Bioconjugate Chem. 1992, 3, 408-413
400
A New Method to Specifically Label Thiophosphorylatable Proteins with Extrinsic Probes. Labeling of Serine-19 of the Regulatory Light Chain of Smooth Muscle Myosin Kevin C. Facemyer and Christine R. Cremo' Washington State University, Department of Biochemistry and Biophysics, Pullman, Washington 99164-4660. Received May 28, 1992 We present a new method to specifically and stably label proteins by attaching extrinsic probes to amino acids that are thiophosphorylated by protein kinases and ATPyS. The method was demonstrated for labeling of a thiophosphorylatable serine of the isolated regulatory light chain of smooth muscle myosin. We stoichiometrically blocked the single thiol (Cys-108) either by forming a reversible intermolecular disulfide bond or by reacting with iodoacetic acid. The protein was stoichiometrically thiophosphorylated at Ser-19 by myosin light chain kinase and ATPyS. The nucleophilic sulfur of the protein phosphorothioate wascoupledatpH7.9and25 OC to the fluorescent haloacetate [3Hl-5-[[2-[(iodoacetyl)amino]ethyllaminolnaphthalene-1-sulfonic acid ( [3HlWDANS)by displacement of the iodide. Typical labeling efficiencies were 70-100%. The labeling was specific for the thiophosphorylated Ser-19, as determined from the sequences of two labeled peptides isolated from a tryptic digest of the labeled protein. t3H1IAEDANS attached to the thiophosphorylated Ser-19 was stable at pH 3-10 at 25 "C, and to boiling in high concentrations of reductant. The labeled light chains were efficiently exchanged for unlabeled regulatory light chains of the whole myosin molecule. The resulting labeled myosin had normal ATPase activities in the absence of actin, indicating that the modification of Ser-19 and the exchange of the labeled light chain into myosin did not significantly disrupt the protein. The labeled myosin partially retained the elevated actin-activated Mg2+-ATPaseactivity which is characteristic of IAEDANS thiophosphorylated myosin. This indicates that labeling of the thiophosphate group with [3H] did not completely disrupt the functional properties of the thiophosphorylated protein in the presence of actin.
INTRODUCTION
The reversible phosphorylation of proteins is a common regulatory mechanism in biological processes. The mechanisms by which phosphorylation modulates protein function are largely unknown, although structural data for a few proteins have begun to answer this question (1, 2). Most phosphorylations are catalyzed by serinelthreonine (3)and tyrosine protein kinases (4)and are reversed by the respective protein phosphatases. These enzymes finely regulate proteins in vivo by virtue of selective modification within consensus amino acid sequences for phosphorylation. In general it appears that most protein kinases will accept the nonphysiological substrate ATPyS (5)in place of ATP (6). Kinases which are known to use ATPyS as asubstrate are phosphorylase kinase (7), CAMPdependent protein kinase (8-11),nuclear protein kinase I1 (121,cGMP dependent protein kinase (12),protein kinase C (13,14), kinase FA(15),heme-regulated protein kinase (16)) myosin light chain kinase (17), calmodulin-dependent protein kinase I1 (18, 19), and EGF-receptor-associated protein kinase (20). A phosphorothioate instead of a phosphate is transferred to the acceptor protein with normal fidelity. In contrast, it appears that thiophosphorylated proteins tend to be poor substrates for phosphatases (6, 7,17,20, 21). Therefore, thiophosphorylated proteins mimic a stable phosphorylated state of a protein.
