THE JWKNAL OF HIOLOGICAL CHEMWHY Vol. ‘51, No. 20, Issue 01’ October “5. pp. 6226-6”31, 1976

Printed in U.S.A.

Methylisocyanate as an Antisickling Agent and Its Reaction with Hemoglobin S*

(Received for publication, April 15, 1976)

CHOONC K. LEE

From the Center for Sickle Cell Disease and Department of Pediatrics and Child Health, College of Medicine, Howard University, Washington, D. C. 20059

Reviewing the reaction of potassium cyanate, an antisickling agent, with a-amino groups of hemoglobin, it was found that the reaction was a slow process and requires a large excess of the reagent. The reason for the slow reaction rate of carbamylation of hemoglobin by cyanate is that cyanate itself does not react with hemoglobin. It is rather isocyanic acid, the reactive species, that reacts with hemoglobin.

Since the pK of isocyanic acid is 3.8, only one out of 4,000 cyanate ions is present as isocyanic acid at physiological pH. Therefore, it appears that a large excess of cyanate ions is required to achieve the carbamylation of hemoglobin S, both in uiuo and in uitro. Furthermore, the pH optimum for carbamylation of carboxyhemoglobin and deoxyhemoglobin is 5.5 and, at pH 7.4, the reaction velocity drops to one-third for carboxyhemoglobin and one-half for deoxyhemoglobin. To seek an approach to reduce the dosage of cyanate and to increase the reaction velocity, an isocyanate derivative, methylisocyanate which is already in the reactive form, was tested for its antisickling activity and its reaction with hemoglobin S. It was found that methylisocyanate had antisickling activity and that only a stoichiometric amount to 2-fold excess of the reagent over hemoglobin S a-amino groups was required to prevent the sickling of erythrocytes. Methylisocyanate-treated sickle erythrocytes showed an increased oxygen affinity compared to untreated. Methylisocyanate reacted with a-amino groups of hemoglobin S and the reaction was complete in less than 1 min. Methylcarbamylated hemoglobin S had a higher minimum gelling concentration than the untreated hemoglobin S. There was no detectable reaction of free sulfhydryl and t-amino groups of hemoglobin S with methylisocyanate. These results indicate that methylisocyanate, and probably other isocyanate derivatives, possesses powerful antisickling activity.

Potassium cyanate was reported to have antisickling effect, in uitro (1). It reacts with a-amino groups of hemoglobins S (1, 2). The carbamylated Hb S’ has higher oxygen affinity than the untreated and reduces the aggregation of Hb S (3). It was found that 0.1 to 1.0 carbamyl group per hemoglobin tetramer was required for this antisickling effect (1). Carbamylation of Hb S at pH 7.4 and 37” with 400-fold excess of cyanate introduced 1.5 carbamyl groups in 5 min. Isocyanic acid, the reactive species of cyanate, was found to be a carbon dioxide analog (2). However, carbon dioxide reacts with hemoglobin much faster than cyanate does, in the range of milliseconds (4). Since carbamylation of Hb S by cyanate is a slow process and requires a large excess of reagent, cyanate administered to the sickle cell patient can be transported to various organs of the body during the reaction time. This leads to reaction with other proteins of the body resulting in possible side effects. The low reactivity of cyanate was examined in order to find a method to increase the reaction velocity of the modification of a-amino

* This work was supported in part by Grant HL-15160 from United States Public Health Service.

‘The abbreviations used are: Hb, hemoglobin; NaCl/PO,, phos- phate-buffered saline, pHMB, p-hydroxymercuribenzoate; MIH, 3- methyl-5-isopropylhydantoin.

groups of Hb S. The reactive species of cyanate is isocyanic acid. Since isocyanic acid has a pK value of 3.8, only one out of 4,000 cyanate ions is isocyanic acid at physiological pH, 7.4. Even though cyanate continuously becomes isocyanic acid during the carbamylation reaction, a large excess of cyanate is required to achieve the reaction. The pK of cyanate will slow down the reaction at pH 7.4.

