Proc. Natl. Acad. Sci. USA Vol. 92, pp. 3444-3448, April 1995 Biochemistry

Heterogeneity in molecular recognition by restriction endonucleases: Osmotic and hydrostatic pressure effects on BamHI, Pvu II, and EcoRV specificity (protein-DNA recognition/star activity/bound water) CLIFFORD R. ROBINSONt AND STEPHEN G. SLIGARt School of Chemical Sciences and Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL 61801 Communicated by Jiri Jonas, University of Illinois, Urbana, IL, December 5, 1994 (received for review August 18, 1994)

ABSTRACT The cleavage specificity of the Pvu II and and hydrostatic-pressure techniques to study the role of water BamHI restriction endonucleases is found to be dramatically in mediating site-specific interactions of the EcoRI endonu- reduced at elevated osmotic pressure. Relaxation in specificity clease with canonical and alternate DNA sequences (16, 18). of these otherwise highly accurate and specific enzymes, To investigate whether this nontraditional role for water that previously termed "star activity," is uniquely correlated with was discovered in EcoRI represents a common feature of osmotic pressure between 0 and 100 atmospheres. No other restriction enzyme systems, we have examined the Pvu II, colligative solvent property exhibits a uniform correlation BamHI, and EcoRV endonucleases. Along with EcoRI, they with star activity for all of the compounds tested. Application are the only members of the type II restriction endonuclease of hydrostatic pressure counteracts the effects of osmotic family to be characterized by high-resolution structural anal- pressure and restores the natural selectivity ofthe enzymes for ysis (4, 19-23). For these three enzymes, cleavage at alternate their canonical recognition sequences. These results indicate sites has been observed previously, either by addition of that water solvation plays an important role in the site- organic solvents to the reaction buffer or by changes in ionic specific recognition of DNA by many restriction enzymes. strength, pH, or divalent cation concentration or identity Osmotic pressure did not induce an analogous effect on the (24-28). Cleavage at alternate sites by these and other restric- specificity of the EcoRV endonuclease, implying that selective tion enzymes has been termed "star activity" by analogy with hydration effects do not participate in DNA recognition in this EcoRI star (EcoRI*) activity (29). system. Hydrostatic pressure was found to have little effect on Each subunit of the Pvu II endonuclease from Proteus the star activity induced by changes in ionic strength, pH, or vulgaris has 157 amino acids, and the homodimer recognizes divalent cation, suggesting that distinct mechanisms may exist symmetric CAGCTG duplex DNA sequences, cleaving both for these observed alterations in specificity. Recent evidence strands in the middle of the site to leave blunt ends (30). Pvu has indicated that BamHI and EcoRI share similar structural II* sites include all single base substitutions at the four internal motifs, while Pvu II and EcoRV belong to a different structural positions (CNGCTG, CANCTG, CAGNTG, and CAGCNG), family. Evidently, the use of hydration water to assist in as well as A or G at the first position and T or C at the sixth site-specific recognition is a motif neither limited to nor position (24). The dimeric BamHI endonuclease from Bacillus defined by structural families. amyloliquefaciens H consists of identical 213-amino acid resi- due subunits, and cleaves GGATCC duplex DNA sequences Type II restriction endonucleases, along with their companion between the two guanine bases on both strands, leaving methyltransferases, make up bacterial restriction/modification four-base 5' overhangs (31). Reported BamHI* sites include systems, which protect bacteria from foreign DNA. They GGAACC, GGCTCC, GAATCC, GGATCN, and comple- typically consist of symmetric homodimers, which recognize mentary sequences. The EcoRV endonuclease derives from symmetric DNA sequences of four to eight bases, and require the restriction/modification system in Escherichia coli. EcoRV only Mg2+ as a cofactor to cleave duplex DNA (1). Despite the is a dimer of identical subunits of 244 amino acids. It recog- vast number of type II enzymes identified to date (nearly 2400) nizes the symmetric GATATC duplex DNA sequences and no homology and little sequence similarity exist (2). Despite cleaves both strands in the middle of the site, leaving blunt ends this lack of homology, recent reports have identified significant (32). EcoRV has been reported to cut at alternate sites with