- Email: [email protected]
Synthesis and properties of sulfhydryl-group-specific reagents containing 125I
ANALYTICAL
BIOCHEMISTRY
(1991)
198,165-173
Synthesis and Properties of Sulfhydryl-Group-Specific Reagents Containing 12511 Rick
J. Krueger,
Department
Received
April
of
Chaomei
Biochemistry
Lin,
and Julie
and School of Biological
A. Frank Sciences, University
Nebraska,
Lincoln,
Nebraska,
68583-0718
1, 1991
groups 12”1-containing compounds that react specifically with sulfhydryl groups were prepared in yields of 30 to 40% on the basis of starting la61 quantity. The synthetic precursors were commercially available heterobifunctional crosslinkers and the peptide L-arginyl-L-tyrosine. Two types of sulfhydryl specific reagents were prepared: 3-(2-pyridyldithio)propionylarginyl-[’221Jmonoiodotyrosine, which permits reversible incorporation of “‘1 at sulfhydryl sites, and 3-maleimidopropionylarginyl-[‘261]monoiodotyrosine, an irreversible labeling reagent. These products were isolated in a highly radiochemically pure form by Cl8 HPLC. The second-order rate constants for the reaction of 3-(2pyridyldithio)propionylarginylmonoiodotyrosine and 3 -maleimidopropionylarginylmonoiodotyrosine with N-acetylcysteine were 28 f 3 M-’ s-l and 154 + 4 M-l s-l, respectively, at pH 7.3. Storage of carrier-free 3(2-pyridyldithio)propionylarginyl-[12sI]monoiodotyrosine and 3-maleimidopropionylarginyl-[‘261]monoiodotyrosine at -80°C at a radioactive concentration of 0.4 mCi/ml resulted in conversion of “‘1 to species that did not react covalently with sulfhydryl groups. This process occurred with first-order kinetics and a t,,, of 5.7 days for the pyridyldithio compound and 7.5 days for the maleimido compound. No conversion was observed during storage at -80°C at radioactive concentrations of 0.02 mCi/ml or less. The labeling properties of these compounds were examined using red blood cell proteins as a test system. 3-(2-Pyridyldithio)propionylarginyl-[ ‘261]monoiodotyrosine and maleimidopropionylarginyl-[ ‘261]monoiodotyrosine reacted preferentially with membrane - associated sulfhydryl
1 Agricultural Research Division, University of Nebraska Journal Series 9543; supported by the University of Nebraska Research Council and NIH Biomedical Research Support Grant 07055. 0003-2697/91 Copyright All rights
of
$3.00 0 1991 hy Academic Press, of reproduction in any form
Q 1991
when Academic
incubated Press,
with
intact
red
blood
cells.
Inc.
Chemical modification of proteins is useful for the incorporation of labels that can be detected with high sensitivity. 1251is a frequently employed label because it can be obtained at low cost in high specific activity and because it can be conveniently incorporated into tyrosyl residues of proteins. Bolton and Hunter (1) have developed chemical modification reagents containing “‘1 that permit the incorporation of a radiolabel at amino groups. These reagents are particularly useful for labeling proteins either devoid of tyrosine or whose activity is modified by standard iodination procedures. Chemical modification reagents that permit the introduction of 1251at sulfhydryl groups have also been described. Holowka described the synthesis and properties of ‘251-labeled N-chloroacetyltyramine (2) and demonstrated that this compound reacts with substantial specificity at sulfhydryl groups. Subsequently, we described the synthesis of carrier-free N-iodoacetyl-[3-‘251]monoiodotyramine (3), which reacts more rapidly with sulfhydryl groups than N-chloroacetyltyramine. Compounds such as these do permit incorporation of 1251 at sulfhydryl sites, but alkylhalides also react with other functional groups found in proteins (4). In addition, the rate of reaction of N-iodoacetyltyramine with sulfhydryl groups is relatively slow compared with reaction rates of sulfhydryl groups with maleimides and dithiopyridines (35-7). The high rate of reaction of maleimides with sulfhydryl groups, coupled with the limited reactivity of maleimides with other functional groups found in proteins, results in a high specificity of sulfhydryl group labeling by maleimides (5). Dithiopyridyl compounds exhibit more favorable reaction properties, with their la165
Inc. reserved.
166
KRUEGER, Step
MN, AND FRANK
1
H,N-(Arg-Tyr)-COO-
f
125
I-
-b
H,N-(Arg-["51]-Tyr)-COOArg-ITyr
Step 2a Arg-ITyr
t
S-S-(CH2)?-!-0-Nb
ci
--+
0
' S-S-(CH,),-C-NH-(Arg-['251]-Tyr)-COOTP-Arg-ITyr
Step Zb 0 0 Arg-ITyr
+
C
N-,,,,.N5
1
\\ 0
FIG. 1.
Synthesis the monoiodopeptide.
of TP-Arg-ITyr The reactions
+
6)
\\ 0
i
Mal-Arg-ITyr
and Mal-Arg-ZTyr. The a-amino and cY-carboxyl groups of the dipeptide are shown. Step 1 is formation resulting in the production of TP-Arg-ITyr and Mal-Arg-ITyr are shown in steps 2a and 2b, respectively.
beling of sulfhydryl groups being one of the most selective reactions of any protein modification reagent (6). These properties have been utilized to study sulfhydry1 groups associated with membrane proteins. Membrane-impermeant maleimides containing 35Shave previously been employed to determine the quantity and topography of sulfhydryl groups associated with red blood cell membranes (8). Spectrophotometric measurement of aryl disulfide reduction has also been used in the study of red blood cell membrane sulfhydryl groups (9). We describe here conditions for the synthesis of carrier-free ‘251-labeled compounds containing pyridyldithio or maleimido functional groups. The reactions are carried out in one vessel using commercially available reagents (Fig. 1). In addition to radioactive iodine, the reactant requirements are: (a) a compound containing a phenol ring with a covalently linked amine and (b) a heterobifunctional crosslinker containing N-hydroxysuccinimide as the amine reactive group and either a pyridyldithio or a maleimido function as the sulfhydryl reactive group. MATERIALS
N-(CH,)2-C-NH-(Arg-['251]-Tyr)-COOm '
AND METHODS
L-Arginyl-L-tyrosine, acetate salt (Arg-Tyr),2 was obtained from Bachem. Chloramine-T, N-ethylmaleimide 2 Abbreviations used: Arg-Tyr, L-arginyl-L-tyrosine; Arg-ITyr, arginyl-L-monoiodotyrosine; Arg-I,Tyr, L-arginyl-L-diiodotyrosine; Mal-Arg-ITyr, 3-maleimidopropionyl-L-arginyl-L-monoiodotyrosine; TP-Arg-ITyr, 3-(2-pyridyldithio)propionyl-L-arginyl-L-monoiodo-
L-
of
(NEM), 3-(2-pyridyldithio)propionic acid N-hydroxysuccinimide ester (TP-NHS ester), and 3-maleimidopropionic acid N-hydroxysuccinimide ester (Mal-NHS ester) were purchased from Aldrich. Na’251 (carrier-free, NEZ033L) was from New England Nuclear. Dithiothreitol (DTT), 5,5’-dithiobis(2-nitrobenzoic acid) (DTNB), N-acetylcysteine, and bovine serum albumin were obtained from Sigma.
Synthesis of [1271]arginylmonoiodotyrosine Nonradioactive Arg-ITyr was prepared by chloramine-T-mediated iodination of Arg-Tyr, and the reaction product was purified by Cl8 HPLC. For the iodination reaction, the following components were added to a 1.5-ml microfuge tube in a total volume of 270 ~1: ArgTyr (15 pmol), Na12?I (5 pmol), Hepes buffer, pH 7.3 (75 pmol). The reaction was initiated by the addition of 15 pmol chloramine-T (30 ~1 aqueous solution). The mixture was incubated for 3 min at room temperature. The reaction was terminated by addition of DTT (15 pmol), and the reaction products were resolved on a 15-pm Bakerbond Cl8 HPLC column (1 X 25 cm). The HPLC solvents used were solvent A, 0.05% trifluoroacetic acid, and solvent B, 0.05% trifluoroacetic acid, 50% acetonityrosine; DTT, dithiothreitol; DTNB, 5,5’-dithiobis(2-nitrobenzoic acid); NEM, N-ethylmaleimide; TP-NHS ester, 3-(2-pyridyldithio)propionic acid N-hydroxysuccinimide ester; Mal-NHS ester, 3-maleimidopropionic acid N-hydroxysuccinimide ester; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
‘261-CONTAINING
SULFHYDRYL-GROUP-SPECIFIC
trile. The separation was achieved with a linear gradient running from 90% solvent A to 70% solvent A over 20 min (flow rate, 4 ml/min). Under these conditions ArgTyr eluted at 6.4 min, Arg-ITyr at 14.6 min, and Arg1,Tyr at 23.1 min. The mono- and diiodinated forms of Arg-Tyr were identified and quantified by their characteristic ultraviolet absorption spectra in acid and base (10). Synthesis
and Characterization
of [lz71] TP-Arg-ITyr
[‘27]TP-Arg-ITyr was prepared by reaction of ArgITyr with TP-NHS ester as follows: 180 hl of 4.4 mM Arg-ITyr aqueous stock solution (0.8 pmol), 120 ~1 of 10 InM TP-NHS ester stock solution in acetone (1.2 pmol), 40 ~1 of 1.0 M Hepes * NaOH, pH 7.3 (40 pmol), and 460 ~1 of water were added to a microfuge tube and incubated for 30 min at room temperature. The reaction products were resolved by Cl8 HPLC (as above) using a linear gradient of 90% solvent A to 40% solvent A over 25 min. The elution profile was monitored at 274 nm. The only major product observed eluted at 25.0 min. Previous analytical scale injections made at intervals during the reaction had indicated an increase in the area of the R, = 25 min peak concomitant with a decrease in the area under the peaks representing Arg-ITyr and TP-NHS ester. The species eluting at 25.0 min was identified as TPArg-ITyr by the following criteria: (a) Ultraviolet absorption spectra in acid and base (10) indicated the presence of monoiodotyrosine. (b) Addition of N-acetylcysteine to a solution of the RT = 25 min material gave a species with an ultraviolet absorption spectrum identical to that of 2-thiopyridinone (7). (c) The Cl8 HPLC mobility of the RT = 25 min species shifted upon addition of N-acetylcysteine, yielding two new components. The Cl8 mobility of one of these components was identical to that of 2-thiopyridinone. No shift in Cl8 retention time or change in ultraviolet absorption occurred when lysine was added to the solution of the RT = 25 min species. The concentration of TP-Arg-ITyr in stock solutions was initially determined spectrophotometrically from the quantity of 2-thiopyridinone produced by addition of excess DTT. The value used for the 2-thiopyridinone extinction coefficient was 7600 M-’ cm-’ at 343 nm (7). An extinction coefficient of 7110 M-’ cm-’ at 284 nm was determined for TP-Arg-ITyr. This value was used for the spectrophotometric determination of TP-ArgITyr concentration in all further studies. Synthesis and Characterization
of [1271]Mal-Arg-ITyr
Nonradioactive Mal-Arg-ITyr was prepared as described above for TP-Arg-ITyr except that Mal-NHS ester was substituted for TP-NHS ester. The reaction products were resolved on Cl8 HPLC using a linear gra-
REAGENTS
167
