- Email: [email protected]
Hybridization as a technique for studying interchain interactions in the catalytic trimers of aspartate transcarbamoylase
ANALYTICAL
163, 188- 195 ( 1987)
BIOCHEMISTRY
Hybridization as a Technique for Studying Interchain Interactions Catalytic Trimers of Aspartate Transcarbamoylasei3*
in the
YING R. YANG AND H. K. SCHACHMAN~ Department of Molecular Biology and the Virus Laboratory Wendell M. Stanley Hall, University of California, Berkeley, California 94720 Received January 27, 1987 Since subunit interactions in regulatory enzymes mediate the ligand-promoted conformationai changes responsible for their allosteric properties, it is necessary to have techniques for determining the effects of ligands and mutational alterations on the strength of the interchain interactions. In aspartate transcarbamoylase from Escherichia coli, the multiple interchain interactions are so linked that it is difficult to study them separately. Therefore, we have focused on the nonahosteric catalytic trimers isolated from the holoenzyme and have used the rate of hybrid formation between native and succinylated protein as a measure of the dissociation of the trimers into single polypeptide chains. Although catalytic trimers exhibit no evident dissociation in sedimentation studies at lo-* M, incubation of mixtures of native and succinylated trimers for long periods of time (days) yielded hybrids which are readily detected by polyacrylamide gel electrophoresis. This sensitive technique was used to demonstrate that the substrate, carbamoylphosphate, and the bisubstrate analog, N-(phosphonacetyl)-L-aspartate, cause a marked strengthening of the interchain interactions, whereas the inhibitor, sodium pyrophosphate, at concentrations as low as 10 mM, promotes dissociation of the trimers. This weakening of the interchain interactions by pyrophosphate facilitated the isolation and purification of functionally competent hybrid trimers by a technique which was much more convenient and provided higher yields than previous, more drastic methods which employed urea or guanidine hydrochloride to cause dissociation of the trimers. The hybridization technique was useful in studying the effects of mutational alterations on the strength of the interchain interactions and the ability of active and inactive mutants to bind pyrophosphate. o 1987 Academic F’res. Inc. KEY WORDS: subunit interactions; ligand effects; hybrid formation; dissociation of oligomerit enzymes; effect of amino acid substitutions.
INTRODUCTION
A aetailecl unaerxanaing ctf the allosteric behavior of oligomeric protej ins in terms of ligand-promoted conformatic mal changes is .*_.*.
,
‘
,.
’ This work was supported by United States Public Health Services Research Grant GM 12159 from the National Institute of General Medical Sciences and by National Science Foundation Research Grant DMB 85-02131. * This paper is dedicated to the memory of Nathan 0. Kaplan who was a pioneer in the use of hybridization in his classical studies of the subunit structure of isozymes. 3 To whom reprint requests should be addressed at Department of Molecular Biology, University of CaIifornia, Berkeley, 229 Stanley Hall, Berkeley, CA 97420. 0003-2697187 $3.00 Copyright @ 1987 by Academic Press, hc. AU x-i&s of reproduction in any fom mawd.
dependent upon knowledge of the strengths of the interchain interactions and the effects of ligands on them. For a regulatory enzyme like Escherichia coli aspartate transcarbamLoylase (ATCase, carbamoylphosphate: 4 Abbreviations used: ATCase, aspartate transcarbamoylase; C, catalytic trimer; R, regulatory dimer; c, catalytic polypeptide chain; r, regulatory polypeptide chain; CN or C,.., native C trimer; Cs or C,, succinylated C trimer; Cr or C&, tetrahydrophthaloylated C trimer; CM, wild-type C trimer; Cx or C,,, mutant C trimer; Cnnt, hybrid trimer containing two native and one tetrahydrophthaloylated chains; THPA, 3,4,5,6-tetrahydrophthalic anhydride; PALA, N-(phosphonacetyl)-L-aspartate; EDTA, ethylenediaminetetraacetate. 188
STUDY
OF
INTERCHAIN
INTERACTIONS
aspartate carbamoyltransferase, EC 2.1.3.2) an evaluation of the changes in the bonding energies between polypeptide chains is particularly difficult because of the large number and different types of noncovalent interchain “bonds” in the holoenzyme. The 12 polypeptide chains in ATCase are organized as two cataytic (C) trimers and three regulatory (R) dimers (l-6). Thus there are three bonds linking the catalytic (c) polypeptide chains within each C trimer and one bond linking the regulatory (r) chains in each R dimer. In addition, the two C trimers are “cross-linked” noncovalently by a variety of interactions involving bonds between c and r chains as well as those between one C trimer and the other (7-12). Because of the multiplicity of these interactions and their intrinsic strengths, ATCase exhibits little tendency to dissociate into subunits even at great dilution (10-l’ M) and very little exchange of either C or R subunits is detectable when radioactive subunits are incubated with wildtype native or reconstituted ATCase ( 13,14). The diverse interactions among the 12 polypeptide chains are doubtlessly linked with the result that the binding of a ligand or an amino acid substitution affects more than one type of bond. Since it is extremely hazardous to attempt partitioning the energies among discrete types of bonds in intact ATCase, we focus here on the isolated nonallosteric C trimers to determine the effects of ligands and mutational alterations on the interactions between the c chains in the C trimers (9,15,16). Although C trimers are very stable in neutral solutions and show no evident dissociation in sedimentation velocity experiments at about 10e8 M, hybrids are readily detected when wild-type C trimers and succinylated (Cs) trimers are incubated together for prolonged periods of time (9). From measurements of the rate of formation of hybrids between unmodified and succinylated trimers, we obtain evidence for the dissociation of C trimers and an indication of the strengths of the interactions between c chains
BY
189
HYBRIDIZATION
under different conditions. The technique is presented here along with preliminary applications to studies on the effects of ligands and mutational alterations on the relative strengths of the interchain interactions. MATERIALS
AND
METHODS
Succinic anhydride and succinate were purchased from Eastman Chemicals, neohydrin and 3,4,5,6-tetrahydrophthalic anhydride (THPA) were obtained from K and K Laboratories, and carrier-free 1251as the sodium salt in NaOH was obtained from Amersham Corp. Carbamoylphosphate as the lithium salt was purchased from Boehringer-Mannheim, L-aspartate from Calbiochem, carbamoylaspartate and all nucleoside triphosphates from Sigma Chemicals, and sodium pyrophosphate and potassium phosphate from Mallinckrodt. DEAEcellulose was purchased from Whatman and DEAE-Sephadex A-50 from Pharmacia Biotechnology. N-(phosphonacetyl)+aspartate (PALA), lot MK45-89-1, was kindly provided by Dr. Robert R. Engle, Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute. Wild-type ATCase was prepared according to the method of Gerhart and Holoubek (17) from E. coli strain TR4363 containing plasmid pPYRB3 (18) which carries the intact pyrB1 operon. Mutant forms of ATCase were purified by the procedure described by Wall and Schachman (19) with slight modifications. C trimers were obtained by dissociating ATCase with neohydrin followed by DEAE-cellulose chromatography (20). ‘251labeled C trimers were prepared by the method of Syvanen et al. (2 1). Succinylation of C trimers was performed by mixing protein with a 3.5 molar ratio of succinic anhydride per lysine residue (4) to yield a derivative with about four succinyl groups per c chain. For the purification of hybrids of different species, the C trimers were acylated with THPA to yield CT as described by Gib-