The pK of a-amino groups of carboxyhemoglobin and deoxyhemoglobin have been reported. Values of the pK for the (Y and p chains were, respectively, 6.95 and 7.05 for carboxyhemo- globin and 7.79 and 6.84 for deoxyhemoglobin (5). Using the average pK value of (Y and @ chains for both carboxyhemo- globin and deoxyhemoglobin, the relative reaction rates of carbamylation of Hb S (1.2 mM) by 0.1 M KNCO at several different pH values were calculated by the method described by Smyth (6). The results showed that the optimum pH for the reactions was 5.5 and, at pH 7.4, the reaction velocity drops to 0.30 and 0.46 for carboxyhemoglobin and deoxyhemoglobin, respectively. It was thought that if the pK requirement of cyanate can be removed from the reaction, the reaction velocity would increase and only a stoichiometric amount of the reagent would be needed. If a reagent, X-N=C=O, where X is an atom or group covalently attached to the nitrogen is

6226

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from

Antisickling Effect of Methylisocyanate 6227

used for the a-amino groun modification of Hb S. the reaction HbS solution at 25”. The solutions were incubated at 25” for 10 min.

may proceed at a fast rate and would. require only a smaller

amount of the reagent to achieve carbamylation of Hb S. Based

on this assumption, several isocyanate derivatives were consid-

ered for the tests of antisickling effect and reaction velocity.

Methylisocyanate was chosen as our example since it is soluble

in water and is thought to be small enough to penetrate the cell

membranes. Since alkylisocyanates are known to be hydro-

lyzed rapidly by water (7) and react with their amines to form

urea derivatives, it was believed that only a small quantity of

the added reagent would react with Hb S.

EXPERIMENTAL PROCEDURE

Materials-Methylisocyanate was purchased from Eastman Chemi- cal Co. and glycyllysine hydrochloride from Calbiochem, respectively. Other reagents used were of reagent grade. Nucleopore polycarbonate filters (2.5 cm in diameter and 5 pm in pore size) were obtained from Nucleopore Corp., Pleasanton, Calif. Sickle blood samples were obtained from sickle cell patients at Howard University Hospital by venipuncture using Na,EDTA or heparin as anticoagulant and refrig- erated at 4’ until used. Blood samples used for experiments were those which demonstrated homozygous Hb S type by cellulose acetate electrophoresis at pH 8.6 in Tris/EDTA/ borate buffer and contained less than 5% of Hb F by alkali denaturation method. The blood samples were used within 24 h after collection. Instruments used in these studies were Beckman 121 H amino acid analyzer, Instrumenta- tion Laboratory model 217 blood gas laboratory system, Beckman ACTA M-VI spectrophotometer, American Optical series 20 micro- scope, and Radiometer pH-stat titration system with TTT 2 titrator. Double-glass-distilled water was used throughout the experiments.

Test for Antisickling Effect of Methylisocyanate-Sickle blood samples were washed three times with NaCI/PO,, pH 7.4 (8), and cen- trifuged at 10,000 x g and 4’ for 10 min. The original vo!ume of the sample was reconstituted with NaCl/PO,. The hemoglc’oin concen- tration of the samples was determined by cyanmethemoglobin method (9). Six milliliters of the washed red cell suspension were placed in a tonometer equilibrated at 37”, and incubated for 10 min with stirring in the laboratory atmosphere in order to oxygenate and to equilibrate the sample at 37”. The laboratory was kept free from cigarette smoke to prevent any binding of carbon monoxide by the samples. A stoichio- metric amount or as much as a 2-fold excess of methylisocyanate over a-amino groups of Hb S (usually 2 to 5 ~1) was added slowly from a Hamilton syringe (10 ~1) to the stirring cell suspension and incubated for 10 min in the laboratory atmosphere. The red cells were then de- oxygenated with hydrated 10% CO, in N, for 10 min. The samples be- came purple-colored by deoxygenation. About 0.3 ml of the deoxygen- ated incubation mixture was transferred into 3 ml of phosphate- buffered 10% formalin solution, pH 7.0, in the deoxygenated atmos- phere. Control samples were prepared similarly but without the addi- tion of methylisocyanate. Five hundred fixed red blood cells were counted under a microscope (940 x magnification) by three individuals who did not know the identity of the samples. The cells were counted in two groups, normal and sickle-shaped cells. Two close vaiues ob- tained from the cell counting were averaged and used in the result.

