The energetic basis of specificity in the Eco RI endonuclease--DNA interaction.

High sequence selectivity in DNA-protein interactions was analyzed by measuring discrimination by Eco RI endonuclease between the recognition site GAATTC and systematically altered DNA sites. Base analogue substitutions that preserve the sequence-dependent conformational motif of the GAATTC site permit deletion of single sites of protein-base contact at a cost of +1 to +2 kcal/mol. However, the introduction of any one incorrect natural base pair costs +6 to +13 kcal/mol in transition state interaction energy, the resultant of the following interdependent factors: deletion of one or two hydrogen bonds between the protein and a purine base; unfavourable steric apposition between a group on the protein and an incorrectly placed functional group on a base; disruption of a pyrimidine contact with the protein; loss of some crucial interactions between protein and DNA phosphates; and an increased energetic cost of attaining the required DNA conformation in the transition state complex. Eco RI endonuclease thus achieves stringent discrimination by both "direct readout" (protein-base contracts) and "indirect readout" (protein-phosphate contacts and DNA conformation) of the DNA sequence.

Structural and thermodynamic strategies for site-specific DNA binding proteins.

BACKGROUND: Site-specific protein-DNA complexes vary greatly in structural properties and in the thermodynamic strategy for achieving an appropriate binding free energy. A better understanding of the structural and energetic engineering principles might lead to rational methods for modification or design of such proteins. RESULTS: A novel analysis of ten site-specific protein-DNA complexes reveals a striking correspondence between the degree of imposed DNA distortion and the thermodynamic parameters of each system. For complexes with relatively undistorted DNA, favorable enthalpy change drives unfavorable entropy change, whereas for complexes with highly distorted DNA, unfavorable DeltaH degrees is driven by favorable DeltaS degrees. We show for the first time that protein-DNA associations have isothermal enthalpy-entropy compensation, distinct from temperature-dependent compensation, so DeltaH degrees and DeltaS degrees do not vary independently. All complexes have favorable DeltaH degrees from direct protein-DNA recognition interactions and favorable DeltaS degrees from water release. Systems that strongly distort the DNA nevertheless have net unfavorable DeltaH degrees as the result of molecular strain, primarily associated with the base pair destacking. These systems have little coupled protein folding and the strained interface suffers less immobilization, so DeltaS degrees is net favorable. By contrast, systems with little DNA distortion have net favorable DeltaH degrees, which must be counterbalanced by net unfavorable DeltaS degrees, derived from loss of vibrational entropy (a result of isothermal enthalpy-entropy compensation) and from coupling between DNA binding and protein folding. CONCLUSIONS: Isothermal enthalpy-entropy compensation implies that a structurally optimal, unstrained fit is achieved only at the cost of entropically unfavorable immobilization, whereas an enthalpically weaker, strained interface entails smaller entropic penalties.

Specific binding by EcoRV endonuclease to its DNA recognition site GATATC.

Restriction endonuclease EcoRV has been reported to be unable to distinguish its specific DNA site, GATATC, from non-specific DNA sites in the absence of the catalytic cofactor Mg2+, and thus to exercise sequence specificity solely in the catalytic step. In contrast, we show here that under appropriate conditions of pH and salt concentration, specific complexes with oligonucleotides containing the GATATC site can be detected by either filter-binding or gel-retardation. Equilibrium binding constants (K(A)) are easily measured by both direct equilibrium and equilibrium-competition methods. The preference for "specific" over "non-specific" binding at pH 7 in the absence of divalent cations is about 1000-fold (per mole of oligonucleotide) or 12,000-fold (per mole of binding sites). Ethylation-interference footprinting shows that the "specific" complex includes strong contacts to the phosphate groups GpApTpApTC. Specific DNA binding is strongly pH-dependent, decreasing about 15-fold for each increase of one pH unit above pH 6, but non-specific binding is not; thus, binding specificity decreases with increasing pH. Gel retardation and filter-binding at pH < or = 7 yield essentially identical values of K(A) for specific-site binding, but at pH > 7 gel retardation significantly underestimates K(A). Specific-site binding is stimulated about 700-fold by Ca2+ (not a cofactor for cleavage), but with non-cleavable 3'-phosphorothiolate and 4'-thiodeoxyribose derivatives whose response to Ca2+ is similar to that of the parent oligonucleotide, Mg2+ stimulates binding only fourfold and twofold, respectively. Thus, binding specificity is not dramatically enhanced by Mg2+. Taking into account discrimination in binding and in the first-order rate constant for phosphodiester bond scission, the overall discrimination exercised against the incorrect site GTTATC is about 10(7)-fold. EcoRV endonuclease is thus not a "new paradigm" for site-specific interaction without binding specificity, but like other type II restriction endonucleases achieves sequence specificity by discriminating both in DNA binding and in catalysis.

