Characterization of a DNA binding protein of bacteriophage PRD1 involved in DNA replication.

Escherichia coli phage PRD1 protein P12, involved in PRD1 DNA replication in vivo, has been highly purified from E. coli cells harbouring a gene XII-containing plasmid. Protein P12 binds to single-stranded DNA as shown by gel retardation assays and nuclease protection experiments. Binding of protein P12 to single-stranded DNA increases about 14% the contour length of the DNA as revealed by electron microscopy. Binding to single-stranded DNA seems to be cooperative, and it is not sequence specific. Protein P12 also binds to double-stranded DNA although with an affinity 10 times lower than to single-stranded DNA. Using the in vitro phage phi 29 DNA replication system, it is shown that protein P12 stimulates the overall phi 29 DNA replication.     

Oligonucleotide synthesis by Escherichia coli dnaG primase in conjunction with phage P22 gene 12 protein

The phage P22 gene 12 protein was found to be like the Escherichia coli dnaB protein in that it stimulated phiX174 DNA synthesis in heat-inactivated extracts of dnaB temperature-sensitive cells (see preceding paper, Wickner, S. (1984) J. Biol. Chem. 259, 14038-14043). phiX174 replication catalyzed by the purified P22 12 protein also by-passed the normal requirement for dnaC protein. However, synthesis still required dnaG primase and the DNA polymerase III holoenzyme components. This DNA synthesis reaction has been reconstituted with purified proteins and found to require P22 12 protein, dnaG protein, DNA polymerase III holoenzyme components, 4 dNTPs, Mg2+, any one of ATP, GTP, UTP, or CTP and single-stranded DNA. The reaction has been dissected into partial reactions: (a) in a prepriming reaction, P22 12 protein binds to single-stranded DNA in an ATP-dependent reaction (Wickner, S. (1984) J. Biol. Chem. 259, 14038-14043); (b) in a priming reaction requiring at least one rNTP and the other dNTPs or rNTPs, dnaG primase catalyzes oligonucleotide synthesis dependent on the P22 12 protein-DNA complex; (c) finally, DNA polymerase III holoenzyme components catalyze DNA elongation of the primer.     

The rep mutation. VI. Purification and properties of the Escherichia coli rep protein, DNA helicase III.

The protein product of the rep gene of Escherichia coli is required for the replication of certain bacteriophage genomes (phi X174, fd, P2) and for the normal replication of E. coli DNA. We have used a specialized transducing phage, lambda p rep+, which complements the defect of rep mutants, to identify the rep protein. The rep protein has been purified from cells infected with lambda p rep+ phage; it has a molecular weight of about 70 000 and appears similar to the protein found in normal cells. Stimulation of phi X174 replicative form DNA synthesis in vitro was observed when highly purified rep protein was supplied to a cell extract derived from phi X-infected E. coli rep cells and supplemented with replicative form DNA. The purified protein has a single-stranded DNA-dependent ATPase activity and is capable of sensitizing duplex DNA to nucleases specific for single-stranded DNA. For this reason we propose the enzyme be called DNA helicase III. We infer that the rep protein uses the energy of hydrolysis of ATP to separate the strands of duplex DNA; the E. coli DNA binding protein need not be present. The rep3 mutant appeared to make a limited amount of active rep protein.     

ATP utilization by rep protein in the catalytic separation of DNA strands at a replicating fork.

Hydrolysis of ATP by rep protein proceeds in the presence of a single-stranded region of DNA 4 residues long, but the true effector for rep ATPase appears to be a replicating fork rather than a random coil. At or near a fork in duplex DNA, rep ATPase action is different from what it is on DNA lacking secondary structure (single-stranded): (i) Km for ATP is lower, (ii) specificity is for ATP and dATP with no action on other nucleoside triphosphates, (iii) sensitivity to certain ATP analogs is reduced, (iv) presence of a DNA-nicking enzyme (e.g. cistron A protein induced by phiX174) is required, and (v) Escherichia coli DNA binding protein facilitates rather than inhibits. During the separation of strands accompanying replication, 2 molecules of nucleoside triphosphate (ATP or dATP) are hydrolyzed for every nucleotide polymerized. Utilization of ATP by rep protein may provide energy for catalytic strand separation at a fork in advance of replication.     