structural similarities among these enzymes (3), and a new single base changes from the canonical site, including GT- classification of restriction enzymes has been proposed on the TATC, AATATC, GANATC, and complementary sequences basis of the location of the scissile bond (4). Recent structural (33, 34). evidence has led to appreciation of structural and functional In this study we test whether the participation of water is a roles for water in site-specific interactions in protein-DNA general phenomenon in restriction enzyme-DNA interactions, complexes (5-8). thereby seeking to advance a general understanding of themes The use of osmotic pressure represents a powerful method in protein-DNA recognition and to provide molecular expla- for identifying and studying the participation of hydration nations for the flawed molecular recognition events that are water in biomolecular systems (9). By examining the correla- manifested as restriction enzyme star activity. tion between functional or structural properties and osmotic pressure, solvent water has been shown to play key roles in biochemical processes, such as substrate binding (10), protein- MATERIALS AND METHODS protein interactions (11), allosteric effects and conformational For each enzyme, a standard buffer was defined in which changes (12), catalysis (13), protein stability (14), and ion cleavage was observed only at the canonical site. Sets of channel formation (15). We have previously utilized osmotic- Abbreviations: DMSO, dimethyl sulfoxide; atm, atmosphere. The publication costs of this article were defrayed in part by page charge tPresent address: Department of Biology, Massachusetts Institute of payment. This article must therefore be hereby marked "advertisement" in Technology, Cambridge, MA 02139 accordance with 18 U.S.C. §1734 solely to indicate this fact. ITo whom reprint requests should be addressed. 3444 Downloaded by guest on September 28, 2021 Biochemistry: Robinson and Sligar Proc. Nati Acad Sci. USA 92 (1995) 3445 reactions were performed in the presence of increasing con- the intensity of the bands produced by cleavage at star sites centrations of the various compounds (osmolytes) added to the divided by the intensity of all bands (canonical site fragments standard buffer. Enzyme reactions were initiated by addition plus star site fragments). A fraction star activity of 0.5 means of enzyme to the solution. Order of addition had no observable that half of the plasmids cut at the canonical sites were also cut effect. Enzymes were incubated with DNA containing canon- at a star site. In all cases in which fraction star activities were ical and alternate ("star") sites in total volumes of 20 p.l at 37°C determined, restriction mapping confirmed that cleavage re- for 4 h and then quenched by addition of EDTA to a final actions proceeded to completion at the canonical sites before concentration of 20 mM. any star sites were cut. Osmotic pressures were determined as Restriction enzymes were purchased from GIBCO/BRL. described previously (16) by using a UIC 70 vapor-phase Purified pUC18 DNA was prepared by using Promega Magic osmometer or by calculation from tabulated data (35). Vis- Mega-Preps, from 1 liter cultures of E. coli grown in medium cosities and dielectric constants were taken from tabulated containing 200 mg of ampicillin. Reaction mixtures contained data (35, 36). 1 unit/,ul of enzyme and 0.5 ,g/,ul of pUC18 DNA, unless otherwise noted. For EcoRV, enzymatic activity at the canon- ical GATATC site was investigated by using purified A DNA RESULTS obtained from GIBCO/BRL. EcoRV* activity was tested by The pUC18 plasmid contains two canonical Pvu II sites using linear or circular pUC18 or A DNA. All three forms of (CAGCTG), at positions 306 and 628. Cleavage at these sites DNA gave equivalent results. yields fragments of 2368 and 318 bp. The most readily cleaved The standard buffers which allowed canonical site cleavages Pvu II* site in pUC18 was CAGCGG at position 112, gener- were: Pvu II, 50 mM Tris HCl, pH 7.4/6 mM MgCl2/50 mM ating fragments of 2174, 318, and 194 bp. The presence of KCl/50 mM NaCl; BamHI, 50 mM Tris HCl, pH 8.0/2 mM osmolytes in the reaction buffer promoted cleavage at alter- 2-mercaptoethanol/10 mM MgCl2/100 mM NaCl; EcoRV, 50 nate sites in pUC18 yielding additional fragments. DMSO, mM Tris HCl, pH 7.5/100 mM NaCl/10 mM MgCl2/10 mM ethylene glycol, glycerol, ethanol, and 2-propanol promoted 2-mercaptoethanol/100 ,tg of bovine serum albumin per ml. Pvu II* activity. Star activity is tightly correlated with osmotic Glycerol, ethylene glycol, dimethyl sulfoxide (DMSO), etha- pressure (Fig. 1A) but correlated less well with dielectric nol, 2-propanol, sucrose, and dextrose were added to the constant (Fig. 2A) or relative viscosity (Fig. 2B). standard buffers to final concentrations of 0-6 M to induce Cleavage at the unique canonical BamHI site in pUC18 osmotic pressures from 0 to 150 atmospheres (atm; 1 atm = (position 429) linearizes the plasmid. There are 11 potential 101.3 kPa). BamHI* sites in pUC18. Addition of DMSO, glycerol, ethylene Reactions at elevated hydrostatic pressures ranging from 1 glycol, sucrose, or dextrose promoted BamHI* site cleavage, to 1600 atm were performed as described (16) in 400-,lI yielding a distinctive band pattern consisting of eight new DNA polypropylene tubes (Cole-Palmer) filled with 0.2 g of zirco- fragments. The fraction of star activity was quantitated from nium oxide beads (Biospec Products, Bartlesville, OK). Solu- these bands and was tightly correlated with osmotic pressure tions of enzyme and DNA were separated by a thin layer of (Fig. 1B) but not well correlated with dielectric constant (Fig. mineral oil (Sigma) until the samples reached the desired 2C) or relative viscosity (Fig. 2D). pressure and temperature, at which time the apparatus was No canonical EcoRV sites exist within pUC18, but the inverted to mix the solutions. Following a 4-h incubation at plasmid contains seven EcoRV* sites. Cleavage at these sites 37°C, the enzymes were inactivated by heating to 65°C and was not promoted by any of the compounds tested except EDTA was added to 20 mM. DMSO, regardless of whether circular or linear pUC18 was The determinations of canonical site activity as a function of used as the substrate. None of the compounds tested appeared hydrostatic pressure were performed by using the standard to inhibit EcoRV canonical site activity, as judged by analysis buffer for each enzyme. For the analysis of hydrostatic pres- of EcoRV cleavage of A DNA (data not shown). sure effects upon star activity induced by osmotic pressure in We sought to apply hydrostatic pressure as a probe of Pvu II and BamHI, glycerol was added to the standard buffer restriction enzyme specificity. First, the effect of elevated in concentrations from 0 to 4.8 M to induce osmotic pressures hydrostatic pressure upon the activity of the three enzymes at from 0 to 120 atm. The analysis of hydrostatic pressure effects their canonical sites was tested under standard buffer condi- upon star activity induced by changes in ionic strength, pH, or tions (Table 1). Pvu II and BamHI were unaffected by increas- divalent cation concentration or identity was performed in the ing the pressure from 0 to 750 atm, showed only small effects following buffers, with closed, circular pUC18 DNA as the at 1000 atm, and still retained most of their activity above 1500 substrate: Pvu II, 10 mM Tris-HCl, pH 8.5/15 mM MgCl2; atm. EcoRV was more sensitive to hydrostatic pressure, but BamHI, 20 mM Tris HCl, pH 8.8/2 mM MnCl2; EcoRV, 50 retained most of its activity up to 1000 atm. mM Tris-HCl, pH 8.5/10 mM MnCl2/10 mM 2-mercaptoetha- When hydrostatic pressure is applied to reactions with nol/100 ,tg of bovine serum albumin per ml. elevated osmotic pressure, a dramatic reversal of the osmotic Analysis of the reaction products was as described (16, 17). pressure effects upon specificity is observed for Pvu II and The fraction of star activity was defined for each reaction as BamHI (Fig. 3). Hydrostatic pressure applied from 1 to 500 1.0 I

FIG. 1. Effects of osmotic pres- sure generated by various sub- stances on the enzyme specificity ._ 0~ of Pvu II (A) or BamHI (B). 0, DMSO; *, ethylene glycol; O, glyc- erol; *, ethanol; X, 2-propanol; +, sucrose; and a, dextrose. Lines rep- resent the best fit by least squares analysis, with the correlation coef- Osmotic pressure, atm ficients indicated. Downloaded by guest on September 28, 2021 3446 Biochemistry: Robinson and Sligar Proc. NatL AcatL Sci. USA 92 (1995)

0.8

:E c) < 0.6 0

.!2cn - 0.4 V., v 0.2 0

Cu .2 1.2 - FIG. 2. Effects of viscosity (B 0 and D) and dielectric constant (A

._ and C) on enzyme specificity. (A Q 0.8 ,, 0.8 and B) Pvu II. (C and D) BamHI. 5: Cu 0, DMSO; *, ethylene glycol; O, 0 Cu- 0.6 Xu 0.6 glycerol; 0, ethanol; X, 2-propanol; +, sucrose; and A\, dextrose. Solid 0 0.4 0 0.4 lines represent best fits for each r cu compound. Dashed lines represent 0.2 : v 0.2 the best fit considering data from 07 0 all compounds as a single data set, 72 74 76 78 0 1 2 3 4 5 6 7 8 with the correlation coefficients in- Dielectric constant, E, Relative viscosity, 71/710 dicated.