Although phosphorothioate and phosphate are structural analogs, they differ chemically in that phosphorothioate contains a nucleophilic sulfur (6). Nucleic acid chemists have made use of this nucleophile by attaching extrinsic probes to internucleotidic phosphorothioate diesters (22) or to single terminal phosphorothioate monoesters (23)of oligodeoxynucleotides. We have used this nucleophilic sulfur as a site to attach extrinsic probes to thiophosphorylatedproteins. This approach to labeling of proteins promises to be generally useful because of the ubiquity of protein phosphorylation, the site specificity afforded by the kinases, and the expected stability of the linkage between the extrinsic probe and the protein. Here we describe this new method and use it to modify Ser-19 on the RLC20 of smooth muscle myosin. RLC2O was chosen as a model to test the labeling method because it is a small protein (19 600 Da) of known sequence with one primary phosphorylation site and a single cysteine. This protein is of interest because phosphorylation of Ser19by myosin light chain kinase (3)is required for activation of smooth muscle contraction (24). We document the conditions and specificity of the labeling method and characterizemyosin in which thiophosphate-labeledRLC20 has been exchanged for the native RLC2O. EXPERIMENTAL PROCEDURES
Reagents. ATPyS (0.1 M solution) and MB-grade glycerol were from Boehringer Mannheim and ultrapure guanidine hydrochloride was from ICN Biochemicals. Abbreviations used: IAA,iodoacetic acid; RLC20, 20-kDa Sucrose was from SchwarzIMann (density-gradient grade). regulatory light chain of chicken gizzard myosin; LC17,17-kDa light chain of chicken gizzard myosin; DTE, dithioerythritol; [3H]- PHI IAEDANS was synthesized and characterized by the method of Hudson and Weber (25) using I3H1IAA (250 IAEDANS, tritiated form of 5-[ [2-[(iodoacetyl)amino]ethyl]mCi/mmol) from NEN. IAEDANS was purchased from aminolnaphthalene-1-sulfonicacid; ATPrS, adenosine, 5'-0-(3thiotriphosphate). Sigma. 1043-1802/92/2903-0408$03.00/0
0 1992 American Chemical Society
Specific Labeling of Thiphosphoryiatable Proteins
Proteins. Gizzards from freshly killed chickens were cleaned, and the connective tissue was removed immediately prior to rapid freezing of the gizzards in liquid nitrogen and storage at -80 "C for up to 6 months. Myosin was purified from the frozen gizzards by the method of Ikebe and Hartshorne (26) and stored on ice for up to 2 weeks. For longer storage, myosin solutions of up to 50% glycerol were made and stored at -20 "C without loss of activity. A mixture of both types of myosin light chains (RLC20 and LC17) was isolated from the myosin by the method of Perrie and Perry (27). EtOH was then added to a final concentration of 82 % to precipitate the RLC2O which was then resuspended in and dialyzed against 50 mM ammonium bicarbonate (pH 7.9). We routinely obtained an 80% yield of RLC2O which appeared to be free of LC17 and myosin heavy chains after analysis by both urea (see Figure 2 legend) and SDS gel electrophoresis. Sucrose (4 mg/mg of protein) was added prior to lyophilization and storage at -80 "C. We determined the to be 3.35 by amino acid analysis, which is in agreement with the value determined by Hathaway and Haeberle (28)and by Sobieszek (29). The protein concentration of PHI IAEDANS-labeled RLC2O was determined by the bicinchoninic acid method (30). Myosin light chain kinase was purified by the method of Adelstein and Klee (31), except that the calmodulin-affinity chromatography step was omitted. Smooth muscle phosphatase SMP-I (32) was a generous gift of Dr. M. Pato. Actin was prepared by the method of Spudich and Watt (33). Oxidation of Cys-108 to Cystine. RLC2O can be intermolecularly cross-linked by lyophilizing a 5 mg/mL protein solution in 50 mM ammonium bicarbonate, pH 7.9. The RLC2O dimers migrate above the monomers on urea gels (data not shown; DTE is omitted from the protein sample buffer). Reduction and Alkylation of Cys-108. Lyophilized RLC2O was dissolved in 50 mM ammonium bicarbonate, pH 7.9, that had been purged with argon, and fresh DTE was added to 5 mM. After 2 h at 25 "C under argon, IAA was added to 20 mM and reacted in the dark for 1h at 25 "C or on ice overnight. Thiophosphorylation. When IAA was used to block Cys-108, RLC2O was dialyzed to remove excess IAA prior to thiophosphorylation. RLC2O was thiophosphorylated in 50 mM ammonium bicarbonate, pH 7.9, 0.2-1.0 mM ATPyS, 1.5 mM CaC12, 5 mM MgClz, 4 pg/mL myosin light chain kinase, 4 pg/mL calmodulin (Sigma) for 1h at 25 "C or on ice overnight. Labeling of Phosphorothioate with [3H]IAEDANS. After thiophosphorylation, samples were dialyzed against 50 mM ammonium bicarbonate, 0.6 M KC1 to remove excess ATPyS. DTE (5 mM) was also included in this buffer if Cys-108had been irreversibly blocked with IAA. The protein was then dialyzed into 50 mM ammonium bicarbonate (previously purged with argon) to remove excesssalt and DTE (if added). L3H1IAEDANSwas added to a 5-10-fold excess over the protein concentration and allowedto react for 3 h a t 25 "C. Unreacted I3H1IAEDANS was removed by Sephadex G-25gel filtration in ammonium bicarbonate, or by precipitating twice from 2.5 M guanidine hydrochloride in 80% EtOH. RESULTS AND DISCUSSION
Our strategy for specificallylabeling thiophosphorylated proteins is shown in Figure 1. In most situations, cysteine sulfhydryls are the only reactive protein side chains under conditions in which the thiophosphate group is labeled. We show two methods to block protein sulfhydryls, either
Bioconjugate Chem., Voi. 3, No. 5, 1992 400
Step A
J
OZ
-OH
9
-S-CHzCR
2
9 1. ATPyS
Step B
2. proteinkinase
excess
9
ICH2CR
ols 9
9
3-P-
9
-0-P-S-
-0-P-S-
OH
OH
9
4
-S-CH&R
0
Step C
d
\ 9
9
-0-P-S-CHzCR' OH
8
-S-CHzCR
6
Step D
1
I
Figure 1. Outline of the method for specifically labeling thiophosphorylated proteins. Step A, blocking of thiol groups; step B, thiophosphorylation with kinase and ATPrS; step C, reaction of phosphorothioate with haloacetate derivative; step D, reduction of reversibly blocked thiol groups. See the text for details.