Preparation of Hemolysate-The untreated sickle cells or methylisocyanate-treated sickle cells were saturated with CO and washed three times with 0.9% NaCl at 4” and centrifuged at 10,000 x g. The cells were resuspended in 0.9% NaCl and the suspension was centrifuged at 20,000 x g for 15 min to have tightly packed red cells. To these cells 1 or 2 volumes of ice-cold water was added, and the mixture was agitated to hemolyze the cells. After a 1 h incubation at 4”, the hemolysate was centrifuged at 35,000 x g for 35 min to separate the red cell membranes and unhemolyzed cells, if any, from the supernatant. The hemoglobin layer was carefully withdrawn with a Pasteur pipette and the solution was then saturated with CO. This solution was dialyzed at 4’ against 0.1 M KC1 saturated with CO with two changes to remove organic phosphates and other low molecular weight sub- stances (2). For the preparation of hemolysate to be used in the study of minimum gelling concentration, blood samples were treated as above except that CO saturation and dialysis steps were omitted.

Determination of Minimum Gelling Concentration-Methylisocya- nate, 2-fold or 4.fold excess over Hb S a-amino groups, was added to

The minimum gelling concentrations of the methylisocyanate-treated and control samples were determined by the method described in Bookchin and Nagel (10) except that test tubes (1.2 x 4.5 cm) were used instead of Erlenmeyer flasks.

Test for Filterability of Sickle Cells-The procedure for the test of filterability or cell membrane deformability of control and methylcar- bamylated red cells followed the procedure described by Baker et al. (11) with minor modifications. Six milliliters of washed sickle cell suspension in NaCl/PO, were placed in a pH-stat at 37” and a 2-fold excess methylisocyanate over Hb S a-amino groups was added. The mixture was incubated at 37”, pH 7.4, for 10 min and washed three times with NaCI/PO,. Control sample was treated similarly but without addition of methylisocyanate. The suspensions were diluted to 0.25%, packed cell volume, with NaCl/PO,. Four milliliters of the cell suspensions were subjected to filtration on Nucleopore polycarbonate filter (2.5 cm in diameter and 5 pm in pore size) under hydrostatic pressure. When 1 ml of the oxygenated suspension is passed through the filter, the filtrate collected was taken and subjected to the determination of the cell concentration along with the original sample. The absorbances of the cell suspensions at 540 nm were measured in Beckman ACTA M-VI spectrophotometer. For the filtration of deox- ygenated samples, 5 ml each of the cell suspensions were first deox- ygenated in a tonometer at 37” for 2 min with 10% CO, in N,. A microscopic examination of the cells revealed that a 2-min incubation of the cell suspension in the above condition was enough to deoxygen- ate the control cells and to make 98% of the cells to be sickle-shaped. The cell suspensions were subjected to filtration as described above, but in a plastic bag filled with the deoxygenating gas. Since the filtration of the deoxygenated control sample was very slow and virtually stopped after 0.8 ml of filtration, only 0.8 ml of the filtrate was used for the absorbance measurement at 540 nm. The ratio of the absorbances of filtrates and original samples were calculated to obtain filterability of each sample.

Determination ojp10 Values-Methylisocyanate-treated and control red cells were recombined with original plasma and subjected to psO determination. The pH value of both samples at pa0 determination was 7.4. For each point of the oxygen saturation curve, the red cell suspension was incubated at 37” for 15 min with a desired oxygen pressure produced by mixing two gasses, 5.55% CO, in N, and 5.65% CO,, 19.59% 0, in N,.

Test for Irreversible Antisickling Effect of Methylisocyanate- Methylisocyanate-treated sickle cells were washed three times with NaCl/PO, or dialyzed against 500 volumes of NaCl/PO, at 4”. The control sample was treated similarly. Deoxygenation, fixation, and counting of these cells were performed as described earlier. Carbon- monoxy Hb S in 0.1 M KC1 which had 1 N-methylcarbamyl residue at NH,-terminal groups was dialyzed overnight against 500 volumes of 0.1 M KC1 saturated with CO at 4”. After the determination of the protein concentration, the sample was subjected to MIH assay. The result was compared with the value of the undialyzed.

