Abbreviations: ADP, adenosine diphosphate; AK, arginine kinase; ATP, adenosine triphosphate; cDNA, complementary DNA; CK, creatine kinase; DNA, deoxyribunucleic acid; DTT, dithiothreitol; EDTA, ethylenediaminetetraacetic acid; HEPES, (N-[2-Hydroxyethyl]piperazine-N'-[2-ethansulfonic acid]); IEF, isoelectric focussing; MES, (2-[N-Morpholino]ethanesulfonic acid); Mi-CK, mitochondrial CK; PAGE; polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; PDS, 4,4'-dithiopyridine; PEG, polyethylene glycol; PMSF, phenylmethylsulfonyl fluoride; psi, pounds per square inch; SDS, sodium dodecyl sulfate; TSA, transition state analog; UV, ultraviolet.
This reaction and those of the rest of the guanidino (or phosphagen) kinase family, including creatine kinase (CK), enable ATP to be quickly regenerated from storage phosphagens during bursts of cellular activity. In addition to this "temporal" buffering, it has been proposed that these enzymes also function in "spatial" buffering, separating sites of cellular energy production from sites of use by a phosphagen shuttle with the guanidino kinases catalyzing opposite reactions at each end of the shuttle (Tombes & Shapiro, 1985; Wallimann et al., 1992).
AK is the only phosphagen kinase in arthropods as CK is in vertebrates (Morrison, 1973; Watts, 1973). Although the substrates are quite different, the enzymes are thought to share a common mechanism of direct, associative in-line gamma-phosphoryl transfer (Hansen & Knowles, 1981). In spite of the diversity of substrates, the guanidino kinases are among the most conserved proteins with the most different sharing ~40% amino acid identity (Babbitt et al., 1986; Dumas & Camonis, 1993; Mühlebach et al., 1994; Suzuki & Furukohri, 1994). Subunits have molecular weights of about 40 kDa. Some arginine kinases are monomers (including that of horseshoe crab), and others are dimers like other invertebrate phosphagen kinases, and like mammalian cytosolic CK. Mammalian mitochondrial CK (Mib-CK) is octameric. It is widely believed that arginine kinase is the modern enzyme closest to the primordial form, based on sequence analysis, existence as monomers, and use of a freely available amino acid substrate.
A wealth of mechanistic data has been accumulated over the last 40 years through "classical" enzymology and spectroscopic techniques (reviewed in Kenyon & Reed, 1983). A stable complex can be formed containing creatine (or arginine), Mg++, ADP, and the nitrate ion, which is thought to mimic the (missing) gamma-phosphoryl in a planar configuration during transfer (Milner-White & Watts, 1971). In the literature, this is referred to as a transition state analog (TSA) complex. When Mg++, ADP and creatine are bound, CK undergoes conformational changes as indicated by EPR (Reed & Cohn, 1972), tryptic susceptibility (Lui & Cunningham, 1966), and X-ray scattering (Forstner et al., 1996). There is conflicting evidence as to whether it is the chemical step that is rate limiting or some other step, perhaps the conformational change (Engelborghs et al., 1975; Rao et al., 1976).
After many reports, over many years, describing crystals diffracting to ~3 Å resolution, the first guanidino kinase structure was reported recently (Fritz-Wolf et al., 1996). The 3 Å structures of apo-Mi-CK with and without bound ATP revealed subunits with a 112-residue alpha-helical N-terminal domain followed by a 277-residue domain with an eight-stranded antiparallel beta-sheet flanked by seven helices. The active site is in the cleft between domains, but its details remain obscure, because of disorder and flexibility in the active site, and because the enzyme has been visualized before the substrate-induced conformational changes (Fritz-Wolf et al., 1996). The work described here is designed to complement the CK results with the development of an experimental system through which it will be possible to study the high resolution structure in its "active" state and perform mutational analysis.