reversibly (left side, Figure 1)or irreversibly (right side, Figure l),in preparation for thiophosphate labeling of the RLC20 from smooth muscle myosin. The native RCL2O (structure 1 in Figure 1) is shown schematically as a vertical line with the thiophosphorylatable Ser-19 at the top and the only cysteine, Cys-108, at the bottom. In step A, Cys-108 was blocked by one of two methods. For reversible thiol blocking, intermolecular disulfide bonds were formed by oxidation with molecular oxygen to produce structure 2. Other reversible methods of blocking are possible, such as methylthiolation with S-methylthio methanesulfonate and related reagents (34). Alternatively, the cysteine was irreversibly modified by reaction with a haloacetyl derivative ICH&(O)R (in our case, IAA) to produce structure 3. It is reasonable to assume that Cys-108 is the labeled amino acid because of the well-known specificity of haloacetates for sulfhydryls
410 Bloconlugate Chem., Vol. 3,No. 5, 1992
Facemyer and Cremo
.-C s 0
Figure 2. Urea gel assay for RLC2O modification. Protein samples were precipitated with 5 volumes of cold acetone and dissolved in 8 M urea, 10 mM DTE (freshly added), 0.2 mM EDTA, 33 mM Tris-glycine, pH 8.6, and applied to a 9% polyacrylamide, 8 M urea, 0.3 M Tris-HC1, pH 8.6, gel (10 cm X 10 cm X 0.75 cm) without a stacking gel. The running buffer was 60 mM Tris-glycine, pH 8.6. Gels were run a t room temperature. Lane 1, unmodified RLC2O (1.5 pg); lane 2, RLC2O (1.5 pg) which is 100% IAA blocked at Cys-108 and 100% thiophosphorylated at Ser-19; Lane 3, smooth muscle myosin standard (30pg) showing RLC2O on top and LC 17 on bottom; Lane 4, RLC2O (1.5 pg) which is 100% IAA blocked a t Cys-108. The faint band below the RLC20 is due to a small amount of proteolytic cleavage which occurs during storage a t 4 "C.
(35). The protein was thiophosphorylated in step B on Ser-19 by myosin light chain kinase, using ATPyS as a substrate, to yield structures 4 and 5. The resonance structure of the phosphorothioatereflects the bond orders suggested by Frey and co-workers (36, 37). Following removal of excess ATP? S, the haloacetate derivative ICH&(O)R' (in our case [3H]IAEDANS)is added in step C. The phosphorothioate group on the protein displaces halogen from the haloacetate to produce the stable structures 6 and 7. It is possible that maleimides may also be used in this step. However, we have not thoroughly explored this possibility as our initial studies indicated that maleimides react nonspecifically. The disulfide bond in structure 6 can be reversed by treatment with reductant such as DTE to yield structure 8. The above reactions with RLCBO can be followed by urea gel electrophoresis (see Figure 2 for gel of samples which represent examples of structures I, 3, and 5 from Figure 1). An increase of a single negative charge causes proteins to migrate faster in this gel system. Thus when IAA was used to block Cys-108 (structure 3) the protein migrated below native RLCBO (structure l),reflecting the stoichiometric addition of a single negative charge. Addition of a negatively charged phosphorothioateto structure 3 to generate structure 5 caused an additional increase in the migration rate in the gel. In our case, 5 (Figure 2) and 7 (not shown) comigrate because a charge is both added and removed during the labeling of the phosphorothioate with [3H]IAEDANS. These gel electrophoresis results indicate that both cysteine labeling and thiophosphorylation both proceed to essentially 100% . However, this assay was not useful to examine the reaction of the haloacetate with the thiophosphorylated serine (step C, Figure 1). The time course of the reaction of thiophosphorylated Ser-19with [3H]IAEDANS was examined by precipitating the protein with acetone at various times after addition of [3H]IAEDANS(Figure 3). The counts per minute in each thiophosphorylated protein pellet were corrected for background by subtracting the counts which were incorporated into an otherwise identical but phosphorylated protein sample. The raw data for one of the time courses is shown in the inset. The maximal labeling achieved under these conditions was 90%. These data show that the reaction is highly specific for the thiophosphate group and nearly stoichiometric labeling can be achieved under these conditions. We found that substoichiometric labeling would occur if the IAA-labeled RLC20 was not treated with DTE prior to labeling with [3H]IAEDANS,
