Saturation-transfer electron paramagnetic resonance detection of anisotropic motion by sickle hemoglobin molecules in the polymer state

The motional behavior of spin-labeled deoxygenated sickle hemoglobin has been studied by using both 9- and 35-GHz saturation-transfer electron paramagnetic resonance (EPR). Using spectral subtraction techniques and saturation-transfer EPR parameter correlation plots, we find that the saturation-transfer EPR spectra for the sickle hemoglobin gel state at high temperature and high hemoglobin concentration cannot be described as a simple superposition of spectra from immobilized hemoglobin plus solution-state hemoglobin but instead suggest that the individual sickle hemoglobin molecules exhibit limited, anisotropic, rotational oscillation within the polymer fiber. The spectra also imply that the symmetry axis for sickle hemoglobin rotational oscillation is approximately coincident with the nitroxide z axis of the covalently attached spin-label. We suggest that this anisotropic rotational motion may be produced by one or two of the known intermolecular contact sites within the sickle hemoglobin fiber acting as strong intermolecular binding sites, and producing "motional alignment" within the fiber; determining the location of the strong binding site should be important in focusing the future development of antisickling agents.     

Intermolecular interactions of oxygenated sickle hemoglobin molecules in cells and cell-free solutions.

We have measured the intermolecular interactions of oxygenated sickle hemoglobin molecules in cells and in cell-free solutions, and have compared the results with similar data for liganded normal adult hemoglobin. The experiments involve the measurement of the spin-lattice relaxation time T1 of protons of solvent water molecules, as a function of an externally applied static magnetic field. From such data, one can derive a correlation time tauc, for each sample, which is a measure of the time taken for a hemoglobin molecule to randomize its orientation due to Brownian motion. Thus tauc is a measure of the freedom of rotational motion, on a molecular or microscopic level, of hemoglobin molecules. Intermolecular interactions will reduce this freedom of motion and lengthen tauc. We find that oxygenated sickle hemoglobin molecules have an additional intermolecular interaction not found for normal hemoglobin. This extra interaction is increased by the presence of either inorganic phosphate or diphosphoglycerate, and is greater for sickle hemoglobin within cells than in cell-free solutions. By comparing the present results with published data on the viscosity of oxygenated sickle and normal hemoglobin, we conclude that, at concentrations comparable to intracellular values, oxygenated sickle hemoglobin molecules form aggregates several tetramers in size. The possibility exists that these aggregates are the earliest stage of fiber formation itself, the physical basis of the sickling phenomena.     

Oxygen equilibrium of emulsified solutions of normal and sickle hemoglobin.

Emulsions containing microdroplets of concentrated solutions of normal or sickle hemoglobin dispersed in a continous oil phase have been prepared, and the aggregation of sickle hemoglobin within microdroplets in a deoxygenated emulsion demonstrated. The equilibrium oxygen saturation of hemoglobin in the emulsions has been measured as a function of the partial pressure of oxygen by a novel spectrophotometric technique which corrects for the scattering of light in the emulsion. The half-saturation oxygen pressure and cooperativity of oxygen binding are substantially greater in concentrated (27 g/dl) solutions of sickle hemoglobin than in solutions of non-aggregating hemoglobin. The shape of the oxygen equilibrium curve of concentrated sickle hemoglobin is qualitatively discussed in terms of a previously proposed model (1).     

Electron paramagnetic resonance probing of macromolecules: a comparison of structure-function relationships in chymotrypsinogen, -chymotrypsin and anhydrochymotrypsin

Chymotrypsinogen, chymotrypsin and anhydrochymotrypsin have been covalently spin-labeled by an analog of bromoacetamide, and the latter two proteins have been labeled by an analog of 1-chloro-3-tosylamido-4-phenyl butanone. The electron paramagnetic resonance spectra of the labeled proteins indicate protein conformational changes accompanying (1) activation of the zymogen and (2) the binding of protons and substrates by the native and anhydro enzymes, and tertiary structural differences between these protein forms which are at once informative and predictable. A spin-label linked to the thioether side-chain of methionine 192 in Chymotrypsinogen may be in contact with a hydrophobic surface. This interaction is lost upon zymogen activation with little change in the isotropic rotational freedom of the nitroxide group. The rotational freedom of the group increases sigmoidally with pH; a spectral dependence upon an ionizing group (pK_a = 8.9) is demonstrated. The binding of indole to the labeled enzyme raises the pK_a of the ionizing group to 10.2. A spin-label linked to histidine 57 in chymotrypsin senses both indole binding and pH changes directly; the same label in anhydrochymotrypsin responds directly only to changes in pH. Neither histidine-labeled derivative exhibits enzymic activity. The electron paramagnetic resonance spectra of these two labeled proteins at high pH indicate a decrease in the motional freedom of the spin label. The spectral data show that the conformational state of the labeled zymogen is not similar to the high-pH conformational state of the labeled enzyme. Furthermore, the pH-dependent conformational transition of labeled chymotrypsin requires neither the serine 195 hydroxyl nor the histidine 57 imidazole, since the transition occurs normally in derivatized and chemically modified protein forms. The chemical reactivity of histidine 57 in anhydrochymotrypsin is evaluated and the catalytic activities of two histidine alkylated enzymes are compared.     