The Autographa californica nuclear polyhedrosis virus p143 gene encodes a DNA helicase

The P143 protein of Autographa californica nuclear polyhedrosis virus is essential for replication of viral DNA. To determine the function of P143, the protein was purified to near homogeneity from recombinant baculovirus-infected cells that overexpress P143. ATPase activity copurified with P143 protein during purification and also during gel filtration at a high salt concentration. The ATPase activity did not require the presence of single-stranded DNA, but was stimulated fourfold by the addition of single-stranded DNA. The ATPase activity of P143 had a K(m) of 60 microM and a turnover of 4.5 molecules of ATP hydrolyzed/s/molecule of enzyme, indicating moderate affinity for ATP and high catalytic efficiency. P143 unwound a 40-nucleotide primer in an ATP-dependent manner, indicating that the enzyme possesses in vitro DNA helicase activity. Based on this result, it seems likely that P143 functions as a helicase in viral DNA replication.     

Primase, the dnaG protein of Escherichia coli. An enzyme which starts DNA chains.

Conversion of the viral DNA of phage G4 to the duplex form provided an opportunity to isolate and determine the function of the dnaG protein, the product of a gene known to be essential for replication of the Escherichia coli chromosome. This stage of G4 DNA replication requires action of three proteins: the E. coli DNA-binding protein, the dnaG protein, and the DNA polymerase III holoenzyme. The dnaG protein has been purified approximately 25,000-fold to near-homogeneity. The native protein contains a single polypeptide of 60,000 daltons. It has been assayed for its activity on G4 DNA in three ways: (a) RNA synthesis, (b) complementation for replication of an extract of a temperature-sensitive dnaG mutant, and (c) priming of DNA replication by DNA polymerase III holoenzyme. The dnaG protein is highly specific for G4 DNA and synthesizes a unique 29-residue RNA primer to be described in the suceeding paper. Other single-stranded and duplex DNA templates are inactive. RNA primer synthesis by the dnaG protein has an apparent Km for ribonucleoside triphosphates near 10 micrometer, and a narrow optimum for Mg2+. The sharp specificity of the dnaG protein in choice of template and the utilization of either deoxyribonucleotides or ribonucleotides to produce a hybrid piece only a few residues long (as described in a succeeding paper) suggests that the dnaG protein previously named RNA polymerase by renamed primase.     

Single-stranded-DNA-dependent ATPase from HeLa cells that stimulates DNA polymerase alpha-primase activity: purification and characterization of the ATPase

A single-stranded DNA-dependent ATPase that cofractionates during the early stages of purification of a multiprotein DNA polymerase alpha complex from HeLa cells has been purified to homogeneity. The ATPase is part of a 16S multienzyme DNA polymerase alpha complex that is fully active in SV40 DNA replication in vitro. The ATPase hydrolyzes ATP to ADP in a reaction that is completely dependent on the presence of DNA. DNA in single-stranded form is strongly preferred as a cofactor, and polydeoxynucleotides with adenine or thymidine residues are highly effective. Glycerol gradient sedimentation showed that the purified ATPase sedimented at an s20,w of 7 S, and polyacrylamide gel electrophoresis under denaturing conditions reveals two polypeptides with relative molecular weights of 83,000 and 68,000. Both of these polypeptides have purine nucleotide binding sites as revealed by photoaffinity cross-linking experiments. ATP binds to the two subunits more efficiently than GTP, and CTP or UTP does not cross-link with the two polypeptides. DNA synthesis catalyzed by purified HeLa cell DNA polymerase alpha-primase is stimulated in the presence of ATPase and ATP at an optimum concentration of 2 mM. Analysis of the DNA product by gel electrophoresis indicates that with poly(dT) but not phage M13 DNA as template the ATPase overcomes a lag and decreases the length of nascent DNA chains synthesized by the DNA polymerase alpha-primase complex.     

T4 phage gene uvsX product catalyzes homologous DNA pairing.