The energetics of the interaction of BamHI endonuclease with its recognition site GGATCC.

The interaction of BamHI endonuclease with DNA has been studied crystallographically, but has not been characterized rigorously in solution. The enzyme binds in solution as a homodimer to its recognition site GGATCC. Only six base-pairs are directly recognized, but binding affinity (in the absence of the catalytic cofactor Mg(2+)) increases 5400-fold as oligonucleotide length increases from 10 to 14 bp. Binding is modulated by sequence context outside the recognition site, varying about 30-fold from the bes t (GTG or TAT) to the worst (CGG) flanking triplets. BamHI, EcoRI and EcoRV endonucleases all have different context preferences, suggesting that context affects binding by influencing the free energy levels of the complexes rather than that of the free DNA. Ethylation interference footprinting in the absence of divalent metal shows a localized and symmetrical pattern of phosphate contacts, with strong contacts at NpNpNpGGApTCC. In the presence of Mg(2+), first-order cleavage rate constants are identical in the two GGA half-sites, are the same for the two nicked intermediates and are unaffected by substrate length in the range 10-24 bp. DNA binding is strongly enhanced by mutations D94N, E111A or E113K, by binding of Ca(2+) at the active site, or by deletion of the scissile phosphate GpGATCC, indicating that a cluster of negative charges at the catalytic site contributes at least 3-4 kcal/mol of unfavorable binding free energy. This electrostatic repulsion destabilizes the enzyme-DNA complex and favors metal ion binding and progression to the transition state for cleavage.

Identification of a putative structural gene for cathepsin D in Caenorhabditis elegans.

Mutants of Caenorhabditis elegans having about 10% of wild-type activity of the aspartyl protease cathepsin D have been isolated by screening. Mutant homozygotes have normal growth rates and no obvious morphological or developmental abnormalities. The mutant gene (cad-1) has been mapped to the right extremity of linkage group II. Heterozygous animals (cad-1/+) show intermediate enzyme levels and animals heterozygous for chromosomal deficiencies of the right extremity of linkage group II have 50% of wild-type activity. Cathepsin D purified from a mutant strain has a lower activity per unit mass of pure enzyme. These data suggest that cad-1 is a structural gene for cathepsin D.

Two-fold symmetry of crystalline DNA-EcoRI endonuclease recognition complexes.

Recognition complexes between EcoRI endonuclease and either of two synthetic oligonucleotides (sequences CGCGAATTCGCG and TCGCGAATTCGCG) crystallize in Space Group P321 with unit cell parameters a = 128 and c = 47 A and a = 118.4 and c = 49.7 A, respectively. Native diffraction data to 3 A resolution have been collected from the form containing the tridecameric sequence. Electrophoretic analyses of dissolved crystals demonstrate that this form contains DNA and protein in a ratio of one double helix per enzyme dimer. The most likely asymmetric unit contents are one 31,000 dalton enzyme subunit and one strand of DNA, yielding VM values of 3.1 A3/dal and 2.8 A3/dal for the forms containing dodecameric and tridecameric DNA, respectively. This implies that the DNA-protein complex possesses two-fold rotational symmetry, which has been incorporated in the crystalline lattice.

Protein-DNA recognition complexes: conservation of structure and binding energy in the transition state.