atm inhibits and ultimately eliminates the Pvu II* and BamHI* and the enzymes or DNA. The simplest interpretation of these activities induced by osmotic pressure from 0 to 100 atm. results is that a population of bound waters, sequestered from Intriguingly, alterations in enzyme specificity induced by bulk solvent, is present in the complexes between the restric- other changes in buffer conditions are not sensitive to hydro- tion enzyme and canonical site DNA but is not present when static pressure. When star activity is induced by decreased the enzyme is associated with the star sites. Release of these ionic strength, increased pH, or changes in divalent cation waters to bulk solvent is promoted by elevated osmotic pres- concentration or identity, hydrostatic pressure up to 500 atm sure in the solution, and enhances the affinity or activity for the inhibits only a small amount (<15%) of the Pvu II* and enzyme at alternate sites. These data are consistent with the BamHI* activities. Above 500 atm no further inhibition is studies of Spolar and Record, who suggested that a coupling observed up to 1400 atm. ForEcoRV, hydrostatic pressures up exists between local folding, water release, and site-specific to 1200 atm did not measurably inhibit the star activity induced binding ofDNA in the EcoRI endonuclease and other protein- by increased pH (from 7.5 to 8.5), decreased ionic strength DNA complexes (39). As reported previously, an analogous (eliminating 100 mM NaCl), or substitution of Mn2+ for Mg2+. effect occurs with EcoRI (16, 17). No such effects were observed with the EcoRV endonuclease, suggesting that hy- DISCUSSION dration waters are not utilized in the same way in the EcoRV- DNA complex. In the cases of BamHI and Pvu II we observe a change in At elevated hydrostatic pressures, the influence of osmotic specificity upon the addition of osmolyte, resulting in cleavage agents on the specificity of Pvu II and BamHI is greatly at alternate, noncanonical sites. The extent of this cleavage is diminished. The natural selectivity of the enzymes observed tightly correlated with osmotic pressure generated by the under standard conditions is restored, even in the presence of addition of neutral solutes or cosolvents. All compounds that an osmotic pressure of 100 atm, by application of hydrostatic were effective at promoting cleavage at star sites induced a pressure (Fig. 3). One possibility is that the effects of hydro- fraction star activity uniformly proportional to the osmotic static pressures to 500 atm are due to a general inhibition of pressure (Fig. 1). The addition of these compounds to the the enzyme; however, overall activity of Pvu II and BamHI at reaction buffer changes other colligative solution properties, the canonical site was unchanged between 0 and 750 atm such as viscosity and dielectric constant, which have been (Table 1). Our interpretation is that the return of selectivity shown to play a role in biomolecular interactions (37, 38). An promoted by hydrostatic pressure is due to specific solvation interpretation of the results as being related to osmotic effects effects which alter the total volume of the system. Hydrostatic requires an analysis of these other properties. Viscosity and pressure has been shown in many cases to preferentially dielectric constant are poorly correlated with the observed hydrate biomacromolecules and complexes (40, 41). A simple fraction star activity over the concentration range tested (Fig. explanation for our results is that waters of solvation, released 2). The variety of neutral solutes and cosolvents tested effec- by elevated osmotic pressure, are restored by hydrostatic tively rules out a specific interaction between the compounds pressure, perhaps due to the participation of discrete waters in Table 1. Inactivation of restriction enzymes by site discrimination by Pvu II and BamHI. The inhibition of