190
YANG AND SCHACHMAN
bons and Schachman (22). Interchain hybrids were formed by incubating mixtures of Cx (C,,,) and Cs (C,) or CT (C& in 50 mM Tris-chloride buffer at pH 7.5 containing 2 mM 2-mercaptoethanol and 0.2 mM EDTA. Each protein was at a concentration of 3.5 mg/ml and the hybridization experiments were performed at 4°C for various lengths of time. Electrophoresis experiments in polyacrylamide gels were performed with a Hoefer Mighty Small slab gel apparatus using the discontinuous buffer system of Jovin et al. (23). The 7.5% gels were stained with Coomassie brilliant blue G-250 in 12.5% trichloracetic acid. In experiments with ‘251-labeled proteins, the various bands were cut from the gels and measurements of radioactivity were performed with a Nuclear Chicago gamma counter. The following mutant forms of ATCase were used in the various studies. ATCasezj,, which is virtually devoid of activity (< 10e5 that of the wild-type enzyme), contains Asp in place of Gly-128 in the c chain (19,24). ATCaseTd5 and ATCaseTd3 are active mutants isolated from two different spontaneous revertants from a pyrimidine auxotrophic strain (25,26). ATCaseTG8 produced by sitedirected mutagenesis contains Gln in place of Lys at position 84 in the c chain (27,28). ATCaseTT I formed by site-directed mutagenesis contains Ala in place of His-l 34 in the c chain (27). ATCaseTT4 formed by site-directed mutagenesis has His in place of Ser-52 in the c chain (Y. R. Yang and H. K. Schachman, unpublished). ATCase? is a double mutant in which Lys-84 was replaced by Gln and His-l 34 by Ala (28). RESULTS AND DISCUSSION
Efect of ligands on the strength of interchain interactions in C trimers. In the initial experiments aimed at determining the number of polypeptide chains in the C subunits of ATCase (4), hybrids between CN and Cs were formed by adding guanidine hydrochloride
to a mixture of the two proteins followed by dialysis to permit random reassociation of the chains into oligomers. Electrophoretic analysis of the reconstituted mixture revealed a four-member set thereby demonstrating that the proteins were trimers. When CN and Cs were incubated overnight in the absence of a denaturant no hybrids were detected, indicating that the C trimers were extremely stable with little tendency to dissociate into individual polypeptide chains. Subsequent experiments under different conditions showed hybrids were formed slowly from C, and Cs and that the rate of hybridization varied with temperature, pH, and ionic strength (9). Moreover, the substrate carbamoylphosphate and the bisubstrate ligand, PALA, caused a very marked decrease in the rate of formation of hybrids (9). Since it was apparent that specific ligands had a significant effect on the strength of the interaction between the chains in C trimers, we conducted additional studies which are illustrated by the electrophoresis patterns in Fig. 1. In all of these experiments the incubation of CN and Cs was performed at 4°C because of the earlier finding (9) that hybrid formation was more rapid at that tempera-
1 2 3 4 5 6 7 8 9101112 FIG. 1. Effect of ligands on the strength of interchain interactions in C trimers. Mixtures of C.,, and C,, each at 3.5 mg/ml in 50 mM T&chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol and 0.2 mM EDTA in the absence or presence of 10 mM ligand were incubated at 4°C for 2 days and then analyzed by polyacrylamide gel electrophoresis as described under Materials and Methods. Lane 1, no l∧ lane 2, PALA; lane 3, carbamoylphosphate; lane 4, L-aspartate; lane 5, succinate; lane 6, carbamoylaspartate; lane 7, potassium phosphate; lane 8, sodium pyrophosphate; lane 9, ATP, lane 10, CTP; lane 11, GTP; and lane 12, UTP.
STUDY OF INTERCHAIN
INTERACTIONS
ture than at 25 ’ C. Lane 1, which represents a control in which no ligands were present, shows the presence of only small amounts of hybrids after 2 days of incubation of CN and Cs in a Tris-chloride buffer at pH 7.5. When PALA or carbamoylphosphate were present (10 mM) there was a significant reduction in the amount of hybrids, lanes 2 and 3, respectively. In contrast, aspartate (lane 4) and succinate (lane 5) had little effect on the rate of hybrid formation. Carbamoylaspartate, a product of the reaction catalyzed by ATCase, apparently strengthened the interchain interactions as seen by the decrease in the amount of hybrids formed during the incubation (lane 6). On the other hand, phosphate, the other product of the reaction between carbamoylphosphate and aspartate, appeared to decrease the strength of the bonds between c chains in the C trimer (lane 7). By far, the greatest weakening of the interchain interactions was caused by pyrophosphate (lane 8) ATP (lane 9), CTP (lane lo), GTP (lane 11) and UTP (lane 12). In the presence of these ligands, hybrid formation was greatly enhanced indicating that their binding to CN ,
.M”,
“^
12345678910 FIG. 2. Formation and purification of interchain hybrids from CN and Cr trimers. Mixtures of C,,,, and C&, each at 3.5 mg/ml, were incubated overnight at 4°C in 50 mM T&chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol, 0.2 mM EDTA, and 10 mM sodium pyrophosphate. The hybrid set containing four species, lanes 1 and 6, was then fractionated by chromatography on a DEAE-Sephadex A-50 column to yield four pooled fractions containing, respectively, C,.. , lane 2; C,,, , lane 3; Cnti, lane 4; and C&, lane 5. Each of the four purified trimers was dialyzed for 36 h at 4°C against 40 mM phosphate buffer at pH 6.1. The deacylated species were then subjected to electrophoresis to yield C.,,, lane 7; deacylated C,,,, lane 8; deacylated C,,, lane 9; and deacylated C, , lane 10.
BY HYBRIDIZATION
191
trimers caused a marked decrease in the strength of the bonds between c chains. The weakening of the interchain interactions by pyrophosphate provides a particularly useful method for the production and isolation of interchain hybrids, which complements the previous techniques using guanidine hydrochloride or urea on the one hand (4) and NaCNS or NaC104 on the other (9,29). Figure 2 shows the hybrid set formed by incubating CN and CT for 1 day at 4°C in the pH 7.5 Tris-chloride buffer containing 10 mM sodium pyrophosphate (lanes 1 and 6). The individual species were separated by chromatography on DEAE-Sephadex A-50 and the patterns for C,,,, C,,,, &, and C& are shown in lanes 2,3,4, and 5, respectively. When these species were dialyzed at pH 6.1 according to the procedure of Gibbons and Schachman (22), the THP-groups were removed and the deacylated proteins (lanes 8, 9, and 10) had the same electrophoretic mobility as the native trimers (lanes 2 and 7). This method for the preparation of interchain hybrids by using sodium pyrophosphate to weaken the interchain bonds in C trimers is clearly preferable to those methods used previously. It is less time consuming and the yields are higher because of the decreased precipitation of protein that occurs when guanidine hydrochloride or NaCNS are used to dissociate the different C trimers. The technique has been used in this laboratory to prepare purified, functional hybrids from native and chemically modified C trimers (30,31) and from C trimers isolated from different mutant forms of ATCase (28,3 1). It should be noted, however, that pyrophosphate is not effective in dissociating the C trimers from some mutants due, presumably, to its inability to bind to the altered proteins (see below).
