutions following a rapid temperature change were examined by several methods. Two of these methods measured viscosity changes in either gently agitated (AGT) or nonagitated (NAGT) samples. The third method utilized a change in turbidity at 735 nm (SDT). By all three methods, a delay period, during which no observable change was detected, followed the temperature change. Gelation, as determined in nonagitated samples by a viscosity- based technique, occurred before or coincided with gelation as determined spectrophotometrically. The slope of the concentration dependence of the delay time is significantly decreased by agitation. Similar monitoring of the depolymerization reaction indicated the persistence of increased viscosity after observation of a marked decrease in turbidity.     

Divalent cations induce protofibril gelation

Soluble fibrin oligomers (protofibrils) undergo phase change merely by adding 1-2 mM Ca2+ or 25-100 microM Zn2+. The cation-induced "protofibrin" clots appear similar to normally formed fibrin gels. Maximal clot turbidity of protofibrin gels increases with cations in a concentration-dependent manner. Magnesium (less than 0.5 mM) is ineffective in inducing protofibril gelation. Turbidity and degree of polymerization (DP) [indirectly expressed as AT (activation time)] appear to be positively correlated, regardless of whether the divalent cation is Ca2+ or Zn2+. Cross sections of Ca2+-induced protofibrin fibers are approximately 6-18-fibrin-monomers-thick. With both Ca2+ and 40 microM Zn2+, fiber cross section increases to 30-50 monomers thick. Negatively stained Zn2+-and Ca2+-induced protofibrin gels exhibit banding periodicity of approximately 240 A, similar to that of normally generated fibrin gels. Regions of lateral merging of individual segments of the protofibrin leads to increased cross section of the fiber and forms a branch required for gelation. These findings indicate that the rate of coagulation and the ultimate thickness of the fibers both relate to lateral associative processes of protofibrils, which are augmented by physiologic concentrations of 2+ and Zn2+.     

Fibers, crystals, and other forms of HbS polymers in deoxygenated sickle erythrocytes

We have examined by electron microscopy the formation of fibers and crystals from sickle hemoglobin within sickle erythrocytes following deoxygenation during capillary storage from 1 to 132 days. Intracellular fibers were found on the first day and throughout the period of study. The fibers exhibited a diameter (mean +/- SD) of 17.4 +/- 0.62 nm and were aligned in the cell with a fiber-to-fiber spacing of 18.6 nm (x-axis) by 22.7 nm (y-axis). Between 65 and 132 days, extracellular hemoglobin crystals developed, with a lattice periodicity of 9.63 +/- 0.6 nm. Fibers and crystals coexist as separate structures. These results suggest that crystal formation upon storage of packed deoxygenated sickle erythrocytes may proceed via a phase of fiber dissolution followed by hemoglobin reassembly into extracellular crystals, rather than by a progressive alignment and direct fusion of existing fibers.     

Enhanced polymerization of recombinant human deoxyhemoglobin beta 6 Glu----Ile.

Polymerization of the deoxy form of sickle cell hemoglobin (Hb S; beta 6 Glu----Val) involves both hydrophobic and electrostatic intermolecular contacts. These interactions drive the mutated molecules into long fibrous rods composed of seven pairs of strands. X-ray crystallography of Hb S and electron microscopy image reconstruction of the fibers have revealed the remarkable complementarity between one of the beta 6 valines of each molecule (the donor site) and an acceptor site at the EF corner of a neighboring tetramer. This interaction constitutes the major lateral contact between the two strands in a pair. To estimate the relative importance of this key hydrophobic contact in polymer formation we have generated a polymerizing Hb with isoleucine at the beta 6 position (beta E6I) by site-directed mutagenesis. The mutated beta chains were produced in Escherichia coli and reassembled into functional tetramers with native alpha chains. Compared to native Hb S, the beta E6I mutant polymerizes faster and with a shortened delay time in 1.8 M phosphate buffer, indicating an increased stability of the nuclei preceding fiber growth. The solubility of the beta E6I mutant Hb is half that of native Hb S. Computer modeling of the donor-acceptor interaction shows that the presence of an isoleucine side chain at the donor site induces increased contacts with the receptor site and an increased buried surface area, in agreement with the higher hydrophobicity of the isoleucine residue. The agreement between the predicted and experimental differences in solubility suggests that the transfer of the beta 6 valine or isoleucine side chain from water to a hydrophobic environment is sufficient to explain the observations.     

