Human RNase H1 activity is regulated by a unique redox switch formed between adjacent cysteines.

Human RNase H1 is active only under reduced conditions. Oxidation as well as N-ethylmaleimide (NEM) treatment of human RNase H1 ablates the cleavage activity. The oxidized and NEM alkylated forms of human RNase H1 exhibited binding affinities for the heteroduplex substrate comparable with the reduced form of the enzyme. Mutants of human RNase H1 in which the cysteines were either deleted or substituted with alanine exhibited cleavage rates comparable with the reduced form of the enzyme, suggesting that the cysteine residues were not required for catalysis. The cysteine residues responsible for the observed redox-dependent activity of human RNase H1 were determined by site-directed mutagenesis to involve Cys(147) and Cys(148). The redox states of the Cys(147) and Cys(148) residues were determined by digesting the reduced, oxidized, and NEM-treated forms of human RNase H1 with trypsin and analyzing the cysteine containing tryptic fragments by micro high performance liquid chromatography-electrospray ionization-Fourier transform ion cyclotron mass spectrometry. The tryptic fragment Asp(131)-Arg(153) containing Cys(147) and Cys(148) was identified. The mass spectra for the Asp(131)-Arg(153) peptides from the oxidized and reduced forms of human RNase H1 in the presence and absence of NEM showed peptide masses consistent with the formation of a disulfide bond between Cys(147) and Cys(148). These data show that the formation of a disulfide bond between adjacent Cys(147) and Cys(148) residues results in an inactive enzyme conformation and provides further insights into the interaction between human RNase H1 and the heteroduplex substrate.

RNase H hydrolyzes RNA in RNA-DNA hybrids (1). RNase H activity appears to be ubiquitous in eukaryotes and bacteria (2)(3)(4)(5)(6)(7). Although RNases H constitute a family of proteins of varying molecular weight, the nucleolytic activity and substrate requirements appear to be similar for the various isotypes. For example, all RNases H studied to date function as endonucleases exhibiting limited sequence specificity and requiring divalent cations (e.g. Mg 2ϩ , Mn 2ϩ ) to produce cleavage products with 5Ј phosphate and 3Ј-hydroxyl termini (8).
Two classes of RNase H enzymes have been identified in mammalian cells (5,9,10). These enzymes were shown to differ with respect to co-factor requirements and were shown to be inhibited by sulfhydryl reagents (10,11). Although the biolog-ical roles of the mammalian enzymes are not fully understood, it has been suggested that mammalian RNase H1 may be involved in replication and that the RNase H2 enzyme may be involved in transcription (12,13).
Recently, two human RNase H genes have been cloned and expressed (11,14,15). RNase H1 is a 286-amino acid protein and is expressed ubiquitously in human cells and tissues (11). The amino acid sequence of human RNase H1 displays strong homology with RNase H1 from yeast, chicken, Escherichia coli, and mouse (11). The human RNase H2 enzyme is a 299-amino acid protein with a calculated mass of 33.4 kDa and has also been shown to be ubiquitously expressed in human cells and tissues (14). 1 Human RNase H2 shares strong amino acid sequence homology with RNase H2 from Caenorhabditis elegans, yeast, and E. coli (14).
The properties of the cloned and expressed human RNase H1 have recently been characterized and many of the properties observed for human RNase H1 are consistent with the E. coli RNase H1 isotype, (e.g. the co-factor requirements, substrate specificity and binding specificity) (16,17). In fact, the carboxyl-terminal portion of human RNase H1 is highly conserved with the amino acid sequence of the E. coli enzyme. The glutamic acid and two aspartic acid residues of the catalytic site, as well as the histidine and aspartic acid residues of the proposed second divalent cation binding site of the E. coli enzyme are conserved in human RNase H1 (18 -21). In addition, the lysine residues within the highly basic ␣-helical substratebinding region of E. coli RNase H1 are also conserved in the human enzyme. Site-directed mutagenesis of the catalytic amino acids and the basic residues of the substrate-binding domain of human RNase H1 showed that these conserved residues are required for activity (22).