dient of 90% solvent A to 60% solvent A over 15 min. The only major component in the Cl8 profile (monitored at 274 nm) eluted at 18.8 min. (Note that RT values indicate real time in the gradient and are not corrected for the volume between the gradient mixing port and the detector.) Previous analytical HPLC analyses demonstrated that this peak increased in area during the reaction as the Arg-ITyr peak decreased in area. The species eluting at 18.8 min was identified as MalArg-ITyr by the following criteria: (a) Ultraviolet absorption spectra in acid and base indicated the presence of monoiodotyrosine. (b) A shoulder was present at 300-350 nm in the ultraviolet absorption spectrum of the RT = 18.8 min component, consistent with the presence of the maleimidopropionyl group. This shoulder disappeared after the addition of N-acetylcysteine (but not lysine). (c) A shift in the Cl8 mobility of the RT = 18.8 min species was observed after addition of Nacetylcysteine (but not lysine). The concentration of Mal-Arg-ITyr in stock solutions was initially determined by reaction of the maleimido group with thiols. A known volume of Mal-Arg-ITyr stock solution was added to a solution of N-acetylcysteine. The decrease in N-acetylcysteine sulfhydryl groups was determined by reaction with DTNB, using the extinction coefficient 13,600 M-’ cm-’ at 412 nm (11). A Mal-Arg-ITyr extinction coefficient of 2980 M-’ cm-’ at 285 nm was determined by this method. Kinetic Analysis of the Reaction of TP-Arg-ITyr and Mal-Arg-ITyr with N-Acetylcysteine and Bovine Serum Albumin All rate constants were determined from reactions carried out at room temperature in 0.25 M Hepes * NaOH, pH 7.3. The sulfbydryl group concentration of N-acetylcysteine and bovine serum albumin stock solutions was determined by reaction with DTNB (7). Bovine serum albumin containing an average of 0.5 sulfhydry1 groups per molecule was prepared as previously described (3). The rate of reaction of TP-Arg-ITyr with N-acetylcysteine and bovine serum albumin was determined spectrophotometrically by measuring the absorbance change at 343 nm associated with production of 2-thiopyridinone (7). Reactions were carried out at a range of N-acetylcysteine and bovine serum albumin concentrations for each TP-Arg-ITyr concentration. Reaction rates were determined from the initial absorbance changes. The rate of Mal-Arg-ITyr reaction with N-acetylcysteine was monitored by using Cl8 HPLC to resolve reaction products at timed intervals after the reaction was initiated. The column eluate was monitored at 285 nm, and data were collected and peak areas determined with a Waters Maxima 820 system. The reaction rate was
168
KRUEGER,
LIN,
determined from the decrease in Mal-Arg-ITyr concentration. Peak areas were converted to nanomole quantities by reference to standard injections of Mal-ArgITyr. The rate of reaction of Mal-Arg-ITyr with bovine serum albumin was determined in an analogous manner except that tracer labeled [1261]Mal-Arg-ITyr was utilized, and 12’1 covalently bound to bovine serum albumin was separated from 125I associated with unreacted MalArg-ITyr by addition of methanol to 80% by volume to precipitate protein. The reaction rate was determined from the 1251content of the bovine serum albumin pellet derived from centrifugation of the 80% methanol fraction. Rate constants were determined either by using integrated second-order rate equations or by measuring initial reaction rates under conditions that were pseudo-first order with respect to TP-Arg-ITyr or MalArg-ITyr at different concentrations of sulfhydryl-containing component. A plot of the pseudo-first-order rate constant vs the concentration of the sulfhydryl group yielded a straight line for all reactions except the TPArg-ITyr reaction with bovine serum albumin, indicating that the TP-Arg-ITyr reaction with N-acetylcysteine and the Mal-Arg-ITyr reactions were second order. Synthesis ITyr
of lz51-Labeled
TP-Arg-ITyr
and Mal-Arg-
‘251-labeled TP-Arg-ITyr and Mal-Arg-ITyr were prepared in both carrier-free and tracer-labeled forms on a scale ranging from 0.5 to 10 nmol starting II. Conditions for carrier-free synthesis at the 0.5-nmol (l.lmCi) level of “‘I- are presented below. The initial reaction step, iodination of Arg-Tyr, was the same for synthesis of both [‘261]TP-Arg-ITyr and [1251]Mal-Arg-ITyr. All steps prior to Cl8 HPLC separation of the reaction products were conducted in a properly ventilated hood approved for use in 1251 labeling reactions. The following components were added to a microfuge tube: 1~1 of 5 mM Arg-Tyr (5 nmol), 2.05 ~1 of carrier-free (0.54-mCi/Fl) Na12’I (0.5 nmol, 1.1 mCi), 0.7 ~1 water, and 0.75 ~1 of 10 mM chloramine-T (7.5 nmol). The tube was incubated for 7 min at room temperature, and then 3.75 ~11of 1 mM DTT (7.5 nmol -SH equivalents) was added. (Addition of reductant is essential for the conversion of nonincorporated 1251 to nonvolatile species prior to removal of the reaction vessel from the hood!) The tube was incubated for 10 min at room temperature. Water (6.25 ~1) was then added to the tube. [‘251]Arg-ITyr prepared in this manner was then used for synthesis of [1251]TP-Arg-ITyr or [12Y]Mal-ArgITyr as described below. Synthesis of [“‘I] TP-Arg-ITyr. Ten microliters (100 nmol) of a 10 mM stock solution (in acetone) of
AND
FRANK
TP-NHS ester was added to the microfuge tube containing [‘251]Arg-ITyr prepared as described above. The tube was incubated for 60 min at room temperature. The reaction products were then resolved by HPLC on a 5-pm Bakerbond Cl8 column (0.46 X 25 cm), using the solvents noted previously (flow rate, 1 ml/min) and a linear gradient of 60% solvent A to 48% solvent A over 24 min. The eluate of the column was monitored at 210 nm (see Fig. 2). Under these conditions TP-Arg-ITyr eluted at 21.6 min. Fractions were collected and 1251content was determined with a solid scintillation counter. The number of moles of TP-Arg-ITyr isolated was determined by comparison of the area of the 21.6-min peak (210-nm trace) from the reaction mixture chromatogram with peak areas from Cl8 analyses of [12’1]TP-Arg-ITyr standard solutions. The specific activity of the product was calculated from these two values. The [1251]TP-Arg-ITyr fraction was immediately placed in the Speed-Vat, and solvent was removed until the volume remaining was approximately 50 ~1. The material was then diluted with 12.5 mM sodium phosphate, 150 mM NaCl, pH 7.4 (PBS), and used immediately, or if the [1251]TP-Arg-ITyr was to be stored prior to use, it was diluted to a radioactive concentration of 0.02 mCi/ ml and stored at -80°C. Synthesis of [““I]Mal-Arg-ITyr. Ten microliters (100 nmol) of a 10 mM stock solution (in acetone) of Mal-NHS ester was added to a microfuge tube containing [1251]Arg-ITyr prepared as described above at the 0.5-nmol (l.l-mCi) I- scale. The tube was incubated for 60 min at room temperature. The reaction products were then resolved on a 5-pm Cl8 column using a linear gradient of 70% solvent A to 60% solvent A over 20 min (see Fig. 3). Under these conditions Mal-Arg-ITyr eluted at 21.8 min. Fractions were collected and processed as described above for [ 1251]TP-Arg-ITyr.
Measurement of Rate of Loss of “‘I from [lz51]TP-ArgITyr and [‘251]MaZ-Arg-ITyr during Storage
Carrier-free 1251-labeled TP-Arg-ITyr and Mal-ArgITyr were prepared as described above. After removal of most of the solvent, both compounds were initially diluted to 0.4 mCi/ml. Aliquots were transferred to microfuge tubes and stored at +4”C or -80°C. A portion of the material was further diluted to 0.02 mCi/ml and stored at -80°C. Aliquots were thawed and analyzed by Cl8 HPLC at intervals up to 21 days after initial storage. The Cl8 column eflluent was monitored for lzsI content with a Beckman 171 radioisotope detector (Fig. 4). The quantity of lz51 associated with [12’I]TP-ArgITyr and [‘251]Mal-Arg-ITyr was determined from the peak areas on the Beckman 171 trace. The rate constant for conversion of 125Ito non-TP-Arg-ITyr (or non-MalAm-ITvr) was determined from a plot of the log _ ~ , suecies _
‘251-CONTAINING
of the percentage (or Mal-Arg-ITyr)
Labeling
of Iz51 associated vs time.
with
SULFHYDRYL-GROUP-SPECIFIC
TP-Arg-ITyr
Red Blood Cells with [1251]TP-Arg-ITyr
TABLE
Property
RESULTS
Synthesis and Properties [‘271]MaZ-Arg-ITyr
of [12’I]TP-Arg-ITyr
[‘271]TP-Arg-ITyr and [‘271]Mal-Arg-ITyr pared by direct reaction of crosslinkers with
and were prea homoge-
1
Physical and Kinetic Properties and Mal-Arg-ITyr
and
Human blood was collected from volunteers into heparinized vacutainer tubes. An aliquot of blood (0.5 ml) was transferred to a polystyrene tube, and the cells were washed five times by centrifugation in 5 ml of PBS (12). The cells were suspended in PBS in a total volume of 1 ml. The hematocrit was approximately 30% and the red blood cell concentration, determined with a hemacytometer, was 2.56 X lo6 cells/pi. To label sulfhydryl groups of “intact” cells, 20 ~1 of washed red blood cell suspension was added to each of two microfuge tubes, followed by addition of 80 ~1 of PBS. To label “shocked” cells, 20 ~1 of washed red blood cell suspension was added to each of two tubes, followed by 80 ~1 of water to osmotically shock the cells. Both sets of tubes were incubated in a shaking water bath at 37°C for 10 min. Twenty microliters of [‘251]-labeled TP-Arg-ITyr in PBS (26 PCi; specific activity, 4.4 X lo5 Ci/mol) was added to one of the intact cell tubes and one of the osmotically shocked cell tubes. Twenty microliters of [‘251]-labeled Mal-Arg-ITyr in PBS (22 &i; specific activity, 4.4 X lo5 Ci/mol) was added to the remaining intact cell and shocked cell tubes. The incubation was continued for 30 min at 37°C. Samples were then centrifuged at 10,000 rpm for 30 s in a microfuge. The supernatants were removed from the intact cell tubes. No pellet was observed in the shocked cell tubes. Ten microliters of 50 mM NEM (in dimethyl sulfoxide) was added to each tube. Seventy microliters of water was then added to each shocked cell tube and 200 ~1 to each intact cell tube. One hundred microliters from each tube was removed and transferred to a separate microfuge tube, to which 100 ~1 of 2~ SDS sample buffer (without reductant) (13) was added. These samples represent unfractionated labeled red blood cell proteins. The remaining loo-p1 fractions were diluted with 900 ~1 of water and subjected to centrifugation at 35,000g for 20 min to yield membrane and soluble protein fractions as described by Abbott and Schachter (8). The membrane protein fraction was dissolved in 100 ~1 of SDS sample buffer (13), which contained no reductant. The soluble protein fraction was precipitated by addition of trichloroacetic acid to 10%. The precipitate was harvested by centrifugation and dissolved in 100 ~1 of SDS sample buffer (without reductant). All samples were heated at 90°C for 2 min and stored at -80°C until analysis by SDS-PAGE (13).