1 /=
/@
y' 0
reaction time (min)
50
la,
200
150
250
reaction time (min)
Figure3. Time course of specificlabeling of thiophosphorylated RLC2O a t 25 "C. RLC2O (0.49 mg/mL; 25 pM in ammonium bicarbonate, pH 7.9,5 mM MgC12) was blocked a t Cys-108 with IAA and then either phosphorylated or thiophosphorylated on serine-19 as described in the methods. At time = 0, [3HlIAEDANS (4765 cpm/nmol) was added to the following final 750 pM. At the concentrations: V, 250 pM; 0 , 500 p M ; indicated times, triplicate aliquots (0.03 mL) were removed and the protein was precipitated with 0.4 mL of ice-cold acetone to remove excess r3H]IAEDANS. The mixture was vortexed and incubated at -20 "C for 15 min prior to centrifuging a t 4 "C in a microfuge (Beckman) for 2 min. The pellets were washed with cold acetone twice, dissolved in 4% sodium dodecyl sulfate, and transferred to scintillation counting vials. For each time point, the data are the averaged counts for the thiophosphorylated minus those for the phosphorylated RLCBO. The inset shows the raw data for the time course a t 250 pM [3H]IAEDANS (v,phosphorylated; V, thiophosphorylated).
.,
suggesting that some IAA remains bound to the protein during dialysis. Although the reactivity of the thiophosphate may differ from protein to protein, Baraniak and Frey (38)have used model compounds to predict that even shielded phosphorothioates that are ion paired with guanidinium or ammonium ions (as may be found in proteins) may have only modest perturbations in electronic distributions and thus may have similar chemical reactivity. Therefore it is likely that thiophosphorylated amino acids of other proteins will show a similar reactivity toward haloacetates as described here (Figure 3). It should be noted that substoichiometriclabeling could occur if the protein is partially phosphorylatedduring the thiophosphorylation reaction (step B, Figure 1) due to contaminating ATP in ATPyS solutions. We tested for this possibility by treating thiophosphorylated RLCBO with light chain phosphatase [under the conditions described by Sellers et al. (39)], which is known to rapidly remove phosphate but not thiophosphate from RLCBO. By urea gel electrophoresis (data not shown) we showed that all of the RLCBO was indeed thiophosphorylated, and was thus unresponsive to phosphatase treatment. This can be explained because our myosin light chain kinase is contaminated with phosphatase (31). Thus, during our thiophosphorylation reaction, the small percentage of RLCBO which was initially phosphorylated (by ATP) was eventually dephosphorylated and then thiophosphorylated. However, in general when it is important to achieve stoichiometric labeling, it may be useful to assay for complete thiophosphorylation prior to thiophosphate labeling. We established directly the specificity of the phosphorothioate labeling by purifying and sequencing the labeled
BioconJugate Chem., Voi. 3, No. 5, 1992 411
Speclfic Labeling of Thiphosphorylatabie Proteins 0
40
20
80
60
,
100
I
I
70 60
3M)
250
50
2
450
$ 300 g
2
='
1
B-
-
I!,,:
I!...................................
..........,........,..*,............**..
450 -
0
300 -
.,..................
8
0
30
...............,.'..*..............:'.'.........'......,.
f . . .
. . . . . . . . . . . . . . . . . . , . . . . e . . .
2 T
D
z.
10
0
1 A
3
w
5
50
0
g
20
C -
a
$ 150 -
150 100
8
3
40
i;
l 150
-s k
200
0 3 X
4 N
5 V
6 F
7 A
8
.......L1
Cycle
Figure 5. Amino acid sequence analysis of phosphorothioatelabeled peptide. Peptides were sequenced with an AB1 Model 470-A gas-phase sequencer with Pulsed Liquid update (Applied Biosystems,Inc.) and AB1 120-A P T H Analyzer. Data acquisition and analyses were performed with the standard program RUN 470-1 (ABI). One nanomole of peptide (4765 counts per cpm/ nmol) from fraction 69 (see Figure 4) was applied to the filter. The initial yield is typically 50% for peptide standards on this instrument. The sample stream was split in half to simultaneously analyze the yield of each amino acid (bars) and the counts per minute corresponding to each cycle (filled circles). The sequence is indicated below the cycle numbers. X indicates that no known amino acid eluted in this cycle. The yield shown for cycle 3 is for serine.