Prior formation of the TSA complex had a marked effect upon crystallization. Tiny aggregates were found in a matrix screen (Jancarik & Kim, 1991) with various salts or polyethylene glycol( PEG)/isopropanol/Mg++ mixtures. No progress was made with salts, but microcrystals appeared 1-3% below the precipitation points of various PEGs >1000 D: 15% w/v for PEG 6000 to ~30% for PEG 1000. Better crystals came with high purity commercial PEGs (Fluka or Hampton Research Inc.), but not with additional PEG purification (Ray & Puvathingal, 1985; Jurnak, 1986). Crystals were obtained at pH 7.5 and with 10 to 20 mg/mL protein at either 4 or 20°C. The largest (0.3 * 0.1 * 0.05 mm3), obtained by microbatch crystallization (Chayen et al., 1992) with 12% PEG 6000 or 23% PEG 1000, diffracted X-rays to ~3 frac;1;2 Å, similar to that reported for CK crystals (McPherson, 1973; Schnyder et al., 1990, 1991).
Fig. 1. SDS polyacrylamide gel following the steps of AK purification
. The gel is heavily overloaded (to show impurities) and stained with Coomassie
blue. Lanes: (1) markers [molecular weights (kDa) are shown on left]; (2)
recovered inclusion bodies (repeated on right as lane 2b at lower contrast);
(3) washed inclusion bodies; (4) following S300 size exclusion chromatography
in denaturing conditions; (5) following S100 size exclusion chromatography
in native conditions.
Table 1. Yields and activities at stages in the purification of recombinant arginine kinase
The recombinant product was the purist enzyme with which we had worked. However, high-resolution isoelectric focusing (IEF) revealed two surprises (Fig. 2).
Fig. 2.Isoelectric focusing gel comparing AK isolated from natural
sources and recombinant AK. Lanes 1: markers, annotated with pI; 2 and
3: AK purified from horseshoe crab; 3: after treatment with alkaline phosphatase;
4 and 5: recombinant AK--note the different fine structure and the shift
consistent with the E73G PCR mutation; 5: after alkaline phosphatase treatment,
showing that one of the bands disappears; the remaining fine bands are
separated by about 0.05 pH units and are consistent with transamination
modifications (see Discussion).
First, the pI was shifted slightly consistent with a single non-conservative random E73G mutation that had occurred during PCR amplification (Strong & Ellington, 1996). Second, the recombinant enzyme showed IEF fine structure, albeit different from that isolated from natural sources. Testing the effects of acid and alkaline phosphatases and cysteine reagents, only one of the eight bands could be accounted for, and was attributed to phosphorylation. The other seven fine bands, separated by about 0.05 pH units, are of unknown origin (see Discussion). Apparently, this microheterogeneity can be tolerated in crystallization.
Fig. 3.Diffraction image of recombinant arginine kinase. The
image is split with different gray scales: left with high saturation to
show the low resolution diffraction; and right with low saturation to show
the weaker high resolution reflections. The edge of the background "water-ring"
is at ~3.8 Å, and the edge of the image plate is at 1.8 Å resolution.
Some reflections can be seen in the corners at up to 1.7 Å. The "Still"
image was taken with a 15-min exposure.
A data set 98% complete to 1.86 Å has been collected with the following statistics: Rmerge = 4.7%, I/sigma(I) = 17 (overall), 3.0 (at 1.86 Å), 232,917 observations of 35,102 unique reflections in P212121. Unit cell dimensions are a = 70.9 Å, b = 80.4 Å, c = 65.4 Å, alpha = beta = gamma = 90°. This gives a VM = 2.24 Å/Da3 (Matthews, 1968), with one subunit per asymmetric unit.
Examination of the recombinant enzyme by high-resolution IEF revealed surprising fine structure (Fig. 2). One band is the product of phosphorylation, but the other seven fine bands that are separated by about 0.05 pH units are of unknown origin. Although N-terminal modification cannot be ruled out, it is likely that the fine structure is similar to that found for other expressed phosphagen kinases (Wood et al., 1995) for which Fourier transform mass spectrometry shows isoforms differing in mass by 1 Dalton, consistent with transamination. The microheterogeneity has not been exhaustively examined, because it does not appear to interfere with the formation of high-resolution crystals.