169
REAGENTS
Absorption maximum (nm) Extinction coefficient at AMAx (M-’ cm-i) Second-order rate constant” Reaction with Nacetylcysteine (M-l
Reaction serum (M-l
S-‘)
of TP-Arg-ITyr
TP-Arg-ITyr
Mal-Arg-ITyr
284
285
7100
2980
28 f 4
154 t
4
with bovine albumin S-l)
First-order rate constant, loss of izI from compound’ (d-i)
25 to 120b
134 + 28
0.12
0.09
D Values shown are the mean + the standard deviation of kinetic measurements carried out in triplicate in 0.1 M Hepes * NaOH, pH 7.3, at 22°C. Kinetic measurements were conducted over the following concentration ranges: TP-Arg-ITyr, 3 to 100 FM; Mal-Arg-ITyr, 5 to 150 pM; and sulfhydryl component, 20 to 500 PM. ‘The rate constant for reaction of TP-Arg-ITyr changed as the ratio of TP-Arg-ITyr to bovine serum albumin was varied. The rate constant increased from a value of 25 M-’ s-* at a TP-Arg-ITyr to bovine serum albumin ratio of 1 to 20 to a value of 120 M-’ s-i at a ratio of 1 to 2. ‘The value represents the rate constant for loss of ‘%I from [iZ61]TP-Arg-ITyr or [‘x61]Mal-Arg-ITyr during storage at -80°C at a concentration of 0.4 mCi/ml.
neous Arg-ITyr preparation. These nonradioactive compounds were prepared on a larger scale than that which could be achieved with 1251so that physical and kinetic properties could be conveniently measured. The TP-Arg-ITyr and Mal-Arg-ITyr were purified by Cl8 HPLC and exhibited the anticipated spectral properties (Table 1). Reaction of these compounds with sulfhydryl groups also resulted in the expected changes in the ultraviolet absorption spectra (see under Materials and Methods). While the TP-Arg-ITyr absorption spectrum was dominated by the pyridyldithio group, monoiodotyrosine absorption features were also present at the level predicted for a 1 to 1 ratio of monoiodotyrosine to pyridyldithio units. The absorption spectrum of MalArg-ITyr was similar to that of Arg-ITyr, except that a shoulder was observed in the 300- to 350-nm region, as would be expected for a compound containing a maleimidopropionyl moiety. The primary motivations for the synthesis of [Y]TP-Arg-ITyr and [‘251]Mal-Arg-ITyr were to provide 1251labeling reagents that would show more favorable reaction properties (higher rate and specificity) toward sulfhydryl groups than alkylhalides such as N-iodoacetyltyramine. The kinetics for reaction of TP-Arg-ITyr and Mal-Arg-ITyr with N-acetylcysteine and bovine serum albumin were therefore examined. Reaction of
170
KRUEGER,
LIN,
-1 I! -
0
5
10
15
20
25
30
TIME (min) FIG. 2. Cl8 HPLCseparation of reactionproducts of carrier-free *‘sZ synthesis of TP-Arg-ZTyr. Synthesis of [1261]TP-Arg-ITyr was carried out at the 0.5nmol (l.l-mCi) scale as described under Materials and Methods. An aliquot representing 0.4% of the reaction mixture was injected onto a 5-pm Cl8 Bakerbond column, and the column effluent was monitored for l*‘I content with a Beckman 171 radioisotope detector (dashed line). The remainder of the reaction mixture was then injected, and the absorbance of the effluent was monitored at 210 nm (solid line). The peak at 21.6 min represented [1251]TP-Arg-ITyr and was collected as indicated by the arrowheads.
Mal-Arg-ITyr with both of these compounds exhibited second-order kinetics over a wide range of sulfhydryl group concentrations. The rate constants for both reactions were similar (Table 1). The reaction of TP-ArgITyr with N-acetylcysteine also gave typical secondorder kinetics and a rate constant approximately one-fifth of that observed with the Mal-Arg-ITyr reaction. The rate constant for the reaction of TP-Arg-ITyr with bovine serum albumin increased as the albumin concentration decreased. We do not at present. understand the basis for this observation.
Synthesis of ‘251-Labekd ITyr
TP-Arg-ITyr
AND
FRANK
mainder of the reaction mixture was then injected onto the Cl8 column, and the effluent, absorbance was monitored at 210 nm. The RT determined from a previous injection of a [‘271]TP-Arg-ITyr standard was 21.6 min. The major radioactive peak eluding at 21.6 min was collected and the yield, based on starting 12’1,and specific activity were determined for this synthesis to be 34% and 1.9 X 10” Ci/mol, respectively. (Note that the accuracy of specific activity measurement for [‘261]TP-ArgITyr is reduced by the presence of a nonradioactive component with RT = 22.0 min that is not well resolved from TP-Arg-ITyr.) Our yields, based on starting I- in the 0.5 to lo-nmol range, for synthesis of TP-Arg-ITyr have been 34 to 40%. Results from a 0.5-nmol (l.l-mCi) scale carrier-free 125I synthesis of Mal-Arg-ITyr (Fig. 3) are similar to those for TP-Arg-ITyr. A [‘271]Mal-Arg-ITyr standard exhibited an RT of 21.8 min under these chromatographic conditions. The yield and specific activity of this fraction were 28% and 2.1 X lo6 Ci/mol, respectively. The yield of Mal-Arg-ITyr, based on starting I-, ranged from 28 to 41% for reactions on the 0.5- to lonmol scale. Immediately after the fractions were collected, they were transferred to the Speed-Vat for solvent removal. During the initial syntheses of carrier-free [1261]TPArg-ITyr and [‘251]Mal-Arg-ITyr, we left samples in the Speed-Vat for approximately 1 h after all solvent had been removed. When this material was dissolved and analyzed by Cl8 HPLC, a significant portion of the 1251 eluted at 4 min, indicating that chemical degradation of samples was occurring. Since lz71samples had not exhibited similar degradation, the breakdown was likely de-
and Mal-Arg-
The initial step in the reaction was chloramine-T-mediated iodination of Arg-Tyr. This was followed by the addition of a limited quantity of reductant to convert unincorporated 1251to nonvolatile forms. The second step is the addition of crosslinker in large excess. In addition to rapidly driving the reaction to completion, the large excess of crosslinker reacts with residual DTT, thereby limiting loss of the desired reaction product. We have conducted syntheses of lz51-labeled TP-ArgITyr and Mal-Arg-ITyr on the 0.5 to lo-nmol I- scale and at 1271 to 1251 ratios ranging from 0 to 100 (conditions described under Materials and Methods). The progress of the reaction can be monitored conveniently by Cl8 HPLC analysis of a small portion of the reaction mixture using a flow monitor to analyze the 12’1content of the effluent. A radiochromatogram for the synthesis of carrier-free [‘251]TP-Arg-ITyr at the 0.5-nmol (l.lmCi) scale is shown in Fig. 2 (dashed line). The re-
2
.02 -
0
00
5
10
15
20
25
30
TIME (min) FIG. 3. Cl8 HPLC separation of reactionproducts of carrier-free “sZ synthesis of Mal-Arg-ZTyr. Synthesis of [‘261]Mal-Arg-ITyr was carried out at the 0.5-nmol (l.l-mCi) scale as described under Materials and Methods. An aliquot representing 0.4% of the reaction mixture was injected onto a 5-pm Cl8 Bakerbond column, and the column effluent was monitored for *%I content with a Beckman 171 radioisotope detector (dashed line). The remainder of the reaction mixture was then injected, and the absorbance of the effluent was monitored at 210 nm (solid line). The peak at 21.8 min represented [‘“I]Mal-Arg-ITyr and was collected as indicated by the arrowheads.
‘251-CONTAINING
‘25--i
100
m ‘0 ;
-I
I
75 E a 0
-
SULFHYDRYL-GROUP-SPECIFIC
50 25 04 0
4
: a
12
TIME
16
20
I
24
28
(min)
FIG. 4. Cl8 HPLC analysis of breakdown products of carrier-free [‘25r]TP-Arg-ZTyr. An aliquot of [iz51]TP-Arg-ITyr, prepared as described in the legend to in Fig. 2, was stored at -80°C for 7 days. The material was then removed from the freezer, thawed, and immediately injected onto a 5-pm Bakerbond Cl8 column. The column was developed with a linear gradient of 85% solvent A to 31% solvent A over 18 min. The column effluent was monitored for 12sI content. The RT value for a [‘271]TP-Arg-ITyr standard (monitored at 210 nm) was 21.0 min.
REAGENTS
171
nents eluting at 3.7 and 13 to 15 min was not altered by during cysteine addition. Loss of “‘1 from TP-Arg-ITyr storage (0.4 mCi/ml, -80°C) exhibited first-order kinetics (t,,, = 5.7 days). A similar analysis was conducted to measure the stability of [1251]Mal-Arg-ITyr (0.4 mCi/ml) at -80°C. The loss of 1251from Mal-Arg-ITyr also exhibited first-order kinetics (t,,, = 7.5 days). Samples of carrier-free [1251]TP-Arg-ITyr and [1251]Mal-Arg-ITyr were also stored at -80°C at a radioactive concentration of 0.02 mCi/ml. No loss of 1251from either compound was observed for storage of up to 21 days. The stability of [1251]TP-Arg-ITyr and [1251]Mal-ArgITyr under conditions employed for labeling proteins was also examined. Samples of TP-Arg-ITyr and MalArg-ITyr (0.02 mCi/ml) were incubated in PBS at 37°C for periods up to 1 h. No degradation was observed for either compound. Labeling Red Blood Cell Sulfhydryl Groups with [‘251]TP-Arg-ITyr and [1251]Mal-Arg-ITyr
pendent on lz51 decay products. The stability of [1251]TP-Arg-ITyr and [1251]Mal-Arg-ITyr under storage and labeling conditions was therefore examined.