Effect of mutational alterations on interchain interactions in C trimers. The availability of two active mutants with one, ATCase745, having significantly less and the other, ATCase743, more cooperativity than the
192
YANG AND SCHACHMAN
wild-type enzyme (26) provided an opportunity to determine whether measurements of the strength of the interchain interactions in the C trimers could provide useful information relevant to the allosteric properties of the holoenzymes. Both ATC~ZXZ,~~ and ATCaseTd3 are revertants of an inactive mutant of wild-type ATCase (25,26). In neither case have the positions of the amino acid substitutions been located. ATCaseTd3 has a significantly higher extinction coefficient at 280 nm than do either ATCaseTd5 or wild-type. ATCase indicating that it contains two additional tyrosines, but it is not known whether ATCaseTd5 is also a second-site revertant. Both mutants, despite their markedly different allosteric properties and ligand-promoted conformational changes, have similarly weakened interactions between c and r chains (26) in comparison to those in wildtype ATCase. As seen in Fig. 3, the rate of dissociation of CT45 trimers was greater than that of wild-
‘lo1 100 90
type C trimers and, conversely, the rate of dissociation of CT43trimers was significantly lower. For these experiments, the mutant trimers were labeled with 12’1 and incubated with an excess of Cs for various lengths of time. The measurements involved slicing the gels and determining the amount of radioactivity associated with the most slowly moving species. This technique, which is much more quantitative than that involving visual inspection of the electrophoresis patterns, shows clearly that the bonds between c chains in the C trimers isolated from the less cooperative mutant are substantially weaker than those in the C trimers from wild-type enzyme. In contrast, the interchain bonds in the C trimers from the more cooperative mutant are stronger than those from the wild-type holoenzyme. For comparative purposes, results are also shown in Fig. 3 for the C trimers from ATCaser3,, an inactive mutant in which Gly at position 128 is replaced by Asp (19). The bonds between c chains in this trimer are clearly stronger than those in either of the other mutants or the wild-type trimer.
Eflect of mutational alterations on the ability ofpyrophosphate to weaken the interchain interactions in C trimers. Since it is known
that pyrophosphate is a strong inhibitor of ATCase (32,33) and that it binds near the interface between adjacent polypeptide chains (34), it seemed to be of interest to determine the effect of pyrophosphate on the interchain bonds in mutant forms of ATCase. In addition to the active mutants and the I I I 10 20 30 40 one inactive mutant described above, there Time, (days) are now available a series of other mutants FIG. 3. Effect of mutational alterations on interchain produced by site-directed mutagenesis (27). interactions in C trimers. A mixture of ‘2sI-labeled C, These mutants, two of which are virtually (0.3 mg/ml; 10s cpm/pg) and C, (3 mg/ml) was incudevoid of enzyme activity, are of particular bated for various times at 4°C in 50 mM T&chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol and interest because the amino acid substitutions 0.2 mM EDTA. Aliquots were removed at specific times are located near the interfaces between adand analyzed by electrophoresis in polyacrylamide gels joining polypeptide chains. Hence it was relas described under Materials and Methods. The amount of radioactivity migrating as C&, was measured and the evant to determine, on the one hand, how results were plotted as percentage C,. remaining as a the interchain bonds in these trimers comhmction oftime. C,,, (O), C&(m), C,,,(O), and CW (A). pared to those in wild-type trimers and, on
STUDY
OF INTERCHAIN
INTERACTIONS
the other hand, whether pyrophosphate weakened the interchain interactions in these inactive mutant trimers as it did in the wildtype trimers. Figure 4 shows the electrophoresis patterns for different trimers (C,,,) in lane 10 mixed with succinylated wild-type trimers (C,,) in lane 1 and incubated together for 1 day and 7 days in the absence and presence of 10 mM sodium pyrophosphate. As seen in Figs. 4A and 4C, hybrids formed more rapidly with CW (lane 4), C,,, (lane 5), and CT76 (lane 9) than with wild-type C trimers (lane 2). In contrast, the rate of hybridization was lower for CT43(lane 3), C13, (lane 6), CYT4(lane 7), and CT68 (lane 8) than for C&. A comparison of the patterns in Fig. 4A with those in Fig. 4B as well as those in Figs. 4C and 4D shows that pyrophosphate in-
123456789lO
12345678910
C
D cr
123456789iO
/
,..r*i
12345678910
FIG. 4. Effect of pyrophosphate on interchain interactions in mutant C trimers. Mixtures of C,,, and C,, each at 3.5 mg/ml, were incubated at 4°C in 50 mM Tris-chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol and 0.2 mM EDTA and then subjected to electrophoresis as described under Materials and Methods. (A) 1 day in the absence of ligand, (B) 1 day of incubation in the presence of 10 mM sodium pyrophosphate, (C) 7 days of incubation in the absence of l&and, (D) 7 days in the presence of 10 mM sodium pyrophosphate. Lane 1, Cs alone; lane 2, wild-type C and Cs; lane 3, CT43and Cs; lane 4, CT45and Cs; lane 5, C,,, and Cs; lane 6, Cz3, and Cs; lane 7, CT,~ and Cs; lane 8, r&s and Cs; lane 9, CT,6 and Cs; and lane 10, wild-type C trimer alone.