Stereospecific Inhibitors of the Gelation of Sickle Hemoglobin

During the last decade there have been major advances in understanding the structure of the gel of deoxyhemoglobin S and the mechanism of its formation. These advances have allowed the development of a new strategy for the inhibition of gelation, i.e., stereo-specific competitive inhibitors of the polymerization process. For this purpose we have used equilibrium solubility measurements to study the effects of amino acids and peptides on deoxyhemoglobin S solubility. We have found that aromatic amino acids and short peptides containing these amino acids significantly increase solubility; longer peptides, however, decrease solubility by an excluded-volume related effect. Recently we have used computer graphics projections of the surface of the hemoglobin molecules in the crystal form to design peptide inhibitors. In addition, we have developed 13_C nuclear magnetic resonance spectroscopic methods to quantitate the gel of deoxyhemoglobin S in hemoglobin solutions and in cells. These spectroscopic methods allow us to study the mechanism of intracellular gelation and to test the effects of inhibitors on sickle cells.     

Introduction of cold light to endoscopy

It is the aim of the paper to describe how, 40 years ago, optic glass fibers were developed, and what has been K. Storz's contribution to the new technology. In 1951 the term "Cold Light" was used the first time for illumination of a French type film- and photoendoscope. In 1957 the gastroenterologist B. Hirschowitz at Ann Arbor, U.S.A. succeeded making glass fibers of high light-guiding properties. In 1961 the Cystoscope Makers Inc (ACMI) at New York using these fibers brought the first flexible gastroscope on the market, still equipped with a conventional electric lamp. But in 1960, the year before, the physicist's of ACMI, J. H. Hett and L. Curtiss built the first cold light endoscope using glass fibers for both light and images conduction. In the following years ACMI equipped all of his endoscopes with this new type of illumination. Not before 1963 did K. Storz and the other German manufacturers produce their first cold light cystoscopes. Not possessing the know-how of glass fiber manufacturing, they had to get their fibers from abroad. K. Storz transmitted the term "cold light", which before had been the label of his French-type endoscopes, to the new glass fiber illumination. He constructed an excellent light source for fiber illumination without having light cables of his own fabrication. That is why his name is intimately connected with cold light illumination. But, nevertheless, the invention of the new glass fiber illumination must be credited to B. Hirschowitz and the physicists of ACMI in the U.S.A.     

Effects of 2,3-diphosphoglycerate on the mechanical properties of erythrocyte membrane

Investigation by Schindler et al and Sheetz and Casaly have indicated that high (approximately 10 mmol/L) concentrations of 2,3- diphosphoglycerate (2,3-DPG) have a destabilizing effect on erythrocyte membrane and the membrane skeleton. We have investigated changes in the membrane mechanical properties that occur at elevated 2,3-DPG levels in both intact cells and ghosts. The membrane shear modulus, viscoelastic recovery time constant, critical force, "plastic" viscosity, and material relaxation time constant were measured by standard micropipette and flow channel techniques. Intact cells showed no change in properties at physiologic ionic strength and 2,3-DPG concentrations of about 20 mmol/L, except for an increase in membrane viscosity resulting from an increased cellular hemoglobin concentration that occurs when the 2,3-DPG concentration is elevated. At ionic strengths 20% below physiologic and 2,3-DPG concentrations of approximately 20 mmol/L, decreases in membrane shear modulus and membrane viscosity were observed. In ghosts, no changes in these properties were observed at a 2,3-DPG concentration of 10 mmol/L and ionic strengths as low as 25% below physiologic, but a decrease in the force required to form tethers (critical force) was observed at physiologic ionic strength. The decrease in membrane shear modulus and viscosity of intact cells and the reduced critical force in ghosts are consistent with the results of other investigators. However, the difference in the effects of 2,3-DPG on ghosts and intact cells indicates that the effects of 2,3-DPG depend strongly on the conditions of the experiment. It appears unlikely that 2,3-DPG affects erythrocyte membrane material properties under physiologic conditions.