Despite these similarities, the structures of the two enzymes differ in a number of important properties. For example, the amino acid sequence of human RNase H1 is ϳ2-fold longer than the E. coli enzyme. The human enzyme contains a 73 amino acid region homologous with the RNA-binding domain of yeast RNase H1 at the amino terminus of the protein, which is separated from the conserved E. coli RNase H1 region by a 62-amino acid spacer region (22)(23)(24). Mutants in which the RNA-binding domain and spacer region of human RNase H1 were deleted showed that the RNA-binding domain was not required for RNase H activity and that this region was responsible for the observed positional preference for cleavage displayed by the enzyme as well as the enhanced binding affinity of the enzyme for various polynucleotides (22). The spacer region, on the other hand, was required for RNase H activity as the deletion of this region resulted in a significant reduction in both k cat and K m for the enzyme.
One biochemical property that has been used to classify RNase H enzymes is the sensitivity to sulfhydryl alkylating reagents such as N-ethylmaleimide (NEM) 2 (10,11,25,26). In general, RNase H1 enzymes are inhibited by NEM and both the E. coli and human enzymes share this property. In the case of E. coli RNase H1, NEM was shown to alkylate all three cysteine residues of the enzyme, although it was determined that alkylation of Cys 13 and Cys 133 was responsible for the observed loss in enzymatic activity (26). Furthermore, site-directed mutagenesis of the cysteine residues of E. coli RNase H1 showed that these residues were not required for endonuclease activity. Finally, the E. coli enzyme was shown to be active under both reduced and oxidized conditions (26). These results suggest that the cysteines are not involved in catalysis but are positioned such that alkylation of the cysteines sterically interferes with substrate binding.
Comparison of the amino acid sequence of Human RNase H1 with the E. coli enzyme indicates that of the five cysteine residues found in human RNase H1, only Cys 148 is conserved (Table I). In fact, this cysteine residue is highly conserved in both prokaryotic and eukaryotic RNases H1. Human RNase H1 contains an additional cysteine adjacent to Cys 148 , and this residue is conserved in RNases H1 from vertebrates (Table I).
In this study we have explored the role of the cysteine residues of Human RNase H1 with respect to the function of the enzyme. We have determined the optimum redox state for the protein as well as the effect of the redox state on the binding and catalytic properties of the enzyme. Finally, we have identified a unique redox switch formed by vicinal cysteine residues.

MATERIALS AND METHODS
Preparation of Human RNase H1-Human RNase H1 was expressed and purified as previously described (11,16). The oxidized form of the enzyme was prepared by resuspending the lyophilized protein in dilution buffer (50 mM Tris, pH 7.5, 50 mM NaCl, 50% glycerol) to a final concentration of 0.5 mg/ml and incubating the enzyme at 25°C for 1-4 h. The non-reduced form of human RNase H1 was prepared by resuspending the lyophilized protein in dilution buffer and adding 20 -50 mM ␤-mercaptoethanol (BME) or 0.1-1 mM tris(2-carboxyethyl)phosphate (TCEP). Human RNase H1 was alkylated with NEM by reducing the protein with TCEP as described, adding 10 -20 mM NEM and incubating for 1 h at 25°C. The reduced and oxidized forms of the enzyme were analyzed by SDS-PAGE.
Construction of Cysteine Mutants-The mutagenesis of human RNase H1 was performed using a PCR-based technique derived from Landt et al. (27). Briefly, two separate PCR were performed using a set of site-directed mutagenic primers and two vector-specific primers (11). For the RNase H1[C147,148A] mutant, the 5Ј-oligodeoxynucleotide used for PCR was CTGATGGCGCTGCTTCCAGTAATGGGCGTTA and the 3Ј-oligodeoxynucleotide was TTACTGGAAGCAGCGCCACTAGTG-TAGACGACG. The PCR primers for RNase H1[C147A] were 5Ј-CTG-ATGGCGCTTGCTCCAGTAATGGGGCTA and 3Ј-TTACTGGAGCAAG-CGCCATCAGTGTAGACGACG. The primers for RNase H1[C148A] were 5Ј-CTGATGGCTGCGCTTCCAGTAATGGGGCTAGA and 3Ј-TTA-CTGGAAGCGCAGCCATCAGTGTAGACG. The primers for RNase H1[C191A] were 5Ј-CATGCAGCCGCTAAAGCCATTGAACAAGCAA and 3Ј-CAATGGCTTTAGCGGCTGCATGAATTTCCGCTCT. Approximately 1 g of human RNase H1 cDNA was used as the template for the first round of amplification of both the amino-and carboxyl-terminal portions of the cDNA corresponding to the mutant site. The fragments were purified by agarose gel extraction (Qiagen). PCR was performed in two rounds consisting of, respectively, 15 and 25 amplification cycles (94°C, 30 s; 55°C, 30 s; 72°C, 180 s). The purified fragments were used as the template for the second round of PCR using the two vectorspecific primers. The final PCR product was purified and cloned into the expression vector pET17b (Novagen) as described previously (11). The incorporation of the desired mutations was confirmed by DNA sequencing.