Determination of Free Sulfhydryl Groups of N-Methylcarbamylated Hb S-Carbonmonoxy Hb S in 0.1 M KCI, treated with 2-fold excess methylisocyanate at pH 7.4 and 37” for 10 min. was dialyzed overnight against 500 volumes of 0.1 M KC1 saturated with CO. The free-SH groups of the hemoglobin were titrated with pHMB according to the method of Boyer (12), and results obtained were compared with those of control.

Preparation of Standard 3.Methyl&sopropylhydantoin-Valine (20 mg) was dissolved in 2 ml of water in a 15-ml ignition tube. The pH was adjusted to 8.0 with 0.1 N NaOH. A 500.fold excess of methylisocyanate was added to the valine solution with stirring. The mixture was kept in a well ventilated hood overnight at room temperature to synthesize N-methylcarbamylvaline. Glacial CH,COOH (2 ml) and 12 N HCl (4 ml) were added to the reaction mixture and the tube was heat-sealed. The mixture was incubated at 100” for 1 h to cyclize N-methylcarbamylvaline to 3-methyl-5-iso- propylhydantoin. Attempts to evaporate the solution to dryness in a rotatory evaporator at 45 to 48” resulted in a viscous solution which slowly crystallized upon drying in a desiccator. A portion of the product was dissolved in water and subjected to amino acid analysis (13). There was no valine detected by this assay. Because of the extremely high hygroscopic nature of MIH, it was not possible to weigh the compound accurately. Element analysis was not conducted. In order to bypass the difficulty of weighing this hygroscopic compound, MIH prepared with 2 pm01 of valine was evaporated to dryness in a

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from

6228 Antisickling Effect of Methylisocyanate

rotatory evaporator and dissolved to 2 ml with water. Aliquots of this solution were subjected to column chromatography, alkahne hydrol- ysis, and amino acid analysis as described later in the text.

Determination of N-Methylcarbamylation of Amino Groups of HbS-The extent of reaction between methylisocyanate and a-amino groups of Hb S was determined by employing hydantoin assay (14) with minor modifications. N-Methylcarbamylation of the NH*-terminal group of Hb S, valine, and its cyclization in 6 N HCl at 100” for 1 h yielded MIH. Hydrolysis of standard MIH with 0.2 N NaOH at 110’ for 22 h in a sealed tube (14) gave 33% recovery of valine based on amino acid analysis. Incubation of MIH with 0.4 N NaOH at 110” for 24, 48, and 72 h yielded 74, 88, and 90% recovery of valine, respectively. Concentration of NaOH higher than 0.8 N formed excessive silica particles m the tube, which prevented the accurate assay of valine. When MIH was passed through a Bio-Rad AG 50 W-X2 column (0.9 x 10 cm) and the effluent was monitored with absorbance at 225 nm, considerable trailing was observed. The first 30 ml of the effluent was collected, evaporated to dryness, and hydrolyzed with 0.4 N NaOH for 48 h at 110’. Based on the amino acid analysis of the above sample and the control sample, which was not passed through the column, it was found that 82% of the compound was recovered after the column chromatography. This recovery value and another value obtained after the hydrolysis of MIH with 0.4 N NaOH for 48 h were used for the calculation of the extent of the reaction between Hb S and methylisocyanate. For the determination of the reaction extent of c-amino groups of Hb S with methylisocyanate, N’-methylcarbamylly- sine was prepared in the following manner: 1 rmol of glycyllysine hydrochloride in 1 ml of water was mixed with a 300-fold excess of methylisocyanate in a well ventilated hood and incubated overnight at room temperature. The solution was adjusted to 6 N HCI using 12 N

HCI, and incubated at 100” for 1 h in a sealed ignition tube. The mixture was evaporated to dryness and subjected to amino acid analysis in a 0.2 N citrate buffer, pH 3.49. The N’-methylcarbamylly- sine was eluted between alanine and valine. Methylisocyanate-treated Hb S was hydrolyzed with 6 N HCl, in uocuo, at 110” for 22 h and subjected to amino acid analysis to determine any modification of the c-amino groups.