Gene uvsX of phage T4 controls genetic recombination and the repair of DNA damage. We have recently purified the gene product, and here describe its properties. The protein has a single-stranded DNA-dependent ATPase activity. It binds efficiently to single- and double-stranded DNAs at 0 degrees C in a cooperative manner. At 30 degree C the double-stranded DNA-protein complex was stable, but the single-stranded DNA-protein complex dissociated rapidly. The instability of the latter complex was reduced by ATP. The protein renatured heat-denatured double-stranded DNA, and assimilated linear single-stranded DNA into homologous superhelical duplexes to produce D-loops. The reaction is stimulated by gene 32 protein when the uvsX protein is limiting. With linear double-stranded DNA and homologous, circular single-stranded DNA, the protein catalyzed single-strand displacement in the 5' to 3' direction with the cooperation of gene 32 protein. All reactions required Mg2+, and all except DNA binding required ATP. We conclude that the uvsX protein is directly involved in strand exchange and is analogous to the recA protein of Escherichia coli. The differences between the uvsX protein and the recA protein, and the role of gene 32 protein in single-strand assimilation and single-strand displacement are briefly discussed.     

Characterization of the phage phi 29 protein p5 as a single-stranded DNA binding protein. Function in phi 29 DNA-protein p3 replication.

The phage phi 29 protein p5, required in vivo in the elongation step of phi 29 DNA replication, was highly purified from Escherichia coli cells harbouring a gene 5-containing plasmid and from phi 29-infected Bacillus subtilis. The protein was characterized as the gene 5 product by amino acid analysis and NH2-terminal sequence determination. The purified protein p5 was shown to bind to single-stranded DNA and to protect it against nuclease degradation. No effect of protein p5 was observed either on the formation of the p3-dAMP initiation complex or on the rate of elongation. However, protein p5 greatly stimulated phi 29 DNA-protein p3 replication at incubation times where the replication in the absence of p5 leveled off.     

Homologous pairing in genetic recombination. Purification and characterization of Escherichia coli recA protein

RecA protein, which is essential for genetic recombination in Escherichia coli, was extensively purified from a strain of E. coli which contained the recA gene cloned in a plasmid (Sancar, A., and Rupp, W. D. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 3144-3148). Using the DNA-dependent ATPase activity of recA protein as an assay, we obtained about 60 mg of purified recA protein from 100 g of cells. Ten micrograms or 1 microgram of the purified protein exhibited only one detectable band with Mr approximately = 40,000 upon sodium dodecyl sulfate-acrylamide gel electrophoresis. More than 99% of the ATPase activity of purified recA protein was dependent on single-stranded DNA. Purified recA protein had no detectable DNase, topoisomerase, or ligase activities. The enzyme was stable for a least a year when stored at 0-4 degrees C. The half-life of the ATPase activity of 25 microM recA protein was 37 min at 51 degrees C. Purified recA protein binds to single-stranded and double-stranded DNA, unwinds duplex DNA by a mechanism that is stimulated by single-stranded DNA or oligonucleotides, and pairs homologous single strands with duplex DNA.     

Overexpression, purification, and characterization of Escherichia coli bacteriophage PRD1 DNA polymerase. In vitro synthesis of full-length PRD1 DNA with purified proteins.

The bacteriophage PRD1 DNA polymerase gene (gene I) has been cloned into the expression vector pPLH101 under the control of the lambda pL promoter. Tailoring of an efficient ribosome binding site in front of the gene by polymerase chain reaction led to a high level heat-inducible expression of the corresponding gene product (P1) in Escherichia coli cells. Expression was confirmed in vivo by complementation of phage PRD1 DNA polymerase gene mutants and in vitro by formation of the genome terminal protein P8-dGMP replication initiation complex. Expressed PRD1 DNA polymerase was purified to apparent homogeneity in an active form. DNA polymerase, 3'-5'-exonuclease, and P8-dGMP replication initiation complex formation activities cosedimented in glycerol gradient with a protein of 65 kDa, the size expected for PRD1 DNA polymerase. The DNA polymerase was active on DNase I-activated calf thymus DNA, poly(dA).oligo(dT) and poly(dA-dT) primer/templates as well as on native phage PRD1 genome. The 3'-5'-exonuclease activity was specific for single-stranded DNA and released mononucleotides. No 5'-3'-exonuclease activity was detected. The inhibitor/activator spectrum of the PRD1 DNA polymerase was also studied. An in vitro replication system with purified components for bacteriophage PRD1 was established. Formation of the P8-dGMP replication initiation complex was a prerequisite for phage DNA replication, which proceeded from the initiation complex and yielded genome length replication products.     

