BritishJournal
of Haernatobgy,
1g7g,42. 225-230.
Kinetics of Red Cell Membrane Phosphorylation:
Altered Affinity of HS Membrane Protein Acceptors S. A. ALI, E. C. GORDON-SMRH AND H. S. SELHI* Department of Haematology, Royal Postgraduate Medical School, London, and *Blood Research Laboratory, Department of Haematology, King’s College Hospital Medical School, London (Received 8 March 1978; accepted for publication 13 October 1978) SUMMARY. Membrane ghosts were prepared from red blood cells of normal and hereditary spherocytosis (HS) subjects. Time dependent phosphorylation studies of the membrane (in the presence of y labelled AT32P) showed that the HS membrane appeared to incorporate less phosphate than the normal membrane but the results were not significantly different. Using the initial linear rate of phosphorylation and varying the ATP concentration, it was found that the Michaelis constant (K,) for the normal membrane was 24.8 PM ATP and V,,, was 0.120nmol phosphate bound/mg membrane protein/min. These values for the HS membrane were found to be 39.0 and 0.118 respectively. In one family the affected mother and son each showed a positive cooperativity effect for similar reciprocal plots. This indicates that phosphorylation of various HS membrane proteins may be an ordered rather than random process and suggests biochemical heterogeneity of the HS condition. The constants, K , and V,,,, however, were found to be similar to the rest of HS subjects studied. Crude extracts of the enzyme protein kinase, which catalyses the membrane phosphorylation, showed no significant difference in the K , and V,,, value in HS compared with that from normal red cells. The kinetic difference in phosphorylation is probably due to abnormalities in the membrane protein($ which accept the phosphate from protein kinase in HS red cells. Hereditary spherocytosis (HS) is a familial haemolytic anaemia inherited as a dominant condition, with characteristically spheroidal red cells. It is thought that the abnormality is associated with the red cell membrane (reviewed by Jacob, 1969). Alteration in CAMP dependent protein kinase have also been reported (Greenquist & Shohet, 1976; Matsumoto et al, 1977).The studies of Zail& Van den Hoek (1975) and Beutler et al(1976), however, were unable to confirm such results. The abnormalities of HS red cell membrane have recently been reviewed by Zail(I977). We report here further work on the phosphorylation of membrane proteins of HS and normal red cell membrane. Correspondence: Dr E. C. Gordon-Smith, Department of Haematology, Royal Postgraduate Medical School, Du Cane Road, London W I ZoHS. 0007- 1048/79/0600-0225$02.00
01979 Blackwell Scientific Publications 22s
2 26
S. A . A l i , E. C. Gordon-Smith and H. S. Selhi METHODS
All chemicals used were of analytical grade from BDH, Poole, England. ATP, CAMP, theophylline, calf thymus histone (type II), bovine serum albumin (USA fraction V) and EGTA were from Sigma, London. [y32P]ATPwas purchased from the Radiochemical Centre, Amersham (c. 2-3 Ci/mmol). Unlabelled ATP was scparated from ADP and inorganic phosphate using Dowex AG-I-X-8 quaternary ammonium resin (ATP eluted with 0.5 M HCI). This was stored at -2o'C after adjusting the pH to 6.8 with Tris. The percentage hydrolysis and purity of the radioactive ATP were checked by elution on the Dowcx resin and by chromatography on Whatman No. I paper (isobutryric acid: water (0.88) : ammonia :EDTA; 100: 56 :4.2 :o.o6) as used by the supplier. Freshly collected blood from nine healthy subjects and nine Type I , splenectomized, HS subjects was taken in CPD. Membrane ghosts were prepared on the same day, usually within 3 h of collection. Preparation ($red cell membraneghosts. The method used was that of Dodge et al (1963) using 5 mM phosphate buffer, p H 8.0. The resulting membrane ghosts were kept at -2oOC and normally used the following day. Each experiment was carried out in duplicate. For each experimcnt an HS and normal membrane sample were used on the same day with the same reagents and controls. About one half of the membrane samples were used for membrane phosphorylation studies, the rest being used for extraction of crude fraction of protein kinase by the method of Rubin et al (1972). Phosphorylalion assay was essentially that of Guthrow at a1 (1972). The ATP concentration for time course experiments was 5 PM. For the kinetic experiments the ATP varied between 2.33 and 20 /AM,keeping the specific activity constant for a particular experiment. The total volume of the reaction mixture was 0.5 mi. It contained the following a t final concentrations ; 2.5 mM; NaF 0.1 mM; stated: ATP 5 p~ (for time course only); CAMP 10 p ~ theophylline EGTA 0.3 m M ; Mg acetate 10mM Na2H PO, 4.1 mM (pH adjusted with dilute phosphoric acid to 6.5). Normally between 1 . 