Improvement of the crystals was a text book case. Although diffraction grade crystals of apo-AK (Berthou et al., 1975), cytosolic apo-CK (McPherson, 1973), and apo-Mi-CK (Schnyder et al., 1990, 1991) have previously been obtained from other sources of enzyme, it was necessary to form the TSA complex for horseshoe crab AK. Formation of the complex is likely stabilizing a flexible protein (Lui & Cunningham, 1966; Reed & Cohn, 1972; Forstner et al., 1996) into a single conformation. This was critical for horseshoe crab AK, as it has been with other enzymes (McPherson, 1982). Increase of crystal quality came with the increase of purity and the elimination of the inherent genetic variability of phosphagen kinases (Hershenson et al., 1986), possible with purification from an insoluble, denatured cloned source. Stabilization of the high-resolution diffraction, as with other proteins (Rodgers, 1994), came with cooling crystals to cryo-temperatures.
Although no direct evidence can be offered, instability of the crystals in the absence of the TSA components is strong circumstantial evidence that it is the complexed form that is crystallized. Arginine kinase is therefore likely to join a select handful of enzymes (Lolis & Petsko, 1990) for which high-resolution structures are available for a transition state analog complex. Attempts are underway to determine phases by isomorphous replacement, but it is also hoped that it will be possible to determine the structure by the molecular replacement method (Rossmann, 1972) using the recent Mi-CK structure (Fritz-Wolf et al., 1996), when the coordinates become available.
The transition state analog (TSA) complex (Milner-White & Watts, 1971) was formed by dialysis against MgCl2 (5 mM), ADP (4 mM), KNO3 (50 mM), and arginine (20 mM), conditions analogous to those that inhibit CK (Gross et al., 1994). A reducing agent (DTT, 1 mM) and an antibacterial (sodium azide; 0.05%) were added. Protein concentration was measured by Bradford assay (Biorad Inc.), calibrated against apo-AK (which could be measured by UV, in the absence of ADP). Crystallization trials started with hanging drop vapor diffusion (Ducruix & Giegé, 1992) and matrix screens (Jancarik & Kim, 1991). Crude optimization was by: (1) vapor diffusion/incomplete factorial trials (Carter, 1992) for combinations of PEG/isopropanol/Mg++; (2) binary search (Stewart & Khimasia, 1994) and grids using the microbatch method (Chayen et al., 1992) and 1-2 µL drops of concentrate protein (>20 mg/mL) with which buffer (various: pH 6.5-10) and crystallizing agent were mixed.
Following induction of expression by standard methods (Strong & Ellington, 1996), the E. coli cells were spun down at 6,000 * g for 15 min using a Beckman JA10 rotor, and resuspended in 40 mL lysis buffer (50 mM Tris pH 8.0, 7 mM DTT, 1 mM PMSF). The suspension was French pressed at 1500 psi with two passes, and then centrifuged to remove cell debris, using a Beckman JA12 rotor at 10,000 * g. The inclusion bodies were washed as follows to remove lipids, DNA, and some of the contaminating proteins: the pellet was alternately (re)suspended in 80 mL of washing buffer "W" (50 mM Tris pH 8.0; 2% Triton X-100; 10 mM EDTA; 1 M urea) and spun down at 5,000 * g in a Beckman JA20 rotor [1 M urea is sufficient to denature some proteins, but not AK, as determined by a difference absorption urea titration (Copeland, 1993), which showed a sigmoidal unfolding transition between 2 and 4.5 M, agreeing with data that was subsequently reported by others (France & Grossman, 1996).] On the final, fourth, cycle of washing, AK was resuspended in storage buffer, "S" (10 mM Tris·HCl pH 8.0; 1 mM EDTA; 1 mM DTT; 10 mM KCl, and 0.02% w/v NaN3). It was then unfolded with a one-hour incubation in 8 M urea (in buffer S) at room temperature. The solution was clarified by centrifugation at 10,000 rpm (8,000 * g) for one hour