Protein Expression and Purification-The plasmid was transfected into E. coli BL21(DE3) (Novagen). The bacteria were grown in M9ZB medium (28) at 37°C and harvested at OD 600 of 0.9. The cells were induced with 0.5 mM isopropyl-1-thio-␤-D-galactopyranoside at 37°C for 2 h. The cells were lysed in 8 M urea solution, and the recombinant protein was partially purified with nickel-nitrilotriacetic acid-agarose (Qiagen).
Synthesis of Oligonucleotides-The oligoribonucleotides were synthesized on a PE-ABI 380B synthesizer using 5Ј-O-silyl-2Ј-O-bis(2-acetoxyethoxy)methyl ribonucleoside phosphoramidites and procedures described elsewhere (29). The oligoribonucleotides were purified by reverse phase HPLC. The DNA oligonucleotides were synthesized on a PE-ABI 380B automated DNA synthesizer and standard phosphoramidite chemistry. The DNA oligonucleotides were purified by precipitation two times out of 0.5 M NaCl with 2.5 volumes of ethyl alcohol.
Preparation of the Heteroduplex-The heteroduplex substrates were prepared in 100 l containing 50 nM unlabeled oligoribonucleotide, 10 5 cpm of 32 P-labeled oligoribonucleotide, 100 nM complementary oligodeoxynucleotide, and hybridization buffer (20 mM Tris, pH 7.5, 20 mM KCl). Reactions were heated at 90°C for 5 min, cooled to 37°C, and 60 units of Prime RNase Inhibitor (5 Prime 3 3 Prime, Boulder, CO) and MgCl 2 at a final concentration of 1 mM were added. Hybridization reactions were incubated 2-16 h at 37°C, and BME was added at final concentration ranging from 0 to 200 mM.
Determination of Initial Rates (V 0 )-The heteroduplex substrates were digested with 0.5 ng of human RNase H1 at 37°C. A 10-l aliquot of the cleavage reaction was removed at time points ranging from 2 to 120 min and quenched by adding 5 l of stop solution (8 M urea and 500 mM EDTA). The aliquots were heated at 90°C for 2 min, resolved in a 12% denaturing polyacrylamide gel, and the substrate and product bands were quantitated on a Amersham Biosciences PhosphorImager. The concentration of the converted product was plotted as a function of time. The initial cleavage rate was obtained from the slope (mole of RNA cleaved per min) of the best-fit line for the linear portion of the plot, which comprises, in general, Ͻ10% of the total reaction and data from at least five time points.
Competition experiments were performed as described for the determination of initial rates with the exception that the hybridization reactions were prepared with 20 nM oligodeoxynucleotide, 10 nM oligoribonucleotide, and hybridization buffer without BME. Oxidized human RNase H1 was added to the hybridization reaction at final concentrations of 0.5 and 2.5 ng of protein. Alternatively, 20 mM BME and NEM-alkylated enzyme was added to the hybridization reaction at final concentrations of 0.5 and 2.5 ng. The hybridization reactions were digested with 250 pg of the reduced form of human RNase H1. The reactions were quenched, analyzed, and quantitated as described for the determination of initial rates. Gel Renaturation Assay-The gel renaturation assay was performed as described previously (6). Briefly, a 12% SDS-polyacrylamide gel containing 300,000 cpm of 32 P-labeled poly(rA)⅐poly(dT) per 13 ϫ 15-cm gel was prepared. Following electrophoresis the SDS was removed by washing the gel with three changes 50 mM Tris, pH 8.0, 1 mM BME, 0.1 mM EDTA, and 25% (v/v) isopropanol for 20 min at 25°C. The isopropanol was removed by washing the gel with two changes of 10 mM Tris, pH 8.0, and 5 mM BME for 15 min at 25°C. The proteins were denatured by soaking the gel for 2 h at 25°C with 50 mM Tris, pH 8.0, 20 mM BME, 10 mm MgCl 2 , 50 mM NaCl, 6 M guanidine HCl, and 10% (v/v) glycerol. The proteins were renatured by washing the gel with three changes of 50 mM Tris, pH 8.0, 20 mM BME, 10 mM MgCl 2 , 50 mM NaCl, 2.5% Nonidet P-40, and 10% (v/v) glycerol for 20 h at 25°C for reduced conditions and without BME for renaturation under oxidized conditions. Soluble radioactivity was washed from the gel with four changes of 5% (v/v) trichloroacedic acid and 1% (v/v) sodium pyrophosphate for 15 min at 25°C. The gel was quantitated on a Amersham Biosciences PhosphorImager.