Protein kinase inhibitor-(6-22)-amide peptide analogs with standard and nonstandard amino acid substitutions for phenylalanine 10. Inhibition of cAMP-dependent protein kinase

The minimal structure in the heat-stable inhibitor protein of cAMP-dependent protein kinase required for a low nanomolar potency of inhibition is the peptide Thr6-Tyr-Ala-Asp-Phe-Ile-Ala-Ser-Gly-Arg-Thr-Gly-Arg-Arg-Asn-Ala-+ ++Ile22-NH2 (PKI-(6-22)-amide). While primary structural determinants for interaction with the protein kinase are distributed throughout the 17 residues of this peptide, we have previously shown that phenylalanine 10 in the NH2-terminal portion is a particularly important determinant for high affinity binding (Glass, D. B., Cheng, H.-C., Mende-Mueller, L., Reed, J., and Walsh, D. A. (1989) J. Biol. Chem. 264, 8802-8810). To investigate this requirement further, peptide analogs of PKI-(6-22)-amide in which various natural and nonstandard amino acids are substituted for phenylalanine 10 have been synthesized and tested for inhibitory potency against the catalytic subunit of the protein kinase. Consistent with the importance of the hydrophobicity of phenylalanine, an alanine 10 substitution analog exhibited a 270-fold decrease in inhibitory potency, whereas the leucine 10 analog lost only 33-fold in activity as compared to the parent peptide PKI-(6-22)-amide. Peptides containing the spatial conformation analogs D-phenylalanine, homophenylalanine, or phenylglycine were 60-120-fold less potent than the parent peptide. Peptides containing various para-substituted phenylalanines at position 10 were only 5-11-fold less potent. One exception to this was (4'-azidophenylalanine 10)PKI-(6-22)-amide, which was nearly equipotent with the parent inhibitor. The most potent analogs were those peptides containing highly aromatic residues at position 10. The 2'-thienylalanine 10, tryptophan (formyl) 10, tryptophan 10, and the 1'-naphthylalanine 10 analogs were 3-fold less potent, equipotent, slightly more potent, and 4-fold more potent than the parent peptide inhibitor, respectively. We conclude that phenylalanine 10 in PKI-(6-22)-amide, and presumably in the native protein inhibitor, interacts through specific hydrophobic and/or aromatic binding to a hydrophobic pocket or cleft near the active site of the protein kinase.     

Myosin regulatory domain orientation in skeletal muscle fibers: application of novel electron paramagnetic resonance spectral decomposition and molecular modeling methods

Reorientation of the regulatory domain of the myosin head is a feature of all current models of force generation in muscle. We have determined the orientation of the myosin regulatory light chain (RLC) using a spin-label bound rigidly and stereospecifically to the single Cys-154 of a mutant skeletal isoform. Labeled RLC was reconstituted into skeletal muscle fibers using a modified method that results in near-stoichiometric levels of RLC and fully functional muscle. Complex electron paramagnetic resonance spectra obtained in rigor necessitated the development of a novel decomposition technique. The strength of this method is that no specific model for a complex orientational distribution was presumed. The global analysis of a series of spectra, from fibers tilted with respect to the magnetic field, revealed two populations: one well-ordered (+/-15 degrees ) with the spin-label z axis parallel to actin, and a second population with a large distribution (+/-60 degrees ). A lack of order in relaxed or nonoverlap fibers demonstrated that regulatory domain ordering was defined by interaction with actin rather than the thick filament surface. No order was observed in the regulatory domain during isometric contraction, consistent with the substantial reorientation that occurs during force generation. For the first time, spin-label orientation has been interpreted in terms of the orientation of a labeled domain. A Monte Carlo conformational search technique was used to determine the orientation of the spin-label with respect to the protein. This in turn allows determination of the absolute orientation of the regulatory domain with respect to the actin axis. The comparison with the electron microscopy reconstructions verified the accuracy of the method; the electron paramagnetic resonance determined that axial orientation was within 10 degrees of the electron microscopy model.     