This paper considers how enzymes that catalyze reactions at specific DNA sites have been engineered to overcome the problem of competitive inhibition by excess nonspecific binding sites on DNA. The formation of a specific protein-DNA recognition complex is discussed from both structural and thermodynamic perspectives, and contrasted with formation of nonspecific complexes. Evidence (from EcoRI and BamHI endonucleases) is presented that a wide variety of perturbations of the DNA substrate alter binding free energy but do not affect the free energy of activation for the chemical step; that is, many energetic factors contribute equally to the recognition complex and the transition-state complex. This implies that the specific recognition complex bears a close resemblance to the transition-state complex, such that very tight binding to the recognition site on the DNA substrate does not inhibit catalysis, but instead provides energy that is efficiently utilized along the path to the transition state. It is suggested that this view can be usefully extended to "noncatalytic" site-specific DNA-binding proteins like transcriptional activators and general transcription factors.

Control of protein synthesis in Escherichia coli: strain differences in control of translational initiation after energy source shift-down.

We have studied the parameters of protein synthesis in a number of Escherichia coli strains after a shift-down from glucose-minimal to succinate-minimal medium. One group of strains, including K-12(lambda) (ATCC 10798) and NF162, showed a postshift translational yield of 50 to 65% and a 2- to 2.5-fold increase in the functional lifetime of general messenger ribonucleic acid. There was no change in the lag time for beta-galactosidase induction in these strains after the shift-down. A second group, including W1 and W2, showed no reduction in translational yield, no change in the functional lifetime of messenger ribonucleic acid, and a 50% increase in the lag time for beta-galactosidase induction. Evidence is presented which indicates that this increased lag time is not the result of a decreased rate of polypeptide chain propagation. A third group of strains, including NF161, CP78, and NF859, showed an intermediate pattern: translational yield was reduced to about 75% of normal, and the messenger ribonucleic acid functional lifetime was increased by about 50%. Calculation of the relative postshift rates of translational initiation gave about 0.2, 1.0, and 0.5, respectively, for the three groups. There was no apparent correlation between the ability to control translation and the genotypes of these strains at the relA, relX, or spoT loci. Measurements of the induction lag for beta-galactosidase during short-term glucose starvation or after a down-shift induced by alpha-methylglucoside indicated that these regimens elicit responses that are physiologically distinct from those elicited by a glucose-to-succinate shift-down.

The enfolding arms of EcoRI endonuclease: role in DNA binding and cleavage.

The N-terminal segments of the EcoRI endonuclease dimer form part of mobile "arms" that encircle DNA in the recognition complex. By treating endonuclease-TCGCGAATTCGCG complexes with proteases, we have prepared a series of deletion derivatives lacking defined segments of the N-terminal region. The 5-12 segment is essential for DNA cleavage and forms one electrostatic interaction (per subunit) with DNA phosphate. These ionic contacts are directly across the double helix from the scissile phosphodiester bonds; they thus may permit the enfolding arms to immobilize DNA in apposition to the catalytic cleft and/or contribute to the unusual "kinked" conformation of DNA in the complex. Sequence specificity is fully retained when 28 residues are deleted from the N-terminus, but the complexes dissociate more rapidly.

The relationship between translational initiation and messenger RNA inactivation in down-shifted Escherichia coli.

The parameters of protein synthesis and functional inactivation of global messenger RNA (mRNA) were examined in a Tic+ strain of Escherichia coli during the 30-min period following a shift-down from glucose-minimal to succinate-minimal medium. The rate of mRNA inactivation and the relative translational initiation frequency were both most severely depressed immediately after the shift-down and increased slowly thereafter. If glucose was restored to the medium at any time after shift-down, mRNA inactivation immediately resumed its normal (preshift) rate and the protein-forming capacity was increased. These changes in mRNA inactivation rate do not reflect an altered mRNA composition in the down-shifted cells. The relative rate of mRNA inactivation was linearly proportional to the relative translational initiation frequency over a 10-fold range of initiation frequencies. Low initiation frequencies represent increased "dwell" of the ribosomes at the initiation site before the commencement of polypeptide chain initiation. We propose that initiating ribosomes protect mRNA from an inactivating endonucleolytic cleavage at or near the ribosome binding site.