BY HYBRIDIZATION
193
creases the rate of hybrid formation in the active mutants, C,, C745, and CT43, as well as in the virtually inactive mutants, C768 and CTT6, and the partially active mutant, CT7,. In contrast, pyrophosphate had essentially no effect on the rate of hybridization of the two inactive mutants C23 I and C7T4. How can we account for the effect of the mutational alterations on the interchain interactions in the C trimers and for the different effects of pyrophosphate in weakening the bonds between the chains of the trimers in the various mutants? Clearly no interpretations are warranted for CT45and CT43since the nature and location of the amino acid replacements are not known. For the other mutants, some speculations are justified but they must be considered as tentative until more quantitative studies with 1251-labeled proteins are conducted. Also, it should be recognized that the hybrids detected in the electrophoresis patterns in Fig. 4 are between mutant trimers and succinylated wild-type trimers. It is possible that such hybrids are less stable than each of the parental proteins and that the extent of hybrid formation is somewhat misleading. This potential pitfall can be overcome in hybridization experiments in which the mutant trimers are incubated with succinylated mutant trimers rather than succinylated wild-type trimers. In C231 the amino acid substitution involves the replacement of GIy-I 28 by a charged, bulky Asp leading to an inactive derivative with a significantly reduced affinity for the substrate, carbamoylphosphate, compared to the wild-type enzyme. It is to be expected, therefore, that the binding of pyrophosphate would be greatly reduced and that addition of 10 mM pyrophosphate would have no effect on the interchain bonds in this mutant. A similar interpretation can be offered for the lack of an effect of pyrophosphate on the interchain interactions in CTY4. In this mutant, Ser-52 is replaced by His leading to an almost complete loss of enzyme activity. Ultraviolet difference spectroscopy
194
YANG AND SCHACHMAN
indicates that this mutant does not bind carbamoylphosphate and it seems likely that pyrophosphate binding is greatly reduced as well. Why the interchain bonds in both of these mutant trimers appear to be strengthened is not known. Although it is tempting to speculate about electrostatic interactions because of the nature of the amino acid replacements, we consider such interpretations premature because local changes in folding resulting from the substitutions might affect other interchain interactions. The addition of sodium pyrophosphate causes a weakening of the interchain bonds in &, which contains Ala substituted for His- 134, in CT68where Lys-84 is replaced by Gln, and the double mutant, CTT6, which has both of these substitutions. Apparently the inhibitor binds to all of these mutants even though two of them, C768 and &, are virtually inactive and the other, C7,, , has only partial activity (V,,, about 5% that of C,.J. It is of interest that the substitution of Ala for His- 134 weakens the interchain bonds in the unliganded trimers, while that of Gln for Lys-84 strengthens the interactions. In the double mutant containing both of these substitutions, the weakening caused by Ala- 134 on one side of the interface between the adjoining chains apparently dominates over the strengthening that occurs when Gln replaces Lys-84. Although definitive interpretations of the effect of amino acid substitutions on the interchain interactions cannot be offered as yet, it seems likely that the application of the hybridization technique to additional mutants will provide invaluable information regarding the interactions between the polypeptide chains in C trimers. It will also be of value in interpreting the results of hydrogen exchange studies (35) and differential scanning calorimetry experiments (36,37) where it is difficult to distinguish between the contributions from interchain and intrachain interactions.
REFERENCES 1. Gerhart, J. C., and Schachman, H. K. (1965) Biochemistry 4, 1054-1062. 2. Weber, K. (1968) Nuture(London) 218,1116-l 119. 3. Wiley, D. C., and Lipscomb, W. N. (1968) Nature (London) 218, 1119- 112 1. 4. Meighen, E. A., Pigiet, V., and Schachman, H. K. (1970) Proc. Natl. Acad. Sci. USA 65,234-24 1. 5. Rosenbusch, J. P., and Weber, K. (1971) J. Biol. Chem. 246, 1644-1657. 6. Cohlberg, J. A., P&et, V. P., Jr., and Schachman, H. K. ( 1972) Biochemistry l&3396-34 11. 7. Schachman, H. K. ( 1974) Harvey Lect. 68,67- 113. 8. Monaco, H. L., Crawford, J. L., and Lipscomb, W. N. (1978) Proc. Nat. Acad. Sci. USA 75, 5276-5280. 9.
10. 11.
12. 13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24.
Bums, D. L., and Schachman, H. K. (1982) .I. Biol. Chem. 257, 12214-12218. Subramani, S., and Schachman, H. K. (1982) J. Biol. Chem. 257, 12219-12223. Honzatko, R. B., Crawford, J. L., Monaco, H. L., Ladner, J. E., Edwards, B. F. P., Evans, D. R., Warren, S. G., Wiley, D. C., Ladner, R. C., and Lipscomb, W. N. (1982) J. Mol. Biol. 160, 219-263. Ke, H.-M., Honzatko, R. B., and Lipscomb, W. N. (1984) Proc. Natl. Acad. Sci. USA 81,4037-4040. Subramani, S., Bothwell, M. A., Gibbons, I., Yang, Y. R., and Schachman, H. K. (1977) Proc. Nutl. Acad. Sci. USA 74,3777-378 1. Bothwell, M. A., and Schachman, H. K. (1980) .I. Biol. Chem. 255, 1962-1970. Bums, D. L., and Schachman, H. K. (1982) J. Biol. Chem. 257,8638-8647. Bums, D. L., and Schachman, H. K. (1982) J. Biol. Chem. 257,8648-8654. Gerhart, J. C., and Holoubek, H. (1967) J. Biol. Chem. 242,2886-2892. Navre, M., and Schachman, H. K. (1983) Proc. Natl. Acad. Sci. USA 80, 1207-12 11. Wall, K. A., and Schachman, H. K. (1979) J. Biol. Chem. 254, 119 17- 11926. Yang, Y. R., Kirschner, M. W., and Schachman, H. K. ( 1978) Methods in Enzymology (Hoffee, P., and Jones, M. E., Eds.), Vol. 15, pp. 35-41, Academic Press, New York. Syvanen, J. M., Yang, Y. R., and Kirschner, M. W. (1973) J. Biol. Chem. 248,3762-3768. Gibbons, I., and Schachman, H. K. (1976) Biochemistry 15, 52-60. Jovin, T., Chrambach, A., and Naughton, M. A. (1964) Anal. B&hem. 9,35 l-369. Wall, K. A., Flatgaard, J. E., Gibbons, I., and Schachman, H. K. (1979) J. Biol. Chem. 254, 11910-l 1916.
STUDY OF INTERCHAIN
INTERACTIONS
25. Gerhart, J. C. (1964) in Brookhaven Symposia in Biology, No. 17, 222-23 1. 26. Vickers, L. P., Compton, J. G., Wall, K. A., Flatgaard, J. E., and Schachman, H. K. ( 1984) J. Biol. Chem. 259, 11027-l 1035. 27. Robey, E. A., Wente, S. R., Markby, D. W., Flint, A., Yang, Y. R., and Schachman, H. K. (1986) Proc. Natl. Acad. Sci. USA 83, 5934-5938. 28. Wente, S. R., and Schachman, H. K. (1987) Proc. Natl. Acad. Sci. USA 84, 31-35. 29. Lahue, R. S., and Schachman, H. K. (1984) J. Biol. Chem. 259, 13906-13913. 30. Iahue, R. S., and Schachman, H. K. (1986) J. Biol. Chem. 261,3079-3084.
BY HYBRIDIZATION
195
3 1. Robey, E. A., and Schachman, H. K. (1985) Proc. Natl. Acad. Sci. USA 82,361-365. 32. Kleppe, K. (1966) Biochem. Biophys. Acta 122, 450-461. 33. Porter, R. W., Modebe, M. O., and Stark, G. R. (1969)J. Biol. Chem. 244, 1846-1859. 34. Honzatko, R. B., and Lipscomb, W. N. (1982) J. Mol. Bioi. 160,265-286. 35. Lennick, M., and Allewell, N. M. (198 1) Proc. Nutl. Acad. Sci. USA 78,6759-6763. 36. Vickers, L. P., Donovan, J. W., and Schachman, H. K. (1978) J. Biol. Chem. 253,8493-8498. 37. Edge, V., Allewell, N. M., and Sturtevant, J. M. (1985) Biochemistry 24,5899-5906.