Erythrocytes were selected as a test system to examine the labeling properties of TP-Arg-ITyr and MalArg-ITyr in a relatively complex array of proteins. Labeling experiments were performed on red blood cells Stability of [‘251]TP-Arg-ITyr and [‘251]MaZ-Arg-ITyr prepared from freshly collected human blood. To test Samples of carrier-free 1251-labeledTP-Arg-ITyr and the relative impermeance of Arg-ITyr derivatives to the Mal-Arg-ITyr were synthesized and purified by Cl8 plasma membrane, one series of labeling reactions was HPLC. The HPLC fractions were transferred to the performed on intact cells and a second series was perSpeed-Vat for solvent removal. When the remaining formed on cells that had been ruptured by osmotic solvent volume was approximately 50 ~1, the samples shock. After a 30-min incubation at 37°C in PBS, intact were removed and diluted with PBS to a concentration cells were pelleted by centrifugation. Excess NEM was of 0.4 mCi/ml, and aliquots were stored at -80°C or added to both intact and shocked cells to quench sulfhy4’C. Samples were removed at intervals after storage, dry1 groups that had not reacted. Half of the material and the distribution of 125Iin the sample was analyzed from each sample was then used for the preparation of by Cl8 HPLC. soluble and membrane protein fractions by centrifugaFor samples stored at 4”C, the percentage of “‘1 co- tion (8). The proportion of 1251associated with each migrating with TP-Arg-ITyr and Mal-Arg-ITyr de- fraction is shown in Table 2. With intact cells, the majorcreased rapidly. After 24 h of storage, approximately ity of 1251(78% for TP-Arg-ITyr and 67% for Mal-Arg50% of the 1251originally present in these compounds ITyr) is associated with the membrane fraction. Conhad been converted to forms with smaller RT values. versely, the majority of the label (92% for TP-Arg-ITyr Carrier-free 1251-labeledTP-Arg-ITyr and Mal-Argand 86% for Mal-Arg-ITyr) is associated with soluble ITyr samples stored at -80°C at 0.4 mCi/ml were de- protein in cells that were osmotically shocked. graded at a considerably slower rate. A radiochromatoTo determine whether specific labeling of exofacial gram for a sample of [‘251]TP-Arg-ITyr that had been protein sulfhydryl groups was occurring during the lastored for 7 days is shown in Fig. 4. The peak at 21 min beling of intact cells by TP-Arg-ITyr and Mal-Argcomigrates with a [‘271]TP-Arg-ITyr standard. More ITyr, labeled intact and broken erythrocyte proteins than 50% of the 125Iis no longer associated with TPwere resolved by SDS-PAGE, and 12’1distribution was Arg-ITyr, and the majority of the non-TP-Arg-ITyr lz51 monitored fluorographically (Fig. 5). Components in is not retarded by the column. Addition of a large excess four apparent molecular weight regions of the gel were of cysteine resulted in a shift in retention time for both heavily labeled in intact cells (lanes 2 and 3) for both the TP-Arg-ITyr and the RT = 17.5 min components, 1251-labeledTP-Arg-ITyr cells and 1251-labeledMal-Argindicating that both of these species were reactive to- ITyr cells. The labeled bands at 80,000, 46,000, and ward sulfhydryl groups. The mobility of the compo- 32,000-34,000 Da are qualitatively consistent with label
172
KRUEGER,
TABLE
2
Distribution of I%1 in Red Blood Cells Labeled [1261]TP-Arg-ITyr and [‘251]Mal-Arg-ITyr Sample [?]TP-Arg-ITyr Intact cells Shocked cells [ixaI]Mal-Arg-ITyr Intact cells Shocked cells
LIN,
“‘1 in soluble fraction (cpm X 10’)
izI
with
in membrane fraction (cpm X 105)
1.25 (22) 57.66 (92)
4.36 (78) 5.02 (8)
1.49 (33) 13.42 (86)
3.00 (67) 2.20 (14)
Note. Intact and shocked human erythrocyte samples were prepared as described under Materials and Methods. Samples were incubated with [‘?]TP-Arg-ITyr (26 &i) or [iZ61]Mal-Arg-ITyr (22 &i) for 30 min in PBS. Intact cells were pelleted by centrifugation, and NEM (500 nmol/tube) was added to each sample. Membrane and soluble protein fractions were then isolated by centrifugation (8) for both broken and intact cell samples. ’ The values in parentheses represent the percentage of the total ‘%I incorporated into the indicated fraction.
incorporation into exofacial sulfhydryl groups of bands 3, 4.5, and 7 as previously described by Abbott and Schachter (8). In addition, an intensely labeled band that comigrated with hemoglobin bands was present at 12,000-16,000 Da. This labeled protein probably is not a membrane protein with an exofacial sulfhydryl group. There is a significant quantitative difference in label incorporation, however, as we observed greater incorporation in the 80,000 band than in the 46,000 band. In contrast, Abbott and Schachter observed 10 times more label incorporation into band 4.5 than into band 3 (8). This could be due to differences in relative reaction rates with different proteins. The labeling reactions carried out by Abbott and Schachter involved essentially complete labeling of sulfhydryl groups by an excess of maleimide labeling reagent, Calculations based on sulfhydryl content values for red blood cells determined by Abbott and Schachter (8) indicate that our reactions were carried out with membrane sulfhydryl groups present in large excess over the labeling reagent. The distribution of lz51 was quite different for samples that had been osmotically shocked prior to reaction with Mal-Arg-ITyr and TP-Arg-ITyr (Fig. 5, lanes 1 and 4), with the exception of a large quantity of label incorporated into the 12,000- to 16,000 M, band. The primary components labeled were at 49,000,39,000, and 29,000 Da. The majority of the label was present in the 29,000- and 16,000-M, bands, which comigrated with carbonic anhydrase and hemoglobin, respectively. The labeling patterns with TP-Arg-ITyr and Mal-Arg-ITyr were qualitatively similar. Addition of excess DTT prior to electrophoresis resulted in complete loss of ‘261from protein in samples that had been labeled with TP-ArgITyr, but not Mal-Arg-ITyr. A portion of the membrane
AND
FRANK
fraction that had been prepared by centrifugation from the intact [‘251]Mal-Arg-ITyr sample was also analyzed by SDS-PAGE, and 12’1-labeledbands at 80,000,46,000, and 32,000-34,000 Da were present in this fraction (data not shown), confirming that these proteins were membrane associated. The 16,000-M, band was also observed at levels greatly reduced from those of the other components. DISCUSSION
The objective of this study was to determine whether high-specific-activity 1251-containing maleimido and pyridyldithio compounds would be more useful than previously described reagents such as N-iodoacetyl-3monoiodotyramine (3) for labeling protein sulfhydryl groups. We found that TP-Arg-ITyr and Mal-Arg-ITyr have rate constants for reaction with sulfhydryl groups much higher than those of N-iodoacetyl-3-monoiodotyramine. In addition, [‘251]TP-Arg-ITyr and [‘251]MalArg-ITyr can be prepared in one reaction vessel in yields of 30 to 40% (based on starting 12’1)from commercially available reactants. Pyridyldithio and maleimido functions were selected for the preparation of ‘251-containing group-specific reagents because they had previously been shown to react rapidly and with high specificity with sulfhydryl groups (4-7). Our results demonstrate that TP-Arg-ITyr and Mal-Arg-ITyr reacted much more rapidly with N-acetylcysteine than did N-iodoacetyltyramine. These rate
FIG. 5.
SDS-PAGE
resolution of [‘261]-hbeled red blood cell pro-
teins. Proteins of intact or osmotically shocked human cell blood cells were labeled with [‘261]Mal-Arg-ITyr or [‘261]TP-Arg-ITyr as described under Materials and Methods. The labeled samples were treated with NEM to alkylate remaining sulthydryl groups. Proteins were than resolved by SDS-PAGE, and the dried gel was exposed to X-ray film for 12 h at -80°C with an intensifying screen. Lanes 1 and 2 are the fluorogram for [12SI]-labeled Mal-Arg-ITyr shocked and intact cells, respectively. Lanes 3 and 4 represent the ‘%I-labeled proteins obtained by treating intact and osmotically shocked cells, respectively, with TP. The mobility of standard proteins is shown on the left, with the molecular masses indicated in kilodaltons.
‘251-CONTAINING
SULFHYDRYL-GROUP-SPECIFIC
constants for TP-Arg-ITyr and Mal-Arg-ITyr at pH 7.3 were 28 and 154 M-’ s-‘, respectively. These values compare quite favorably with the value of 0.7 M-’ s-l obtained at pH 8.0 for N-iodoacetyltyramine (3). Note that rate constant for the reaction of N-iodoacetyltyramine at pH 7.3 would be predicted to be even lower than that measured at pH 8.0, because the thiolate anion is thought to be the reactive species in this reaction (5). In addition to the favorable reaction kinetics of reaction of TP-Arg-ITyr with sulfhydryl groups, this labeling reaction is reversible. The labeled reaction product is an alkyl-alkyl disulfide between a protein cysteinyl residue and 3-TP-Arg-ITyr. Addition of disulfide reductants to the reaction product results in release of the label from the protein and regeneration of the cysteinyl residue in its sulfhydryl form. During storage of carrier-free TP-Arg-ITyr and MalArg-ITyr at high radioactive concentration, we observed relatively rapid loss of 1251from TP-Arg-ITyr and Mal-Arg-ITyr. Kinetic analysis of 1251at -80°C indicated that these reactions followed a first-order kinetics and had t1,2 values of 5.7 and 7.5 days for TP-ArgITyr and Mal-Arg-ITyr, respectively. The degradation reaction was even more rapid when the radiolabeled compounds were stored at 4°C. Dilution of TP-ArgITyr and Mal-Arg-ITyr to a radioactive concentration of 0.02 mCi/ml essentially eliminated the loss of ‘25I from these compounds during storage at -80°C. We therefore routinely dilute samples to this level for storage and then concentrate the material to the desired level under vacuum immediately before use in sulfhydry1 group labeling reactions. The strategy that we employed for synthesis of 1251containing sulfhydryl-group-specific reagents requires that one of the reactants contain a phenolic ring and an amino group that is reasonably reactive toward N-hydroxysuccinimides. We selected the dipeptide Arg-Tyr as this component for our initial studies because we anticipated that it would be relatively impermeant to cell membranes and therefore would provide a means for specific labeling of cell surface sulfhydryl groups, such as those associated with membrane-bound receptors
173
REAGENTS
and transporters. Our labeling experiments with red blood cells indicate that TP-Arg-ITyr and Mal-ArgITyr react preferentially with exofacial membrane sulfhydryl groups of intact cells. 1251incorporation into a largely different set of proteins occurred when cells were osmotically shocked prior to the addition of labeling reagent. The findings that 1251incorporation into proteins by TP-Arg-ITyr was completely reversed by the addition of DTT and that the labeling pattern observed with Mal-Arg-ITyr was qualitatively quite similar to that seen with TP-Arg-ITyr suggest that these reagents display a high specificity of labeling toward sulfhydryl groups. The favorable reaction kinetics and ease of synthesis of TP-Arg-ITyr and Mal-Arg-ITyr from readily available precursors represent significant advances over previously described methods for the labeling of sulfhydryl groups with lz51 (2,3). ACKNOWLEDGMENT The authors phlebotomies.
thank
Professor
Ann
Strom
for performing
painless
REFERENCES 1. Bolton, 538. 2. Holowka,
A. E., and Hunter, D. (1981)
And.
W. M. (1973) Biochm.
Biochem.
J. 133,529-
117,390-397.
3. Lin, C., Mihal, K. A., and Krueger, R. J. (1989) Anal. Biochem. 179,389-395. 4. Means, G. E., and Feeny, R. E. (1971) Chemical Modification of Proteins, pp. 105-110, Holden-Day, San Francisco. 5. Lundblad, R. L., and Noyes, C. M. (1984) Chemical Reagents for Protein Modification, Vol. 1, pp. 55-93, CRC Press, Boca Raton, FL. 6. Brocklehurst, K. (1979) Znt. J. Biochem. 7. Pedersen, A. O., and Jacobsen, J. (1980) 291-295.
10, 259-274. Eur. J. Biochem.
106,
8. Abbott, R. E., and Schachter, D. (1976) J. Biol. Chem. 251,71767183. 9. Haest, C. W. M., Plasa, G., Kamp, D., and Deuticke, B. (1978) Biochim. Biophys. Acta 509,21-32. 10. Edlehoch,
H. (1962)
J. Biol.
Chem.
237,2778-2787.
11. Ellman, G. L. (1958) Arch. Biochem. Biophys. 74,443-450. 12. May, J. M. (1988) J. Biol. Chem. 263, 13,63513,640. 13. Laemmli,
U. K. (1970)
Nature
227,680-685.