5 and 2.0 x lo6 dpm of AT32P were used per 2.5 nmol total ATP. The tubes were preincubated for 1 0 min a t 35°C before the addition of either 0. I mg membrane protein or 1 . 2 mg histone where appropriate. When present, protein kinase was used a t 0.3 m g of the crude fraction per sample. The reaction was stopped a t the required time by addition of 200 pl of ice-cold 0.63 % BSA followed immediately by 4 ml of ice-cold 7.5% trichloracetic acid (TCA). The protein was sedimented a t 2500 rpm using an MSE Mistral 4L centrifuge a t 4°C for 20 min. The pellets 4 ml of 5% T C A was then added and the were dissolved in 200 pl of I M NaOH (0°C). suspension mixed thoroughly. 2 ml of 10% sucrose solution in 7.5% T C A was then gently pumped into the bottom of the conical glass centrifuge tube ( ~ ml j capacity). This was ccntrifugcd as previously. The supernatant was aspirated and discarded. The tubes were then inverted and allowed to drain far 10 min. Finally, the sidesofthe tube were wiped and 200 pl of I M NaOH added to dissolve the pellet completely. 50 p1 of this was taken for counting, in 10 ml of dioxanc based scintillant, using an Intertechnique SL 40 scintillation counter. Quench correction was carried out by 'spiking' the contents of the vial with known specific activity AT3*P and then recounting. Protein determination was by a modified method of Lowry et a1 (1951).
Kinetics of Red Cell Membrane Phosphorylation
227
RESULTS Fig I shows the time dependent phosphorylation of normal and HS membrane ghosts. The reaction was found to be linear for at least 5 min. A slow rate of dephosphorylation was noted after the maximal level of phosphorylation was reached. In some experiments a clear dephosphorylation (equal in rate to the phosphorylation) and rephosphorylation was noted in the plateau region of the curve shown, but in most experiments this phosphorylation-dephosphorylation process was not as obvious. These variations gave a large standard deviation in the plateau region. For this reason a significant difference (at the 5% level) in phosphorylation of normals and HS, in the plateau region, was not observed. T
T
T I Y X T
T
e I
I
I
I
I
I
1
I
10
20
30
40
50
60
70
80
9’0
Time of incubation (min)
FIG I . Phosphorylation of normal ( 0 ) and HS (m) membrane ghosts with time. One SD is shown.
Initial linear rates of phosphorylation, using different concentrations of ATP, were ensured by stopping the phosphorylation reaction a t I , 3 , ; and 10 min as described. The range of ATP used was from 2 . 3 3 to 20 ,UM. Fig 2 shows the double reciprocal plot for reaction velocity and substrate concentration. The K , for the normal red cell membrane ghosts was found to be 24.8 ,UM and the V,,, 0.120 nmol phosphate bound/mg membrane protein/min. For the HS membrane the values were 39.0 ,UM and 0.I 1 8 nmol phosphate bound/mg membrane protein/min respectively. A familiar HS group of mother and son showed different kinetics from that shown in Fig 2. A positive cooperativity was noticed for this group as shown in Fig 3 . However, the K , and V,,, values are not very different from the rest of the HS membrane ghost studied. As shown in Fig 4, the
S . A . Ali, E. C.Gordon-Smith and H . S . Selhi
228 160-
140-
L 0)
a
E
g
80-
a
0.1
0.2
0.3
0.4
(ATP pM)-'
FIG2. Lineweaver-Burk plotsof reciprocal of velocity of phosphorylation of membrane ghosts with reciprocal of ATP concentration used. The lines were calculated by multiple linear regression analysis. Normal membrane (0);HS membrane (a). I-
(ATP p.4-l
FIG3 . Lineweaver-Burk plots for membrane ghosts, A familial group of mother and son showing son positive cooperativity effect. A mean of four determinations in duplicate is shown. Mother (0);
(3.