"Ultraviolet footprinting" accurately maps sequence-specific contacts and DNA kinking in the EcoRI endonuclease-DNA complex.

The "UV footprinting" technique has been used to detect contacts between EcoRI endonuclease and its recognition sequence at single nucleotide resolution. Comparison of the UV-footprinting results to the published crystal structure of the EcoRI endonuclease-DNA complex allows us to determine how UV light detects protein-DNA contacts. We find that kinking of the DNA helix in the complex greatly enhances the UV photoreactivity of DNA at the site of the kink. In contrast to kinking, contacts between the endonuclease and the DNA bases inhibit the UV photoreactivity of DNA. Similar analysis of a proteolytically modified endonuclease that exhibits the same sequence specificity as wild-type enzyme but that does not cleave DNA supports these conclusions. Furthermore, detection of enhanced photoreactivity at the same kink in the modified enzyme-DNA complex allows us to conclude that the loss of cleavage activity by the modified endonuclease is not due to its failure to kink DNA.

Facilitated distortion of the DNA site enhances EcoRI endonuclease-DNA recognition.

We have measured the binding of EcoRI endonuclease to a complete set of purine-base analogue sites, each of which deletes one functional group that forms a hydrogen bond with the endonuclease in the canonical GAATTC complex. For five of six functional group deletions, the observed penalty in binding free energy is +1.3 to +1.7 kcal/mol. For two of these cases (replacement of adenine N7 with carbon) a single protein-base hydrogen bond is removed without deleting an interstrand Watson-Crick hydrogen bond or causing structural "adaptation" in the complex. This observation establishes that the incremental energetic contribution of one protein-base hydrogen bond is about -1.5 kcal/mol. By contrast, deletion of the N6-amino group of the inner adenine in the site improves binding by -1.0 kcal/mol because the penalty for deleting a protein-base hydrogen bond is outweighed by facilitation of the required DNA distortion ("kinking") in the complex. This result provides direct evidence that the energetic cost of distorting a DNA site can make an unfavorable contribution to protein-DNA binding.

Chiral phosphorothioates as probes of protein interactions with individual DNA phosphoryl oxygens: essential interactions of EcoRI endonuclease with the phosphate at pGAATTC.

The contact between EcoRI endonuclease and the "primary clamp" phosphate of its recognition site pGAATTC is absolutely required for recognition of the canonical and all variant DNA sites. We have probed this contact using oligonucleotides containing single stereospecific (Rp)- or (Sp)- phosphorothioates (Ps). At the GAApTTC position, where the endonuclease interacts with only one phosphoryl oxygen at the central DNA kink, Rp-Ps inhibits and Sp-Ps stimulates binding and cleavage [Lesser et al. (1992) J. Biol. Chem. 267, 24810-24818]: in contrast, at the pGAATTC position both diastereomers inhibit binding. For single-strand substitution, the penalty in binding free energy (delta delta G0bind) is slightly greater for Sp-Ps (+ 0.9 kcal/mol) than for Rp-Ps (+ 0.7 kcal/mol). Binding penalties are approximately additive for double-strand substitution (Rp,Rp-Ps or Sp,Sp-Ps). Neither Ps diastereomer in one DNA strand affects the first-order rate constants for cleavage in the unmodified DNA strand, and only Sp-Ps inhibits the cleavage rate constant (3-fold) in the modified DNA strand. Thus, the second-order cleavage rate (including binding and catalysis) is inhibited 14-fold by Sp-Ps and 45-fold by Sp,Sp-Ps. In the canonical complex, the phosphate at pGAATTC is completely surrounded by protein and each nonbridging phosphoryl oxygen receives two hydrogen bonds from the endonuclease, such that in either orientation the increased bond length of P-S- inhibits binding. However, the pro-Sp oxygen interacts with residues that are connected (by proximity or inter-side-chain hydrogen bonding) to side chains with essential roles in catalysis, so cleavage is preferentially inhibited when these side chains are slightly displaced by the Sp-Ps diastereomer.