Trypsin Digestion and Mass Spectral Analysis of Human RNase H1 Proteins-Trypsin digestion of human RNase H1 proteins was prepared in 30 l containing 2 M human RNase H1, 0.67 M urea, 50 mM Tris-HCl, and 0.9 mM CaCl 2 and a (trypsin:RNase H1) ratio of 1:75 (w/w) (32,33). Digestion reactions were incubated for 2 h at 65°C. Immediately after removing the samples from the hot water bath, 3 l of Me 2 SO was added to the mixture to enhance the solubility of hydrophobic peptides. 3 Samples with the reducing agent were prepared as above except with the addition of 5 mM TCEP. The reaction was allowed to proceed at room temperature for 1 h. In selected experiments, NEM (10 mM final concentration, shaken at room temperature for 3 h) was introduced at this point to irreversibly "cap" the free sulfhydryl groups before adding trypsin.
HPLC-ESI-FITCR Mass Spectrometry-A Zorbax C18 0.32 ϫ 150-mm capillary silica column (Micro-Tech Scientific, Sunnyvale, CA) was employed on a Micro-Tech Ultra-Plus II HPLC system and directly coupled to the mass spectrometer. The mobile phases were 1% formic acid, 10% Me 2 SO (mobile phase A), and 1% formic acid, 10% Me 2 SO in acetonitrile (mobile phase B). Samples containing 25 l of the human RNase H1 tryptic digest solution were injected onto the HPLC column, equilibrated with 99% A and 1% B at 4 l/min, and eluted with 99% B and 1% A, also at 4 l/min.
Experiments were performed on a modified Bruker Daltonics (Billerica, MA) Apex II 94e electrospray ionization-Fourier transform ion cyclotron (ESI-FTICR) mass spectrometer (35) with an actively shielded 9.4-tesla superconducting magnet. HPLC-ESI-FTICR mass spectra were acquired at 6-s intervals and subsequently processed using the ICR2LS software package (Pacific Northwest National Laboratory, Richmond, WA).

RESULTS
The enzymatic activity of human RNase H1 under oxidized and reduced conditions is shown in Table IIA. Oxidation of human RNase H1 resulted in the ablation of cleavage activity. The initial cleavage rate (V 0 ) observed for the enzyme under reduced conditions was greater than 3 orders of magnitude faster than the rate observed for the oxidized enzyme. The loss of the enzymatic activity resulting from the oxidation of human RNase H1 was observed to be reversible (Table IIA). The enzymatic activity for Human RNase H1 was regenerated to the level of activity observed for the reduced form when the oxidized enzyme was incubated with 20 mM BME for 10 min. Furthermore, the enzyme activity was readily regenerated without requiring gradual reduction of the protein through gradient methods such as dialysis suggesting that regeneration of the enzyme was rapid and cooperative. The initial cleavage rate for human RNase H1 increased as a function of the con-centration of the reducing agent BME (Fig. 1). The enzyme was most active at BME concentrations Ն20 mM, and no loss in enzymatic activity was observed at 200 mM BME. Finally, analysis of the oxidized and reduced forms of human RNase H1 by SDS-PAGE showed that both forms migrated as monomers on the gel (data not shown).
The oxidized form of human RNase H1 was observed to competitively inhibit the endoribonuclease activity of the reduced form of the enzyme (Table IIB). These experiments were performed under single-turnover kinetics with the concentration of the reduced form of human RNase H1 in excess of the substrate concentration and with the concentration of the oxidized form of the enzyme in excess of the reduced enzyme concentration. The V 0 for the reduced form of the enzyme was 2-fold faster than the cleavage rate observed for the reduced form of the enzyme in the presence of 2-fold excess oxidized human RNase H1. In the presence of 10-fold excess oxidized enzyme, the initial cleavage rate for the reduced form of human RNase H1 was below the detection limit of the assay. Initial cleavage rates were also determined under multiple-turnover kinetics with the substrate concentration in excess of the enzyme concentration and with the concentration of the oxidized enzyme in 10-fold excess over the reduced form of human RNase H1. Competition experiments under multiple-turnover conditions showed no reduction in the cleavage rate compared with the reduced form of human RNase H1 in the absence of oxidized enzyme (data not shown).