Kinetics of Red Cell Membrane Phosphorylation 300r 250
I
T
I -
1
I
I
I
I
0.20
0.10
I
0.30
I
I
)
I
0.40
(ATP p d '
FIG 4. Lineweaver-Burk plots for phosphorylation of calf thymus histone using extracted protein kinase from normal (0)and HS (0) red blood cell membrane.
TABLE
I Km Vmax @mu/ A TP) (pmul Pi buundlmg proteinlmin)
~
Normal membrane HS membrane HS membrane, familial group of mother and son Extracted protein kinase, normal Extracted protein kinase, HS
24.8
0.120
39.0
0.118
38.0
0.I I 0
20.3
0.027
19.4
0.025
double reciprocal plot for normal and HS extracted protein kinase (using calf-thymus histone as the phosphate acceptor) shows no significant difference a t P=O.OIlevel. The K,,, and V,,, values for membrane and extracted protein kinase using normal and HS red cells are shown in Table I.
DISCUSSION Measurement of initial rates was adopted for studying the phosphorylation mechanism in normal and HS membranes because the maximal level of phosphorylation in the two types of membrane was found to be statistically insignificant. This was due to the variable results in the time course region when 5 PM ATP was used. However, as shown in Fig I , it was observed that the initial rate of phosphorylation was linear. Although more than one membrane protein may be phosphorylated (Rubin, I975), the initial part of the curve of Fig I is similar to one substrateenzyme type of reaction. Fig 2 shows that a difference in K, value for ATP exists for the phosphorylated form of HS
230
S. A . Ali, E. C . Gordon-Smith and H. S . Selhi
and normal membrane. This indicates that a difference in affinity for phosphate exists in the two types of membranes. Greenquist & Shohet (1976) have reported a differcnce in protein kinasc activity in HS and normal mcmbrane. When the enzyme is present free in solution as a crude preparation, no significant difference (P=o.os) in K, and V,,, was noted for the HS and normal protein kinase. It is assumed in this study that the relative amounts of the enzyme, protein kinase, extracted from the two types of membranes is proportional to the rest of the proteins eluted during the extraction procedure. This is supported by preliminary SDS PAGE studies on the extracted protein kinase. However, even if this were not true, no difference in K , value should be observed if the enzyme has similar affinity for ATP. The results indicate that the difference in kinetics of phosphorylation of HS and normal membrane protein(s) is probably due to the phosphate acceptor. T h e positive cooperativity behaviour of a single familial group is important in two respects. Firstly, it shows that the phosphorylation of the various peptides revealed by SDS PAGE (Rubin, 1975) may not be a random process but one in which cooperation is present between various phosphate acceptors. This cooperative behaviour was found to be absent in the crude protein kinase fraction, using histone as a phosphate acceptor. Secondly, this behaviour demonstrates that the category known as Type I HS is most probably a heterogcneous group of disorders. REFERENCES BEUTLER, E., GUINTO,E. & JOHNSON,C. (1976) Human red cell protein kinase in normal subjects and patients with hereditary spherocytosis, sickle cell disease, and autoimmune hemolytic anemia. Blood, 48,887-898. D o ~ c tJ.T., , MITCHELL, C. & HANAHAN, D.J. (1963) The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Archives of Biochemistry and Biophysics, 100, I1~130. GHCENQIJIST, A.C. & SHOHET,S.B. (1976) Yhosphorylation in erythrocyte membranes from abnormally shaped cells. Blood, 48,877-886. GUTHROW, E.C., J R , A L I E NJ.E. , & RASMUSSEN, H. (1972) Phosphorylation of an endogenous menibrane protein by an endogenous, membrane-associated cyclic adenosine 3‘,5’-monophosphatedependent protein kinase in human erythrocyte ghosts. Journal of Biological Chemistry, 247. 8145-8153. JACOB, H.S. (1969) The defective red cell in hereditary spherocytosis. Annual R e v i e w .f Medicine, 20, 41-46. LOWRY, O.H.. ROSEBROUGH, N.J.,FARR,A.L. & RAND A L L , R.J. (1951) Protein measurement with the
