J. Mol. Biol. (1991) 219, 161-163
Crystallization
of Human Thymidylate
Synthase
Celia A. Schiffer’3*, V. Jo Davisson113t, Daniel V. Santi1y3 and Robert M. Stroud’*2$ 1 Department
of Biochemistry
and Biophysics
and
*Graduate Group in Biophysics, and 3 Department of Pharmaceutical Chemistry University of California San Francisco, CA 94143-0448, U.S.A. (Received 25 July 1990; accepted 2 February 1991) Human thymidylate synthase has been crystallized in the absence of ligands and diffracts beyond 3.0 A. The protein was cloned and expressed in Escherichia coli and then crystallized from ammonium sulfate in the presence of /?-mercaptoethanol at a variety of pH values. The crystals are trigonal in the space-group P3,21; the unit cell dimensions are a = b = 967 A, r:=84.1 A. Keywords: thymidylate
synthase; crystallography;
1. Introduction
solution containing 6.5 mg protein/ml, buffer with 20 mix-j?-mercaptoethanol
$ Author to whom all correspondence should be addressed.
TS, thymidylate
synthasr.
161 (M)22--28:~6/91/1(W,1614)3 $03.0010
in Tris.HCl
and 1.5 Mammonium sulfate at pH 6.5 to 8-O. This is the first example of TS crystallized without either phosphate or substrate in the medium. Addition of phosphate to the suspensions cracked the crystals, suggesting a possible phosphate-induced conformational change. The space-group P3,21 was distinguished from its enantiomorph P3221 by the molecular replacement (C.A.S., unpublished results) solution with unit cell dimensions, determined by diffractometry as a = h = 96.7 A, c = 841 A. There is one monomer per
7 Present address: Depart,ment of Medicinal (lhemistry and Pharmacognosy, Purdue University, RHPH, West Lafayette, Ix 47907, U.S.A. used:
cancer
A knowledge of the structure of human TS may be important for rational design of chemotherapeutically useful inhibitors of this enzyme. First, it would provide information important in the design of inhibitors useful in the treatment of human neoplastic diseases. Second, it would serve as a needed template to compare the host TS with the enzyme from human pathogens; this will elucidate key differences between the host and micro-organism TS, and aid the design of inhibitors that selectively target the invading organism. It has not been feasible to obtain the amounts of human TS needed for crystallography from natural sources. Recently, human TS has been cloned and expressed in E. coli (Davisson et al., 1989). Here, we describe the cryst,allization and space-group determination of recombinant human TS. Human TS was cloned and expressed in E. coli and the enzyme was purified and assayed as described (Davisson et aZ., 1989). Crystals were grown in hanging drops at room temperature from a
Thymidylate synthase (TS$) catalyzes the con2’-deoxyuridylate version of (dUMP) and 5,10-methylenetetrahydrofolate to 2’-deoxythymidylate (dTMP) and dihydrofolate. Since TS catalyzes a step in the sole de nova pathway for dTMP synthesis, it is an important target for potential chemotherapeutic agents. There is a large body of knowledge on the structure and function of TS. Many details of the mechanism and inhibition have been elucidated (for a review, see Santi & Danenberg, 1984; Finer-Moore et al., 1990). The primary sequences of TS from 17 sources are known and their alignment shows that TS is the most highly conserved of all enzymes (Perryman et al., 1986; Perry et al., 1990). The X-ray structures of the Lactobacillus casei TS-Pi (Hardy et aZ., 1987) and TS-dUMP complexes have been determined and refined to 2.3 A (1 f! = 0.1 nm) resolution (J. S. Finer-Moore, unpublished results). The structure of a binary complex of the Escherichia coli enzyme with phosphate (Perry et al., 1990) and a ternary complex with dUMP and a cofactor analog, N’0-propargyl-5,8-dideazafolate (CB3717), have also been determined (Montfort et aZ., 1990).