Defective replication activity of a dominant-lethal dnaB gene product from Escherichia coli.

dnaB protein of Escherichia coli is an essential replication protein. A missense mutant has been obtained which results in replacement of an arginine residue with cysteine at position 231 of the protein (P. Shrimankar, L. Shortle, and R. Maurer, unpublished data). This mutant displays a dominant-lethal phenotype in strains that are heterodiploid for dnaB. Biochemical analysis of the altered form of dnaB protein revealed that it was inactive in replication in several purified enzyme systems which involve specific and nonspecific primer formation on single-stranded DNAs, and in replication of plasmids containing the E. coli chromosomal origin. Inactivity in replication appeared to be due to its inability to bind to single-stranded DNA. The altered dnaB protein was inhibitory to the activity of wild type dnaB protein in replication by sequestering dnaC protein which is also required for replication. By contrast, it was not inhibitory to dnaB protein in priming of single-stranded DNA by primase in the absence of single-stranded DNA binding protein. Sequestering of dnaC protein into inactive complexes may relate to the dominant-lethal phenotype of this dnaB mutant.     

Isolation and characterization of a DNA primase from human mitochondria

A family of enzymatic activities isolated from human mitochondria is capable of initiating DNA replication on single-stranded templates. The principal enzymes include at least a primase and DNA polymerase gamma and require that rNTPs as well as dNTPs be present in the reaction mixture. Poly(dC) and poly(dT), as well as M13 phage DNA, are excellent templates for the primase activity. A single-stranded DNA containing the cloned origin of mitochondrial light-strand synthesis can be a more efficient template than M13 phage DNA alone. Primase and DNA polymerase activities were separated from each other by sedimentation in a glycerol density gradient. Using M13 phage DNA as template, these mitochondrial enzymes synthesize RNA primers that are 9 to 12 nucleotides in size and are covalently linked to nascent DNA. The formation of primers appears to be the rate-limiting step in the replication process. Replication of M13 DNA is sensitive to N-ethylmaleimide and dideoxynucleoside triphosphates, but insensitive to rifampicin, alpha-amanitin, and aphidicolin.     

Replication factor C from the hyperthermophilic archaeon Pyrococcus abyssi does not need ATP hydrolysis for clamp-loading and contains a functionally conserved RFC PCNA-binding domain

The molecular organization of the replication complex in archaea is similar to that in eukaryotes. Only two proteins homologous to subunits of eukaryotic replication factor C (RFC) have been detected in Pyrococcus abyssi (Pab). The genes encoding these two proteins are arranged in tandem. We cloned these two genes and co-expressed the corresponding recombinant proteins in Escherichia coli. Two inteins present in the gene encoding the small subunit (PabRFC-small) were removed during cloning. The recombinant protein complex was purified by anion-exchange and hydroxyapatite chromatography. Also, the PabRFC-small subunit could be purified, while the large subunit (PabRFC-large) alone was completely insoluble. The highly purified PabRFC complex possessed an ATPase activity, which was not enhanced by DNA. The Pab proliferating cell nuclear antigen (PCNA) activated the PabRFC complex in a DNA-dependent manner, but the PabRFC-small ATPase activity was neither DNA-dependent nor PCNA-dependent. The PabRFC complex was able to stimulate PabPCNA-dependent DNA synthesis by the Pabfamily D heterodimeric DNA polymerase. Finally, (i) the PabRFC-large fraction cross-reacted with anti-human-RFC PCNA-binding domain antibody, corroborating the conservation of the protein sequence, (ii) the human PCNA stimulated the PabRFC complex ATPase activity in a DNA-dependent way and (iii) the PabRFC complex could load human PCNA onto primed single-stranded circular DNA, suggesting that the PCNA-binding domain of RFC has been functionally conserved during evolution. In addition, ATP hydrolysis was not required either for DNA polymerase stimulation or PCNA-loading in vitro.     