Reaction Rate of N-Methylcarbamylation of Hb S a-Amino Groups by Methylisocyanate-Since alkylisocyanates are known to decompose rapidly in water (7) and the decomposition rate of methylisocyanate is not known, the second order rate constant of N-methylcarbamylation of Hb S a-amino groups by methylisocyanate was not determined. The reaction rate was monitored by MIH assay. Six milliliters of dialyzed carbonmonoxy Hb S solution (0.8 mM in 0.1 M KCI) were equilibrated at 37” in a reaction vessel of pH stat-system, TTT2 of Radiometer. The

FIG. 1. Morphology of control and N-methylcarbamylated red cells (Hb S type) in deoxygenated condition. Red cells (Hb S type) sus- pended in NaCI/PO, , pH 7.4, were incubated at 37” for 10 min with 2- fold excess of methylisocyanate over Hb S a-amino groups. The red cells were then deoxygenated with hydrated 10% COz in Nz for 10

pH was adjusted to 7.4 with 0.2 N NaOH. Two-fold excess of methylisocyanate over Hb S a-amino groups was added to the Hb S solution with a Hamilton syringe (10 ~1) to initiate the reaction. Titrant used was 0.2 N NaOH. Aliquots of the reaction mixture (0.5 ml) were removed at 15, 30, and 45 s and 1.5- and 2-min intervals and placed in 5 ml of ice-cold 5% trichloroacetic acid to terminate the reaction. The precipitated hemoglobin was washed five times with 5 ml of ice-cold 5% trichloroacetic acid solution and two times with ethyl ether. The precipitate was dried in a warm water bath and then dissolved in 2 ml of 50% acetic acid. The product was subjected to the determination of N-methylcarbamylation of Hb S a-amino groups by the procedure given above.

RESULTS

Effect of Methylisocyanate on the Sickling of Erythrocytes- Deoxygenated red cells (Hb S type) are shown in Fig. 1. Most of the red cells in the control sample were in sickle shape (Fig. la). With P-fold excess of methylisocyanate over Hb S a-amino groups, most of the cells retained the normal morphology upon deoxygenation (Fig. lb). No hemolysis of red cells was observed in this experiment. Methylisocyanate did not reverse the sickling of irreversibly sickled cells. Increasing the amount of methylisocyanate, there was an increase of the number of N-methylcarbamyl groups incorporated into Hb S with a concomitant decrease of the number of sickle-shaped cells (Table I).

Determination of Minimum Gelling Concentration-The minimum gelling concentrations of control and methylisocya- nate-treated Hb S are shown in Table II. The minimum gelling concentration of the control Hb S was 24.0 * 0.0 g/100 ml, which is comparable with published data (10). Treatment of Hb S samples with methylisocyanate increased the minimum gelling concentration up to 31.5 f 0.9 g/100 ml, suggesting that N-methylcarbamylation of Hb S reduces the aggregation of Hb S molecules.

Effect of Methylisocyanate on Oxygen Affinity of Hemoglo- bin S-Fig. 2 shows oxygen dissociation curves of control and N-methylcarbamylated red cells, which has 0.4 N-methylcar- bamyl residues at Hb S a-amino groups, for the determination

min. About 0.3 ml of the deoxygenated incubation mixture was trans- ferred into 3 ml of phosphate-buffered 10% formalin solution, pH 7.0, in the deoxygenated atmosphere (b). Control sample (a) was prepared similarly but without the addition of methylisocyanate. The magnification for photography was x 1000.

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from

Antisickling Effect of Methylisocyanate 6229

TABLE I

Effect of methylisocyanate on sickle cells

Sickle cells were washed three times with NaCl/PO, and the original volume was reconstituted with NaCl/PO,. Hemoglobin concentration was 1.3 mu. Incorporation of N-methylcarbamyl group to the a-amino groups of hemoglobin S was determined by modified hydantoin assay

as described in the text.