Fohn phenol reagentJourna1 nfBiological Chemirtry, 193,265-275. MATSUMOTO, N., YAWATA, Y. &JA(:oB, H.S. (1977) Association of decreased membrane protein phosphorylation with red blood cell spherocytosis. Blood, 499 233-239. RUBIN,C.S. (1975) Adenosine 3’: 5’-monophosphateregulated phosphor y la tion of erythrocyte m embrane proteins. Separation of membrane-associated cyclic adenosine 3’ : 5'-mono phosphatedependent protein kinase from its endogenous substrates._luurnal of Biological Chemistry, 250,9044-yo52. RUBIN,C.S., ERLICHMAN, J. & ROSEN,O.M. (1972) Molecular forms and subunit composition of a cyclic adenosine 3’,5’-monophosphate-dependentprotein kinase purified from bovine heart muscle.Journal ofBiological Chemistry, 247, 36-44. ZAIL,S.S. (1977) The erythrocyte membrane abnormality of hereditary spherocytosis. (Annotation). BritishJournal of Haematology, 37,305-3 10. ZAIL,S.S. & V A N DEN HOEK,A.K. (1975) Studies on protein kinase activity and binding of adenosine 3’s-monophosphate by membranes of hereditary spherocytosis erythrocytes. Biochemical and Biophysical Research Communications, 66. 1078-1086.
of Haernatobgy,
1g7g,42. 225-230.
Kinetics of Red Cell Membrane Phosphorylation:
Altered Affinity of HS Membrane Protein Acceptors S. A. ALI, E. C. GORDON-SMRH AND H. S. SELHI* Department of Haematology, Royal Postgraduate Medical School, London, and *Blood Research Laboratory, Department of Haematology, King’s College Hospital Medical School, London (Received 8 March 1978; accepted for publication 13 October 1978) SUMMARY. Membrane ghosts were prepared from red blood cells of normal and hereditary spherocytosis (HS) subjects. Time dependent phosphorylation studies of the membrane (in the presence of y labelled AT32P) showed that the HS membrane appeared to incorporate less phosphate than the normal membrane but the results were not significantly different. Using the initial linear rate of phosphorylation and varying the ATP concentration, it was found that the Michaelis constant (K,) for the normal membrane was 24.8 PM ATP and V,,, was 0.120nmol phosphate bound/mg membrane protein/min. These values for the HS membrane were found to be 39.0 and 0.118 respectively. In one family the affected mother and son each showed a positive cooperativity effect for similar reciprocal plots. This indicates that phosphorylation of various HS membrane proteins may be an ordered rather than random process and suggests biochemical heterogeneity of the HS condition. The constants, K , and V,,,, however, were found to be similar to the rest of HS subjects studied. Crude extracts of the enzyme protein kinase, which catalyses the membrane phosphorylation, showed no significant difference in the K , and V,,, value in HS compared with that from normal red cells. The kinetic difference in phosphorylation is probably due to abnormalities in the membrane protein($ which accept the phosphate from protein kinase in HS red cells. Hereditary spherocytosis (HS) is a familial haemolytic anaemia inherited as a dominant condition, with characteristically spheroidal red cells. It is thought that the abnormality is associated with the red cell membrane (reviewed by Jacob, 1969). Alteration in CAMP dependent protein kinase have also been reported (Greenquist & Shohet, 1976; Matsumoto et al, 1977).The studies of Zail& Van den Hoek (1975) and Beutler et al(1976), however, were unable to confirm such results. The abnormalities of HS red cell membrane have recently been reviewed by Zail(I977). We report here further work on the phosphorylation of membrane proteins of HS and normal red cell membrane. Correspondence: Dr E. C. Gordon-Smith, Department of Haematology, Royal Postgraduate Medical School, Du Cane Road, London W I ZoHS. 0007- 1048/79/0600-0225$02.00