Cysteine 111 affects coupling of single-stranded DNA binding to ATP hydrolysis in the herpes simplex virus type-1 origin-binding protein

Herpes simplex virus type-1 origin-binding protein (UL9 protein) initiates viral replication by unwinding the origins. It possesses sequence-specific DNA-binding activity, single-stranded DNA-binding activity, DNA helicase activity, and ATPase activity that is strongly stimulated by single-stranded DNA. We have examined the role of cysteines in its action as a DNA helicase. The DNA helicase and DNA-dependent ATPase activities of UL9 protein were stimulated by reducing agent and specifically inactivated by the sulfhydryl-specific reagent N-ethylmaleimide. To identify the cysteine responsible for this phenomenon, a conserved cysteine in the vicinity of the ATP-binding site (cysteine 111) was mutagenized to alanine. UL9C111A protein exhibits defects in its DNA helicase and DNA-dependent ATPase activities and was unable to support origin-specific DNA replication in vivo. A kinetic analysis indicates that these defects are due to the inability of single-stranded DNA to induce high affinity ATP binding in UL9C111A protein. The DNA-dependent ATPase activity of UL9C111A protein is resistant to N-ethylmaleimide, while its DNA helicase activity remains sensitive. Accordingly, sensitivity of UL9 protein to N-ethylmaleimide is due to at least two cysteines. Cysteine 111 is involved in coupling single-stranded DNA binding to ATP-binding and subsequent hydrolysis, while a second cysteine is involved in coupling ATP hydrolysis to DNA unwinding.     

Bacteriophage T4 gene 41 protein, required for the synthesis of RNA primers, is also a DNA helicase

Bacteriophage T4 gene 41 protein is one of the two phage proteins previously shown to be required for the synthesis of the pentaribonucleotide primers which initiate the synthesis of new chains in the T4 DNA replication system. We now show that a DNA helicase activity which can unwind short fragments annealed to complementary single-stranded DNA copurifies with the gene 41 priming protein. T4 gene 41 is essential for both the priming and helicase activities, since both are absent after infection by T4 phage with an amber mutation in gene 41. A complete gene 41 product is also required for two other activities previously found in purified preparations of the priming activity: a single-stranded DNA-dependent GTPase (ATPase) and an activity which stimulates strand displacement synthesis catalyzed by T4 DNA polymerase, the T4 gene 44/62 and 45 polymerase accessory proteins, and the T4 gene 32 helix-destabilizing protein (five-protein reaction). The 41 protein helicase requires a single-stranded DNA region adjoining the duplex region and begins unwinding at the 3' terminus of the fragment. There is a sigmoidal dependence on both nucleotide (rGTP, rATP) and protein concentration for this reaction. 41 Protein helicase activity is stimulated by our purest preparation of the T4 gene 61 priming protein, and by the T4 gene 44/62 and 45 polymerase accessory proteins. The direction of unwinding is consistent with the idea that 41 protein facilitates DNA synthesis on duplex templates by destabilizing the helix as it moves 5' to 3' on the displaced strand.     

TWINKLE Has 5' -> 3' DNA helicase activity and is specifically stimulated by mitochondrial single-stranded DNA-binding protein

Mutations in TWINKLE cause autosomal dominant progressive external ophthalmoplegia, a human disorder associated with multiple deletions in the mitochondrial DNA. TWINKLE displays primary sequence similarity to the phage T7 gene 4 primase-helicase, but no specific enzyme activity has been assigned to the protein. We have purified recombinant TWINKLE to near homogeneity and demonstrate here that TWINKLE is a DNA helicase with 5' to 3' directionality and distinct substrate requirements. The protein needs a stretch of 10 nucleotides of single-stranded DNA on the 5'-side of the duplex to unwind duplex DNA. In addition, helicase activity is not observed unless a short single-stranded 3'-tail is present. The helicase activity has an absolute requirement for hydrolysis of a nucleoside 5'-triphosphate, with UTP being the optimal substrate. DNA unwinding by TWINKLE is specifically stimulated by the mitochondrial single-stranded DNA-binding protein. Our enzymatic characterization strongly supports the notion that TWINKLE is the helicase at the mitochondrial DNA replication fork and provides evidence for a close relationship of the DNA replication machinery in bacteriophages and mammalian mitochondria.     

Functional aspects of Escherichia coli rep helicase in unwinding and replication of DNA

The gene for Escherichia coli rep helicase (rep protein) was subcloned in a pBR plasmid and the protein overproduced in cells transformed with the hybrid DNA. The effect of purified enzyme on strand unwinding and DNA replication was investigated by electron microscopy. The templates used were partial duplexes of viral DNA from bacteriophage fd::Tn5 and reannealed DNA from bacteriophage Mu. The experiments with the two DNA species show DNA unwinding uncoupled from replication. The single-stranded phage fd::Tn5 DNA with the inverted repeat of transposon Tn5 could be completely replicated in the presence of the E. coli enzymes rep helicase, DNA binding protein I, RNA polymerase and DNA polymerase III holoenzyme. A block in the unwinding step increases secondary initiation events in single-stranded parts of the template, as DNA polymerase III holoenzyme cannot switch across the stem structure of the transposon.     