[CH,NCO]/ [Hb a-NH,]

Number of N-methylcarbamyl groups incorporated per

Hb S tetramer

0.0 0.0 87.6 1.0 0.4 47.5 1.5 0.6 34.1 2.0 0.8 27.2

TABLE II

Minimum gelling concentration of control and methylisocyanate-treated hemoglobin S

Minimum gelling concentrations were determined by a method similar to that of Bookchin and Nagel (10).

Minimum gelling concentration [CH,NCO]/[Hb n-NH,]

Experiment 1 Experiment 2

g/100 ml

0.0 24.0 23.9 2.0 28.9 27.0 4.0 32.3 30.6

0 10 20 30 40

Pod 1 mm Hg

FIG. 2. Oxygen dissociation curve of control and N-methylcar- bamylated red cells (Hb S type). Red cells (Hb S type) in NaCl/PO,, pH 7.4, were incubated at 37” for 10 min with a stoichiometric amount of methylisocyanate over Hb S a-amino groups. Control sample was treated similarly but without addition of methylisocyanate. For each point of the oxygen saturation curve, the red cell suspensions were incubated at 37” for 15 min with a desired oxygen pressure produced by mixing two gases, 5.55% CO1 in N, and 5.65% CO,, 19.59% O2 in Nf. O--O, control cells; 04, methylisocyanate-treated cells.

of P-50 values. The p6,, of the control cells was 30.0 mm Hg while that of N-methylcarbamylated cells was 27.5 mm Hg. The reduced psO value of the N-methylcarbamylated red cells suggests that the a-amino groups of the Hb S a-chain was modified by methylisocyanate. The a-amino groups of hemo- globin (Y chains are known to contribute about 25% to the

alkaline Bohr effect (15). Increase of oxygen affinity of Hb S by methylcarbamylation should reduce Hb S deoxygenation, resulting in the reduction of sickling of erythrocytes.

Test for Filterability of Sickle Cells-Filtration of control and methylisocyanate-treated samples in the oxygenated con- dition gave filtrates which had the same concentration of red cells as in the original samples. The filtrate of the deoxygen- ated control sample contained 0.0% of the red cells. Filtration of the treated cells in the deoxygenated condition resulted in permeation of 80.2 h 5.8% bf the cells through the polycarbon- ate filter in the given experimental condition. These results indicate that the membrane deformability of the sickle cells were restored to a great extent by N-methylcarbamylation.

Test for Irreversible Antisickling Effect of Methyliso- cyan&e-The washed or dialyzed red cells previously treated with methylisocyanate retained their normal morphology upon deoxygenation. The MIH assay of a Hb S sample, which had one N-methylcarbamyl group of its a-amino groups, prior to and after dialysis gave identical results suggesting that the dialysis did not remove the N-methylcarbamyl groups from Hb S. These observations indicate that the antisickling effect of methylisocyanate is permanent and that there was a stable bond formation between Hb S a-amino groups and methyliso- cyanate, like in the case of potassium cyanate (1).

Determination of Free Sulfhydryl Groups of N-Methylcar- bamylated Hb S-The titratable free sulfhydryl groups in methylisocyanate-treated and control Hb S sample were 1.7 and 1.6 residues/hemoglobin tetramer, respectively. These results suggest that there was no stable bond formation between Hb S-free sulfhydryl groups and methylisocyanate.

Test for Modification of Hb 5’ C-Amino Groups by Methyl- isocyanate-Amino acid analysis of the methylisocyanate- treated Hb S hydrolysate showed no detectable N-methylcar- bamylysine, indicating that there was no modification of Hb S t-amino groups by methylisocyanate.

Reaction Rate of N-Methylcarbamylation of Hb S by Methylisocyanate-Carbonmonoxy Hb S treated with 2-fold excess methylisocyanate over Hb S a-amino groups (Fig. 3) showed that the reaction was complete within 45 s after the initiation, and that one N-methylcarbamyl group was intro- duced per Hb S molecule. This incorporation was at the a-amino groups of Hb S. After the addition of 3.0 ~1 of methylisocyanate (2-fold excess over a-amino groups) to the carbonmonoxy Hb S in the given experimental condition, a continuous decrease in the pH was observed. The reason for the pH drop in the dilute hemoglobin solution is probably due to carbonic acid formation with CO*, a decomposition product of methylisocyanate. Some of the other decomposition product, methylamine, could escape to the air since the unprotonated form of methylamine has a boiling point of -6”. The pH decreased to 7.2 at 37 s after the addition of methylisocyanate and remained constant, suggesting that the reaction was stopped at this point. This observation supports the data obtained by amino acid analysis.