noted by others (41). The sequence of the minor labeled peptide (fraction 41 from Figure 4B) and ATXNVF, where X indicates that an unknown amino acid eluted at this position (data not shown). These data confirm that the labeling of structure 5 (Figure 1)by i3H1IAEDANSwas specific for the thiophosphorylated serine residue. It was important to demonstrate that the linkage between the protein and the 13HlIAEDANSin structure 6 was stable to reducing conditions (step D, Figure 1).We detected no loss of i3H1IAEDANSfrom the protein after treatment with 1mM 8-mercaptoethanol for 120 h at 37 "C or after boiling for 30 min with 110 mM 6-mercaptoethanol (data not shown). This is particularly important to those who may need to analyze labeled proteins by SDS gel electrophoresis, which often requires heating the protein with high concentrations of reductant. In addition we found that the I3H1IAEDANS remained coupled to the protein after incubation at pH 3.0 to pH 10.5 for 23 h a t 25 OC (data not shown). We did not test the stability of the linkage outside this pH range, but it is likely that strong base will cleave the C-S bond (5). This data indicates that the l3H1IAEDANS-labeled protein is stable under most physiological conditions. We provide evidence that the thiophosphate-labeled RLC2O remained in the native conformation by obtaining efficient exchange of thiophosphate-labeled RLC2O for unlabeled RLCBO of the whole smooth muscle myosin molecule. The resulting labeled myosin had normal K+, NH4+-EDTA-, and Ca2+-ATPaseactivities in the absence of actin (data not shown). This indicates that the modification of Ser-19 and the treatment required for exchange of the labeled light chain into myosin did not significantly disrupt the myosin. This also allowed us to specifically examine the effect of the thiophosphate labeling upon the actin-activated Mg2+-ATPaseactivity of myosin which reflects its ability to generate force in the muscle (24).
412
Facemyer and Cremo
Bioconjugate Chem., Vol. 3, No. 5, 1992
Table I. Actin-Activated Mg*+-ATPaseActivities actin-activated RLC2O used % of myosin MgZt-ATPase activity sites (nmol/min per mg) myosin during RLC2O exchangedb -TPd +TPd samde exchange" 0.74 25 control none 0.78 23 exchanged unmodified control 50 2.9 14 exchanged thiophosphate labeled I
a We used the method of Dr. K. Trybus (unpublished results) to exchange light chains into whole myosin. Two mg/mL myosin and 1.6mg/mL RLC2O (20-foldmolar excess over myosin, either unlabeled or [3H]IAEDANS-thiophosphate-labeledwith IAA on Cys-108) in 0.5 M NaC1,5 mM EDTA, 10 mM DTE, 10 mM sodium phosphate, pH 7.5,l mM ATP, was incubated at 42 OC for 30 min. To stabilize the binding of the RLC2O to the myosin, MgClz was added to 20 mM at 4 OC. The samples were dialyzed at 4 "C into 15 mM Tris, pH 7.5, 1 mM DTE, 10 mM MgClZ, and the filamentous myosin was centrifuged. After the free light chains in the supernatant were removed, the pellet was resuspended and the centrifugation was repeated. This procedure was repeated once more and the pellet was resuspended in 15 mM Tris, pH 7.5 (at 25 "C), 5 mM MgC12,O.l mM EGTA prior to thiophosphorylation as described in the methods. The control was treated identically except that it was kept on ice and RLC2O was not added. * The extent of exchange of the RLC2O was calculated after determining the protein concentration by the bicinchoninic acid method (30) and counting the solution for [SHIIAEDANS. A malachite green based spectroscopicmethod (49) was used to follow the generation of inorganic phosphate in the presence of 1mg/mL myosin, 1mg/mL actin in 15 mM Tris, pH 7.5 (at 25 "C), 5 mM MgC12,O.l mM EGTA, 30 mM KCl with 1mM ATP at 25 "C. d+TP is thiophosphorylated and -TP is not thiophosphor ylated.