Protease Ti, a new ATP-dependent protease in Escherichia coli, contains protein-activated ATPase and proteolytic functions in distinct subunits

In addition to protease La (the lon gene product), Escherichia coli contains another ATP-dependent protease, Ti. This enzyme (approximately 340 kDa) is composed of two components, both of which are required for proteolysis. Both have been purified to homogeneity by conventional procedures using [3H]casein as the substrate. The ATP-stabilized component, A, has a subunit molecular weight of 80,000 upon gel electrophoresis in the presence of sodium dodecyl sulfate, but it behaves as a dimer (140 kDa) upon gel filtration. Component P, which is relatively heat stable, is inactivated by diisopropyl fluorophosphate and can be labeled with [3H] diisopropyl fluorophosphate. It has a subunit size of 23 kDa, but the isolated component behaves as a complex (260 kDa) of 10-12 subunits. The isoelectric point of component A is 7.0 and that of P is 8.2, and their amino acid compositions differ considerably. The purified enzyme has an ATPase activity that is stimulated 2-4-fold by casein and other protein substrates but not by nonhydrolyzed proteins. Component A also shows ATPase activity which can be stimulated by casein. Addition of component P (which lacks ATPase activity) inhibits basal ATP hydrolysis by A and makes this ATPase more responsive to casein. Although component P contains the serine active site for proteolysis, it shows no proteolytic activity in the absence of component A, Mg2+, and ATP or dATP. Other nucleoside triphosphates are not hydrolyzed and do not support proteolysis. Protease Ti has a Km for ATP of 210 microM for hydrolysis of both casein and ATP. Casein increases the Vmax for ATP without affecting the Km. A Mg2+ concentration of 5 mM is necessary for half-maximal rates of ATP and casein hydrolysis. Ca2+ and Mn2+ partially support these activities. Thus, protease Ti shares many unusual properties with protease La (e.g. coupled ATP and protein hydrolysis and protein-activated ATPase), but these functions in protease Ti are associated with distinct subunits that modify each other's activities.     

DNA-dependent adenosinetriphosphatase C1 from mouse FM3A cells has DNA helicase activity.

In our previous study, we identified four chromatographically distinct DNA-dependent ATPases, B, C1, C2, and C3, in mouse FM3A cells (Tawaragi, Y., Enomoto, T., Watanabe, Y., Hanaoka, F., and Yamada, M. (1984) Biochemistry 23, 529-533). The DNA-dependent ATPase C1 has been purified and characterized in detail. A divalent cation and a polynucleotide cofactor were required for the ATPase activity. Poly(dT), single-stranded circular DNA, and heat-denatured DNA were very effective. Almost no ATPase activity was observed with S1 nuclease-treated native DNA. ATPase C1 hydrolyzed ATP only among the ribo- and deoxyribonucleoside triphosphates tested, and this fact distinguished ATPase C1 from ATPases B, C2, and C3, because the latter enzymes are capable of hydrolyzing both ATP and dATP. The purified DNA-dependent ATPase C1 fraction was shown to have a DNA helicase activity that was dependent on hydrolysis of ATP. The helicase activity and DNA-dependent ATPase activity cosedimented at 5.2 S on glycerol gradient centrifugation. Both activities showed similar preferences for nucleoside 5'-triphosphates and similar requirements for divalent cations. The DNA helicase activity was inhibited by the addition of single-stranded DNAs that served as cofactor for the ATPase activity. The efficiency of a single-stranded DNA to inhibit DNA helicase activity correlated well with the capacity of the DNA to serve as cofactor for DNA-dependent ATPase activity. The helicase was shown to migrate along the DNA strand in the 5' to 3' direction, which is the same direction of migration of the mouse DNA helicase B (Seki, M., Enomoto, T., Yanagisawa, J., Hanaoka, F., and Ui, M. (1988) Biochemistry 27, 1766-1771).