J. Biochem. 84, 395-402 (1978)

Ribonuclease H from Rat Liver

II. Partial Purification and Characterization of Cytosol H1

Fumio TASHIRO 2 and Yoshio UENO

Department of Microbial Chemistry, Faculty of Pharmaceutical Sciences, Tokyo University of Science, Ichigaya, Shinjuku-ku, Tokyo 162

Received for publication, March 11, 1978

We have detected in rat liver cytosol three (termed C-1, C-2, and C-3) which cleaved

the RNA moiety of RNA-DNA hybrid. These enzymes were separated from each other by

DEAE-Sephadex and Sephadex G-200 chromatography. C-1 and C-2 specifically act on the

RNA moiety of RNA-DNA hybrid, while C-3 degrades single-stranded RNA as well as the

RNA of the hybrid. The molecular weights of C-1, C-2, and C-3 are about 110,000, 35,000 and

110,000 daltons, respectively, and their activities are absolutely dependent on divalent cations

such as Mg2+ and Mn2+. Cleavage by C-1 and C-2 is endonucleolytic, producing mostly

oligonucleotides and a small amount of mononucleotides which possess 3•L-hydroxyl termini. It seems likely that C-2 is originally present in the nucleus and is released into cytosol because

of its loose binding to the nuclear components. As for biochemical properties, C-1 is very

similar to the cytosol initially reported by Roewekamp and Sekeris, and C-2

is very similar to the nuclear ribonuclease H reported by us in the preceding paper.

Ribonuclease H of rat liver have been isolated by region of RNA-DNA hybrid occur in rat liver Roewekamp and Sekeris (1) from cytosol, and by cytosol. The possible relationship between the us from nuclei as described in the preceding paper cytosol and the nuclear ribonuclease H is also (2). However, the relationship between the cytosol discussed. and nuclear enzymes and their possible roles in cells are still not clear. MATERIALS AND METHODS In this communication, it is demonstrated that at least three different enzymes degrading the RNA Buffers-The composition of buffers A and B is described in the accompanying paper (2). Buffer 1 This work was supported in part by grants from the C: 250 mm sucrose, 25 mm KCl, 10 mat MgCl2, Ministry of Education, Science and Culture (1977) and and 50 mm Tris-HCl (pH 7.5). Buffer D: 10 mm the Ministry of Public Health and Welfare (1977), Japan. Tris-HCl (pH 7.9), 1 mat MgCl2, 0.25 mm EDTA, 2 To whom request for reprint should be addressed. 10% glycerol, and 0.5 mat dithiothreitol. Enzymes: Ribonuclease H or RNA-DNA-hybrid ribo Chemicals-All reagents were as described in nucleotidohydrolase [EC 3.1.4.34], DNA-dependent the accompanying paper (2). RNA polymerase or nucleosidetriphosphate: RNA Substrates-[14C]RNA-DNA hybrid, [14C]- nucleotidyltransferase [EC 2.7.7.6], RNA-[3H]DNA hybrid, 3H-labelled T7 DNA, phosphodiesterase or oligonucleate 5'-nucleotidohy drolase [EC 3.1.4.1]. [3H]poly(rU)-poly(rA) and [3H]poly(rU)-poly(dA) vol. 84, No. 2, 1978 395 396 F. TASHIRO and Y. UENO were prepared as described in the preceding paper lysosome fractions were removed by low-speed (2). centrifugation. Cytosol was then prepared by Standard Assay Conditions-[14C]RNA-DNA ultracentrifugation at 105,000 x g for 1.5 h. The hybrid (2,000-2,500 cpm) was incubated with clear supernatant was stored at -80°C until use. in 300 pl of assay mixture containing 50 mm

Tris-HCl (pH 7.9) and 0.5 mm dithiothreitol; the RESULTS assay mixtures were adjusted to contain 10 mm (NH4)2S04 and 17 mm MgCl2 for C-2 and 10 mm Purification of Ribonuclease H-The enzyme (NH4)2S04 and 3 mm MgCl2 for C-3. After activity was precipitated from the cytosol fraction incubation at 37°C for 30 min, acid-soluble mate- with solid (NH4)2SO4 at a saturation of 30% to rial was determined as described in the preceding 60%. The precipitate was dissolved in buffer D paper (2). and dialyzed against the same buffer. The dialy Enzyme activity is defined in the preceding sate was centrifuged at 31,000 x g for 1 h, the paper (2), and was determined under the standard supernatant was applied to a DEAE-Sephadex assay conditions described above. column, and the enzymes were eluted with a linear Preparation of Cytosol-Cytosol was usually gradient from 0 M to 0.5 M NH4Cl in buffer D. prepared from three Wistar male rats by the method The ribonuclease H activity was resolved into of Roewekamp and Sekeris (1) with slight modifi two peaks (A and B) as indicated in Fig. 1. cations. The perfused livers were weighed, minced The fraction corresponding to peak A (fraction and homogenized in 3 vol. of buffer C in a Potter number 5-13) and peak B (fraction number 40- Elvehjem homogenizer. Nuclei and mitochodria 57) were each combined, and solid (NH4)2SO4

Fig. 1. DEAE-Sephadex chromatography of ribonuclease H. The crude extract obtained by ammonium sulphate fractionation (30-60% saturation) was dissolved in 20 ml of buffer D and dialyzed against the same buffer. The dialysate was centrifuged at 31,000 x g for 1 h. The supernatant was applied to DEAE- Sephadex column previously equilibrated with buffer D (bed volume 51 ml, column size 1.8 •~ 20 cm, flow rate 24 ml/h). The column was washed with buffer D, then with a linear gradient from 0 to 0.5 M NH4C1 in buffer D (100 ml+ 100 ml). Fractions of 4.5 ml were collected and assayed for ribonuclease H and ribonuclease activity. Ribonuclease activity was assayed with heat-de- natured ["C]RNA-DNA hybrid as substrate by determining the acid-soluble radioactivity. --- , Absorbance at 280 nm; -, NH4C1 concentration; •ü

, ribonuclease H activity; •œ, ribonuclease activity.