Direct measurement of the internal viscosity of sickle erythrocytes as a function of cell density

The rotational dynamics of TEMPAMINE can be used to study directly the intracellular environment. The extracellular signal from TEMPAMINE is broadened away by the use of potassium ferricyanide which does not enter the cell. The EPR signal which results when 1 mM TEMPAMINE, 120 mM ferricyanide, and erythrocytes are mixed together arises from TEMPAMINE only in the intracellular aqueous space. The relative viscosity measured by the motion of TEMPAMINE in various control environments is: water at 37 degrees C = 1; human plasma at 37 degrees C = 1.1; internal aqueous environment of washed erythrocytes or whole blood at 37 degrees C = 4.92 +/- 0.32. Erythrocytes can be fractionated by density. In sickle-cell anemia (SS), the percentage of cells we find with density greater than 1.128 g/ml is 15-40%, in normals (AA) and sickle trait (AS) 1%. By direct spin-label measurements with TEMPAMINE we show, for the first time, that the relative internal viscosity (eta mu) of these dense erythrocytes is markedly elevated and density-dependent. Our results show that (1) eta mu increases with increasing cell density; (2) eta mu obtained from sickle cells is higher than eta mu obtained from normal cells at a given density, and this effect is greater at 37 degrees C than at 20 degrees C; (3) eta mu is proportional to MCHC, but eta mu in erythrocytes is higher than eta mu obtained from in vitro preparations of hemoglobin S at equivalent concentrations. We conclude that the relative internal viscosity of erythrocytes is affected by three factors: the state of cell hydration, the amount of hemoglobin polymer present, and the potential interactions of the cell membrane with intracellular hemoglobin.     

A new bifunctional spin-label suitable for saturation-transfer EPR studies of protein rotational motion

A new bifunctional spin-label (BSL) has been synthesized that can be immobilized on the surface of proteins, allowing measurement of rotational motion of proteins by saturation-transfer electron paramagnetic resonance (STEPR). The spin-label contains a photoactivatable azido moiety, a cleavable disulfide, and a nitroxide spin with restricted mobility relative to the rest of the label. The label reacts with surface lysine residues modified with beta-mercaptopropionate. Bifunctional attachment is achieved by photoactivation of the azido group. Any spin-label that remains monofunctionally attached after photolysis is removed by reduction of the disulfide. Only bifunctionally attached BSL remains on the protein. Hemoglobin was used to test the utility of the BSL in STEPR by comparison with hemoglobin modified with maleimide spin-label (MSL), a commonly used standard for the STEPR technique. MSL is a monofunctional spin-label which is fortuitously immobilized by local protein structure within hemoglobin. The BSL labeling of hemoglobin did not significantly affect the quaternary structure of hemoglobin as determined by gel filtration chromatography. The conventional EPR spectra of the mono- and bifunctionally attached BSL-hemoglobin were similar to the MSL-hemoglobin spectrum, indicating that both forms of BSL were rigidly bound to hemoglobin. In contrast, the spectrum obtained by reaction of modified hemoglobin lysine residues with MSL indicated that these labels were highly mobile. The monofunctionally attached BSL was mobilized upon octyl glucoside addition whereas bifunctionally attached BSL was only slightly mobilized, suggesting that hydrophobic interactions immobilize the monofunctionally attached label on hemoglobin. The response of STEPR spectra of mono- and bifunctionally attached BSL-hemoglobin to changes in hemoglobin rotational correlation time was similar to the MSL-hemoglobin over the range of 10(-5)-10(-3) s. The spectra of bifunctionally attached BSL indicated slightly less motion than corresponding spectra for MSL or monofunctionally attached BSL. The new BSL is a good reporter of protein rotation and does not require unique protein structures for its immobilization on the protein. Thus, the BSL should be more generally applicable for STEPR studies of membrane protein rotation than existing monofunctional spin-labels.     