in JA20 rotor, and filtered through 0.22 µm membrane filter. AK was partially purified in the unfolded state by size exclusion chromatography (2.5 * 120 cm Sephacryl S-300 column) in 6 M urea/buffer S. [Crystals of urea sometimes formed with 8 M solutions, and 6 M is above the unfolding transition (see above).] AK was diluted to 0.25 mg/mL and refolded by sequential dialysis against 4 M, 2 M, 0.5 M, and 0 M urea in refolding buffer "R" (50 mM Tris·HCl pH 8.0; 1 mM EDTA; 100 mM beta-mercaptoethanol) with four-hour incubations at 4°C. Unsuccessful attempts were made to increase the yield of refolding by using temperature leaps during incubation (Xie & Wetlaufer, 1996). Purification following refolding was by anion exchange and size exclusion chromatographies. Overall, it makes little difference which is first, but explaining the low yield by mis-folded aggregates (see Discussion) suggests that size exclusion chromatography should be first, the opposite of all but our last preparations. Size exclusion chromatography was with a 2.5 * 120 cm Sephacryl S-100 column, gravity fed, and monitored with an in-line UV monitor. Anion exchange chromatography followed extensive dialysis and concentration by pressure ultrafiltration, and was with a 1.5 * 10 cm DEAE-sepharose CL-6B column, eluting with a 10 to 60 mM KCl gradient in a 10 mM Tris buffer pH 8.0 containing 1 mM EDTA, 1 mM DTT, and 0.02% NaN3. The product was concentrated by ultrafiltration starting with a large volume pressure system, and then down to ~1 mL by centrifugal filters. The TSA complex (Milner-White & Watts, 1971) was formed by dialysis against 5 mM MgCl2, 4 mM ADP, 50 mM KNO3, and 20 mM arginine using the membrane from a 10 kDa centrifugal microconcentrator. Activities were measured by the enzyme linked assay (see above). Concentrations prior to TSA complex formation were measured by UV absorption, as above, except that an extinction coefficient of 0.60 mL/cm/mg was assumed for unfolded AK.
The optimized protocol for crystallization uses a grid of 8 µL hanging drops with starting concentrations of 10 mg/mL AK, 53 mM MgCl2, 2 mM ADP, 25 mM KNO3, 10 mM arginine, 0.5 mM DTT, 2.5 mM sodium azide, 25 mM HEPES (pH 7.5), and 11, 12, or 13% w/v PEG 6000. The drops are equilibrated against a reservoirs of 15, 16, 17, 18, 19, or 20% PEG 6000 and incubated for 7 to 14 days at 4°C.
The diffraction data were collected from a crystal with dimensions of ~1.0 * 0.3 * 0.25 mm3 using 1.5418 Å X-rays from a Rigaku RU200 HB generator run at 40 kV * 100 mA with a graphite monochromator and a 0.3 mm collimator. Fifteen-minute exposures were collected with the R-Axis II imaging plate system 90 mm from the crystal, and processed with the programs BioTex (Molecular Structure Corp.), XDisplayF (Minor, 1993) and Denzo (Otwinowski, 1990).
Berthou J, Rérat C, Rérat B, Gadet A, Fourme R, Renaud M, Dubord C, Pradel L-A, Roustan C, van Thoai N. 1975. Preliminary X-ray diffraction studies on lobster ATP: L-arginine phosphotransferase. J Mol Biol 95:331 -333.
Blow DM, Chayen NE, Lloyd LF, Saridakis E. 1994. Control of nucleation of protein crystals. Protein Sci 3:1638- 1643.
Brzozowski AM, Tolley SP. 1994. Poly(ethylene) glycol monomethyl ethers--An alternative to poly(ethylene) glycols in protein crystallization. Fifth International Conference on Crystallization of Biological Macromolecules, San Diego, CA.
Burgess AN, Liddell JM, Cook W, Tweedle RM, Swan IDA. 1978. Creatine kinase. A new crystal form providing evidence of subunit structural homogeneity. J Mol Biol 123: 691-695.
Carter CW. 1992. Design of crystallization experiments and protocols. In: Ducruix A, Giegé R, eds. Crystallization of nucleic acids and proteins. Oxford: IRL Press.
Chayen NE, Stewart PDS, Blow DM. 1992. Microbatch crystallization under oil--A new technique allowing many small-volume crystallization trials. J Crystal Growth 122:176-180 .