5 Abbreviation
enzyme; protein structure;
0 1991 Academic, Press Limited
162
C. A. Nchifler
et al
Figure 1. A precession photograph showing the [O, I, - I] zone of diffraction shown extends to 3.6 A resolution and the crystals diffract to 2% A resolution.
asymmetric unit and the crystals diffract to beyond 2.8 A resolution. A precession photograph of the [0, 1, - l] zone is shown in Figure 1. Observations (36,811) were collected from two crystals and the reflections merged with the XENGEN package (Howard et ab., 1985). X-ray crystallographic intensities were recorded on a Nicolet IPC area detector, with a three-circle goniometer, using CuKa X-rays. The source was a 200 pm focus, Rigaku 200 RV rotating target tube with a graphite monochromator. Data frames were processed with the software of Blum et al. (1987). Intensities were reduced as described by Montfort et al. (1990). From the high primary sequence homology (Takeishi et al., 1985), it is expected that the structure of human TS will follow the same general fold as that found for the structures from L. casei and E. co& (Perry et al., 1990). The amino-terminal in any other TS sequence, without counterpart sequences, is 27 residues longer than L. casei TS and contains eight proline residues. Two eukaryoticspecific insertions at positions 117-128(lc9&101~), and 145-154(1c156-157) exist in the human TS sequence. The first insertion is near the active site? probably in a similar position as the small domain of L. casei TS. The second insertion is on the edge of the dimer interface. From a homology model based on the crystal structure of L. casei TS (C.A.S. and results), the human TS R.M.S., unpublished monomer has two cysteine residues that occur at t The numbering system used for amino acid residues of TS refers to the sequence of the human enzyme followed by the corresponding homologous sequences of the reference L. casei (lc) TS residues in parentheses.
from csrystnls of human TS. The pattern
positions 199(1c202) and 210(1c213). These residues are expected to line on adjacent p-strands close enough potentially to form disulfide bonds. Also: cysteine residues 18O(lc183) in each monomer are juxtaposed across the dimer interface in a manner that might allow for formation of a disulfide linkage. However, non-reducing SDS/polyacrylamide gel electrophoresis of dissolved crystals did not reveal evidence for a disulfide-linked dimer (data not shown). Refinement of the X-ray structure may elucidate whether a disulfide bond forms during crystallization. The proximity of two different pairs of sulfhydryl groups suggests the possibility of cross-linking under oxidizing conditions. The refinement of the crystal structure to the data will provide the first three-dimensional view of an unliganded mammalian TS and provide a template for anti-cancer drug design. C.A.S. thanks V. Ramalingam, William Montfort, Janet Finer-Moore, Kathy Perry and Chris Carrerras for their help and advice. This research was supported by the National Institutes of Health grants R01-CA41323 to J.S.F. and R.M.S., GM24485 to R.M.S. and CA14394 to D.V.S. V.J.D. was supported by the Damon RunyonWalter Winchell Cancer Research fund for a postdoctoral fellowship.
References Blum, M., Metcalf, P., Harrison, S. C. & Wiley. I). (1. (1987). J. Appl. Crystallogr. 20, 235-242. Bradford, M. M. (1976). And Biochem. 72, 248-254. Davisson, V. J., Siraworaporn, W. & Santi, D. V. (1989). J. Biol. Chem. 264, 9145-9148. Finer-Moore, J. S., Montfort, W. R. & Stroud, R. M. (1990). Riochemistry, 29. 6977-6986.
Communications Hardy, L. W., Finer-Moore, J. S., Montfort, W. R., Jones, M. O., Santi. D. V. & Stroud, R. M. (1987). Science, 235, 448-455. Howard, A. J., ?u’eilsen, C. & Xuong, ru’g, H. (1985). Methods Enzymol. 115, 252-270. Montfort, W. R., Perry, K. M., Fauman, E. B., FinerMoore, J. S., Maley, G. F., Maley, F. & Stroud, R. M. (1990). Biochemistry, 29, 6964-6977. Perry, K. M., Fauman, E. B., Finer-Moore, J. S., Montfort, W. R., Maley, C. F., Maley. F. & Stroud, R. M. (1990). Proteins, 8, 315-333.
163
Perryman, S. M., Rossana, C., Deng, T.. Vanin, E. F. & Johnson, L. F. (1986). Mol. Biol. Evol. 3, 313-321. Santi, D. V. & Danenberg, P. V. (1984). In Folates and Pterins, vol. 1, Chemistry and Biochemistry of Folates (Blakley, R. L. t Benkovic, S. J., eds), pp. 345-398, John Wiley and Sons, New York. Takeishi. K.. Saneda, S., Ayusawa, D., Shimizu, Ii., (iotoh, 0. & Seno. T. (1985). Nucl. Acids Res. 13, 203562043.