Coordinate ion pair formation between EcoRI endonuclease and DNA.

The free energy of the binding reaction between EcoRI restriction endonuclease and a specific cognate dodecadeoxynucleotide (d(CGCGAATTCGCG)) has contributions from both electrostatic and nonelectrostatic components. These contributions were dissected by measuring the effects of varying salt concentration on the equilibrium binding constant and applying the thermodynamic analyses of Record et al. (Record, M. T., Jr., Lohman, T. M., and deHaseth, P. L. (1976) J. Mol. Biol. 107, 145-158). Endonuclease mutation S187 (Arg 187 to Ser) (Greene, P. J., Gupta, M., Boyer, H. W., Brown, W. E., and Rosenberg, J. M. (1981) J. Biol. Chem. 256, 2143-2153) did not significantly affect the nonelectrostatic component but did perturb the electrostatic contribution to the binding energy (we are numbering the amino acid residues according to the DNA sequence). The former was determined by extrapolating the linear portion of the salt dependence curve (0.125 to 0.25 M KCl) to 1 M ionic strength, with the same result for both wild type and S187 endonucleases at both pH 6.0 and 7.4 (-8.5 +/- 1.5 kcal/mol or greater than 50% of the total binding free energy). The slopes of these same curves yield estimates of eight ionic interactions between wild type endonuclease and the DNA at both pH values. By contrast, binding of EcoRI-S187 to dodecanucleotide involves six charge-charge interactions at pH 6.0. Only two ionic interactions are observed at pH 7.4. This was unexpected since gel permeation chromatography demonstrated that the recognition complex for both wild type and S187 proteins contains an enzyme dimer and a DNA duplex. EcoRI-S187 endonuclease retains wild type DNA sequence specificity, and the rate of the phosphodiester hydrolysis step is also unchanged. Thus, electrostatic interactions are functionally separable from sequence recognition and strand cleavage. Our results also establish that arginine 187 plays a key role in the electrostatic function and suggest that it might be located at the DNA-protein interface. The disproportionate loss of ion pairs at pH 7.4 can be rationalized by a model which suggests that six conformationally mobile ionic groups on the protein act in a coordinated manner during the interaction with DNA.

Stereoselective interaction with chiral phosphorothioates at the central DNA kink of the EcoRI endonuclease-GAATTC complex.

We have probed the contacts between EcoRI endonuclease and the central phosphate of its recognition site GAApTTC, using synthetic oligonucleotides containing single stereospecific Rp- or Sp-phosphorothioates (Ps). These substitutions produce subtle stereospecific effects on EcoRI endonuclease binding and cleavage. An Sp-Ps substitution in one strand of the DNA duplex improves binding free energy by -1.5 kcal/mol, whereas the Rp-Ps substitution has an unfavorable effect (+0.3 kcal/mol) on binding free energy. These effects derive principally from changes in the first order rate constants for dissociation of the enzyme-DNA complexes. The first order rate constants for strand scission are also affected, in that a strand containing Sp-Ps substitution is cleaved 2 to 3 times more rapidly than a strand containing a normal prochiral phosphate, whereas a strand containing Rp-Ps substitution is cleaved about 3 times slower than normal. As a result, single-strand substitutions produce pronounced asymmetry in the rates of cleavage of the two DNA strands, and this effect is exaggerated in an Rp,Sp-heteroduplex. Ethylation-interference footprinting indicates that none of the Ps substitutions cause any major change in contacts between endonuclease and DNA phosphates. When an Sp-Ps localizes P = O in the DNA major groove, a hydrogen-bonding interaction with the backbone amide-NH of Gly116 of the endonuclease is improved relative to that with a prochiral phosphate having intermediate P-O bond order and delocalized charge.

N-termini of EcoRI restriction endonuclease dimer are in close proximity on the protein surface.