163, 188- 195 ( 1987)
BIOCHEMISTRY
Hybridization as a Technique for Studying Interchain Interactions Catalytic Trimers of Aspartate Transcarbamoylasei3*
in the
YING R. YANG AND H. K. SCHACHMAN~ Department of Molecular Biology and the Virus Laboratory Wendell M. Stanley Hall, University of California, Berkeley, California 94720 Received January 27, 1987 Since subunit interactions in regulatory enzymes mediate the ligand-promoted conformationai changes responsible for their allosteric properties, it is necessary to have techniques for determining the effects of ligands and mutational alterations on the strength of the interchain interactions. In aspartate transcarbamoylase from Escherichia coli, the multiple interchain interactions are so linked that it is difficult to study them separately. Therefore, we have focused on the nonahosteric catalytic trimers isolated from the holoenzyme and have used the rate of hybrid formation between native and succinylated protein as a measure of the dissociation of the trimers into single polypeptide chains. Although catalytic trimers exhibit no evident dissociation in sedimentation studies at lo-* M, incubation of mixtures of native and succinylated trimers for long periods of time (days) yielded hybrids which are readily detected by polyacrylamide gel electrophoresis. This sensitive technique was used to demonstrate that the substrate, carbamoylphosphate, and the bisubstrate analog, N-(phosphonacetyl)-L-aspartate, cause a marked strengthening of the interchain interactions, whereas the inhibitor, sodium pyrophosphate, at concentrations as low as 10 mM, promotes dissociation of the trimers. This weakening of the interchain interactions by pyrophosphate facilitated the isolation and purification of functionally competent hybrid trimers by a technique which was much more convenient and provided higher yields than previous, more drastic methods which employed urea or guanidine hydrochloride to cause dissociation of the trimers. The hybridization technique was useful in studying the effects of mutational alterations on the strength of the interchain interactions and the ability of active and inactive mutants to bind pyrophosphate. o 1987 Academic F’res. Inc. KEY WORDS: subunit interactions; ligand effects; hybrid formation; dissociation of oligomerit enzymes; effect of amino acid substitutions.
INTRODUCTION
A aetailecl unaerxanaing ctf the allosteric behavior of oligomeric protej ins in terms of ligand-promoted conformatic mal changes is .*_.*.
,
‘
,.
’ This work was supported by United States Public Health Services Research Grant GM 12159 from the National Institute of General Medical Sciences and by National Science Foundation Research Grant DMB 85-02131. * This paper is dedicated to the memory of Nathan 0. Kaplan who was a pioneer in the use of hybridization in his classical studies of the subunit structure of isozymes. 3 To whom reprint requests should be addressed at Department of Molecular Biology, University of CaIifornia, Berkeley, 229 Stanley Hall, Berkeley, CA 97420. 0003-2697187 $3.00 Copyright @ 1987 by Academic Press, hc. AU x-i&s of reproduction in any fom mawd.
dependent upon knowledge of the strengths of the interchain interactions and the effects of ligands on them. For a regulatory enzyme like Escherichia coli aspartate transcarbamLoylase (ATCase, carbamoylphosphate: 4 Abbreviations used: ATCase, aspartate transcarbamoylase; C, catalytic trimer; R, regulatory dimer; c, catalytic polypeptide chain; r, regulatory polypeptide chain; CN or C,.., native C trimer; Cs or C,, succinylated C trimer; Cr or C&, tetrahydrophthaloylated C trimer; CM, wild-type C trimer; Cx or C,,, mutant C trimer; Cnnt, hybrid trimer containing two native and one tetrahydrophthaloylated chains; THPA, 3,4,5,6-tetrahydrophthalic anhydride; PALA, N-(phosphonacetyl)-L-aspartate; EDTA, ethylenediaminetetraacetate. 188
STUDY
OF
INTERCHAIN
INTERACTIONS
aspartate carbamoyltransferase, EC 2.1.3.2) an evaluation of the changes in the bonding energies between polypeptide chains is particularly difficult because of the large number and different types of noncovalent interchain “bonds” in the holoenzyme. The 12 polypeptide chains in ATCase are organized as two cataytic (C) trimers and three regulatory (R) dimers (l-6). Thus there are three bonds linking the catalytic (c) polypeptide chains within each C trimer and one bond linking the regulatory (r) chains in each R dimer. In addition, the two C trimers are “cross-linked” noncovalently by a variety of interactions involving bonds between c and r chains as well as those between one C trimer and the other (7-12). Because of the multiplicity of these interactions and their intrinsic strengths, ATCase exhibits little tendency to dissociate into subunits even at great dilution (10-l’ M) and very little exchange of either C or R subunits is detectable when radioactive subunits are incubated with wildtype native or reconstituted ATCase ( 13,14). The diverse interactions among the 12 polypeptide chains are doubtlessly linked with the result that the binding of a ligand or an amino acid substitution affects more than one type of bond. Since it is extremely hazardous to attempt partitioning the energies among discrete types of bonds in intact ATCase, we focus here on the isolated nonallosteric C trimers to determine the effects of ligands and mutational alterations on the interactions between the c chains in the C trimers (9,15,16). Although C trimers are very stable in neutral solutions and show no evident dissociation in sedimentation velocity experiments at about 10e8 M, hybrids are readily detected when wild-type C trimers and succinylated (Cs) trimers are incubated together for prolonged periods of time (9). From measurements of the rate of formation of hybrids between unmodified and succinylated trimers, we obtain evidence for the dissociation of C trimers and an indication of the strengths of the interactions between c chains
BY
189
HYBRIDIZATION
under different conditions. The technique is presented here along with preliminary applications to studies on the effects of ligands and mutational alterations on the relative strengths of the interchain interactions. MATERIALS
AND
METHODS
Succinic anhydride and succinate were purchased from Eastman Chemicals, neohydrin and 3,4,5,6-tetrahydrophthalic anhydride (THPA) were obtained from K and K Laboratories, and carrier-free 1251as the sodium salt in NaOH was obtained from Amersham Corp. Carbamoylphosphate as the lithium salt was purchased from Boehringer-Mannheim, L-aspartate from Calbiochem, carbamoylaspartate and all nucleoside triphosphates from Sigma Chemicals, and sodium pyrophosphate and potassium phosphate from Mallinckrodt. DEAEcellulose was purchased from Whatman and DEAE-Sephadex A-50 from Pharmacia Biotechnology. N-(phosphonacetyl)+aspartate (PALA), lot MK45-89-1, was kindly provided by Dr. Robert R. Engle, Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute. Wild-type ATCase was prepared according to the method of Gerhart and Holoubek (17) from E. coli strain TR4363 containing plasmid pPYRB3 (18) which carries the intact pyrB1 operon. Mutant forms of ATCase were purified by the procedure described by Wall and Schachman (19) with slight modifications. C trimers were obtained by dissociating ATCase with neohydrin followed by DEAE-cellulose chromatography (20). ‘251labeled C trimers were prepared by the method of Syvanen et al. (2 1). Succinylation of C trimers was performed by mixing protein with a 3.5 molar ratio of succinic anhydride per lysine residue (4) to yield a derivative with about four succinyl groups per c chain. For the purification of hybrids of different species, the C trimers were acylated with THPA to yield CT as described by Gib-