Competition experiments were also performed with NEMalkylated human RNase H1. Here, the concentration of the reduced from of human RNase H1 was in excess of the substrate concentration, and the concentration of the NEM alkylated protein was in excess of the reduced enzyme concentration. The V 0 for the reduced form of the enzyme was greater TABLE II V 0 for the reduced and oxidized forms of human RNase H1 A: initial rate measurements for human RNase H1 under oxidized and reduced conditions was determined as described under "Materials and Methods." The V 0 values are an average of three measurements with estimated errors of the coefficient of variation Ͻ10%. B: competition experiments were performed as described under "Materials and Methods." The heteroduplex substrate was incubated with the oxidized form of human RNase H1 prior to adding the reduced form of the enzyme. The concentration of the reduced form of human RNase H1 enzyme was in excess of the substrate concentration. The concentration of the oxidized form of human RNase H1 was 10-fold in excess of the reduced enzyme. The initial rate for the reduced form of human RNase H1 enzyme alone and in the presence of the oxidized form of the enzyme was determined as described under "Materials and Methods." C: the competition experiments were performed as described in B except that excess NEM-labeled human RNase H1 was used as the competing protein. (0:1) 9200 (2:1) 2500 (10:1) ϽDetectable limit a a The detection limit of the assay corresponds to Ͻ1% of the heteroduplex substrate cleaved over 60 min. than 3-fold faster than the cleavage rate observed for the reduced form of the enzyme in the presence of 2-fold excess NEM-alkylated human RNase H1 (Table IIC). In the presence of 10-fold excess NEM alkylated enzyme, the initial cleavage rate for the reduced form of human RNase H1 was 15-fold slower than the cleavage for the reduced enzyme alone. Again, competition experiments under multiple-turnover conditions showed no reduction in the cleavage rate compared with the reduced form of human RNase H1 in the absence of the NEM alkylated enzyme (data not shown).
To identify the cysteine residues responsible for the observed loss the cleavage activity under oxidized conditions, the initial cleavage rate for the human RNase H1 deletion mutant H1[⌬1-73] was determined. This deletion mutant was missing the first 73 amino acids from the NH 2 terminus of the protein and therefore lacked the C18 and C48 residues ( Fig. 2A). The cleavage rate for the H1[⌬1-73] mutant was compared with the wild-type human RNase H1 enzyme in the presence of increasing BME concentration (Fig. 1). Similar to the wild-type enzyme, the H1[⌬1-73] mutant was inactive under oxidized conditions, and the cleavage rate of the H1[⌬1-73] mutant increased with increasing concentration of BME.
Four mutants of human RNase H1 were prepared in which the three remaining cysteines were substituted with alanine (e.g. C147A, C148A, C191A, and C147A/C148A) ( Fig. 2A). The cleavage activity of these mutants under both reduced and oxidized conditions was determined using a gel renaturation assay (Fig. 3). Here the enzymes are separated under denaturing conditions on an SDS-polyacrylamide gel containing 32 Plabeled poly(rA)⅐poly(dT) substrate. The denaturing agents were washed away, and the embedded substrates were degraded by the renatured enzymes at the position the enzymes migrated on the gel. Renaturation was performed in either the presence of BME (reduced conditions) or in the absence of reducing reagent (oxidized conditions). All four mutants degraded the heteroduplex substrate under reduced conditions (Fig. 3). In contrast, under oxidized conditions only the C147A, C148A, and C147A/C148A mutants exhibited cleavage activities comparable with the activities observed for the mutants under reduced conditions. The C191A mutant exhibited significantly lower cleavage activity under oxidized conditions when compared with the cleavage activity observed for this mutant under reduced conditions.
To confirm the roles of Cys 147 and Cys 148 , the cysteine residues of human RNase H1 were analyzed using HPLC-ESI-FTICR mass spectrometry (32,33). Trypsin digestion was performed on the oxidized form of human RNase H1, the reduced form of the enzyme treated with NEM (reduced/NEM), and the oxidized enzyme treated with NEM (oxidized/NEM). Trypsin hydrolyzes proteins at the COOH terminus of lysine and arginine, and a tryptic digest map of human RNase H1 was generated based on the predicted molecular weights of the tryptic fragments using the PAWS software (Genomic Solutions, Ann Arbor, MI). Scans from the mass spectrometer were analyzed for the cysteine containing fragments from each of the digestion conditions (e.g. oxidized, reduced/NEM, and oxidized/NEM).
