Hemoglobins that polymerize are not uncommon among amphibians, reptiles, and elasmobranchs, but are rare among teleosts and humans. They are useful models for the study of protein structure, function, and molecular evolutionary relationships. The freshwater teleost Hoplias malabarica (I), the marine teleost Lophius americanus (2), and the clown fish, Amphiprion (3) are the only bony fishes reported to have hemoglobins that polymerize. Hoplias and Lophius polymerize by forming intermolecular disulfide bonds, while the polymerization mechanism for the clown fish is, as yet, unknown. Among humans, hemoglobins Ta-Li, Mississippi, and Pot-to Alegre (4, 5, 6) have a single mutation involving the substitution of a cysteine residue at different loci on the globin chain. We now report that the ocean pout has a single hemoglobin with six sullhydryl groups per tetramer and polymerizes, in vitro, after oxidation with potassium ferricyanide (K,Fe(CN),) or cupric chloride (CuClJ. Procedures for hemoglobin purification by DE-52 cellulose chromatography, molecular weight estimations by gel filtration on G100 columns, and urea gel electrophoresis for globin subunit analysis have been described (2). Hemoglobin was incubated with a twoto threefold molar excess of CuCIZ for 15 min in an ice bucket. The excess copper was removed by dialysis or by gel filtration from Sephadex G-25 columns equilibrated with 0.1 M K-PO., buffer, pH 7.4. Thereafter 20-40 mg of the hemoglobin was applied to and eluted from G100 columns. One-ml fractions were collected and the absorbance of each measured at 540 nm. The approximate polymer molecular weight was estimated from the elution volume of the peak tube. Hemoglobin solutions (0.52.0 g/%) were also oxidized by adding l-3 mg of potassium ferricyanide per ml of hemoglobin. Iodoacetamide (IAA)-alkylated hemoglobin ( 15 rmoles/mg hemoglobin) was dialyzed overnight, centrifuged, and subsequently treated with CuC12. In a separate xperiment, a 60-fold molar excess of beta-mercaptoethanol (B-ME) was added to CuCl*-treated hemoglobin, incubated for 1 h, and dialyzed for 3 h before G100 chromatography. Untreated hemoglobin and hemoglobin treated with CuCIZ alone served as controls and were dialyzed for the same period. The gel filtration profile of known molecular weight proteins is shown in Figure 1 A. The inset graph is a plot of the log of the molecular weight of each standard versus the elution volume of the peak tube. The linear relationship indicates that the peak tube elution volume provides a reasonable stimate of the protein molecular weight between 16,000 and 150,000 Da. (The molecular weight fractionation range for Sephadex G-100 is 4 X 10’ and 1.5 X 105.) .The effect of CuCl* on pout hemoglobin is shown in Figure 1 B. In the absence of the cation, the peak tube eluted at a volume comparable to human hemoglobin, suggesting anormal molecular weight for pout hemoglobin of approximately 60,000. The addition of CuC12 results in a polymer with a minimum molecular mass of 100 to 155 kDa, which corresponds to aggregates of, perhaps, two to three tetramers. Cupric-chloride-induced polymer formation may be inhibited by prior incubation with IAA (Fig. 1C). Hemoglobin alkylated with IAA elutes at the same volume as normal pout and human hemoglobin, but in the absence of IAA, cupric-chloride-induced