01979 Blackwell Scientific Publications 22s
2 26
S. A . A l i , E. C. Gordon-Smith and H. S. Selhi METHODS
All chemicals used were of analytical grade from BDH, Poole, England. ATP, CAMP, theophylline, calf thymus histone (type II), bovine serum albumin (USA fraction V) and EGTA were from Sigma, London. [y32P]ATPwas purchased from the Radiochemical Centre, Amersham (c. 2-3 Ci/mmol). Unlabelled ATP was scparated from ADP and inorganic phosphate using Dowex AG-I-X-8 quaternary ammonium resin (ATP eluted with 0.5 M HCI). This was stored at -2o'C after adjusting the pH to 6.8 with Tris. The percentage hydrolysis and purity of the radioactive ATP were checked by elution on the Dowcx resin and by chromatography on Whatman No. I paper (isobutryric acid: water (0.88) : ammonia :EDTA; 100: 56 :4.2 :o.o6) as used by the supplier. Freshly collected blood from nine healthy subjects and nine Type I , splenectomized, HS subjects was taken in CPD. Membrane ghosts were prepared on the same day, usually within 3 h of collection. Preparation ($red cell membraneghosts. The method used was that of Dodge et al (1963) using 5 mM phosphate buffer, p H 8.0. The resulting membrane ghosts were kept at -2oOC and normally used the following day. Each experiment was carried out in duplicate. For each experimcnt an HS and normal membrane sample were used on the same day with the same reagents and controls. About one half of the membrane samples were used for membrane phosphorylation studies, the rest being used for extraction of crude fraction of protein kinase by the method of Rubin et al (1972). Phosphorylalion assay was essentially that of Guthrow at a1 (1972). The ATP concentration for time course experiments was 5 PM. For the kinetic experiments the ATP varied between 2.33 and 20 /AM,keeping the specific activity constant for a particular experiment. The total volume of the reaction mixture was 0.5 mi. It contained the following a t final concentrations ; 2.5 mM; NaF 0.1 mM; stated: ATP 5 p~ (for time course only); CAMP 10 p ~ theophylline EGTA 0.3 m M ; Mg acetate 10mM Na2H PO, 4.1 mM (pH adjusted with dilute phosphoric acid to 6.5). Normally between 1 . 5 and 2.0 x lo6 dpm of AT32P were used per 2.5 nmol total ATP. The tubes were preincubated for 1 0 min a t 35°C before the addition of either 0. I mg membrane protein or 1 . 2 mg histone where appropriate. When present, protein kinase was used a t 0.3 m g of the crude fraction per sample. The reaction was stopped a t the required time by addition of 200 pl of ice-cold 0.63 % BSA followed immediately by 4 ml of ice-cold 7.5% trichloracetic acid (TCA). The protein was sedimented a t 2500 rpm using an MSE Mistral 4L centrifuge a t 4°C for 20 min. The pellets 4 ml of 5% T C A was then added and the were dissolved in 200 pl of I M NaOH (0°C). suspension mixed thoroughly. 2 ml of 10% sucrose solution in 7.5% T C A was then gently pumped into the bottom of the conical glass centrifuge tube ( ~ ml j capacity). This was ccntrifugcd as previously. The supernatant was aspirated and discarded. The tubes were then inverted and allowed to drain far 10 min. Finally, the sidesofthe tube were wiped and 200 pl of I M NaOH added to dissolve the pellet completely. 50 p1 of this was taken for counting, in 10 ml of dioxanc based scintillant, using an Intertechnique SL 40 scintillation counter. Quench correction was carried out by 'spiking' the contents of the vial with known specific activity AT3*P and then recounting. Protein determination was by a modified method of Lowry et a1 (1951).