A tryptic fragment containing the Cys 147 and Cys 148 residues was identified for the oxidized-, reduced/NEM-, and oxidized/ NEM-treated human RNase H1 (Fig. 2B). Analysis of the mass spectrometry scans for the fragment Asp 131 -Arg 153 generated under oxidized conditions revealed a signal between scans 100 and 114 corresponding to a mass of 2519.999 Ϯ 0.006 Da, which was consistent with the calculated mass of 2520.003 Da for this fragment minus two hydrogen atoms and was consistent with the two cystines forming a disulfide bond (Fig. 4a). No signal was observed for the reduced form of the protein (Fig. 4a). A search of the mass spectrometry scans for the Asp 131 -Arg 153 fragment from the reduced/NEM treated enzyme revealed a signal corresponding to a mass of 2772.118 Ϯ 0.006 Da (Fig.   FIG. 1. The effect   The cleavage activity of the C147A, C148A, C191A, and C147A/C148A mutants of human RNase H1 was determined by gel renaturation assay as described under "Materials and Methods." The mutants were renatured in the presence of 20 mM BME (reduced) and absence of BME (oxidized). The white bands correspond to the absence of 32 P-labeled heteroduplex substrate as a result of human RNase H1 digestion. 4b). The observed mass was in excellent agreement with the calculated mass of 2772.114 Da for a double NEM-labeled Asp 131 -Arg 153 fragment. Again, no signal was observed for either the oxidized or reduced forms of the Asp 131 -Arg 153 fragment (Fig. 4b). Finally, analysis of the mass spectrometry scans for the Asp 131 -Arg 153 fragment from the oxidized/NEM-treated RNase H1 showed a signal corresponding to a mass consistent with the calculated mass for the peptide minus two hydrogen atoms and no signal for either the reduced-or double NEMlabeled fragments (Fig. 4c). DISCUSSION Human RNase H1 was shown to be active only under reduced conditions. Oxidation of human RNase H1 resulted in the ablation of the cleavage activity (Table IIA). The cleavage rate increased with increasing BME concentrations and in fact human RNase H1 was observed to be most active under fully reduced conditions (e.g. BME concentrations as high as 200 mM), suggesting that the active conformation of the enzyme does not contain disulfide bonds (Fig. 1). Analysis of the oxidized form of the enzyme by SDS-polyacrylamide gel electrophoresis indicated a single protein with a molecular weight consistent with the monomeric enzyme (data not shown). Taken together these data suggest that if the observed redoxdependent cleavage activity of human RNase H1 was the result of a disulfide bond, the disulfide bridge was intramolecular in nature.
A competition assay was designed to elucidate the role of the cysteine residues with respect to the function of the enzyme. Specifically, whether oxidation affected the binding or the catalytic properties of the enzyme. Here, the activity of the active reduced form of the enzyme was determined in the presence of an excess concentration of the inactive oxidized enzyme under conditions in which both forms of the enzyme were competing for limiting substrate. The oxidized form of human RNase H1 was shown to competitively inhibit the cleavage activity of the reduced form of the enzyme (Table IIB). Furthermore, the activity of the reduced form of human RNase H1 was inhibited by ϳ50% when an equal concentration of the oxidized enzyme was added suggesting that the oxidized form of the protein bound to the substrate with similar affinity as the reduced protein. Therefore, these data suggest that the oxidation of Human RNase H1 does not affect the ability of the enzyme to bind to the substrate, but that oxidation results in a conformation that can no longer catalyze the hydrolysis of the RNA.
To ensure that the observed inhibition of the reduced form of Human RNase H1 was not due to nonspecific protein-protein interactions between the oxidized and reduced forms of the enzyme (e.g. allosteric interactions or obstruction of the active site), competition experiments were performed under substrate saturating conditions. No reduction in the enzymatic activity of the reduced-from of the enzyme was observed with 10-fold excess oxidized enzyme under these conditions (data not shown). Therefore, the inhibition of the reduced form of Human RNase H1 by the oxidized form does not appear to be the result protein-protein interactions.