BIOCHEMISTRY
(1991)
198,165-173
Synthesis and Properties of Sulfhydryl-Group-Specific Reagents Containing 12511 Rick
J. Krueger,
Department
Received
April
of
Chaomei
Biochemistry
Lin,
and Julie
and School of Biological
A. Frank Sciences, University
Nebraska,
Lincoln,
Nebraska,
68583-0718
1, 1991
groups 12”1-containing compounds that react specifically with sulfhydryl groups were prepared in yields of 30 to 40% on the basis of starting la61 quantity. The synthetic precursors were commercially available heterobifunctional crosslinkers and the peptide L-arginyl-L-tyrosine. Two types of sulfhydryl specific reagents were prepared: 3-(2-pyridyldithio)propionylarginyl-[’221Jmonoiodotyrosine, which permits reversible incorporation of “‘1 at sulfhydryl sites, and 3-maleimidopropionylarginyl-[‘261]monoiodotyrosine, an irreversible labeling reagent. These products were isolated in a highly radiochemically pure form by Cl8 HPLC. The second-order rate constants for the reaction of 3-(2pyridyldithio)propionylarginylmonoiodotyrosine and 3 -maleimidopropionylarginylmonoiodotyrosine with N-acetylcysteine were 28 f 3 M-’ s-l and 154 + 4 M-l s-l, respectively, at pH 7.3. Storage of carrier-free 3(2-pyridyldithio)propionylarginyl-[12sI]monoiodotyrosine and 3-maleimidopropionylarginyl-[‘261]monoiodotyrosine at -80°C at a radioactive concentration of 0.4 mCi/ml resulted in conversion of “‘1 to species that did not react covalently with sulfhydryl groups. This process occurred with first-order kinetics and a t,,, of 5.7 days for the pyridyldithio compound and 7.5 days for the maleimido compound. No conversion was observed during storage at -80°C at radioactive concentrations of 0.02 mCi/ml or less. The labeling properties of these compounds were examined using red blood cell proteins as a test system. 3-(2-Pyridyldithio)propionylarginyl-[ ‘261]monoiodotyrosine and maleimidopropionylarginyl-[ ‘261]monoiodotyrosine reacted preferentially with membrane - associated sulfhydryl
1 Agricultural Research Division, University of Nebraska Journal Series 9543; supported by the University of Nebraska Research Council and NIH Biomedical Research Support Grant 07055. 0003-2697/91 Copyright All rights
of
$3.00 0 1991 hy Academic Press, of reproduction in any form
Q 1991
when Academic
incubated Press,
with
intact
red
blood
cells.
Inc.
Chemical modification of proteins is useful for the incorporation of labels that can be detected with high sensitivity. 1251is a frequently employed label because it can be obtained at low cost in high specific activity and because it can be conveniently incorporated into tyrosyl residues of proteins. Bolton and Hunter (1) have developed chemical modification reagents containing “‘1 that permit the incorporation of a radiolabel at amino groups. These reagents are particularly useful for labeling proteins either devoid of tyrosine or whose activity is modified by standard iodination procedures. Chemical modification reagents that permit the introduction of 1251at sulfhydryl groups have also been described. Holowka described the synthesis and properties of ‘251-labeled N-chloroacetyltyramine (2) and demonstrated that this compound reacts with substantial specificity at sulfhydryl groups. Subsequently, we described the synthesis of carrier-free N-iodoacetyl-[3-‘251]monoiodotyramine (3), which reacts more rapidly with sulfhydryl groups than N-chloroacetyltyramine. Compounds such as these do permit incorporation of 1251 at sulfhydryl sites, but alkylhalides also react with other functional groups found in proteins (4). In addition, the rate of reaction of N-iodoacetyltyramine with sulfhydryl groups is relatively slow compared with reaction rates of sulfhydryl groups with maleimides and dithiopyridines (35-7). The high rate of reaction of maleimides with sulfhydryl groups, coupled with the limited reactivity of maleimides with other functional groups found in proteins, results in a high specificity of sulfhydryl group labeling by maleimides (5). Dithiopyridyl compounds exhibit more favorable reaction properties, with their la165
Inc. reserved.
166
KRUEGER, Step
MN, AND FRANK
1
H,N-(Arg-Tyr)-COO-
f
125
I-
-b
H,N-(Arg-["51]-Tyr)-COOArg-ITyr
Step 2a Arg-ITyr
t
S-S-(CH2)?-!-0-Nb
ci
--+
0
' S-S-(CH,),-C-NH-(Arg-['251]-Tyr)-COOTP-Arg-ITyr
Step Zb 0 0 Arg-ITyr
+
C
N-,,,,.N5
1
\\ 0
FIG. 1.
Synthesis the monoiodopeptide.
of TP-Arg-ITyr The reactions
+
6)
\\ 0
i
Mal-Arg-ITyr
and Mal-Arg-ZTyr. The a-amino and cY-carboxyl groups of the dipeptide are shown. Step 1 is formation resulting in the production of TP-Arg-ITyr and Mal-Arg-ITyr are shown in steps 2a and 2b, respectively.
beling of sulfhydryl groups being one of the most selective reactions of any protein modification reagent (6). These properties have been utilized to study sulfhydry1 groups associated with membrane proteins. Membrane-impermeant maleimides containing 35Shave previously been employed to determine the quantity and topography of sulfhydryl groups associated with red blood cell membranes (8). Spectrophotometric measurement of aryl disulfide reduction has also been used in the study of red blood cell membrane sulfhydryl groups (9). We describe here conditions for the synthesis of carrier-free ‘251-labeled compounds containing pyridyldithio or maleimido functional groups. The reactions are carried out in one vessel using commercially available reagents (Fig. 1). In addition to radioactive iodine, the reactant requirements are: (a) a compound containing a phenol ring with a covalently linked amine and (b) a heterobifunctional crosslinker containing N-hydroxysuccinimide as the amine reactive group and either a pyridyldithio or a maleimido function as the sulfhydryl reactive group. MATERIALS
N-(CH,)2-C-NH-(Arg-['251]-Tyr)-COOm '
AND METHODS
L-Arginyl-L-tyrosine, acetate salt (Arg-Tyr),2 was obtained from Bachem. Chloramine-T, N-ethylmaleimide 2 Abbreviations used: Arg-Tyr, L-arginyl-L-tyrosine; Arg-ITyr, arginyl-L-monoiodotyrosine; Arg-I,Tyr, L-arginyl-L-diiodotyrosine; Mal-Arg-ITyr, 3-maleimidopropionyl-L-arginyl-L-monoiodotyrosine; TP-Arg-ITyr, 3-(2-pyridyldithio)propionyl-L-arginyl-L-monoiodo-
L-
of
(NEM), 3-(2-pyridyldithio)propionic acid N-hydroxysuccinimide ester (TP-NHS ester), and 3-maleimidopropionic acid N-hydroxysuccinimide ester (Mal-NHS ester) were purchased from Aldrich. Na’251 (carrier-free, NEZ033L) was from New England Nuclear. Dithiothreitol (DTT), 5,5’-dithiobis(2-nitrobenzoic acid) (DTNB), N-acetylcysteine, and bovine serum albumin were obtained from Sigma.
Synthesis of [1271]arginylmonoiodotyrosine Nonradioactive Arg-ITyr was prepared by chloramine-T-mediated iodination of Arg-Tyr, and the reaction product was purified by Cl8 HPLC. For the iodination reaction, the following components were added to a 1.5-ml microfuge tube in a total volume of 270 ~1: ArgTyr (15 pmol), Na12?I (5 pmol), Hepes buffer, pH 7.3 (75 pmol). The reaction was initiated by the addition of 15 pmol chloramine-T (30 ~1 aqueous solution). The mixture was incubated for 3 min at room temperature. The reaction was terminated by addition of DTT (15 pmol), and the reaction products were resolved on a 15-pm Bakerbond Cl8 HPLC column (1 X 25 cm). The HPLC solvents used were solvent A, 0.05% trifluoroacetic acid, and solvent B, 0.05% trifluoroacetic acid, 50% acetonityrosine; DTT, dithiothreitol; DTNB, 5,5’-dithiobis(2-nitrobenzoic acid); NEM, N-ethylmaleimide; TP-NHS ester, 3-(2-pyridyldithio)propionic acid N-hydroxysuccinimide ester; Mal-NHS ester, 3-maleimidopropionic acid N-hydroxysuccinimide ester; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
‘261-CONTAINING
SULFHYDRYL-GROUP-SPECIFIC
trile. The separation was achieved with a linear gradient running from 90% solvent A to 70% solvent A over 20 min (flow rate, 4 ml/min). Under these conditions ArgTyr eluted at 6.4 min, Arg-ITyr at 14.6 min, and Arg1,Tyr at 23.1 min. The mono- and diiodinated forms of Arg-Tyr were identified and quantified by their characteristic ultraviolet absorption spectra in acid and base (10). Synthesis
and Characterization
of [lz71] TP-Arg-ITyr
[‘27]TP-Arg-ITyr was prepared by reaction of ArgITyr with TP-NHS ester as follows: 180 hl of 4.4 mM Arg-ITyr aqueous stock solution (0.8 pmol), 120 ~1 of 10 InM TP-NHS ester stock solution in acetone (1.2 pmol), 40 ~1 of 1.0 M Hepes * NaOH, pH 7.3 (40 pmol), and 460 ~1 of water were added to a microfuge tube and incubated for 30 min at room temperature. The reaction products were resolved by Cl8 HPLC (as above) using a linear gradient of 90% solvent A to 40% solvent A over 25 min. The elution profile was monitored at 274 nm. The only major product observed eluted at 25.0 min. Previous analytical scale injections made at intervals during the reaction had indicated an increase in the area of the R, = 25 min peak concomitant with a decrease in the area under the peaks representing Arg-ITyr and TP-NHS ester. The species eluting at 25.0 min was identified as TPArg-ITyr by the following criteria: (a) Ultraviolet absorption spectra in acid and base (10) indicated the presence of monoiodotyrosine. (b) Addition of N-acetylcysteine to a solution of the RT = 25 min material gave a species with an ultraviolet absorption spectrum identical to that of 2-thiopyridinone (7). (c) The Cl8 HPLC mobility of the RT = 25 min species shifted upon addition of N-acetylcysteine, yielding two new components. The Cl8 mobility of one of these components was identical to that of 2-thiopyridinone. No shift in Cl8 retention time or change in ultraviolet absorption occurred when lysine was added to the solution of the RT = 25 min species. The concentration of TP-Arg-ITyr in stock solutions was initially determined spectrophotometrically from the quantity of 2-thiopyridinone produced by addition of excess DTT. The value used for the 2-thiopyridinone extinction coefficient was 7600 M-’ cm-’ at 343 nm (7). An extinction coefficient of 7110 M-’ cm-’ at 284 nm was determined for TP-Arg-ITyr. This value was used for the spectrophotometric determination of TP-ArgITyr concentration in all further studies. Synthesis and Characterization
of [1271]Mal-Arg-ITyr
Nonradioactive Mal-Arg-ITyr was prepared as described above for TP-Arg-ITyr except that Mal-NHS ester was substituted for TP-NHS ester. The reaction products were resolved on Cl8 HPLC using a linear gra-
REAGENTS
167