Either phosphorylation or thiophosphorylation of Ser19 increases the actin-activated Mg2+-ATPaseactivity of smooth muscle myosin (24,42). As shown in Table I, the activity of an untreated myosin sample increased 34-fold upon thiophosphorylation. Myosin that had been exposed to the exchange conditions in the presence of unlabeled light chains showed a similar level of activation upon thiophosphorylation (Table I), indicating that the exchange procedure does not alter this critical property of the protein. It has been shown previously that exchange with Cys-108-modified light chains does not alter the activity of smoothmuscle myosin (43). After the exchange procedure in the presence of I3H1IAEDANS-thiophosphate-modified RLC2O (and IAA on Cys-108),50% of the unmodified in situ light chains had been exchanged with modified ones (Table I). This preparation partially retained the elevated actin-activated Mg2+-ATPase activity (2.9 nmol/min per mg) which is characteristic of thiophosphorylated myosin. This indicates that at least in this case, removal of the negative charge of the thiophosphate by modifying it with PHIIAEDANS did not completely disrupt the actin-dependent functional properties of the protein. Further data is needed to establish the consequences and significance of modifications of this type. Application of our strategy to other proteins may be limited. Specific labeling may not be achieved if a protein contains haloacetate reactive groups which may not be effectively blocked (44). In addition, this technique may not be useful to modify proteins which have many thiols that must be reversibly blocked. For example, we have had limited success applying the technique to thiophosphorylated smooth muscle myosin (45) which has many accessible thiols (46). In this case it was important to use a reversible method of thiol blocking, because thiol modification greatly altered the activity of the protein. We found that S-methylthiomethanesulfonate served as
an efficient thiol blocker, but we were able to return only limited activity to the protein after reduction with DTE. We had more success reversing the effects of 5,5'-dithiobis(2-nitrobenzoicacid) (Ellman's reagent), whichis a charged thiol blocker. These data suggest that activity-critical methylthio groups may become buried in the enzyme and are not accessible to DTE, as has been found for other proteins (47, 48). We are currently testing Bu3P as an alternative reducing agent, which has the potential to access more hydrophobic sites on the protein (48). Despite these potential limitations, it is likely that our thiophosphate labeling approach will be useful for a variety of proteins. In summary, we have demonstrated a new technique to specifically label thiophosphorylatable amino acids on proteins with extrinsic probes. Specifically, we have applied the technique to label a thiophosphorylatable serine residue of the RLCPO of smooth muscle myosin. We have focused on utilizing the thiophosphate labeling method to modify a serine; however, the method may also work for modifying threonines and tyrosines. The labeling reaction proceeds within hours at slightly alkaline pH and is highly specific for the thiophosphate moiety, and the product is stable to a wide pH range and to the presence of high concentrations of reductant. We feel that this approach to protein labeling is potentially useful for many proteins. By using various kinds of extrinsic probes, such as fluorophores, spin labels, electron-dense derivatives, and various antigens, this method may provide a new way to examine the structural aspects and functional implications of protein phosphorylation. ACKNOWLEDGMENT
Dr. G. Munske performed the sequence analyses at the WSU Laboratory for Bioanalysis and Biotechnology. We thank M. Tibeau for technical assistance. We thank Dr. K. Nakamaye and Dr. L. McLaughlin, who gave advice and encouragement at the initiation of this project, Dr. K. Trybus for communicating her procedure for the exchange of RLC20 into smooth muscle myosin prior to publication, and Dr. M. Pato for providing valuable assistance concerning RLC2O phosphatases. LITERATURE CITED (1) Hurley, J. H., Dean, A. M., Sohl, J. L., Koshland, D. E. J.,
and Stroud, R. M. (1990) Regulation of an Enzyme by Phosphorylation a t the Active Site. Science 249,1012-1016. (2) Sprang, S. R., Acharya, K. R., Goldsmith, E. J., Stuart, D. I., Varvill, K., Fletterick, R. J., Madsen, N. B., and Johnson, L. N. (1988) Structural changes in glycogen Phosphorylase induced by phosphorylation. Nature 336, 215-221. (3) Edelman, A. M., Blumenthal, D. K., and Krebs, E. G. (1987) Protein Serine/Threonine Kinases. Annu. Rev. Biochem. 56, 567-613. (4) Hunter, T., and Cooper, J. A. (1985) Protein-Tyrosine Kinases. Annu. Rev. Biochem. 54, 897-930. ( 5 ) Goody, R. S.,and Eckstein, F. (1971) Thiophosphate Analogs 93, of Nucleoside Di and Triphosphates. J. Am. Chem. SOC. 6252-6257. (6) Eckstein, F. (1985) Nucleoside Phosphorothioates. Annu. Rev. Biochem. 54,367-402. ( 7 ) Gratecos, D., and Fischer, E. H. (1974) Adenosine 5'-0-(3thiotriphosphate) in the Control of Phosphorylase Activity. Biochem. Biophys. Res. Commun. 58,960-967. (8) Moffett, R. B., and Webb, T. E. (1983) Characterization of a Messenger RNA Transport Protein. Biochim. Biophys. Acta 740, 231-242.