190
YANG AND SCHACHMAN
bons and Schachman (22). Interchain hybrids were formed by incubating mixtures of Cx (C,,,) and Cs (C,) or CT (C& in 50 mM Tris-chloride buffer at pH 7.5 containing 2 mM 2-mercaptoethanol and 0.2 mM EDTA. Each protein was at a concentration of 3.5 mg/ml and the hybridization experiments were performed at 4°C for various lengths of time. Electrophoresis experiments in polyacrylamide gels were performed with a Hoefer Mighty Small slab gel apparatus using the discontinuous buffer system of Jovin et al. (23). The 7.5% gels were stained with Coomassie brilliant blue G-250 in 12.5% trichloracetic acid. In experiments with ‘251-labeled proteins, the various bands were cut from the gels and measurements of radioactivity were performed with a Nuclear Chicago gamma counter. The following mutant forms of ATCase were used in the various studies. ATCasezj,, which is virtually devoid of activity (< 10e5 that of the wild-type enzyme), contains Asp in place of Gly-128 in the c chain (19,24). ATCaseTd5 and ATCaseTd3 are active mutants isolated from two different spontaneous revertants from a pyrimidine auxotrophic strain (25,26). ATCaseTG8 produced by sitedirected mutagenesis contains Gln in place of Lys at position 84 in the c chain (27,28). ATCaseTT I formed by site-directed mutagenesis contains Ala in place of His-l 34 in the c chain (27). ATCaseTT4 formed by site-directed mutagenesis has His in place of Ser-52 in the c chain (Y. R. Yang and H. K. Schachman, unpublished). ATCase? is a double mutant in which Lys-84 was replaced by Gln and His-l 34 by Ala (28). RESULTS AND DISCUSSION
Efect of ligands on the strength of interchain interactions in C trimers. In the initial experiments aimed at determining the number of polypeptide chains in the C subunits of ATCase (4), hybrids between CN and Cs were formed by adding guanidine hydrochloride
to a mixture of the two proteins followed by dialysis to permit random reassociation of the chains into oligomers. Electrophoretic analysis of the reconstituted mixture revealed a four-member set thereby demonstrating that the proteins were trimers. When CN and Cs were incubated overnight in the absence of a denaturant no hybrids were detected, indicating that the C trimers were extremely stable with little tendency to dissociate into individual polypeptide chains. Subsequent experiments under different conditions showed hybrids were formed slowly from C, and Cs and that the rate of hybridization varied with temperature, pH, and ionic strength (9). Moreover, the substrate carbamoylphosphate and the bisubstrate ligand, PALA, caused a very marked decrease in the rate of formation of hybrids (9). Since it was apparent that specific ligands had a significant effect on the strength of the interaction between the chains in C trimers, we conducted additional studies which are illustrated by the electrophoresis patterns in Fig. 1. In all of these experiments the incubation of CN and Cs was performed at 4°C because of the earlier finding (9) that hybrid formation was more rapid at that tempera-
1 2 3 4 5 6 7 8 9101112 FIG. 1. Effect of ligands on the strength of interchain interactions in C trimers. Mixtures of C.,, and C,, each at 3.5 mg/ml in 50 mM T&chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol and 0.2 mM EDTA in the absence or presence of 10 mM ligand were incubated at 4°C for 2 days and then analyzed by polyacrylamide gel electrophoresis as described under Materials and Methods. Lane 1, no l∧ lane 2, PALA; lane 3, carbamoylphosphate; lane 4, L-aspartate; lane 5, succinate; lane 6, carbamoylaspartate; lane 7, potassium phosphate; lane 8, sodium pyrophosphate; lane 9, ATP, lane 10, CTP; lane 11, GTP; and lane 12, UTP.
STUDY OF INTERCHAIN
INTERACTIONS
ture than at 25 ’ C. Lane 1, which represents a control in which no ligands were present, shows the presence of only small amounts of hybrids after 2 days of incubation of CN and Cs in a Tris-chloride buffer at pH 7.5. When PALA or carbamoylphosphate were present (10 mM) there was a significant reduction in the amount of hybrids, lanes 2 and 3, respectively. In contrast, aspartate (lane 4) and succinate (lane 5) had little effect on the rate of hybrid formation. Carbamoylaspartate, a product of the reaction catalyzed by ATCase, apparently strengthened the interchain interactions as seen by the decrease in the amount of hybrids formed during the incubation (lane 6). On the other hand, phosphate, the other product of the reaction between carbamoylphosphate and aspartate, appeared to decrease the strength of the bonds between c chains in the C trimer (lane 7). By far, the greatest weakening of the interchain interactions was caused by pyrophosphate (lane 8) ATP (lane 9), CTP (lane lo), GTP (lane 11) and UTP (lane 12). In the presence of these ligands, hybrid formation was greatly enhanced indicating that their binding to CN ,
.M”,
“^
12345678910 FIG. 2. Formation and purification of interchain hybrids from CN and Cr trimers. Mixtures of C,,,, and C&, each at 3.5 mg/ml, were incubated overnight at 4°C in 50 mM T&chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol, 0.2 mM EDTA, and 10 mM sodium pyrophosphate. The hybrid set containing four species, lanes 1 and 6, was then fractionated by chromatography on a DEAE-Sephadex A-50 column to yield four pooled fractions containing, respectively, C,.. , lane 2; C,,, , lane 3; Cnti, lane 4; and C&, lane 5. Each of the four purified trimers was dialyzed for 36 h at 4°C against 40 mM phosphate buffer at pH 6.1. The deacylated species were then subjected to electrophoresis to yield C.,,, lane 7; deacylated C,,,, lane 8; deacylated C,,, lane 9; and deacylated C, , lane 10.
BY HYBRIDIZATION
191
trimers caused a marked decrease in the strength of the bonds between c chains. The weakening of the interchain interactions by pyrophosphate provides a particularly useful method for the production and isolation of interchain hybrids, which complements the previous techniques using guanidine hydrochloride or urea on the one hand (4) and NaCNS or NaC104 on the other (9,29). Figure 2 shows the hybrid set formed by incubating CN and CT for 1 day at 4°C in the pH 7.5 Tris-chloride buffer containing 10 mM sodium pyrophosphate (lanes 1 and 6). The individual species were separated by chromatography on DEAE-Sephadex A-50 and the patterns for C,,,, C,,,, &, and C& are shown in lanes 2,3,4, and 5, respectively. When these species were dialyzed at pH 6.1 according to the procedure of Gibbons and Schachman (22), the THP-groups were removed and the deacylated proteins (lanes 8, 9, and 10) had the same electrophoretic mobility as the native trimers (lanes 2 and 7). This method for the preparation of interchain hybrids by using sodium pyrophosphate to weaken the interchain bonds in C trimers is clearly preferable to those methods used previously. It is less time consuming and the yields are higher because of the decreased precipitation of protein that occurs when guanidine hydrochloride or NaCNS are used to dissociate the different C trimers. The technique has been used in this laboratory to prepare purified, functional hybrids from native and chemically modified C trimers (30,31) and from C trimers isolated from different mutant forms of ATCase (28,3 1). It should be noted, however, that pyrophosphate is not effective in dissociating the C trimers from some mutants due, presumably, to its inability to bind to the altered proteins (see below).