dient of 90% solvent A to 60% solvent A over 15 min. The only major component in the Cl8 profile (monitored at 274 nm) eluted at 18.8 min. (Note that RT values indicate real time in the gradient and are not corrected for the volume between the gradient mixing port and the detector.) Previous analytical HPLC analyses demonstrated that this peak increased in area during the reaction as the Arg-ITyr peak decreased in area. The species eluting at 18.8 min was identified as MalArg-ITyr by the following criteria: (a) Ultraviolet absorption spectra in acid and base indicated the presence of monoiodotyrosine. (b) A shoulder was present at 300-350 nm in the ultraviolet absorption spectrum of the RT = 18.8 min component, consistent with the presence of the maleimidopropionyl group. This shoulder disappeared after the addition of N-acetylcysteine (but not lysine). (c) A shift in the Cl8 mobility of the RT = 18.8 min species was observed after addition of Nacetylcysteine (but not lysine). The concentration of Mal-Arg-ITyr in stock solutions was initially determined by reaction of the maleimido group with thiols. A known volume of Mal-Arg-ITyr stock solution was added to a solution of N-acetylcysteine. The decrease in N-acetylcysteine sulfhydryl groups was determined by reaction with DTNB, using the extinction coefficient 13,600 M-’ cm-’ at 412 nm (11). A Mal-Arg-ITyr extinction coefficient of 2980 M-’ cm-’ at 285 nm was determined by this method. Kinetic Analysis of the Reaction of TP-Arg-ITyr and Mal-Arg-ITyr with N-Acetylcysteine and Bovine Serum Albumin All rate constants were determined from reactions carried out at room temperature in 0.25 M Hepes * NaOH, pH 7.3. The sulfbydryl group concentration of N-acetylcysteine and bovine serum albumin stock solutions was determined by reaction with DTNB (7). Bovine serum albumin containing an average of 0.5 sulfhydry1 groups per molecule was prepared as previously described (3). The rate of reaction of TP-Arg-ITyr with N-acetylcysteine and bovine serum albumin was determined spectrophotometrically by measuring the absorbance change at 343 nm associated with production of 2-thiopyridinone (7). Reactions were carried out at a range of N-acetylcysteine and bovine serum albumin concentrations for each TP-Arg-ITyr concentration. Reaction rates were determined from the initial absorbance changes. The rate of Mal-Arg-ITyr reaction with N-acetylcysteine was monitored by using Cl8 HPLC to resolve reaction products at timed intervals after the reaction was initiated. The column eluate was monitored at 285 nm, and data were collected and peak areas determined with a Waters Maxima 820 system. The reaction rate was
168
KRUEGER,
LIN,
determined from the decrease in Mal-Arg-ITyr concentration. Peak areas were converted to nanomole quantities by reference to standard injections of Mal-ArgITyr. The rate of reaction of Mal-Arg-ITyr with bovine serum albumin was determined in an analogous manner except that tracer labeled [1261]Mal-Arg-ITyr was utilized, and 12’1 covalently bound to bovine serum albumin was separated from 125I associated with unreacted MalArg-ITyr by addition of methanol to 80% by volume to precipitate protein. The reaction rate was determined from the 1251content of the bovine serum albumin pellet derived from centrifugation of the 80% methanol fraction. Rate constants were determined either by using integrated second-order rate equations or by measuring initial reaction rates under conditions that were pseudo-first order with respect to TP-Arg-ITyr or MalArg-ITyr at different concentrations of sulfhydryl-containing component. A plot of the pseudo-first-order rate constant vs the concentration of the sulfhydryl group yielded a straight line for all reactions except the TPArg-ITyr reaction with bovine serum albumin, indicating that the TP-Arg-ITyr reaction with N-acetylcysteine and the Mal-Arg-ITyr reactions were second order. Synthesis ITyr
of lz51-Labeled
TP-Arg-ITyr
and Mal-Arg-
‘251-labeled TP-Arg-ITyr and Mal-Arg-ITyr were prepared in both carrier-free and tracer-labeled forms on a scale ranging from 0.5 to 10 nmol starting II. Conditions for carrier-free synthesis at the 0.5-nmol (l.lmCi) level of “‘I- are presented below. The initial reaction step, iodination of Arg-Tyr, was the same for synthesis of both [‘261]TP-Arg-ITyr and [1251]Mal-Arg-ITyr. All steps prior to Cl8 HPLC separation of the reaction products were conducted in a properly ventilated hood approved for use in 1251 labeling reactions. The following components were added to a microfuge tube: 1~1 of 5 mM Arg-Tyr (5 nmol), 2.05 ~1 of carrier-free (0.54-mCi/Fl) Na12’I (0.5 nmol, 1.1 mCi), 0.7 ~1 water, and 0.75 ~1 of 10 mM chloramine-T (7.5 nmol). The tube was incubated for 7 min at room temperature, and then 3.75 ~11of 1 mM DTT (7.5 nmol -SH equivalents) was added. (Addition of reductant is essential for the conversion of nonincorporated 1251 to nonvolatile species prior to removal of the reaction vessel from the hood!) The tube was incubated for 10 min at room temperature. Water (6.25 ~1) was then added to the tube. [‘251]Arg-ITyr prepared in this manner was then used for synthesis of [1251]TP-Arg-ITyr or [12Y]Mal-ArgITyr as described below. Synthesis of [“‘I] TP-Arg-ITyr. Ten microliters (100 nmol) of a 10 mM stock solution (in acetone) of
AND
FRANK
TP-NHS ester was added to the microfuge tube containing [‘251]Arg-ITyr prepared as described above. The tube was incubated for 60 min at room temperature. The reaction products were then resolved by HPLC on a 5-pm Bakerbond Cl8 column (0.46 X 25 cm), using the solvents noted previously (flow rate, 1 ml/min) and a linear gradient of 60% solvent A to 48% solvent A over 24 min. The eluate of the column was monitored at 210 nm (see Fig. 2). Under these conditions TP-Arg-ITyr eluted at 21.6 min. Fractions were collected and 1251content was determined with a solid scintillation counter. The number of moles of TP-Arg-ITyr isolated was determined by comparison of the area of the 21.6-min peak (210-nm trace) from the reaction mixture chromatogram with peak areas from Cl8 analyses of [12’1]TP-Arg-ITyr standard solutions. The specific activity of the product was calculated from these two values. The [1251]TP-Arg-ITyr fraction was immediately placed in the Speed-Vat, and solvent was removed until the volume remaining was approximately 50 ~1. The material was then diluted with 12.5 mM sodium phosphate, 150 mM NaCl, pH 7.4 (PBS), and used immediately, or if the [1251]TP-Arg-ITyr was to be stored prior to use, it was diluted to a radioactive concentration of 0.02 mCi/ ml and stored at -80°C. Synthesis of [““I]Mal-Arg-ITyr. Ten microliters (100 nmol) of a 10 mM stock solution (in acetone) of Mal-NHS ester was added to a microfuge tube containing [1251]Arg-ITyr prepared as described above at the 0.5-nmol (l.l-mCi) I- scale. The tube was incubated for 60 min at room temperature. The reaction products were then resolved on a 5-pm Cl8 column using a linear gradient of 70% solvent A to 60% solvent A over 20 min (see Fig. 3). Under these conditions Mal-Arg-ITyr eluted at 21.8 min. Fractions were collected and processed as described above for [ 1251]TP-Arg-ITyr.
Measurement of Rate of Loss of “‘I from [lz51]TP-ArgITyr and [‘251]MaZ-Arg-ITyr during Storage
Carrier-free 1251-labeled TP-Arg-ITyr and Mal-ArgITyr were prepared as described above. After removal of most of the solvent, both compounds were initially diluted to 0.4 mCi/ml. Aliquots were transferred to microfuge tubes and stored at +4”C or -80°C. A portion of the material was further diluted to 0.02 mCi/ml and stored at -80°C. Aliquots were thawed and analyzed by Cl8 HPLC at intervals up to 21 days after initial storage. The Cl8 column eflluent was monitored for lzsI content with a Beckman 171 radioisotope detector (Fig. 4). The quantity of lz51 associated with [12’I]TP-ArgITyr and [‘251]Mal-Arg-ITyr was determined from the peak areas on the Beckman 171 trace. The rate constant for conversion of 125Ito non-TP-Arg-ITyr (or non-MalAm-ITvr) was determined from a plot of the log _ ~ , suecies _
‘251-CONTAINING
of the percentage (or Mal-Arg-ITyr)
Labeling
of Iz51 associated vs time.
with
SULFHYDRYL-GROUP-SPECIFIC
TP-Arg-ITyr
Red Blood Cells with [1251]TP-Arg-ITyr
TABLE
Property
RESULTS
Synthesis and Properties [‘271]MaZ-Arg-ITyr
of [12’I]TP-Arg-ITyr
[‘271]TP-Arg-ITyr and [‘271]Mal-Arg-ITyr pared by direct reaction of crosslinkers with
and were prea homoge-
1
Physical and Kinetic Properties and Mal-Arg-ITyr
and
Human blood was collected from volunteers into heparinized vacutainer tubes. An aliquot of blood (0.5 ml) was transferred to a polystyrene tube, and the cells were washed five times by centrifugation in 5 ml of PBS (12). The cells were suspended in PBS in a total volume of 1 ml. The hematocrit was approximately 30% and the red blood cell concentration, determined with a hemacytometer, was 2.56 X lo6 cells/pi. To label sulfhydryl groups of “intact” cells, 20 ~1 of washed red blood cell suspension was added to each of two microfuge tubes, followed by addition of 80 ~1 of PBS. To label “shocked” cells, 20 ~1 of washed red blood cell suspension was added to each of two tubes, followed by 80 ~1 of water to osmotically shock the cells. Both sets of tubes were incubated in a shaking water bath at 37°C for 10 min. Twenty microliters of [‘251]-labeled TP-Arg-ITyr in PBS (26 PCi; specific activity, 4.4 X lo5 Ci/mol) was added to one of the intact cell tubes and one of the osmotically shocked cell tubes. Twenty microliters of [‘251]-labeled Mal-Arg-ITyr in PBS (22 &i; specific activity, 4.4 X lo5 Ci/mol) was added to the remaining intact cell and shocked cell tubes. The incubation was continued for 30 min at 37°C. Samples were then centrifuged at 10,000 rpm for 30 s in a microfuge. The supernatants were removed from the intact cell tubes. No pellet was observed in the shocked cell tubes. Ten microliters of 50 mM NEM (in dimethyl sulfoxide) was added to each tube. Seventy microliters of water was then added to each shocked cell tube and 200 ~1 to each intact cell tube. One hundred microliters from each tube was removed and transferred to a separate microfuge tube, to which 100 ~1 of 2~ SDS sample buffer (without reductant) (13) was added. These samples represent unfractionated labeled red blood cell proteins. The remaining loo-p1 fractions were diluted with 900 ~1 of water and subjected to centrifugation at 35,000g for 20 min to yield membrane and soluble protein fractions as described by Abbott and Schachter (8). The membrane protein fraction was dissolved in 100 ~1 of SDS sample buffer (13), which contained no reductant. The soluble protein fraction was precipitated by addition of trichloroacetic acid to 10%. The precipitate was harvested by centrifugation and dissolved in 100 ~1 of SDS sample buffer (without reductant). All samples were heated at 90°C for 2 min and stored at -80°C until analysis by SDS-PAGE (13).
