ya ARN ANS malt ee a emt RS emA eae , yh Errors in Human Hemoglobin as a Function of Age att Hemoglobin was separated from nonhemoglobin proteinsfirst by molecular seiving over Sephadex G-200 (columnsize, 1.9 x 120 cm; buffer, 50 mm sodium phosphate, pH 5.8, saturated with CO; flow rate, 6 mi/h. Fractions containing the majority of the carbon monoxyhemoglobin (60-90 ml) were pooled and placed on a column (3.7 x 100 cm) of carboxymethyl cellulose (Whatman CM 23), which was washed according to manufacturer’s instruc- tions, and equilibrated with 50 mm sodium phosphate buffer, pH 5.8 (9 parts NaH,PO, and ! part Na,HPO,). Carbon monoxyhemoglobin was eluted using a nonlinear gradient of 1.7 Lof 50 mm sodium phosphate buffer, pH 5.8, in a 2-liter reagent bottle, and 1.0 1 of 50 mm Na,PO,, in a J-liter reagent bottle (both solutions were bubbled briefly with CO); the flow rate was 1.5 ml/min. Carbon monoxyhemoglobin eluting between pH 6.7 and 6.95 was pooled and concentrated by pressure dialysis to a volume ofless than 20 ml to remove much ofthe salt and then diluted to 40 mi with distilled water; then the pH was adjusted to 6.0 by adding 0.5 m H,PQ,, and the carbon monoxyhemoglobin was converted to methemoglobin by making the solution 1 ma in K,Fe(CN),. The second chromatography was performed on one of two columns of carboxymethyl cellulose (2.5 x 100 cm) using a gradient identical (one gradient supply for two columns) to the previous one but without CO, Methemoglobin eluting between pH 7.05 and 7.35 was concentrated,’ diluted, and pH adjusted as before. Methemoglobin was converted to cyanmethemoglobin by making the solution J matin KCN and the cyanmethemoglobin was chromatographed on one of four columns (1.8 x 100 cm) of carboxymethyl cellulose using a gradientidentical to the previous one but now also containing 10-' mM KCN. Cyanmethemoglobin eluting between pH 6.7 and 7.0 was concentrated by pressure dialysis and used to prepare globin [10], which in a few cases was further separated into «- and B-chains[I]. Globin or chains were hydrolyzed in 6 N HCl for 21 h at H10°C. 2% of each hydrolysate was used to determine by amino acid analysis {11} the quantity of globin or chain in each sample. For reference markers, tracer amounts of L-leucine-“C and L-isoleucine-4,5H (New England Nuclear) were added to the remainder of the hydrolysate before it was chromatographed on a preparative ion exchange column (1.9 x 60 cm) of 8% crosslinked sulfonated styrene divinylbenzene copolymer (Beckman Type 150A) _ to separate and recover the isoleucine in the hydrolysate. The resin was equilibrated with 0.2 m sodium citrate buffer, pH 3.25 (Beckman), and the amino acids were eluted with 0.2 M sodium citrate buffer, pH 4.25 (Beckman); the flow rate was 68 ml/h. 3-min fractions of the eluate were collected and the radiotracers were used to locate fractions containing isoleucine but excluding —-Sshtzbrt . * Aye