Modification of isolated α and β chains of human hemoglobin by the metal tetrasulfonated phthalocyanines. Studies on hybrid phthalocyanine-heme intermediates

α and β chains of hemoglobin have been modified with cobalt(II) tetrasulfonated phthalocyanine in place of heme. They display properties very similar to those of iron(II) phthalocyanine modified α and β chains. Mixed together they form tetrameric cobalt(II) phthalocyanine hemoglobin.Incorporation of Co(II)L into α and β globins results in stabilization of the protein structure, which is shown by a marked increase in its helicity content. Cobalt phthalocyanine substituted α and β chains are able to combine reversibly with oxygen giving more stable oxygenated species than their native analogues. The rate of both processes is lower in the case of the modified α chain. Recombination of the phthalocyanine α and β chains with the alternate heme containing chains give tetrameric hybrid hemoglobins. These comprise two phthalocyanine modified subunits and two heme containing subunits. The helicity content of the tetrameric hybrid hemoglobin calculated for one subunit is lower that the arithmetic mean of helicities for its isolated subunits. This suggests a destabilizing chain-chain interaction within the tetramer. Unlike in the separated subunits, oxygen binding by hybrid hemoglobins is irreversible. Deoxygenation by argon bubbling leads to the formation of inactive species which in oxygen atmosphere undergo irreversible oxidation with destruction of the complex.     

Mixed gelation theory. Kinetics, equilibrium and gel incorporation in sickle hemoglobin mixtures.

This paper outlines a theoretical formalism for describing the gelling behavior of sickle cell hemoglobin in mixtures with other hemoglobin and non-hemoglobin proteins. Experimental applications are reported for hybridized and unhybridized mixtures of HbS (sickle hemoglobin), HbA (adult hemoglobin), HbF (fetal hemoglobin), and HbC Harlem. The theory is a general one based on a modification of the sol—gel phase equilibrium equation to take into account the varying tendencies of different hemoglobin species to promote gelation, and specific hemoglobin interactions are encoded in gelling coefficients which quantify gelling capability. Gelling coefficients for the hemoglobin species dealt with here are evaluated by measuring incorporation into the polymer phase in S-A, S-F, and S-C_H mixtures. Given this information, the theory is found to provide accurate prodictions for the equilibrium gelling behavior of the calibrating pairs themselves when they are hybridized or unhybridized, for gelation kinetics in diverse mixtures of these species taken two, three and four at a time, for the anomalous equilibrium and kinetic gelling behavior of A- C_H mixtures, and it also accounts for a variety of results previously published by others. Apparently, given the gelling coefficients for any mutant hemoglobin, one can compute gelling behavior (equilibrium, kinetics, incorporation, etc.) in any specified mixture with any other known hemoglobin(s). The gelling coefficients for any mutant hemoglobin depend upon, and therefore provide information about, gel interactions at the mutant site. From the gelling coefficients one can also obtain the change in free energy of interaction in the gel due to the altered residue. Experimental approaches are described which allow an analysis for the gelling coefficients of any mutant hemoglobin to be performed in a few hours.     

Distribution of IgG subclass antibodies specific for Plasmodium falciparum glutamate-rich-protein molecule in sickle cell trait children with asymptomatic infections

Polymorphism in the beta-globin gene (hemoglobin S) has been associated with protection against severe forms of malaria. In a cross-sectional study, 180 young Gabonese children with and without sickle cell trait and harboring asymptomatic Plasmodium falciparum infections, were assessed for the responses to recombinant protein containing the conserved region of glutamate-rich protein (GLURP). We reported increased age-dependence of antibody prevalence and levels of total IgG (p<0.0001), IgG1 (p=0.009), and IgG3 (p<0.03) antibodies to GLURP with a cut-off at 5 years of age. Whatever the hemoglobin type, cytophilic antibodies (IgG1 and IgG3) were prevalent, but GLURP-specific IgG4 antibodies were detected at significantly (p<0.05) lower levels in HbAS children. We showed that the distribution of non-cytophilic IgG antibodies differs according to the hemoglobin type and to the malaria antigens tested. This may have possible implication for the clearance of malaria parasites and for protection against severe malaria.     

Nitric oxide transport on sickle cell hemoglobin: Where does it bind?