Consistent with previous studies, alkylation of human RNase H1 with NEM resulted in the loss in cleavage activity (16). The NEM-labeled enzyme was also shown to competitively inhibit the cleavage activity of human RNase H1, suggesting that the NEM labeled enzyme was capable of binding to the heteroduplex substrate (Table IIC). These observations are also consistent with E. coli RNase H1, which was shown to be inactivated by NEM (25). Site-directed mutagenesis of the E. coli enzyme revealed that of the three cysteine residues found in the protein, the alkylation of Cys 13 and Cys 133 with NEM was responsible for the loss in cleavage activity (26). X-ray analysis of E. coli RNase H1 revealed that both cysteines were solvent-exposed and situated close to both the catalytic site of the enzyme as well as the phosphate backbone of the heteroduplex substrate (38). Consequently, the alkylation of these residues with NEM was likely interfering with either metal coordination at the catalytic site or improper positioning of the catalytic site on substrate. The Cys 13 residue as well as the amino acids that make up the catalytic site is conserved in the human enzyme and therefore NEM may be ablating human RNase H1 activity in a similar manner.
Mutants of human RNase H1 were prepared to determine which cysteines were responsible for the observed loss in cleavage activity under oxidized conditions. These mutants included a previously described deletion mutant H1[⌬1-73] in which the first 73 amino acids from the NH 2 terminus of the enzyme was deleted (22), effectively eliminating the Cys 18 and Cys 46 residues, as well as the alanine substituted mutants C147A, C148A, C147A/C148A, and C191A ( Fig. 2A). The cleavage activity of H1[⌬1-73] exhibited a similar response to the BME concentration as the wild-type enzyme (Fig. 1). The maximum cleavage rate for the H1[⌬1-73] mutant was ϳ2-fold faster than the maximum cleavage rate observed for the human RNase H1. The faster cleavage rate for the H1[⌬1-73] mutant is consistent with previous observations and suggests that Cys 18 and Cys 48 were not contributing to the observed ablation of the cleavage activity of the oxidized form the human RNase H1 (22). Similarly, the C191A mutant appeared to be significantly less active in the gel renaturation assay under oxidized conditions compared with the reduced conditions (Fig. 3). The C147A, C148A, and C147A/C148A mutants, one the other hand, exhibited comparable cleavage activities under both reduced and oxidized conditions, suggesting that the oxidation of Cys 147 and Cys 148 residues was responsible for the loss of RNase H1 activity under oxidized conditions (Fig. 3). Finally, the deletion or substitution of the cysteine residues resulted in a catalytically active enzyme under reduced conditions, sug- gesting that similar to E. coli RNase H1, the cysteine residues of human RNase H1 are not required for cleavage activity (26).
The redox states of the Cys 147 and Cys 148 residues for the oxidized and reduced forms of human RNase H1 were determined by treating the enzymes with trypsin and analyzing the cysteine containing tryptic fragments by HPLC-ESI-FTICR mass spectrometry (32,33). The oxidized form of human RNase H1 was also treated with NEM (oxidized/NEM) prior to trypsin digestion to eliminate the possibility of potential disulfide bonds between tryptic fragments. A tryptic fragment (Asp 131 -Arg 153 ) containing the Cys 147 and Cys 148 residues was identified for both the oxidized and reduced forms of the enzyme (Fig.  2B). Analysis of the Asp 131 -Arg 153 fragment from the reduced form of human RNase H1 treated with NEM revealed a double NEM-labeled peptide, indicating that the Cys 147 and Cys 148 residues were accessible to the alkylating reagent (Fig. 4b). The Asp 131 -Arg 153 fragment of the oxidized form of human RNase H1 exhibited a mass consistent with the calculated mass for the peptide minus two hydrogens suggesting the formation of a disulfide bridge between the Cys 147 and Cys 148 residues (Fig.  4a). Consistent with this observation was the absence of the NEM-labeled Asp 131 -Arg 153 fragment for the oxidized/NEM form of human RNase H1, suggesting that no sulfhydryl moieties were present within the peptide (Fig. 4c). The lack of NEM label for this peptide also suggests that the disulfide bond between the adjacent cystine residues was present in the intact protein and therefore was not a tryptic peptide specific structure. Finally, larger peptides containing the Cys 147 and Cys 148 residues were also identified which were the products of one or more missed tryptic digestions. In all cases, the observed mass for these larger peptides was consistent with the formation of a disulfide bond between Cys 147 and Cys 148 , i.e. the calculated mass for the missed cleaved peptides minus two hydrogens.