Effect of mutational alterations on interchain interactions in C trimers. The availability of two active mutants with one, ATCase745, having significantly less and the other, ATCase743, more cooperativity than the
192
YANG AND SCHACHMAN
wild-type enzyme (26) provided an opportunity to determine whether measurements of the strength of the interchain interactions in the C trimers could provide useful information relevant to the allosteric properties of the holoenzymes. Both ATC~ZXZ,~~ and ATCaseTd3 are revertants of an inactive mutant of wild-type ATCase (25,26). In neither case have the positions of the amino acid substitutions been located. ATCaseTd3 has a significantly higher extinction coefficient at 280 nm than do either ATCaseTd5 or wild-type. ATCase indicating that it contains two additional tyrosines, but it is not known whether ATCaseTd5 is also a second-site revertant. Both mutants, despite their markedly different allosteric properties and ligand-promoted conformational changes, have similarly weakened interactions between c and r chains (26) in comparison to those in wildtype ATCase. As seen in Fig. 3, the rate of dissociation of CT45 trimers was greater than that of wild-
‘lo1 100 90
type C trimers and, conversely, the rate of dissociation of CT43trimers was significantly lower. For these experiments, the mutant trimers were labeled with 12’1 and incubated with an excess of Cs for various lengths of time. The measurements involved slicing the gels and determining the amount of radioactivity associated with the most slowly moving species. This technique, which is much more quantitative than that involving visual inspection of the electrophoresis patterns, shows clearly that the bonds between c chains in the C trimers isolated from the less cooperative mutant are substantially weaker than those in the C trimers from wild-type enzyme. In contrast, the interchain bonds in the C trimers from the more cooperative mutant are stronger than those from the wild-type holoenzyme. For comparative purposes, results are also shown in Fig. 3 for the C trimers from ATCaser3,, an inactive mutant in which Gly at position 128 is replaced by Asp (19). The bonds between c chains in this trimer are clearly stronger than those in either of the other mutants or the wild-type trimer.
Eflect of mutational alterations on the ability ofpyrophosphate to weaken the interchain interactions in C trimers. Since it is known
that pyrophosphate is a strong inhibitor of ATCase (32,33) and that it binds near the interface between adjacent polypeptide chains (34), it seemed to be of interest to determine the effect of pyrophosphate on the interchain bonds in mutant forms of ATCase. In addition to the active mutants and the I I I 10 20 30 40 one inactive mutant described above, there Time, (days) are now available a series of other mutants FIG. 3. Effect of mutational alterations on interchain produced by site-directed mutagenesis (27). interactions in C trimers. A mixture of ‘2sI-labeled C, These mutants, two of which are virtually (0.3 mg/ml; 10s cpm/pg) and C, (3 mg/ml) was incudevoid of enzyme activity, are of particular bated for various times at 4°C in 50 mM T&chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol and interest because the amino acid substitutions 0.2 mM EDTA. Aliquots were removed at specific times are located near the interfaces between adand analyzed by electrophoresis in polyacrylamide gels joining polypeptide chains. Hence it was relas described under Materials and Methods. The amount of radioactivity migrating as C&, was measured and the evant to determine, on the one hand, how results were plotted as percentage C,. remaining as a the interchain bonds in these trimers comhmction oftime. C,,, (O), C&(m), C,,,(O), and CW (A). pared to those in wild-type trimers and, on
STUDY
OF INTERCHAIN
INTERACTIONS
the other hand, whether pyrophosphate weakened the interchain interactions in these inactive mutant trimers as it did in the wildtype trimers. Figure 4 shows the electrophoresis patterns for different trimers (C,,,) in lane 10 mixed with succinylated wild-type trimers (C,,) in lane 1 and incubated together for 1 day and 7 days in the absence and presence of 10 mM sodium pyrophosphate. As seen in Figs. 4A and 4C, hybrids formed more rapidly with CW (lane 4), C,,, (lane 5), and CT76 (lane 9) than with wild-type C trimers (lane 2). In contrast, the rate of hybridization was lower for CT43(lane 3), C13, (lane 6), CYT4(lane 7), and CT68 (lane 8) than for C&. A comparison of the patterns in Fig. 4A with those in Fig. 4B as well as those in Figs. 4C and 4D shows that pyrophosphate in-
123456789lO
12345678910
C
D cr
123456789iO
/
,..r*i
12345678910
FIG. 4. Effect of pyrophosphate on interchain interactions in mutant C trimers. Mixtures of C,,, and C,, each at 3.5 mg/ml, were incubated at 4°C in 50 mM Tris-chloride buffer, pH 7.5, containing 2 mM 2-mercaptoethanol and 0.2 mM EDTA and then subjected to electrophoresis as described under Materials and Methods. (A) 1 day in the absence of ligand, (B) 1 day of incubation in the presence of 10 mM sodium pyrophosphate, (C) 7 days of incubation in the absence of l&and, (D) 7 days in the presence of 10 mM sodium pyrophosphate. Lane 1, Cs alone; lane 2, wild-type C and Cs; lane 3, CT43and Cs; lane 4, CT45and Cs; lane 5, C,,, and Cs; lane 6, Cz3, and Cs; lane 7, CT,~ and Cs; lane 8, r&s and Cs; lane 9, CT,6 and Cs; and lane 10, wild-type C trimer alone.