169
REAGENTS
Absorption maximum (nm) Extinction coefficient at AMAx (M-’ cm-i) Second-order rate constant” Reaction with Nacetylcysteine (M-l
Reaction serum (M-l
S-‘)
of TP-Arg-ITyr
TP-Arg-ITyr
Mal-Arg-ITyr
284
285
7100
2980
28 f 4
154 t
4
with bovine albumin S-l)
First-order rate constant, loss of izI from compound’ (d-i)
25 to 120b
134 + 28
0.12
0.09
D Values shown are the mean + the standard deviation of kinetic measurements carried out in triplicate in 0.1 M Hepes * NaOH, pH 7.3, at 22°C. Kinetic measurements were conducted over the following concentration ranges: TP-Arg-ITyr, 3 to 100 FM; Mal-Arg-ITyr, 5 to 150 pM; and sulfhydryl component, 20 to 500 PM. ‘The rate constant for reaction of TP-Arg-ITyr changed as the ratio of TP-Arg-ITyr to bovine serum albumin was varied. The rate constant increased from a value of 25 M-’ s-* at a TP-Arg-ITyr to bovine serum albumin ratio of 1 to 20 to a value of 120 M-’ s-i at a ratio of 1 to 2. ‘The value represents the rate constant for loss of ‘%I from [iZ61]TP-Arg-ITyr or [‘x61]Mal-Arg-ITyr during storage at -80°C at a concentration of 0.4 mCi/ml.
neous Arg-ITyr preparation. These nonradioactive compounds were prepared on a larger scale than that which could be achieved with 1251so that physical and kinetic properties could be conveniently measured. The TP-Arg-ITyr and Mal-Arg-ITyr were purified by Cl8 HPLC and exhibited the anticipated spectral properties (Table 1). Reaction of these compounds with sulfhydryl groups also resulted in the expected changes in the ultraviolet absorption spectra (see under Materials and Methods). While the TP-Arg-ITyr absorption spectrum was dominated by the pyridyldithio group, monoiodotyrosine absorption features were also present at the level predicted for a 1 to 1 ratio of monoiodotyrosine to pyridyldithio units. The absorption spectrum of MalArg-ITyr was similar to that of Arg-ITyr, except that a shoulder was observed in the 300- to 350-nm region, as would be expected for a compound containing a maleimidopropionyl moiety. The primary motivations for the synthesis of [Y]TP-Arg-ITyr and [‘251]Mal-Arg-ITyr were to provide 1251labeling reagents that would show more favorable reaction properties (higher rate and specificity) toward sulfhydryl groups than alkylhalides such as N-iodoacetyltyramine. The kinetics for reaction of TP-Arg-ITyr and Mal-Arg-ITyr with N-acetylcysteine and bovine serum albumin were therefore examined. Reaction of
170
KRUEGER,
LIN,
-1 I! -
0
5
10
15
20
25
30
TIME (min) FIG. 2. Cl8 HPLCseparation of reactionproducts of carrier-free *‘sZ synthesis of TP-Arg-ZTyr. Synthesis of [1261]TP-Arg-ITyr was carried out at the 0.5nmol (l.l-mCi) scale as described under Materials and Methods. An aliquot representing 0.4% of the reaction mixture was injected onto a 5-pm Cl8 Bakerbond column, and the column effluent was monitored for l*‘I content with a Beckman 171 radioisotope detector (dashed line). The remainder of the reaction mixture was then injected, and the absorbance of the effluent was monitored at 210 nm (solid line). The peak at 21.6 min represented [1251]TP-Arg-ITyr and was collected as indicated by the arrowheads.
Mal-Arg-ITyr with both of these compounds exhibited second-order kinetics over a wide range of sulfhydryl group concentrations. The rate constants for both reactions were similar (Table 1). The reaction of TP-ArgITyr with N-acetylcysteine also gave typical secondorder kinetics and a rate constant approximately one-fifth of that observed with the Mal-Arg-ITyr reaction. The rate constant for the reaction of TP-Arg-ITyr with bovine serum albumin increased as the albumin concentration decreased. We do not at present. understand the basis for this observation.
Synthesis of ‘251-Labekd ITyr
TP-Arg-ITyr
AND
FRANK
mainder of the reaction mixture was then injected onto the Cl8 column, and the effluent, absorbance was monitored at 210 nm. The RT determined from a previous injection of a [‘271]TP-Arg-ITyr standard was 21.6 min. The major radioactive peak eluding at 21.6 min was collected and the yield, based on starting 12’1,and specific activity were determined for this synthesis to be 34% and 1.9 X 10” Ci/mol, respectively. (Note that the accuracy of specific activity measurement for [‘261]TP-ArgITyr is reduced by the presence of a nonradioactive component with RT = 22.0 min that is not well resolved from TP-Arg-ITyr.) Our yields, based on starting I- in the 0.5 to lo-nmol range, for synthesis of TP-Arg-ITyr have been 34 to 40%. Results from a 0.5-nmol (l.l-mCi) scale carrier-free 125I synthesis of Mal-Arg-ITyr (Fig. 3) are similar to those for TP-Arg-ITyr. A [‘271]Mal-Arg-ITyr standard exhibited an RT of 21.8 min under these chromatographic conditions. The yield and specific activity of this fraction were 28% and 2.1 X lo6 Ci/mol, respectively. The yield of Mal-Arg-ITyr, based on starting I-, ranged from 28 to 41% for reactions on the 0.5- to lonmol scale. Immediately after the fractions were collected, they were transferred to the Speed-Vat for solvent removal. During the initial syntheses of carrier-free [1261]TPArg-ITyr and [‘251]Mal-Arg-ITyr, we left samples in the Speed-Vat for approximately 1 h after all solvent had been removed. When this material was dissolved and analyzed by Cl8 HPLC, a significant portion of the 1251 eluted at 4 min, indicating that chemical degradation of samples was occurring. Since lz71samples had not exhibited similar degradation, the breakdown was likely de-
and Mal-Arg-
The initial step in the reaction was chloramine-T-mediated iodination of Arg-Tyr. This was followed by the addition of a limited quantity of reductant to convert unincorporated 1251to nonvolatile forms. The second step is the addition of crosslinker in large excess. In addition to rapidly driving the reaction to completion, the large excess of crosslinker reacts with residual DTT, thereby limiting loss of the desired reaction product. We have conducted syntheses of lz51-labeled TP-ArgITyr and Mal-Arg-ITyr on the 0.5 to lo-nmol I- scale and at 1271 to 1251 ratios ranging from 0 to 100 (conditions described under Materials and Methods). The progress of the reaction can be monitored conveniently by Cl8 HPLC analysis of a small portion of the reaction mixture using a flow monitor to analyze the 12’1content of the effluent. A radiochromatogram for the synthesis of carrier-free [‘251]TP-Arg-ITyr at the 0.5-nmol (l.lmCi) scale is shown in Fig. 2 (dashed line). The re-
2
.02 -
0
00
5
10
15
20
25
30
TIME (min) FIG. 3. Cl8 HPLC separation of reactionproducts of carrier-free “sZ synthesis of Mal-Arg-ZTyr. Synthesis of [‘261]Mal-Arg-ITyr was carried out at the 0.5-nmol (l.l-mCi) scale as described under Materials and Methods. An aliquot representing 0.4% of the reaction mixture was injected onto a 5-pm Cl8 Bakerbond column, and the column effluent was monitored for *%I content with a Beckman 171 radioisotope detector (dashed line). The remainder of the reaction mixture was then injected, and the absorbance of the effluent was monitored at 210 nm (solid line). The peak at 21.8 min represented [‘“I]Mal-Arg-ITyr and was collected as indicated by the arrowheads.
‘251-CONTAINING
‘25--i
100
m ‘0 ;
-I
I
75 E a 0
-
SULFHYDRYL-GROUP-SPECIFIC
50 25 04 0
4
: a
12
TIME
16
20
I
24
28
(min)
FIG. 4. Cl8 HPLC analysis of breakdown products of carrier-free [‘25r]TP-Arg-ZTyr. An aliquot of [iz51]TP-Arg-ITyr, prepared as described in the legend to in Fig. 2, was stored at -80°C for 7 days. The material was then removed from the freezer, thawed, and immediately injected onto a 5-pm Bakerbond Cl8 column. The column was developed with a linear gradient of 85% solvent A to 31% solvent A over 18 min. The column effluent was monitored for 12sI content. The RT value for a [‘271]TP-Arg-ITyr standard (monitored at 210 nm) was 21.0 min.
REAGENTS
171
nents eluting at 3.7 and 13 to 15 min was not altered by during cysteine addition. Loss of “‘1 from TP-Arg-ITyr storage (0.4 mCi/ml, -80°C) exhibited first-order kinetics (t,,, = 5.7 days). A similar analysis was conducted to measure the stability of [1251]Mal-Arg-ITyr (0.4 mCi/ml) at -80°C. The loss of 1251from Mal-Arg-ITyr also exhibited first-order kinetics (t,,, = 7.5 days). Samples of carrier-free [1251]TP-Arg-ITyr and [1251]Mal-Arg-ITyr were also stored at -80°C at a radioactive concentration of 0.02 mCi/ml. No loss of 1251from either compound was observed for storage of up to 21 days. The stability of [1251]TP-Arg-ITyr and [1251]Mal-ArgITyr under conditions employed for labeling proteins was also examined. Samples of TP-Arg-ITyr and MalArg-ITyr (0.02 mCi/ml) were incubated in PBS at 37°C for periods up to 1 h. No degradation was observed for either compound. Labeling Red Blood Cell Sulfhydryl Groups with [‘251]TP-Arg-ITyr and [1251]Mal-Arg-ITyr
pendent on lz51 decay products. The stability of [1251]TP-Arg-ITyr and [1251]Mal-Arg-ITyr under storage and labeling conditions was therefore examined.