Kinetics of Red Cell Membrane Phosphorylation
227
RESULTS Fig I shows the time dependent phosphorylation of normal and HS membrane ghosts. The reaction was found to be linear for at least 5 min. A slow rate of dephosphorylation was noted after the maximal level of phosphorylation was reached. In some experiments a clear dephosphorylation (equal in rate to the phosphorylation) and rephosphorylation was noted in the plateau region of the curve shown, but in most experiments this phosphorylation-dephosphorylation process was not as obvious. These variations gave a large standard deviation in the plateau region. For this reason a significant difference (at the 5% level) in phosphorylation of normals and HS, in the plateau region, was not observed. T
T
T I Y X T
T
e I
I
I
I
I
I
1
I
10
20
30
40
50
60
70
80
9’0
Time of incubation (min)
FIG I . Phosphorylation of normal ( 0 ) and HS (m) membrane ghosts with time. One SD is shown.
Initial linear rates of phosphorylation, using different concentrations of ATP, were ensured by stopping the phosphorylation reaction a t I , 3 , ; and 10 min as described. The range of ATP used was from 2 . 3 3 to 20 ,UM. Fig 2 shows the double reciprocal plot for reaction velocity and substrate concentration. The K , for the normal red cell membrane ghosts was found to be 24.8 ,UM and the V,,, 0.120 nmol phosphate bound/mg membrane protein/min. For the HS membrane the values were 39.0 ,UM and 0.I 1 8 nmol phosphate bound/mg membrane protein/min respectively. A familiar HS group of mother and son showed different kinetics from that shown in Fig 2. A positive cooperativity was noticed for this group as shown in Fig 3 . However, the K , and V,,, values are not very different from the rest of the HS membrane ghost studied. As shown in Fig 4, the
S . A . Ali, E. C.Gordon-Smith and H . S . Selhi
228 160-
140-
L 0)
a
E
g
80-
a
0.1
0.2
0.3
0.4
(ATP pM)-'
FIG2. Lineweaver-Burk plotsof reciprocal of velocity of phosphorylation of membrane ghosts with reciprocal of ATP concentration used. The lines were calculated by multiple linear regression analysis. Normal membrane (0);HS membrane (a). I-
(ATP p.4-l
FIG3 . Lineweaver-Burk plots for membrane ghosts, A familial group of mother and son showing son positive cooperativity effect. A mean of four determinations in duplicate is shown. Mother (0);
(3.
Kinetics of Red Cell Membrane Phosphorylation 300r 250
I
T
I -
1
I
I
I
I
0.20
0.10
I
0.30
I
I
)
I
0.40
(ATP p d '
FIG 4. Lineweaver-Burk plots for phosphorylation of calf thymus histone using extracted protein kinase from normal (0)and HS (0) red blood cell membrane.
TABLE
I Km Vmax @mu/ A TP) (pmul Pi buundlmg proteinlmin)
~
Normal membrane HS membrane HS membrane, familial group of mother and son Extracted protein kinase, normal Extracted protein kinase, HS
24.8
0.120
39.0
0.118
38.0
0.I I 0
20.3
0.027
19.4
0.025
double reciprocal plot for normal and HS extracted protein kinase (using calf-thymus histone as the phosphate acceptor) shows no significant difference a t P=O.OIlevel. The K,,, and V,,, values for membrane and extracted protein kinase using normal and HS red cells are shown in Table I.