We have recently reported that nitric oxide inhalation in individuals with sickle cell anemia increases the level of NO bound to hemoglobin, with the development of an arterial-venous gradient, suggesting delivery to the tissues. A recent model suggests that nitric oxide, in addition to its well-known reaction with heme groups, reacts with the β-globin chain cysteine 93 to form S-nitrosohemoglobin (SNO-Hb) and that SNO-Hb would preferentially release nitric oxide in the tissues and thus modulate blood flow. However, we have also recently determined that the primary NO hemoglobin adduct formed during NO breathing in normal (hemoglobin A) individuals is nitrosyl (heme)hemoglobin (HbFe^IINO), with only a small amount of SNO-Hb formation. To determine whether the NO is transported as HbFe^IINO or SNO-Hb in sickle cell individuals, which would have very different effects on sickle hemoglobin polymerization, we measured these two hemoglobin species in three sickle cell volunteers before and during a dose escalation of inhaled NO (40, 60, and 80 ppm). Similar to our previous observations in normal individuals, the predominant species formed was HbFe^IINO, with a significant arterial-venous gradient. Minimal SNO-Hb was formed during NO breathing, a finding inconsistent with significant transport of NO using this pathway, but suggesting that this pathway exists. These results suggest that NO binding to heme groups is physiologically a rapidly reversible process, supporting a revised model of hemoglobin delivery of NO in the peripheral circulation and consistent with the possibility that NO delivery by hemoglobin may be therapeutically useful in sickle cell disease.     

A path analysis of the causal relationships between red cell glycolytic intermediates, ADP and ATP in sickle cell anemia

The statistical relationships among the glycolytic intermediates (GI)) of the Embden-Meyerhof pathway, adenine nucleotides (ANs) and various hematological measures were estimated for 34 sickle cell anemia patients. Heterogeneity in linear and quadratic regressions of hemoglobin and hematocrit, both singly and jointly, on the GI and AN variables implied 1) that any single formula to standardize optical density measures of the GIs and ANs on a per gram hemoglobin or per liter cell water basis would not uniformly remove hemoglobin and hematocrit effects: 2) that ignoring significant hematological effects could bias the estimates of correlation among GIs and ANs; and 3) that hemoglobin and hematocrit measures do not reflect the same source of variability. The correlations among the GIs and ANs, after adjustment for hematological variability, were analyzed by path analysis to determine which of five proposed path models for cause and effect relationships were compatible with the data. AMP had a greater influence on ADP (coefficient of determination (CD) = 23%) than all the GIs together, while G6P and ADP influenced ATP variability the most (CD = 33% and 12%). The contributions of unknown factors to ADP and ATP variability were large for all models (CD = 56--77%) possibly due to stress of sickle cell disease. The path model with AMP and the four GIs (G6P, F6P, FDP, DHAP) influencing ADP variation, and the same GIs and ADP influencing ATP was the model most compatible with the data.     

Antioxidant and pro-oxidant effects of red wine and its fractions on Cu(II) induced LDL oxidation evaluated by absorbance and chemiluminescence measurements

We have recently reported that nitric oxide inhalation in individuals with sickle cell anemia increases the level of NO bound to hemoglobin, with the development of an arterial-venous gradient, suggesting delivery to the tissues. A recent model suggests that nitric oxide, in addition to its well-known reaction with heme groups, reacts with the β-globin chain cysteine 93 to form S-nitrosohemoglobin (SNO-Hb) and that SNO-Hb would preferentially release nitric oxide in the tissues and thus modulate blood flow. However, we have also recently determined that the primary NO hemoglobin adduct formed during NO breathing in normal (hemoglobin A) individuals is nitrosyl (heme)hemoglobin (HbFe^IINO), with only a small amount of SNO-Hb formation. To determine whether the NO is transported as HbFe^IINO or SNO-Hb in sickle cell individuals, which would have very different effects on sickle hemoglobin polymerization, we measured these two hemoglobin species in three sickle cell volunteers before and during a dose escalation of inhaled NO (40, 60, and 80 ppm). Similar to our previous observations in normal individuals, the predominant species formed was HbFe^IINO, with a significant arterial-venous gradient. Minimal SNO-Hb was formed during NO breathing, a finding inconsistent with significant transport of NO using this pathway, but suggesting that this pathway exists. These results suggest that NO binding to heme groups is physiologically a rapidly reversible process, supporting a revised model of hemoglobin delivery of NO in the peripheral circulation and consistent with the possibility that NO delivery by hemoglobin may be therapeutically useful in sickle cell disease.     