the hydrostatic pressure overall enzymatic activity at the canonical sites observed at hydrostatic pressures in excess of 1500 atm is more likely to Percent of ambient pressure activity at represent dissociation of the protein-DNA or dimeric protein- various pressures* protein complexes (or both), as has been observed previously Enzyme 750 atm 1000 atm 1600 atm (40, 42, 43). Strikingly, hydrostatic pressure has very different effects upon star activity that is induced by in ionic Pvu II changes 100 98 90 strength, pH, or divalent cations. This contrast between the BamHI 100 96 79 responses of the two types of star activity could indicate EcoRV 94 82 41 distinct origins for the changes in specificity promoted by the *Values have errors of approximately 7 percentage points. two methods. Downloaded by guest on September 28, 2021 Biochemistry: Robinson and Sligar Proc. NatL Acad Sci. USA 92 (1995) 3447

0.8 B 0.6 it, C13 0.4 FIG. 3. Effect of hydrostatic ._ pressure on altered specificity in- C ee 0.2 duced by osmotic pressure for Pvu II (A) or BamHI (B). 0, Ambient pressure; 0, 100 atm; *, 200 atm;

0 W. K, 300 atm; X, 400 atm; and A, 500 0 20 40 60 80 100 120 140 atm. Lines represent the best fit by Osmotic pressure, atm least squares analysis. We can speculate that for Pvu II and BamHI, elevated results presented in this communication indicate that the role osmotic pressures result in conformational changes accompa- of differential hydration in recognition is a widely used motif, nied by release of bound waters, which are reversed by neither limited to nor defined by the emerging subclasses of hydrostatic pressure. For the other class of changes which type-II restriction endonucleases, and one which is not readily induced star activity, the small change in Pvu II and BamHI detected by structural analysis. star activity observed with pressure could indicate that a While the existence of structurally or functionally significant limited conformation change is induced by the ionic/pH/ waters in these systems is clear, we can as yet only speculate on cation changes (and reversible by hydrostatic pressure) but that their exact locations and roles. One possibility is that bound most of the effects are electrostatic. waters are directly involved in contacts between the protein In addition to identifying a role for bound water in recog- and DNA, as observed in the tip repressor complex with nition by Pvu II and BamHI, the results presented here argue cognate DNA (6, 7). Although information on the presence of that fundamental differences exist in the mechanisms of site waters in the protein-DNA interface of the Pvu II and EcoRV discrimination among type II restriction enzymes. EcoRI has structures was not available at the time of this writing, two previously been shown to display the same behavior as Pvu II waters do mediate base-specific contacts in the EcoRI-DNA and BamHI (16, 17). For all three enzymes, increases in star complex (45). EcoRV binds to noncognate DNA with modest activity accompany increases in osmotic pressure, while hy- affinity in a relatively loose conformation believed to be drostatic pressure restores specificity. In contrast, EcoRV* necessary for facilitated diffusion of the enzyme along the activity was not induced by a simple application of osmotic DNA helix (32). When shifting from noncognate to cognate pressure. We conclude that a differential solvation is linked to DNA, the protein tightens its grip, completely surrounding the the recognition of DNA by EcoRI, BamHI, and Pvu II, but not DNA helix and making specific contacts in the major and EcoRV. Furthermore, the response of EcoRV to hydrostatic minor grooves. This transition results in an increase in the pressures is also different from the other three restriction buried surface area of >1800 A2, including 230 A2 from new enzymes. For Pvu II and BamHI, hydrostatic pressure not only