The N-terminal region of EcoRI endonuclease is essential for cleavage yet is invisible in the 2.5 A crystal structure of endonuclease-DNA complex [Kim, Y., Grable, J. C., Love, R., Greene, P. J., Rosenberg, J. M. (1990) Science 249, 1307-1309]. We used site-directed fluorescence spectroscopy and chemical cross-linking to locate the N-terminal region and assess its flexibility in the absence and presence of DNA substrate. The second amino acid in each subunit of the homodimer was replaced with cysteine and labeled with pyrene or reacted with bifunctional cross-linkers. The broad absorption spectra and characteristic excimer emission bands of pyrene-labeled muteins indicated stacking of the two pyrene rings in the homodimer. Proximity of N-terminal cysteines was confirmed by disulfide bond formation and chemical cross-linking. The dynamics of the N-terminal region were determined from time-resolved emission anisotropy measurements. The anisotropy decay had two components: a fast component with rotational correlation time of 0.3-3 ns representing probe internal motions and a slow component with 50-100 ns correlation time representing overall tumbling of the protein conjugate. We conclude that the N-termini are close together at the dimer interface with limited flexibility. Binding of Mg2+ cofactor or DNA substrate did not affect the location or flexibility of the N-terminal region as sensed by pyrene fluorescence and cross-linking, indicating that substrate binding is not accompanied by folding or unfolding of the N-terminus.

Solution conformation of EcoRI restriction endonuclease changes upon binding of cognate DNA and Mg2+ cofactor.

EcoRI endonuclease has two tryptophans at positions 104 and 246 on the protein surface. A single tryptophan mutant containing Trp246 and a single cysteine labeling site at the N-terminus was used to determine the position of the N-terminus in the protein structure. The N-termini of EcoRI endonuclease are essential for tight binding and catalysis yet are not resolved in any of the crystal structures. Resonance energy transfer was used to measure the distance from Trp246 donor to IAEDANS or MIANS acceptors at Cys3. The distance is 36 A in apoenzyme, decreasing to 26 A in the DNA complex. Molecular modeling suggests that the N-termini are located at the dimer interface formed by the loops comprising residues 221-232. Protein conformational changes upon binding of cognate DNA and cofactor Mg(2+) were monitored by tryptophan fluorescence of the single tryptophan mutant and wild-type endonuclease. The fluorescence decay of Trp246 is a triple exponential with lifetimes of 7, 3.5, and 0.7 ns. The decay-associated spectra of the 7- and 3.5-ns components have emission maxima at approximately 345 and approximately 338 nm in apoenzyme, which shift to approximately 340 and approximately 348 nm in the DNA complex. The fluorescence quantum yield of the single tryptophan mutant drops 30% in the DNA complex, as compared to 10% for wild-type endonuclease. Fluorescence changes of Trp104 upon binding of DNA were inferred by comparison of the decay-associated spectra of wild type and single tryptophan mutant. Fluorescence changes are related to changes in proximity and orientation of quenching functional groups in the tryptophan microenvironments, as seen in the crystal structures.

Proteases of the nematode Caenorhabditis elegans.

Crude homogenates of the soil nematode Caenorhabditis elegans exhibit strong proteolytic activity at acid pH. Several kinds of enzyme account for much of this activity: cathepsin D, a carboxyl protease which is inhibited by pepstatin and optimally active toward hemoglobin at pH 3; at least two isoelectrically distinct thiol proteases (cathepsins Ce1 and Ce2) which are inhibited by leupeptin and optimally active toward Z-Phe-Arg-7-amino-4-methylcoumarin amide at pH 5; and a thiol-independent leupeptin-insensitive protease (cathepsin Ce3) with optimal activity toward casein at pH 5.5. Cathepsin D is quantitatively most significant for digestion of macromolecular substrates in vitro, since proteolysis is inhibited greater than 95% by pepstatin. Cathepsin D and the leupeptin-sensitive proteases act synergistically, but the relative contribution of the leupeptin-sensitive proteases depends upon the protein substrate.