Effects of amino acids on gelation kinetics and solubility of sickle hemoglobin

Ten amino acids have been studied for their effects on the gelation of sickle hemoglobin using the recently developed assay of Hofrichter, Ross and Eaton. By monitoring kinetics and using high speed sedimentation, the rate and extent of gelation are directly measured. Of the amino acids tested, only phenylalanine significantly inhibited the gelation of sickle hemoglobin. The systematic study of the effects of additives, such as amino acids, on gelation serves as a basis for the study of potential non-covalent inhibitors of sickling.     

Inhibition of invitro sickling by liposome-mediated transport of amino acids into intact human red blood cells

To investigate the role of phenylalanine and tryptophane as potential antisickling agents in intact human SS-red blood cells a liposomal transport system was employed to transfer phenyl-alanine or tryptophane into intact SS-red blood cells. Aromatic amino acids and short peptides containing phenylalanine have been demonstrated to increase the minimum gelling concentration and solubility of deoxy-hemoglobin S in aqueous solution. However, these compounds do not cross the red blood cell membrane under usual incubation conditions. Incorporation of phenylalanine or tryptophane into intact SS-red blood cells $\text{via}$ liposomal transport system markedly inhibited the $\text{in}\text{vitro}$ sickling of deoxy-hemoglobin S. These findings raise the possibility that a nontoxic liposomal transport system which facilitates incorporation of antisickling agents into intact SS-RBC may have significant therapeutic implications in the treatment of sickle cell disease.     

The hepatic PP1 glycogen-targeting subunit interaction with phosphorylase a can be blocked by C-terminal tyrosine deletion or an indole drug

The inhibition of hepatic glycogen-associated protein phosphatase-1 (PP1-G(L)) by glycogen phosphorylase a prevents the dephosphorylation and activation of glycogen synthase, suppressing glycogen synthesis when glycogenolysis is activated. Here, we show that a peptide ((280)LGPYY(284)) comprising the last five amino acids of G(L) retains high-affinity interaction with phosphorylase a and that the two tyrosines play crucial roles. Tyr284 deletion abolishes binding of phosphorylase a to G(L) and replacement by phenylalanine is insufficient to restore high-affinity binding. We show that a phosphorylase inhibitor blocks the interaction of phosphorylase a with the G(L) C-terminus, suggesting that the latter interaction could be targeted to develop an anti-diabetic drug.     

Reduced water exchange in sickle cell anemia red cells: a membrane abnormality

We have measured the diffusional water permeability of sickle cell anemia red blood cells under isotonic conditions using pulsed nuclear magnetic resonance (NMR) techniques. We have found that the equilibrium diffusional permeability for sickle cells is about 1.61.10(-3) cm/s, or about 60% of the value measured for normal cells. This abnormality is not related to the heterogeneity generally found in cell populations in sickle red cells with different mean corpuscular hemoglobin concentrations. We speculate that the abnormality of water exchange under isotonic conditions in sickle cells reflects an alteration of membrane proteins responsible for water exchange, possibly caused by oxidation of Band 3 proteins.     

Aggregation of normal and sickle hemoglobin in high concentration phosphate buffer

Sickle cell disease is caused by a mutant form of hemoglobin, hemoglobin S, that polymerizes under hypoxic conditions. The extent and mechanism of polymerization are thus the subject of many studies of the pathophysiology of the disease and potential treatment strategies. To facilitate such studies, a model system using high concentration phosphate buffer (1.5 M-1.8 M) has been developed. To properly interpret results from studies using this model it is important to understand the similarities and differences in hemoglobin S polymerization in the model compared to polymerization under physiological conditions. In this article, we show that hemoglobin S and normal adult hemoglobin, hemoglobin A, aggregate in high concentration phosphate buffer even when the concentration of hemoglobin is below the solubility defined for polymerization. This phenomenon was not observed using 0.05 M phosphate buffer or in another model system we studied that uses dextran to enhance polymerization. We have used static light scattering, dynamic light scattering, and differential interference contrast microscopy to confirm aggregation of deoxygenated and oxygenated hemoglobins below their solubility and have shown that this aggregation is not observable using turbidity measurements, a common technique for assessing polymerization. We have also shown that the aggregation increases with increasing temperature in the range of 15 degrees -37 degrees C and that it increases as the concentration of phosphate increases. These studies contribute to the working knowledge of how to properly apply studies of hemoglobin S polymerization that are conducted using the high phosphate model.