Copeland RA. 1993. Methods for protein analysis. New York: Chapman & Hall.
Ducruix A, Giegé R. 1992. Methods of crystallization. In: Ducruix A, Giegé R, eds. Crystallization of nucleic acids and proteins. Oxford: IRL Press.
Dumas C, Camonis J. 1993. Cloning and sequence analysis of the cDNA for arginine kinase of lobster muscle. J Biol Chem 268:21599 -21606.
Engelborghs Y, Marsh A, Gutfreund H. 1975. A quenched-flow study of the reaction catalysed by creatine kinase. Biochem J 151:47- 50.
Forstner M, Kriechbaum M, Laggner MP, Wallimann T. 1996. Changes of creatine kinase structure upon ligand binding as seen by small-angle scattering. J Mol Struct 383:217.
France RM, Grossman SH. 1996. Denaturation and urea gradient gel electrophoresis of arginine kinase: Evidence for a collapsed-state conformation. Arch Biochem Biophys 326:93- 99.
Fritz-Wolf K, Schnyder T, Wallimann T, Kabsch W. 1996. Structure of mitochondrial creatine kinase. Nature 381:341- 345.
Gilliland GL, Sjölin L, Olsson G. 1983. Crystallization and preliminary X-ray diffraction data of two crystal forms of bovine heart creatine kinase. J Mol Biol 170:791- 793.
Gross M, Furter-Graves EM, Wallimann T, Eppenberger HM, Furter R. 1994. The Tryptophan residues of mitochondrial creatine kinase: Roles of Trp-233, Trp-206, and Trp-264 in active-site and quaternary structure formation. Protein Sci 3 :1058-1068.
Hansen DE, Knowles JR. 1981. The stereochemical course of the reaction catalyzed by creatine kinase. J Biol Chem 256:5967- 5969.
Hershenson S, Helmers N, Desmueles P, Stroud R. 1986. Purification and crystallization of creatine kinase from rabbit skeletal muscle. J Biol Chem 261: 3732-3736.
Jancarik J, Kim S-H. 1991. Sparse matrix sampling: A screening method for crystallization of proteins. J Appl Crystallogr 24: 409-11.
Jurnak F. 1986. Effect of chemical impurities in polyethylene glycol on macromolecular crystallization. J Crystal Growth 76: 577-582.
Kenyon GL, Reed GH. 1983. Creatine kinase: Structure-activity relationships. Adv Enzymol 54:367- 426.
Laemmli UK. 1970. Cleavage of structural proteins during assembly of the head bacteriophage T4. Nature 227:680- 685.
Lolis E, Petsko G. 1990. Transition-state analogues in protein crystallography: Probes of the structural source of enzyme catalysis. Annu Rev Biochem 59:597- 630.
Lui NST, Cunningham L. 1966. Cooperative effects of substrates and substrate analogs on the conformation of creatine phosphokinase. Biochemistry 5: 144-149.
Matthews BW. 1968. Solvent content of protein crystals. J Mol Biol 33: 491-497.
McPherson A. 1982. Preparation and analysis of protein crystals. Malabar, FL: Kreiger.
McPherson A, Koszelak S, Axelrod H, Day J, Williams R, Robinson L, McGrath M, Cascio D. 1986. An experiment regarding crystallization of soluble proteins in the presence of beta-octyl glucoside. J Biol Chem 261:1969- 1975.
McPherson A Jr. 1973. A preliminary crystallographic investigation of rabbit muscle creatine kinase. J Mol Biol 82: 83-86.
Milner-White EJ, Watts DC. 1971. Inhibition of adenosine 5'-triphosphate-creatine phosphotransferase by substrate-anion complexes--Evidence for the transition-state organization of the catalytic site. Biochem J 122:727- 740.
Minor W. 1993. XDisplayF Program. West Lafayette: Purdue University.
Morrison JF. 1973. Arginine kinase and other invertebrate guanidino kinases. In: Boyer PD, ed. The enzymes, Vol. 8.. New York: Academic Press. pp 457-486.