DISCUSSION Measurement of initial rates was adopted for studying the phosphorylation mechanism in normal and HS membranes because the maximal level of phosphorylation in the two types of membrane was found to be statistically insignificant. This was due to the variable results in the time course region when 5 PM ATP was used. However, as shown in Fig I , it was observed that the initial rate of phosphorylation was linear. Although more than one membrane protein may be phosphorylated (Rubin, I975), the initial part of the curve of Fig I is similar to one substrateenzyme type of reaction. Fig 2 shows that a difference in K, value for ATP exists for the phosphorylated form of HS
230
S. A . Ali, E. C . Gordon-Smith and H. S . Selhi
and normal membrane. This indicates that a difference in affinity for phosphate exists in the two types of membranes. Greenquist & Shohet (1976) have reported a differcnce in protein kinasc activity in HS and normal mcmbrane. When the enzyme is present free in solution as a crude preparation, no significant difference (P=o.os) in K, and V,,, was noted for the HS and normal protein kinase. It is assumed in this study that the relative amounts of the enzyme, protein kinase, extracted from the two types of membranes is proportional to the rest of the proteins eluted during the extraction procedure. This is supported by preliminary SDS PAGE studies on the extracted protein kinase. However, even if this were not true, no difference in K , value should be observed if the enzyme has similar affinity for ATP. The results indicate that the difference in kinetics of phosphorylation of HS and normal membrane protein(s) is probably due to the phosphate acceptor. T h e positive cooperativity behaviour of a single familial group is important in two respects. Firstly, it shows that the phosphorylation of the various peptides revealed by SDS PAGE (Rubin, 1975) may not be a random process but one in which cooperation is present between various phosphate acceptors. This cooperative behaviour was found to be absent in the crude protein kinase fraction, using histone as a phosphate acceptor. Secondly, this behaviour demonstrates that the category known as Type I HS is most probably a heterogcneous group of disorders. REFERENCES BEUTLER, E., GUINTO,E. & JOHNSON,C. (1976) Human red cell protein kinase in normal subjects and patients with hereditary spherocytosis, sickle cell disease, and autoimmune hemolytic anemia. Blood, 48,887-898. D o ~ c tJ.T., , MITCHELL, C. & HANAHAN, D.J. (1963) The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Archives of Biochemistry and Biophysics, 100, I1~130. GHCENQIJIST, A.C. & SHOHET,S.B. (1976) Yhosphorylation in erythrocyte membranes from abnormally shaped cells. Blood, 48,877-886. GUTHROW, E.C., J R , A L I E NJ.E. , & RASMUSSEN, H. (1972) Phosphorylation of an endogenous menibrane protein by an endogenous, membrane-associated cyclic adenosine 3‘,5’-monophosphatedependent protein kinase in human erythrocyte ghosts. Journal of Biological Chemistry, 247. 8145-8153. JACOB, H.S. (1969) The defective red cell in hereditary spherocytosis. Annual R e v i e w .f Medicine, 20, 41-46. LOWRY, O.H.. ROSEBROUGH, N.J.,FARR,A.L. & RAND A L L , R.J. (1951) Protein measurement with the
Fohn phenol reagentJourna1 nfBiological Chemirtry, 193,265-275. MATSUMOTO, N., YAWATA, Y. &JA(:oB, H.S. (1977) Association of decreased membrane protein phosphorylation with red blood cell spherocytosis. Blood, 499 233-239. RUBIN,C.S. (1975) Adenosine 3’: 5’-monophosphateregulated phosphor y la tion of erythrocyte m embrane proteins. Separation of membrane-associated cyclic adenosine 3’ : 5'-mono phosphatedependent protein kinase from its endogenous substrates._luurnal of Biological Chemistry, 250,9044-yo52. RUBIN,C.S., ERLICHMAN, J. & ROSEN,O.M. (1972) Molecular forms and subunit composition of a cyclic adenosine 3’,5’-monophosphate-dependentprotein kinase purified from bovine heart muscle.Journal ofBiological Chemistry, 247, 36-44. ZAIL,S.S. (1977) The erythrocyte membrane abnormality of hereditary spherocytosis. (Annotation). BritishJournal of Haematology, 37,305-3 10. ZAIL,S.S. & V A N DEN HOEK,A.K. (1975) Studies on protein kinase activity and binding of adenosine 3’s-monophosphate by membranes of hereditary spherocytosis erythrocytes. Biochemical and Biophysical Research Communications, 66. 1078-1086.