Edited by R. Huber
Crystallization
of Human Thymidylate
Synthase
Celia A. Schiffer’3*, V. Jo Davisson113t, Daniel V. Santi1y3 and Robert M. Stroud’*2$ 1 Department
of Biochemistry
and Biophysics
and
*Graduate Group in Biophysics, and 3 Department of Pharmaceutical Chemistry University of California San Francisco, CA 94143-0448, U.S.A. (Received 25 July 1990; accepted 2 February 1991) Human thymidylate synthase has been crystallized in the absence of ligands and diffracts beyond 3.0 A. The protein was cloned and expressed in Escherichia coli and then crystallized from ammonium sulfate in the presence of /?-mercaptoethanol at a variety of pH values. The crystals are trigonal in the space-group P3,21; the unit cell dimensions are a = b = 967 A, r:=84.1 A. Keywords: thymidylate
synthase; crystallography;
1. Introduction
solution containing 6.5 mg protein/ml, buffer with 20 mix-j?-mercaptoethanol
$ Author to whom all correspondence should be addressed.
TS, thymidylate
synthasr.
161 (M)22--28:~6/91/1(W,1614)3 $03.0010
in Tris.HCl
and 1.5 Mammonium sulfate at pH 6.5 to 8-O. This is the first example of TS crystallized without either phosphate or substrate in the medium. Addition of phosphate to the suspensions cracked the crystals, suggesting a possible phosphate-induced conformational change. The space-group P3,21 was distinguished from its enantiomorph P3221 by the molecular replacement (C.A.S., unpublished results) solution with unit cell dimensions, determined by diffractometry as a = h = 96.7 A, c = 841 A. There is one monomer per
7 Present address: Depart,ment of Medicinal (lhemistry and Pharmacognosy, Purdue University, RHPH, West Lafayette, Ix 47907, U.S.A. used:
cancer
A knowledge of the structure of human TS may be important for rational design of chemotherapeutically useful inhibitors of this enzyme. First, it would provide information important in the design of inhibitors useful in the treatment of human neoplastic diseases. Second, it would serve as a needed template to compare the host TS with the enzyme from human pathogens; this will elucidate key differences between the host and micro-organism TS, and aid the design of inhibitors that selectively target the invading organism. It has not been feasible to obtain the amounts of human TS needed for crystallography from natural sources. Recently, human TS has been cloned and expressed in E. coli (Davisson et al., 1989). Here, we describe the cryst,allization and space-group determination of recombinant human TS. Human TS was cloned and expressed in E. coli and the enzyme was purified and assayed as described (Davisson et aZ., 1989). Crystals were grown in hanging drops at room temperature from a
Thymidylate synthase (TS$) catalyzes the con2’-deoxyuridylate version of (dUMP) and 5,10-methylenetetrahydrofolate to 2’-deoxythymidylate (dTMP) and dihydrofolate. Since TS catalyzes a step in the sole de nova pathway for dTMP synthesis, it is an important target for potential chemotherapeutic agents. There is a large body of knowledge on the structure and function of TS. Many details of the mechanism and inhibition have been elucidated (for a review, see Santi & Danenberg, 1984; Finer-Moore et al., 1990). The primary sequences of TS from 17 sources are known and their alignment shows that TS is the most highly conserved of all enzymes (Perryman et al., 1986; Perry et al., 1990). The X-ray structures of the Lactobacillus casei TS-Pi (Hardy et aZ., 1987) and TS-dUMP complexes have been determined and refined to 2.3 A (1 f! = 0.1 nm) resolution (J. S. Finer-Moore, unpublished results). The structure of a binary complex of the Escherichia coli enzyme with phosphate (Perry et al., 1990) and a ternary complex with dUMP and a cofactor analog, N’0-propargyl-5,8-dideazafolate (CB3717), have also been determined (Montfort et aZ., 1990).