Mühlebach S, Gross M, Wirz T, Wallimann T, Perriard JC, Wyss M. 1994. Sequence homology and structure predictions of the creatine kinase isoenzymes. Mol Cell Biochem 133 :245- 262.
Otwinowski Z. 1990. Data collection and processing. In: Henrick K, Moss DS, Tickle IJ, eds. Proceedings of the CCP4 study weekend. Daresbury, England: Daresbury Laboratory. pp 73-82.
Patel S, Cudney B, McPherson A. 1995. Polymeric precipitants for the crystallization of macromolecules. Biochem Biophys Res Commun 207: 819-829.
Rao BDN, Buttlaire DH, Cohn M. 1976. 31P NMR studies of the arginine kinase reaction. J Biol Chem 251:6981-6986.
Ray WJ, Puvathingal J. 1985. A simple procedure for removing contaminating aldehydes and peroxides from aqueous solutions of polyethylene glycols and of nonionic detergents that are based on the polyoxyethylene linkage. Ann Biochem 146:307.
Reed GH, Cohn M. 1972. Structural changes induced by substrates and anions at the active site of creatine kinase. J Biol Chem 247: 3073-3081.
Rodgers DW. 1994. Cryocrystallography. Structure 2:1135- 1140.
Rossmann MG. 1972. The molecular replacement method. New York: Gordon and Breach.
Schnyder T, Sargent DF, Richmond TJ, Eppenberger HM, Wallimann T. 1990. Crystallization and preliminary X-ray analysis of two different forms of mitochondrial creatine kinase from chicken cardiac muscle. J Mol Biol 216:809- 812.
Schnyder T, Winkler H, Gross H, Eppenberger HM, Wallimann, T. 1991. Crystallization of mitochondrial creatine kinase: Growing of large protein crystals and electron microscopic investigation of microcrystals consisting of octamers. J Biol Chem 266:5318- 5322.
Stewart PDS, Khimasia M. 1994. Predispensed gradient matrices--A new rapid method of finding crystallization conditions. Acta Crystallogr D50: 441-442.
Strong S. 1995. Molecular studies of arginine kinase isoforms. Ph.D. Dissertation, Florida State University.
Strong SJ, Ellington WR. 1993. Horseshoe crab sperm contain a unique isoform of arginine kinase that is present in midpiece and flagellum. J Exp Zool 267:563- 571.
Strong SJ, Ellington WR. 1995. Isolation and sequence analysis of the gene for arginine kinase from the chelicerate arthropod, Limulus polyphemus: Insights into catalytically important residues. Biochim Biophys Acta 1246:197-200.
Strong SJ, Ellington WR. 1996. Expression of horseshoe crab arginine kinase in Escherichia coli and site directed mutations of the reactive cysteine peptide. Comp Biochem Physiol 113B: 809-816.
Suzuki T, Furukohri T. 1994. Evolution of phosphagen kinase: Primary structure of glycocyamine kinase and arginine kinase from invertebrates. J Mol Biol 237:353- 357.
Tombes RM, Shapiro BM. 1985. Metabolite channeling: A phosphorylcreatine shuttle to mediate high energy phosphate transport between sperm mitochondrion and tail. Cell 41:325- 344.
Wallimann T, Wyss M, Brdiczka D, Nicolay K, Eppenberger HM. 1992. Intracellular compartmentation, structure and function of creatine kinase isoenzymes: The "phospho-creatine circuit" for cellular energy homeostasis. Biochem J 281 :21-40 .
Watts DC. 1973. Creatine kinase (adenosine 5'-triphosphate-creatine phosphotransferase). In: Boyer PD, ed. The enzymes, Vol. 8. New Yo rk: Academic Press. pp 383 -455.
Wood TW, Chen LH, Kelleher NL, Little DP, Kenyon GL, McLafferty FW. 1995. Direct sequence data from heterogeneous creatine kinase (43 kDa) by high-resolution tandem mass spectrometry. Biochemistry 34:16251- 16254.
Xie Y, Wetlaufer DB. 1996. Control of aggregation
in protein refolding: The temperature-leap tactic. Protein Sci 5:
517-523.