protein-protein contacts (21). Since an increase in buried reverses the effects of osmotic pressure on specificity, it also surface area is often accompanied by release of bound water has a small but measurable impact on star activity induced by (10-12, 39, 46), elevated osmotic pressures could serve to changes in ionic strength, pH, or divalent cation. This obser- enhance the affinity of this complex. Pvu II, BamHI, and vation is also true for EcoRI (16, 18). By contrast, in the EcoRI bind with low affinity to noncognate DNA (2) and EcoRV system, hydrostatic pressure has no observable effect make specific contacts primarily in the major groove. For these upon ionic strength-, pH-, or cation-induced star activity, enzymes, recognition might be facilitated by formation of a supporting the hypothesis that fundamental differences exist relatively "loose" complex (47, 48) or one in which confor- between the mechanisms of molecular recognition in these mational heterogeneity is required for optimal binding (49). restriction enzyme-DNA complexes. Release of water from the complex could result in an altered, EcoRV differs substantially from other type II restriction tighter, configuration conferring a different specificity. The enzymes in that in the absence of the "catalytic cofactor" observation of discrete waters in the EcoRI-DNA interface Mg2+, it binds with equal affinity to cognate and noncognate (45) is consistent with this model. DNA. Binding specificity (and cleavage) is only achieved in the Another possibility is that hydrostatic and osmotic pressures presence of Mg2+ (or Mn2+ or other divalent cations; refs. 34 are influencing the conformational equilibrium of the two and 44). Other type II enzymes can discriminate cognate from monomer subunits of these dimeric enzymes and their com- noncognate sites with high selectivity (103-105) in the absence plexes with cognate and noncognate sequences. For example, of magnesium (2). Thus EcoRV may rely upon interaction with in many oligomeric proteins, pressure-induced dissociation the divalent cation to help direct its specificity, while the other leads to conformational drifting of the monomers (50, 51). enzymes utilize bound waters for a similar function. Upon reassociation, these proteins or enzymes can display Despite the lack of sequence homology among restriction subtly different characteristics and activities (52). enzymes, recent x-ray crystallographic studies of the EcoRI, Solvent molecules might serve to orient the two enzyme Pvu II, BamHI, and EcoRV endonucleases have indicated that monomers with respect to each other or direct the conforma- EcoRI and BamHI belong to the same structural class, while tion of individual enzyme subunits. This idea is supported by Pvu II and EcoRV appear to be members of a distinct the structural evidence. The Pvu II structure is similar to structural family (3, 4, 19-23). Another feature which may EcoRV only in its DNA-binding domain; the dimerization define these classes of enzymes is the cleavage position within domains of the two proteins are very different (20). The their recognition sequences. EcoRI and BamHI cut after the structure of BamHI is most similar to EcoRI in the dimeriza- first base in a six-base site, leaving four-base 5' overhangs, tion domain (22). while Pvu II and EcoRV cut both stands in the middle of a Water can also play a role in controlling the conformation six-base site, leaving blunt ends. Given the apparent correla- of DNA by affecting the susceptibility of a particular sequence tion between structure and functional properties, it might have to deformation, binding, or cleavage. DNA conformation is been expected that DNA-recognition motifs would be also extremely sensitive to solution conditions, and changes in preserved within these putative enzyme families. However, the solvation can have dramatic effects upon its deformability (53, Downloaded by guest on September 28, 2021 3448 Biochemistry: Robinson and Sligar Proc. Natl. Acad Sci. USA 92 (1995) 54). In the EcoRV-cognate DNA complex, unstacking of the 15. Zimmerberg, J. & Parsegian, V. A. (1986) Nature (London) 323, central 2 bp results in a sharp bend of the helix and a narrowing 36-38. of the major groove, while the minor groove is widened. This 16. Robinson, C. R. & Sligar, S. G. (1994) Biochemistry 33, 3787- by 3793. distortion of the B-form helix axis, coupled with binding 17. Robinson, C. R. & Sligar, S. G. (1993) J. Mol. Biol. 234, 302-306. EcoRV in the minor groove, may displace the minor groove 18. Robinson, C. R. & Sligar, S. G. (1995) Methods Enzymol., in hydration waters (55). Osmotic pressure would then serve to press. facilitate site-specific binding. In contrast, Pvu II and EcoRI do 19. Kim, Y., Grable, J. C., Love, R., Greene, P. J. & Rosenberg, J. M. not distort the DNA helix axis and may therefore preserve (1990) Science 249, 1307-1309. hydration of the minor groove. Release of these waters could 20. Athanasiadis, A., Vlassi, M., Kotsifaki, D., Tucker, P. A., Wilson, alter the preferred conformation of cognate and noncognate K. S. & Kokkinidis, M. (1994) Nat. Struct. Biol. 1, 469-475. DNA, altering enzyme specificity. 21. Winkler, F. K., Banner, D. W., Oefner, C., Tsernoglou, D., Brown, R. S., Heathman, S. P., Bryan, R. K., Martin, P. D., It has been proposed that the effect of cosolvents to enhance Petratos, K. & Wilson, K. S. (1993) EMBO J. 12, 1781-1795. cleavage at noncanonical sites simply results from increased 22. Newman, M., Strzelecka, T., Dorner, L. F., Schildkraut, I. & affinity of the enzymes for DNA at all positions due to global Aggarwal, A. K. (1994) Nature (London) 368, 660-664. dehydration or excluded volume effects (56). Three lines of 23. Cheng, X., Balendirian, K., Schildkraut, I. & Anderson, J. E. evidence argue against this explanation. (i) In both the Pvu II (1994) EMBO J. 13, 3927-3935. and BamHI systems reported here and in the EcoRI system 24. Nasri, M. & Thomas, D. (1987) NucleicAcids Res. 15, 7677-7687. reported previously (16), the rate of cleavage at the canonical 25. George, J. & Chirikjian, J. G. (1982) Proc. Natl. Acad. Sci. USA site decreases as osmotic pressure increases, while cleavage at 79, 2432-2436. 26. George, J., Blakesley, R. W. & Chirikjian, J. G. (1981) J. Biol. the star site increases. This shift in specificity would not be Chem. 255, 6521-6524. expected if the enzymes were simply displaying enhanced 27. Kolesnikov, V. A., Zinoviev, V. V., Yashina, L. N., Karginov, binding or cleavage at all sites. (ii) No decrease in activity is V. A., Baclanov, M. M. & Malygin, E. G. (1981) FEBS Lett. 132, observed at the canonical site under standard buffer condi- 101-104. tions when hydrostatic pressure is applied. This implies that the 28. Malyguine, E., Vannier, P. & Yot, P. (1980) Gene 8, 163-177. mechanism by which hydrostatic pressure reverses the effects 29. Polisky, B., Greene, P., Garfin, D. E., McCarthy, B. J., Goodman, of cosolvents does not involve dissociation of the complex. (iii) H. M. & Boyer, H. W. (1975) Proc. Natl. Acad. Sci. USA 72, Although under standard conditions EcoRV discriminates 3310-3314. 30. Gingeras, T. R., Greenough, L., Schildkraut, L. & Roberts, R. I. between canonical and alternate sites to a similar degree as the (1981) Nucleic Acids Res. 9, 4525-4536. other three enzymes, its specificity is unaffected by the addi- 31. Roberts, R. J., Wilson, G.A. & Young, F. E. (1977) Nature tion of neutral solutes or cosolvents. If these compounds were (London) 265, 82-84. simply lowering the dissociation constant at all sites by a few 32. Vipond, I. B. & Halford, S. E. (1993) Mol. Microbiol. 