Erythrocytes were selected as a test system to examine the labeling properties of TP-Arg-ITyr and MalArg-ITyr in a relatively complex array of proteins. Labeling experiments were performed on red blood cells Stability of [‘251]TP-Arg-ITyr and [‘251]MaZ-Arg-ITyr prepared from freshly collected human blood. To test Samples of carrier-free 1251-labeledTP-Arg-ITyr and the relative impermeance of Arg-ITyr derivatives to the Mal-Arg-ITyr were synthesized and purified by Cl8 plasma membrane, one series of labeling reactions was HPLC. The HPLC fractions were transferred to the performed on intact cells and a second series was perSpeed-Vat for solvent removal. When the remaining formed on cells that had been ruptured by osmotic solvent volume was approximately 50 ~1, the samples shock. After a 30-min incubation at 37°C in PBS, intact were removed and diluted with PBS to a concentration cells were pelleted by centrifugation. Excess NEM was of 0.4 mCi/ml, and aliquots were stored at -80°C or added to both intact and shocked cells to quench sulfhy4’C. Samples were removed at intervals after storage, dry1 groups that had not reacted. Half of the material and the distribution of 125Iin the sample was analyzed from each sample was then used for the preparation of by Cl8 HPLC. soluble and membrane protein fractions by centrifugaFor samples stored at 4”C, the percentage of “‘1 co- tion (8). The proportion of 1251associated with each migrating with TP-Arg-ITyr and Mal-Arg-ITyr de- fraction is shown in Table 2. With intact cells, the majorcreased rapidly. After 24 h of storage, approximately ity of 1251(78% for TP-Arg-ITyr and 67% for Mal-Arg50% of the 1251originally present in these compounds ITyr) is associated with the membrane fraction. Conhad been converted to forms with smaller RT values. versely, the majority of the label (92% for TP-Arg-ITyr Carrier-free 1251-labeledTP-Arg-ITyr and Mal-Argand 86% for Mal-Arg-ITyr) is associated with soluble ITyr samples stored at -80°C at 0.4 mCi/ml were de- protein in cells that were osmotically shocked. graded at a considerably slower rate. A radiochromatoTo determine whether specific labeling of exofacial gram for a sample of [‘251]TP-Arg-ITyr that had been protein sulfhydryl groups was occurring during the lastored for 7 days is shown in Fig. 4. The peak at 21 min beling of intact cells by TP-Arg-ITyr and Mal-Argcomigrates with a [‘271]TP-Arg-ITyr standard. More ITyr, labeled intact and broken erythrocyte proteins than 50% of the 125Iis no longer associated with TPwere resolved by SDS-PAGE, and 12’1distribution was Arg-ITyr, and the majority of the non-TP-Arg-ITyr lz51 monitored fluorographically (Fig. 5). Components in is not retarded by the column. Addition of a large excess four apparent molecular weight regions of the gel were of cysteine resulted in a shift in retention time for both heavily labeled in intact cells (lanes 2 and 3) for both the TP-Arg-ITyr and the RT = 17.5 min components, 1251-labeledTP-Arg-ITyr cells and 1251-labeledMal-Argindicating that both of these species were reactive to- ITyr cells. The labeled bands at 80,000, 46,000, and ward sulfhydryl groups. The mobility of the compo- 32,000-34,000 Da are qualitatively consistent with label
172
KRUEGER,
TABLE
2
Distribution of I%1 in Red Blood Cells Labeled [1261]TP-Arg-ITyr and [‘251]Mal-Arg-ITyr Sample [?]TP-Arg-ITyr Intact cells Shocked cells [ixaI]Mal-Arg-ITyr Intact cells Shocked cells
LIN,
“‘1 in soluble fraction (cpm X 10’)
izI
with
in membrane fraction (cpm X 105)
1.25 (22) 57.66 (92)
4.36 (78) 5.02 (8)
1.49 (33) 13.42 (86)
3.00 (67) 2.20 (14)
Note. Intact and shocked human erythrocyte samples were prepared as described under Materials and Methods. Samples were incubated with [‘?]TP-Arg-ITyr (26 &i) or [iZ61]Mal-Arg-ITyr (22 &i) for 30 min in PBS. Intact cells were pelleted by centrifugation, and NEM (500 nmol/tube) was added to each sample. Membrane and soluble protein fractions were then isolated by centrifugation (8) for both broken and intact cell samples. ’ The values in parentheses represent the percentage of the total ‘%I incorporated into the indicated fraction.
incorporation into exofacial sulfhydryl groups of bands 3, 4.5, and 7 as previously described by Abbott and Schachter (8). In addition, an intensely labeled band that comigrated with hemoglobin bands was present at 12,000-16,000 Da. This labeled protein probably is not a membrane protein with an exofacial sulfhydryl group. There is a significant quantitative difference in label incorporation, however, as we observed greater incorporation in the 80,000 band than in the 46,000 band. In contrast, Abbott and Schachter observed 10 times more label incorporation into band 4.5 than into band 3 (8). This could be due to differences in relative reaction rates with different proteins. The labeling reactions carried out by Abbott and Schachter involved essentially complete labeling of sulfhydryl groups by an excess of maleimide labeling reagent, Calculations based on sulfhydryl content values for red blood cells determined by Abbott and Schachter (8) indicate that our reactions were carried out with membrane sulfhydryl groups present in large excess over the labeling reagent. The distribution of lz51 was quite different for samples that had been osmotically shocked prior to reaction with Mal-Arg-ITyr and TP-Arg-ITyr (Fig. 5, lanes 1 and 4), with the exception of a large quantity of label incorporated into the 12,000- to 16,000 M, band. The primary components labeled were at 49,000,39,000, and 29,000 Da. The majority of the label was present in the 29,000- and 16,000-M, bands, which comigrated with carbonic anhydrase and hemoglobin, respectively. The labeling patterns with TP-Arg-ITyr and Mal-Arg-ITyr were qualitatively similar. Addition of excess DTT prior to electrophoresis resulted in complete loss of ‘261from protein in samples that had been labeled with TP-ArgITyr, but not Mal-Arg-ITyr. A portion of the membrane
AND
FRANK
fraction that had been prepared by centrifugation from the intact [‘251]Mal-Arg-ITyr sample was also analyzed by SDS-PAGE, and 12’1-labeledbands at 80,000,46,000, and 32,000-34,000 Da were present in this fraction (data not shown), confirming that these proteins were membrane associated. The 16,000-M, band was also observed at levels greatly reduced from those of the other components. DISCUSSION
The objective of this study was to determine whether high-specific-activity 1251-containing maleimido and pyridyldithio compounds would be more useful than previously described reagents such as N-iodoacetyl-3monoiodotyramine (3) for labeling protein sulfhydryl groups. We found that TP-Arg-ITyr and Mal-Arg-ITyr have rate constants for reaction with sulfhydryl groups much higher than those of N-iodoacetyl-3-monoiodotyramine. In addition, [‘251]TP-Arg-ITyr and [‘251]MalArg-ITyr can be prepared in one reaction vessel in yields of 30 to 40% (based on starting 12’1)from commercially available reactants. Pyridyldithio and maleimido functions were selected for the preparation of ‘251-containing group-specific reagents because they had previously been shown to react rapidly and with high specificity with sulfhydryl groups (4-7). Our results demonstrate that TP-Arg-ITyr and Mal-Arg-ITyr reacted much more rapidly with N-acetylcysteine than did N-iodoacetyltyramine. These rate
FIG. 5.
SDS-PAGE
resolution of [‘261]-hbeled red blood cell pro-
teins. Proteins of intact or osmotically shocked human cell blood cells were labeled with [‘261]Mal-Arg-ITyr or [‘261]TP-Arg-ITyr as described under Materials and Methods. The labeled samples were treated with NEM to alkylate remaining sulthydryl groups. Proteins were than resolved by SDS-PAGE, and the dried gel was exposed to X-ray film for 12 h at -80°C with an intensifying screen. Lanes 1 and 2 are the fluorogram for [12SI]-labeled Mal-Arg-ITyr shocked and intact cells, respectively. Lanes 3 and 4 represent the ‘%I-labeled proteins obtained by treating intact and osmotically shocked cells, respectively, with TP. The mobility of standard proteins is shown on the left, with the molecular masses indicated in kilodaltons.
‘251-CONTAINING
SULFHYDRYL-GROUP-SPECIFIC
constants for TP-Arg-ITyr and Mal-Arg-ITyr at pH 7.3 were 28 and 154 M-’ s-‘, respectively. These values compare quite favorably with the value of 0.7 M-’ s-l obtained at pH 8.0 for N-iodoacetyltyramine (3). Note that rate constant for the reaction of N-iodoacetyltyramine at pH 7.3 would be predicted to be even lower than that measured at pH 8.0, because the thiolate anion is thought to be the reactive species in this reaction (5). In addition to the favorable reaction kinetics of reaction of TP-Arg-ITyr with sulfhydryl groups, this labeling reaction is reversible. The labeled reaction product is an alkyl-alkyl disulfide between a protein cysteinyl residue and 3-TP-Arg-ITyr. Addition of disulfide reductants to the reaction product results in release of the label from the protein and regeneration of the cysteinyl residue in its sulfhydryl form. During storage of carrier-free TP-Arg-ITyr and MalArg-ITyr at high radioactive concentration, we observed relatively rapid loss of 1251from TP-Arg-ITyr and Mal-Arg-ITyr. Kinetic analysis of 1251at -80°C indicated that these reactions followed a first-order kinetics and had t1,2 values of 5.7 and 7.5 days for TP-ArgITyr and Mal-Arg-ITyr, respectively. The degradation reaction was even more rapid when the radiolabeled compounds were stored at 4°C. Dilution of TP-ArgITyr and Mal-Arg-ITyr to a radioactive concentration of 0.02 mCi/ml essentially eliminated the loss of ‘25I from these compounds during storage at -80°C. We therefore routinely dilute samples to this level for storage and then concentrate the material to the desired level under vacuum immediately before use in sulfhydry1 group labeling reactions. The strategy that we employed for synthesis of 1251containing sulfhydryl-group-specific reagents requires that one of the reactants contain a phenolic ring and an amino group that is reasonably reactive toward N-hydroxysuccinimides. We selected the dipeptide Arg-Tyr as this component for our initial studies because we anticipated that it would be relatively impermeant to cell membranes and therefore would provide a means for specific labeling of cell surface sulfhydryl groups, such as those associated with membrane-bound receptors
173
REAGENTS
and transporters. Our labeling experiments with red blood cells indicate that TP-Arg-ITyr and Mal-ArgITyr react preferentially with exofacial membrane sulfhydryl groups of intact cells. 1251incorporation into a largely different set of proteins occurred when cells were osmotically shocked prior to the addition of labeling reagent. The findings that 1251incorporation into proteins by TP-Arg-ITyr was completely reversed by the addition of DTT and that the labeling pattern observed with Mal-Arg-ITyr was qualitatively quite similar to that seen with TP-Arg-ITyr suggest that these reagents display a high specificity of labeling toward sulfhydryl groups. The favorable reaction kinetics and ease of synthesis of TP-Arg-ITyr and Mal-Arg-ITyr from readily available precursors represent significant advances over previously described methods for the labeling of sulfhydryl groups with lz51 (2,3). ACKNOWLEDGMENT The authors phlebotomies.
thank
Professor
Ann
Strom
for performing
painless
REFERENCES 1. Bolton, 538. 2. Holowka,
A. E., and Hunter, D. (1981)
And.
W. M. (1973) Biochm.
Biochem.
J. 133,529-
117,390-397.
3. Lin, C., Mihal, K. A., and Krueger, R. J. (1989) Anal. Biochem. 179,389-395. 4. Means, G. E., and Feeny, R. E. (1971) Chemical Modification of Proteins, pp. 105-110, Holden-Day, San Francisco. 5. Lundblad, R. L., and Noyes, C. M. (1984) Chemical Reagents for Protein Modification, Vol. 1, pp. 55-93, CRC Press, Boca Raton, FL. 6. Brocklehurst, K. (1979) Znt. J. Biochem. 7. Pedersen, A. O., and Jacobsen, J. (1980) 291-295.
10, 259-274. Eur. J. Biochem.
106,
8. Abbott, R. E., and Schachter, D. (1976) J. Biol. Chem. 251,71767183. 9. Haest, C. W. M., Plasa, G., Kamp, D., and Deuticke, B. (1978) Biochim. Biophys. Acta 509,21-32. 10. Edlehoch,
H. (1962)
J. Biol.
Chem.
237,2778-2787.
11. Ellman, G. L. (1958) Arch. Biochem. Biophys. 74,443-450. 12. May, J. M. (1988) J. Biol. Chem. 263, 13,63513,640. 13. Laemmli,
U. K. (1970)
Nature
227,680-685.