5 Abbreviation
enzyme; protein structure;
0 1991 Academic, Press Limited
162
C. A. Nchifler
et al
Figure 1. A precession photograph showing the [O, I, - I] zone of diffraction shown extends to 3.6 A resolution and the crystals diffract to 2% A resolution.
asymmetric unit and the crystals diffract to beyond 2.8 A resolution. A precession photograph of the [0, 1, - l] zone is shown in Figure 1. Observations (36,811) were collected from two crystals and the reflections merged with the XENGEN package (Howard et ab., 1985). X-ray crystallographic intensities were recorded on a Nicolet IPC area detector, with a three-circle goniometer, using CuKa X-rays. The source was a 200 pm focus, Rigaku 200 RV rotating target tube with a graphite monochromator. Data frames were processed with the software of Blum et al. (1987). Intensities were reduced as described by Montfort et al. (1990). From the high primary sequence homology (Takeishi et al., 1985), it is expected that the structure of human TS will follow the same general fold as that found for the structures from L. casei and E. co& (Perry et al., 1990). The amino-terminal in any other TS sequence, without counterpart sequences, is 27 residues longer than L. casei TS and contains eight proline residues. Two eukaryoticspecific insertions at positions 117-128(lc9&101~), and 145-154(1c156-157) exist in the human TS sequence. The first insertion is near the active site? probably in a similar position as the small domain of L. casei TS. The second insertion is on the edge of the dimer interface. From a homology model based on the crystal structure of L. casei TS (C.A.S. and results), the human TS R.M.S., unpublished monomer has two cysteine residues that occur at t The numbering system used for amino acid residues of TS refers to the sequence of the human enzyme followed by the corresponding homologous sequences of the reference L. casei (lc) TS residues in parentheses.
from csrystnls of human TS. The pattern
positions 199(1c202) and 210(1c213). These residues are expected to line on adjacent p-strands close enough potentially to form disulfide bonds. Also: cysteine residues 18O(lc183) in each monomer are juxtaposed across the dimer interface in a manner that might allow for formation of a disulfide linkage. However, non-reducing SDS/polyacrylamide gel electrophoresis of dissolved crystals did not reveal evidence for a disulfide-linked dimer (data not shown). Refinement of the X-ray structure may elucidate whether a disulfide bond forms during crystallization. The proximity of two different pairs of sulfhydryl groups suggests the possibility of cross-linking under oxidizing conditions. The refinement of the crystal structure to the data will provide the first three-dimensional view of an unliganded mammalian TS and provide a template for anti-cancer drug design. C.A.S. thanks V. Ramalingam, William Montfort, Janet Finer-Moore, Kathy Perry and Chris Carrerras for their help and advice. This research was supported by the National Institutes of Health grants R01-CA41323 to J.S.F. and R.M.S., GM24485 to R.M.S. and CA14394 to D.V.S. V.J.D. was supported by the Damon RunyonWalter Winchell Cancer Research fund for a postdoctoral fellowship.
References Blum, M., Metcalf, P., Harrison, S. C. & Wiley. I). (1. (1987). J. Appl. Crystallogr. 20, 235-242. Bradford, M. M. (1976). And Biochem. 72, 248-254. Davisson, V. J., Siraworaporn, W. & Santi, D. V. (1989). J. Biol. Chem. 264, 9145-9148. Finer-Moore, J. S., Montfort, W. R. & Stroud, R. M. (1990). Riochemistry, 29. 6977-6986.
Communications Hardy, L. W., Finer-Moore, J. S., Montfort, W. R., Jones, M. O., Santi. D. V. & Stroud, R. M. (1987). Science, 235, 448-455. Howard, A. J., ?u’eilsen, C. & Xuong, ru’g, H. (1985). Methods Enzymol. 115, 252-270. Montfort, W. R., Perry, K. M., Fauman, E. B., FinerMoore, J. S., Maley, G. F., Maley, F. & Stroud, R. M. (1990). Biochemistry, 29, 6964-6977. Perry, K. M., Fauman, E. B., Finer-Moore, J. S., Montfort, W. R., Maley, C. F., Maley. F. & Stroud, R. M. (1990). Proteins, 8, 315-333.
163
Perryman, S. M., Rossana, C., Deng, T.. Vanin, E. F. & Johnson, L. F. (1986). Mol. Biol. Evol. 3, 313-321. Santi, D. V. & Danenberg, P. V. (1984). In Folates and Pterins, vol. 1, Chemistry and Biochemistry of Folates (Blakley, R. L. t Benkovic, S. J., eds), pp. 345-398, John Wiley and Sons, New York. Takeishi. K.. Saneda, S., Ayusawa, D., Shimizu, Ii., (iotoh, 0. & Seno. T. (1985). Nucl. Acids Res. 13, 203562043.
Edited by R. Huber