DISCUSSION

The discovery of potassium cyanate as an antisickling agent (1) offered a new approach for the treatment of sickle cell disease and the understanding of the Hb S molecule. The basis for the antisickling effect of potassium cyanate was found to be mainly due to the carbamylation of the Hb S a-amino groups (3). The a-amino groups of LY chains are known to contribute about 25% to the alkaline Bohr effect (15). Carbamylation of

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from

Antisickling Effect of Methylisocyanate

1.0 - Lpo- -

0.8 -Jc- 0.6 - 0

0 0.5 1.0 1.5 2.0 3.0

TIME [MINUTES]

FIG. 3. Reaction of a-amino groups of Hb S with methylisocyanate. Carbonmonoxy Hb S (final concentration, 3.36 rn~ in a-amino group) was mixed with methylisocyanate (final concentration, 6.72 mM) at 37” and pH 7.4 in a pH-stat system, TTTZ, Radiometer. At the indicated intervals, aliquots of’ the reaction mixture (0.5 ml) were transferred into 5 ml of’ cold 5% trichloroacetic acid to terminate the reaction. The precipitated protein was subjected to hydantoin assay in order to determine the extent of the reaction at each point as described in the text.

these groups reduces the alkaline Bohr effect, thereby increas- ing the oxygen affinity of the Hb S molecule. Oxygenated Hb S molecules do not aggregate and thus prevent the sickling of erythrocytes. The carbamylation of the /3 chain a-amino groups, in uiuo, was 2 times less frequent than that of (Y chain (16): Therefore, the contribution of the @ chain carbamylation in the antisickling effect would be relatively small. However, the /3 chain-carbamylated Hb S showed about 23% increase in minimum gelling concentration (3). Since the antisickling effect of potassium cyanate is due to the modification of the Hb S a-amino groups, it was desirable to find a reagent which reacts fast with Hb S and requires a smaller amount of the reagent to achieve the modification. Methylisocyanate, one of the isocyanate derivatives which are in reactive form, was investigated as an example for the above purpose. It was found that this reagent has powerful antisickling activity and reacts fast with Hb S o-amino groups. The antisickling activity of methylisocyanate was irreversible, like that of potassium cyanate (1). Since there were increases of oxygen affinity of the sickle cells and minimum gelling concentration of Hb S after incubation with methylisocyanate, it is believed that the reagent reacted with both 01 and p chain NH,-terminal as cyanate does. Since methylisocyanate is an analog of isocyanic acid, the antisickling effect of methylisocyanate appears to be similar to that of potassium or sodium cyanate (1, 3) in many respects. The difference between methylisocyanate and potas- sium or sodium cyanate is that smaller amount of methylisocyanate is required for the antisickling effect and methylisocyanate possesses a higher reactivity than potassium or sodium cyanate. Since methylisocyanate is a lachrymator, it may not be advisable for use in uiuo. The use of this reagent in vitro without an additional process, like dialysis, may depend on the toxicity of methylamine and 1,3-dimethylurea in a methylisocyanate-treated medium. Methylisocyanate will decompose to methylamine in an aqueous solution. Methylam-

ine could react with methylisocyanate to form 1,3-dimeth- ylurea. Any possible side effects by these byproducts in the red cell suspension after reinfusion have yet to be investigated. Should the concentrations of methylamine and 1,3-dimeth- ylurea in methylisocyanate-treated sickle cell suspension be so high as to result in side effects in the sickle cell patient, then these byproducts should be removed from the red cell suspen- sion, either by washing or by dialysis before the reinfusion. Methylamine has a boiling point of -6” and pK of 10.6. The evaporation rate of methylamine in the red cell suspension at pH 7.4 and 37” is currently under investigation in this laboratory.