The redox states of the remaining cysteine residues were also analyzed by HPLC-ESI-FTICR mass spectrometry. Tryptic fragments Val 11 -Arg 19 , Thr 37 -Arg 47 , and Ala 185 -Lys 192 were identified that contained, respectively, Cys 18 , Cys 46 , and Cys 191 (Fig. 2B). The oxidized form of the enzyme treated with NEM showed all three cysteine residues labeled with NEM (data not shown). In addition, a peptide mass corresponding the crosslinked fragments Val 11 -Arg 19 and Thr 37 -Arg 47 was also identified for the oxidized/NEM protein. The presence of both a cross-linked Val 11 -Arg 19 /Thr 37 -Arg 47 fragment and single NEM-labeled Val 11 -Arg 19 and Thr 37 -Arg 47 fragments suggests a partial or transient disulfide linkage between Cys 18 and Cys 46 . Clearly, this observed transient disulfide linkage was not contributing to redox dependent cleavage activity of human RNase H1 given the fact that the H1[⌬1-73] mutant, which did not contain the Cys 18 and Cys 46 residues, was also shown to be inactive under oxidized conditions. The formation of a vicinal disulfide bridge between adjacent cysteines has been shown in peptides to result in a structure consisting of a novel eight-membered ring with either a cis-or trans-conformation (39,40). This structure requires considerable distortion of the peptide backbone for its formation, and therefore, the occurrence of vicinal linkages are rare in proteins. One example is the quinoprotein methanol dehydrogenase from Methylobacterium extorquens (41). X-ray analysis of quinoprotein methanol dehydrogenase revealed a disulfide linkage between Cys 103 and Cys 104 . In this case, the formation of the disulfide bridge produced a eight-membered ring with a cis-configuration and a non-planar linking peptide bond. Again, a significant distortion of the peptide backbone was observed with this structure. A perturbation of this nature in the human RNase H1 structure could account for the observed loss of cleavage activity for the enzyme under oxidized conditions given the predicted proximity of the Cys 147 and Cys 148 residues to the catalytic site. In this case the deformation of the of the structure of human RNase H1 could lead to interference with either metal coordination at the catalytic site or improper positioning of the catalytic site onto the substrate.
A redox switch involving adjacent cysteines has been shown in other proteins. For example, oxidation of the Cys 558 and Cys 559 residues of mercuric reductase resulted in an inactive enzyme whereas reduction of the protein restored the catalytic activity of the enzyme (42). Recently, a redox switch involving two adjacent cysteines was designed into RNase A (43). Sitedirected mutagenesis was used to introduce the cysteine residues at sites critical to catalysis (A5C/A6C) and within a structurally flexible portion of the enzyme (S15C/S16C). The mutation A5C/A6C resulted in an RNase A enzyme that was active under reduced conditions and inactive under oxidized conditions. This study demonstrated that adjacent cysteines are capable of forming a disulfide bridge that results in a local structural perturbation that renders the enzyme inactive.
Human RNase H1 has evolved a cysteine at position 147 that results in a redox switch between Cys 147 and Cys 148 . The observed redox modulation of Human RNase H1 may play a regulatory role in the biological processes involving the enzyme either through oxidative stress or other redox regulating mechanisms. In fact, the active form of the ribonuclease inhibitor family of proteins does not contain disulfide bonds (44,45). Oxidation of ribonuclease inhibitor proteins, which has been shown to involve the formation of a disulfide bond between adjacent cysteines, not only inactivates the protein but also targets the protein for proteolytic degradation within the cytosol. In addition, certain signal transduction pathways that control gene expression have been shown in vivo to be regulated by a redox mechanism (46, 47). These regulatory mechanisms involve either the formation of cystine sulfides, disulfides, or the isomerization of cystine disulfides. Finally, the optimum redox conditions for the cleavage activity of other eukaryotic RNases H1 have not been determined, but the presence of the Cys 147 and Cys 148 residues within these enzyme suggest that RNases H1 from eukaryotes likely contain a redox switch at these positions. Obviously, more work is required before any of these mechanisms can be applied to the regulation of human RNase H1. Nevertheless, this observation opens new avenues to explore the effect of local structural perturbation on substrate interactions and the roles of such processes in the biological regulation of the enzyme.