9, 225-231. orders of. magnitude, EcoRV would be expected to show an 33. Luke, P. A., McCallum, S. A. & Halford, S. E. (1987) in Gene enhanced level of star activity as well. Amplification and Analysis, ed. Chirikjian, J. G. (Elsevier, New York), Vol. 5, pp. 183-205. We thank Professor Gregorio Weber, Professor Richard Gumport, 34. Taylor, J. D. & Halford, S. E. (1989) Biochemistry 28, 6198-6207. Dr. Eric Fisher, Jason Rockhill, and Dr. Anne Skaja Robinson for 35. Wolf, A. V., Brown, M. G. & Prentiss, P. C. (1987) in CRC helpful discussions and advice; Professor Steven Zimmerman for the Handbook ofChemistry and Physics: 1986-1987, ed. Weast, R. T. use of his vapor pressure osmometer; and Ms. Aretta Weber for (CRC, Boca Raton, FL), pp. D227-D271. assistance in manuscript preparation. This work was supported by 36. Dean, J. A. (1985) in Lange's Handbook ofChemistry, ed. Lange, of and N. A. (McGraw-Hill, St. Louis), Vol. 13, pp. 10-97-10-115. National Institutes Health Grants GM31756 GM33775 (S.G.S.) 37. Dzingeleski, G. D. & Wolfenden, R. (1993) Biochemistry 32, and by National Institutes of Health National Research Service Award 9143-9147. in Molecular Biophysics Grant GM08276 (C.R.R.). 38. Beece, D., Eisenstein, L., Frauenfelder, H., Good, D., Marden, M. C., Reinisch, L., Reynolds, A. H., Sorensen, L. B. & Yue, 1. Fisher, E. W. & Gumport, R. I. (1995) in The Encyclopedia of K. T. (1980) Biochemistry 19, 5147-5157. Molecular Biology and Biotechnology: Fundamentals and Appli- 39. Spolar, R. S. & Record, M. T., Jr. (1994) Science 263, 777-784. cations, ed. Meyers, R. A. (VCH, Weinhein, Germany), in press. 40. Weber, G. (1993) J. Phys. Chem. 97, 7108-7115. 2. Bennet, S. P. & Halford, S. E. (1989) Curr. Top. Cell. Regul. 30, 41. Weber, G. & Drickamer, H. G. (1983) Quart. Rev. Biophys. 16, 57-104. 89-112. 3. Riddihough, G. (1994) Nature (London) 370, 78. 42. Silva, J. L. & Siliveira, C. F. (1993) Protein Sci. 2, 945-950. 4. Newman, M., Strzelecka, T., Dorner, L. F., Schildkraut, I. & 43. Peng, X., Jonas, J. & Silva, J. L. (1993) Proc. Natl. Acad. Sci. USA Aggarwal, A. K. (1994) Structure 2, 439-452. 90, 1776-1780. 5. Ha, J.-H., Capp, M. W., Hehenwalter, M. D., Baskerville, M. & 44. Vermote, C. L. M. & Halford, S. E. (1992) Biochemistry 31, Record, M. T., Jr. (1992) J. Mol. Biol. 228, 252-264. 6082-6089. 6. Lawson, C. L. & Carey, J. (1993) Nature (London) 366, 178-182. 45. Rosenberg, J. (1991) Curr. Opin. Struct. Biol. 1, 104-113. 7. Otwinowski, Z., Schevitz, R. W., Zhang, R.-G., Lawson, C. L., 46. Rand, R. P. (1992) Science 256, 618. Joachimiak, A., Marmorstein, R. Q., Luisi, B. F. & Sigler, P. B. 47. Douzou, P. (1994) Proc. Natl. Acad. Sci. USA 91, 1657-1661. (1988) Nature (London) 335, 321-329. 48. Hoffman, B. M. & Ratner, M. A. (1987) J. Am. Chem. Soc. 109, 8. Feng, J. A., Johnson, R. C. & Dickerson, R. E. (1994) Science 6237-6243. 263, 348-355. 49. Weber, G. (1992) Protein Interactions (Chapman & Hall, New 9. Parsegian, V. A., Rand, R. P., Fuller, N. L. & Rau, D. C. (1986) York), p. 293. Methods Enzymol. 127, 400-437. 50. King, L. & Weber, G. (1986) Biochemistry 25, 3632-3637. 10. Rand, R. P., Fuller, N. L., Butko, P., Francis, G. & Nicholls, P. 51. Ruan, K. & Weber, G. (1993) Biochemistry 32, 6295-6301. (1993) Biochemistry 32, 5925-5929. 52. Silva, J. L. & Weber, G. (1993) Annu. Rev. Phys. Chem. 44, 11. Kornblatt, J. A., Kornblatt, M. J., Hui Bon Hoa, G. & Mauk, 89-113. A. G. (1993) Biophys. J. 65, 1059-1065. 53. Kimura, T., Asai, T. & Imai, M. (1989) Mol. Gen. Genet. 219, 12. Colombo, M. F., Rau, D. C. & Parsegian, V. A. (1992) Science 69-78. 256, 655-659. 54. Terry, B. J. & Modrich, P. (1983) J. Biol. Chem. 253, 9820-9825. 13. Di Primo, C., Sligar, S. G., Hui Bon Hoa, G. & Douzou, P. (1992) 55. Liepinsh, E., Otting, G. & Wuthrich, K. (1992) NucleicAcids Res. FEBS Lett. 312, 252-254. 20, 6549-6557. 14. Santoro, M. M., Liu, Y., Khan, S. M. A., Hou, L.-X. & Bolen, 56. Nasri, M. & Thomas, D. (1986) Biomed. Biochim. Acta 45, D. W. (1992) Biochemistry 31, 5278-5283. 997-1005. Downloaded by guest on September 28, 2021