Isocyanic acid, the reactive tautomer of cyanate, was found to be a carbon dioxide analog (2), and the specificity of isocyanic acid for a-amino groups of hemoglobin was explained on the basis of the structural analogy between carbon dioxide and isocyanic acid. This report shows that the antisickling reagents need not be analogs of CO, in order to react with a-amino groups of Hb S and to prevent sickling of erythrocytes. Methylisocyanate, which is not a structural analog of carbon dioxide, reacted with a-amino groups of hemoglobin by virtue of its reactivity. At this point, the scope of screening new antisickling reagents broadens.

It would be of great interest to find reactive compounds which are small enough to penetrate the erythrocyte mem- brane and react with Hb S a-amino groups. Other isocyanate derivatives may also react fast with Hb S a-amino groups, and may, if they can penetrate the erythrocyte membrane, possess antisickling activity.

The structural and elemental analysis of the product of the reaction of methylisocyanate with valine followed by cycliza- tion has not been performed at this time. However, the chemistry of hydantoin synthesis (13) suggests that the com- pound is most likely to be 3-methyl-5-isopropylhydantoin. The preparation of pure MIH was difficult because of its extremely high hygroscopic nature. It was also difficult to hydrolyze MIH to valine because of the stabilizing effect on the hydantoin ring by the methyl group at position-3. We found that methylisocyanate reacted with several amino acids and may be used for the determination of NH,-terminal groups of proteins.

Acknowledgmerzts-I thank Dr. R. B. Scott for his support and Dr. M. Mann for his help in providing laboratory space for this investigation. The expert technical assistance of Miss Monica Slack and Mr. B. C. Kim is gratefully acknowledged. I am also grateful to Drs. D. L. Rucknagel and 0. Castro for their valuable assistance in obtaining blood samples and to Mr. M. Y. Park for his service in preparing the photographs.

REFERENCES

1. Cerami, A., and Manning, J. M. (1971) Proc. N&l. Acad. Sci. U.S.A. 68, 1180-1183

2. Lee, C. K., and Manning, J. M. (1973) J. Biol. Chem. 248, 5861-5865

3. Nigen, A. M., Njikam, N., Lee, C. K., and Manning, J. M. (1974) J. Biol. Chem. 249, 6611-6616

4. Kernohan, J. C., and Roughton, F. J. W. (1968) J. Physiol. 197, 345-361

5. Garner, M. H., Bogardt, R. A., Jr., and Gurd, F. R. N. (1975) J. Biol. Chem.250, 4398-4404

6. Smyth, D. G. (1967) J. Biol. Chem. 242, 1579-1591 7. Brown, W. E., and Weld, F. (1973) Biochemistry 12, 828-834 8. Dulbecco, R., and Vogt, M. (1954) J. Erp. Med. 99, 167182 9. Drabkin, D. L. (1946) J. Biol. Chem. 164, 703-723

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from

Antisickling Effect of Methylisocyanate 6231

10. Bookchin, R. M., and Nagel, R. L. (1971) J. Mol. Biol. 60,263-270 14. Stark, G. R., and Smyth, D. G. (1963)5. Biol. Chem. 238,214-226 11. Baker, R., Powars, D., and Haywood, L. J. (1974) Biochem. 15. Kilmartin, J. V., and Rossi-Bernardi, L. (1969) Nature 222,

Biophys. Res. Commun. 59, 548-556 1243-1246 12. Boyer, P. D. (1954) J. Am. Chem. Sot. 76, 4331-4337 16. Njikam, N., Jones, W. M., Nigen, A.M., Gillette, P. N., Williams, 13. Spackman, D. H., Stein, W. H., and Moore, S. (1958)Anal. Chem. R. C. Jr., and Manning, J. M. (1973) J. Biol. Chem. 248,

30, 1190-1206 8052-8056

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from

C K LeeMethylisocyanate as an antisickling agent and its reaction with hemoglobin S.

1976, 251:6226-6231.J. Biol. Chem. 

  http://www.jbc.org/content/251/20/6226Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/251/20/6226.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on October 30, 2017

http://ww

w.jbc.org/

Dow

nloaded from