BY HYBRIDIZATION
193
creases the rate of hybrid formation in the active mutants, C,, C745, and CT43, as well as in the virtually inactive mutants, C768 and CTT6, and the partially active mutant, CT7,. In contrast, pyrophosphate had essentially no effect on the rate of hybridization of the two inactive mutants C23 I and C7T4. How can we account for the effect of the mutational alterations on the interchain interactions in the C trimers and for the different effects of pyrophosphate in weakening the bonds between the chains of the trimers in the various mutants? Clearly no interpretations are warranted for CT45and CT43since the nature and location of the amino acid replacements are not known. For the other mutants, some speculations are justified but they must be considered as tentative until more quantitative studies with 1251-labeled proteins are conducted. Also, it should be recognized that the hybrids detected in the electrophoresis patterns in Fig. 4 are between mutant trimers and succinylated wild-type trimers. It is possible that such hybrids are less stable than each of the parental proteins and that the extent of hybrid formation is somewhat misleading. This potential pitfall can be overcome in hybridization experiments in which the mutant trimers are incubated with succinylated mutant trimers rather than succinylated wild-type trimers. In C231 the amino acid substitution involves the replacement of GIy-I 28 by a charged, bulky Asp leading to an inactive derivative with a significantly reduced affinity for the substrate, carbamoylphosphate, compared to the wild-type enzyme. It is to be expected, therefore, that the binding of pyrophosphate would be greatly reduced and that addition of 10 mM pyrophosphate would have no effect on the interchain bonds in this mutant. A similar interpretation can be offered for the lack of an effect of pyrophosphate on the interchain interactions in CTY4. In this mutant, Ser-52 is replaced by His leading to an almost complete loss of enzyme activity. Ultraviolet difference spectroscopy
194
YANG AND SCHACHMAN
indicates that this mutant does not bind carbamoylphosphate and it seems likely that pyrophosphate binding is greatly reduced as well. Why the interchain bonds in both of these mutant trimers appear to be strengthened is not known. Although it is tempting to speculate about electrostatic interactions because of the nature of the amino acid replacements, we consider such interpretations premature because local changes in folding resulting from the substitutions might affect other interchain interactions. The addition of sodium pyrophosphate causes a weakening of the interchain bonds in &, which contains Ala substituted for His- 134, in CT68where Lys-84 is replaced by Gln, and the double mutant, CTT6, which has both of these substitutions. Apparently the inhibitor binds to all of these mutants even though two of them, C768 and &, are virtually inactive and the other, C7,, , has only partial activity (V,,, about 5% that of C,.J. It is of interest that the substitution of Ala for His- 134 weakens the interchain bonds in the unliganded trimers, while that of Gln for Lys-84 strengthens the interactions. In the double mutant containing both of these substitutions, the weakening caused by Ala- 134 on one side of the interface between the adjoining chains apparently dominates over the strengthening that occurs when Gln replaces Lys-84. Although definitive interpretations of the effect of amino acid substitutions on the interchain interactions cannot be offered as yet, it seems likely that the application of the hybridization technique to additional mutants will provide invaluable information regarding the interactions between the polypeptide chains in C trimers. It will also be of value in interpreting the results of hydrogen exchange studies (35) and differential scanning calorimetry experiments (36,37) where it is difficult to distinguish between the contributions from interchain and intrachain interactions.
REFERENCES 1. Gerhart, J. C., and Schachman, H. K. (1965) Biochemistry 4, 1054-1062. 2. Weber, K. (1968) Nuture(London) 218,1116-l 119. 3. Wiley, D. C., and Lipscomb, W. N. (1968) Nature (London) 218, 1119- 112 1. 4. Meighen, E. A., Pigiet, V., and Schachman, H. K. (1970) Proc. Natl. Acad. Sci. USA 65,234-24 1. 5. Rosenbusch, J. P., and Weber, K. (1971) J. Biol. Chem. 246, 1644-1657. 6. Cohlberg, J. A., P&et, V. P., Jr., and Schachman, H. K. ( 1972) Biochemistry l&3396-34 11. 7. Schachman, H. K. ( 1974) Harvey Lect. 68,67- 113. 8. Monaco, H. L., Crawford, J. L., and Lipscomb, W. N. (1978) Proc. Nat. Acad. Sci. USA 75, 5276-5280. 9.
10. 11.
12. 13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24.
Bums, D. L., and Schachman, H. K. (1982) .I. Biol. Chem. 257, 12214-12218. Subramani, S., and Schachman, H. K. (1982) J. Biol. Chem. 257, 12219-12223. Honzatko, R. B., Crawford, J. L., Monaco, H. L., Ladner, J. E., Edwards, B. F. P., Evans, D. R., Warren, S. G., Wiley, D. C., Ladner, R. C., and Lipscomb, W. N. (1982) J. Mol. Biol. 160, 219-263. Ke, H.-M., Honzatko, R. B., and Lipscomb, W. N. (1984) Proc. Natl. Acad. Sci. USA 81,4037-4040. Subramani, S., Bothwell, M. A., Gibbons, I., Yang, Y. R., and Schachman, H. K. (1977) Proc. Nutl. Acad. Sci. USA 74,3777-378 1. Bothwell, M. A., and Schachman, H. K. (1980) .I. Biol. Chem. 255, 1962-1970. Bums, D. L., and Schachman, H. K. (1982) J. Biol. Chem. 257,8638-8647. Bums, D. L., and Schachman, H. K. (1982) J. Biol. Chem. 257,8648-8654. Gerhart, J. C., and Holoubek, H. (1967) J. Biol. Chem. 242,2886-2892. Navre, M., and Schachman, H. K. (1983) Proc. Natl. Acad. Sci. USA 80, 1207-12 11. Wall, K. A., and Schachman, H. K. (1979) J. Biol. Chem. 254, 119 17- 11926. Yang, Y. R., Kirschner, M. W., and Schachman, H. K. ( 1978) Methods in Enzymology (Hoffee, P., and Jones, M. E., Eds.), Vol. 15, pp. 35-41, Academic Press, New York. Syvanen, J. M., Yang, Y. R., and Kirschner, M. W. (1973) J. Biol. Chem. 248,3762-3768. Gibbons, I., and Schachman, H. K. (1976) Biochemistry 15, 52-60. Jovin, T., Chrambach, A., and Naughton, M. A. (1964) Anal. B&hem. 9,35 l-369. Wall, K. A., Flatgaard, J. E., Gibbons, I., and Schachman, H. K. (1979) J. Biol. Chem. 254, 11910-l 1916.
STUDY OF INTERCHAIN
INTERACTIONS
25. Gerhart, J. C. (1964) in Brookhaven Symposia in Biology, No. 17, 222-23 1. 26. Vickers, L. P., Compton, J. G., Wall, K. A., Flatgaard, J. E., and Schachman, H. K. ( 1984) J. Biol. Chem. 259, 11027-l 1035. 27. Robey, E. A., Wente, S. R., Markby, D. W., Flint, A., Yang, Y. R., and Schachman, H. K. (1986) Proc. Natl. Acad. Sci. USA 83, 5934-5938. 28. Wente, S. R., and Schachman, H. K. (1987) Proc. Natl. Acad. Sci. USA 84, 31-35. 29. Lahue, R. S., and Schachman, H. K. (1984) J. Biol. Chem. 259, 13906-13913. 30. Iahue, R. S., and Schachman, H. K. (1986) J. Biol. Chem. 261,3079-3084.
BY HYBRIDIZATION
195
3 1. Robey, E. A., and Schachman, H. K. (1985) Proc. Natl. Acad. Sci. USA 82,361-365. 32. Kleppe, K. (1966) Biochem. Biophys. Acta 122, 450-461. 33. Porter, R. W., Modebe, M. O., and Stark, G. R. (1969)J. Biol. Chem. 244, 1846-1859. 34. Honzatko, R. B., and Lipscomb, W. N. (1982) J. Mol. Bioi. 160,265-286. 35. Lennick, M., and Allewell, N. M. (198 1) Proc. Nutl. Acad. Sci. USA 78,6759-6763. 36. Vickers, L. P., Donovan, J. W., and Schachman, H. K. (1978) J. Biol. Chem. 253,8493-8498. 37. Edge, V., Allewell, N. M., and Sturtevant, J. M. (1985) Biochemistry 24,5899-5906.