Speclfic Labeling of Thiphosphorylatable Proteins
(9) O’Brian, C. A., Roczniak, S. O., Bramson, H. N., Baraniak, J., Stec, W. J., and Kaiser, E. T. (1982) A Kinetic Study of Interaction of (R& and (S,)-Adenosine Cyclic 3’,5’-Phosphorothioates with Type I1 Bovine Cardiac Muscle Adenosine Cyclic 3’,5’-Phosphate Dependent Protein Kinase. Biochemistry 21, 4371-4376. (10) Watabe, S., Tsuchiya, T., and Hartshorne, D. J. (1989) Phosphorylation of Paramyosin. Comp. Biochem. Physiol. 9 4 ~813-821. , (11) Sun, I. Y.-C., Johnson, M., and Allfrey, V. G. (1980) Affinity Purification of Newly Phosphorylated Protein Molecules. J . Biol. Chem. 255, 742-747. (12) Palvimo, J., Linnala-Kankkunen, A., and Maenpaa, P. H. (1985) Thiophosphorylation and Phosphorylation of Chromatin Proteins from Calf Thymus InVitro. Biochem.Biophys. Res. Commun. 126, 103-108. (13) Sarafian, T., Pradel, L.-A,, Henry, J.-P., Aunis, D., and Bader, M.-F. (1991)The Participation of Annexin I1 (Calpactin I) in Calcium-evoked Exocytosis Requires Protein Kinase C. J . Cell. Biol. 114, 1135-1147. (14) Giambalvo, C. (1988) Protein Kinase C and Dopamine Release. Biochem. Pharm. 37, 4001-4007. (15) Vandenheede, J. R., Yang, S.-D., Merlevede, W., Jurgensen, S., and Chock, B. (1985) Kinase FA-mediated Regulation of Rabbit Skeletal Muscle Protein Phosphatase. J. Biol. Chem. 260, 10512-10516. (16) Ranu, R. S. (1986)Regulation of Protein Synthesis in Rabbit Reticulocyte Lysates. FEBS Lett. 208, 117-122. (17) Sherry, J. M. F., Gorecka, A. A., M. O., Dabrowska, R., and Hartshorne, D. J. (1978)Roles of Calcium and Phosphorylation in the Regulation of the Activity of Gizzard Myosin. Biochemistry 17, 4411-4418. (18) Ikeda, A., Okuno, S., and Fujisawa, H. (1991) Studies on Activity of the Generation of Ca2+/Calmodulin-independent Calmodulin-dependent Protein Kinase I1 by Autophosphorylation. J. Biol. Chem. 266, 11582-11588. (19) Nichols, R. A., Sihra, T. S., Czernik, A. J., Nairn, A. C., and Greengard, P. (1990) Calcium/calmodulin-dependentProtein Kinase I1Increases Glutamate and Noradrenaline Release from Synaptosomes. Nature. 343, 647-651. (20) Cassel, D., and Glaser, L. (1982) Resistance to Phosphatase of Thiophosphorylation Epidermal Growth Gactor Receptor in A431 Membranes. Proc. Natl. Acad. Sci. U S A . 79,22312235. (21) Nichols,R. A., Sihra, T. S., Czernik,A. J., Nairn, A. C., and Greengard, P. (1990) Calcium/calmodulin-dependentprotein kinase I1 increases glutamate and noradrenaline release from synaptosomes. Nature 343, 647-651. (22) Hodges, R. R., Conway, N. E., andMcLaughlin, L. W. (1989) “Post-Assay” Covalent Labeling of Phosphorothioate-Containing Nucleic Acids with Multiple Fluorescent Markers. Biochemistry 28, 261-267. (23) Cosstick, R., McLaughlin, L. W., and Eckstein, F. (1984) Fluorescent labelling of tRNA and oligodeoxynucleotides using T4 RNA ligase. Nucleic Acids Res. 12, 1791-1810. (24) Hartshorne, D. J. (1987) Biochemistry of the Contractile Process in Smooth Muscle, 2nd ed. (L. R. Johnson, Ed.) pp 423-480, Raven Press, New York. (25) Hudson, E. N., and Weber, G. (1973) Synthesis and Characterization of Two Fluorescent Sulfhydryl Reagents. Biochemistry 12, 4154-4161. (26) Ikebe, M., and Hartshorne, D. J. (1985) Effects of Ca2+on the Conformation and Enzymatic Activity of Smooth Muscle Myosin. J . Biol. Chem. 260, 13146-13153. (27) Perrie, W. T., and Perry, S. V. (1970) An Electrophoretic Study of the Low-Molecular-Weight Components of Myosin. Biochem. J. 119, 31-38. (28) Hathaway, D. R., and Haeberle, J. R. (1983) Selective Purification of the 20,000-Da Light Chains of Smooth Muscle Myosin. Anal. Biochem. 135, 37-43.
Bioconjugate Chem., Vol. 3, No. 5, 1992
413
(29) Sobieszek, A. (1988) Bulk Isolation of the 20,000-Da Light Chain of Smooth Muscle Myosin: Separation of the Unphosphorylated and Phosphorylated Species. Anal. Biochem. 172, 43-50. (30) Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia,A. K., Gartner, F. H., Provenzano, M. D., Fugimoto, E. K., Goeke, N. M., Olson, B. J., and Klenk, D. C. (1985) Measurement of Protein Using Bicinchoninic Acid. Anal. Biochem. 150, 7685. (31) Adelstein, R. S., and Klee, C. B. (1982) Purification of Smooth Muscle Myosin Light Chain Kinase. Methods Enzymol. 85, 298-308. (32) Pato, M., and Adelstein, R. S. (1982) Purification of Smooth Muscle Phosphatases. Methods Enzymol. 85, 308-315. (33) Spudich, J. A., and Watt, S. (1971) The Regulation of Rabbit Skeletal Muscle Contraction. J. Biol. Chem. 246,4866-4871. (34) Smith, D. J., Maggio, E. T., and Kenyon, G. L. (1975) Simple Alkanethiol Groups for Temporary Blocking of Sulfhydryl Groups on Enzymes. Biochemistry 14, 766-771. (35) Means, G. E., and Feeney, R. E. (1971) Chemical Modification of Proteins, Holden-Day, Inc., San Francisco. (36) Frey, P. A., and Sammons, R. D. (1985) Bond Order and Charge Delocalization in Nucleoside Phosphorothioates. Science 228, 541-545. (37) Frey, P. A., Reimschussel, W., and Paneth, P. (1986) Phosphorus-Sulfur Bond Order in Phosphorothioate Anions. J. Am. Chem. SOC.108, 1720-1722. (38) Baraniak, J., and Frey, P. A. (1988) Effect of Ion Pairing on Bond Order and Charge Localization in Alkyl Phosphorothioates. J. Am. Chem. SOC.110, 4059-4060. (39) Sellers, J. R., Pato, M. D., and Adelstein, R. S. (1981) Reversible Phosphorylation of Smooth Muscle Myosin, Heavy Meromyosin, and Platelet Myosin. J . Biol. Chem.256,1313713142. (40) Matsuda, G. (1983) The Light Chains of Muscle Myosin: It’s Structure, Function, and Evolution. Adu. Biophys. 16, 185-218. (41) Murakami, N., Healy-Louie, G., and Elzinga, M. (1990) Amino Acid Sequence around the Serine Phosphorylated by Casein Kinase I1 in Brain Myosin Heavy Chain. J. Biol. Chem. 265, 1041-1047. (42) Kenney, R. E., Hoar, P. E., and Kerrick, G. L. (1990) The Relationship between ATPase Activity, Isometric Force, and Myosin Light-chain Phosphorylation and Thiophosphorylation in Skinned Smooth Muscle Fiber Bundles from Chicken Gizzard. J . Biol. Chem. 8642-8649. (43) Morita, J.-I., Takashi, R., and Ikebe, M. (1991) Exchange of the Fluorescence-Labeled 20000-Dalton Light Chain of Smooth Muscle Myosin. Biochemistry 30, 9539-9545. (44) Gurd, F. R. N. (1967) Methods Enzymol. 11, 532-541. (45) Facemyer, K. C., and Cremo, C. R. (1991) Specific Labeling of Serine-19of the Regulatory Light Chain of Chicken Gizzard Myosin. Biophys. J. 59, 436a. (46) Chandra, T. S., Nath, N., Suzuki, H., and Seidel, J. C. (1985) Modification of Thiols of Gizzard Myosin Alters ATPase Activity, Stability of Myosin Filaments, and the 6-10 S Conformational Transition. J. Biol. Chem. 260, 202-207. (47) Bednar, R. A., McCaffrey, C., and Shan, K. (1991) Introduction of Unnatural Amino Acids into Chalcone Isomerase. Bioconjugate Chem. 2, 211-216. (48) Nishimura, J. S., Kenyon, G. L., and Smith, D. J. (1975) Reversible Modification of the Sulfhydryl Groups of Escherichia coli Succinic Thiokinase with Methanethiolating ReAcid), p-Hydroxymeragents, 5,5’-Dithio-bis-(2-Nitrobenzoic curibenzoate, and Ethylmercurithiosalicylate. Arch. Biochm. Biophys. 170, 461-467. (49) Lanzetta, P. A., Alvarez, L. J., Reinach, P. S., and Candia, 0. A. (1979) An Improved Assay for Nanomole Amounts of Inorganic Phosphate. Anal. Biochem. 100,95-97.