J. Biochem. CYTOSOL RIBONUCLEASE H FROM RAT LIVER 397

(0.42 g/ml) was added to afford precipitates, which TABLE I. Summary of the purification procedure for were collected by centrifugation at 100,000 x g for cytosol ribonuclease H. The reaction mixture for the 1 h. The precipitates were chromatographed on ribonuclease H is described in " MATERIALS AND Sephadex G-200 column as described in the METHODS." The amount of protein was determined by the method of Lowry et al. (5). preceding paper (2). Figure 2-A shows the elution profile of peak B ribonuclease H activity on Sephadex G-200 chromatography. The enzyme activity was sepa rated into the two peaks, termed C-I and C-2. The molecular weights of C-1 and C-2 were esti mated to be about 110,000 and 35,000, respectively. When heat-denatured [14C]RNA-DNA hybrid (100•Ž, 5 min) was used as substrate, no acid- soluble radioactivity was detected, indicating that enzyme C-1 and C-2 act exclusively on the RNA region of RNA-DNA hybrid. The ribonuclease activity contained in peak B fraction on DEAE- Sephadex chromatography was not detected on Sephadex G-200 chromatography. This is because the ribonuclease probably has a very small molec ular weight or is inactivated during the purification. Figure 2-B shows the elution profile of peak A on Sephadex G-200 chromatography. The enzyme was resolved into four peaks of ribonuclease

Fig. 2. Sephadex G-200 chromatography of ribonuclease H. The fractions corresponding to

peaks A and B ribonuclease H activity were each combined. Each enzyme was precipitated by the addition of solid (NH4)2SO4 (0.42 g/ml). The precipitate was collected by centrifugation at 100,000 •~ g for I h, dissolved in 3 ml of buffer A containing 0.5 M NH4C1, and dialyzed against the same buffer for 4 h. The dialysate was applied to a Sephadex G-200 column previously equi librated with buffer A containing 0.5 M NH4C1 (bed volume 182 ml, column size 2 •~ 58 cm, flow rate 9 ml/h, fraction volume 2.8 ml). (A) and (B) show the elution profiles of peaks B and A ribo H activity of Fig. 1, respectively. The letters a, b, c, d, e, and f represent the elution

positions of blue dextran, r-globulin, bovine serum albumin, ovalbumin, chymotrypsinogen and myoglobin, respectively. ----, Absorbance at 280 nm; •ü, ribonuclease H activity; •œ, ribonuclease activity.

Vol. 84, No. 2, 1978 398 F. TASHIRO and Y. UENO

H activity. The main peak enzyme, termed C-3, Fig. 3-A, the enzyme was eluted at approximately has a molecular weight of about 110,000. This 0.05 M NH4Cl as a sharp peak. enzyme acts on heat-denatured [14C]RNA-DNA The fractions having C-2 activity were sub hybrid (100•Ž, 5 min) as well as [14C]RNA-DNA jected to phosphocellulose chromatography as hybrid as shown in Fig. 7. The three minor peaks shown in Fig. 3-B. A summary of the purification were not studied further. procedure is presented in Table I. The fractions having C-1 activity were rechro Each enzyme from the final purification step matographed on DEAE-Sephadex. As shown in was stored at -20•Ž in 50% glycerol.

Fig. 3. A) DEAE-Sephadex rechromatography of C-1. B) Phosphocellulose chromatography of C-2. C-I obtained by Sephadex G-200 chromatography was applied to a DEAE-Sephadex column previously equilibrated with buffer D (bed volume 26 ml, column size 1.6 x 10 cm, flow rate 30 ml/h, fraction volume 2.8 ml). The column was eluted with a linear gradient from 0 to 0.5 M NH4Cl in buffer D (45 ml+45 ml). C-2 separated by Sephadex G-200 chromatography was applied to a column off phosphocellulose previously equilibrated with buffer B containing 0.05 M NH4Cl (bed volume 6.8 ml, column size 1.2 x 6 cm, flow rate 25 ml/h, fraction volume 2.8 ml) and eluted with a linear gradient from 0.05 to 1.0 M NH4Cl in buffer B (45 ml+45 ml). ----, Absorbance at 280 nm; - , NH4Cl concentration; •ü, ribonuclease H activity.

Fig. 4. Dependency of ribonuclease H reaction on MgCl2 and MnCl2. The experiments were carried out under the standard assay conditions with various concentrations of MgC12 (A) and MnC12 (B). Amounts of C-l, C-2, and C-3 used were 10.9, 7.5, and 10.6 units, respectively. 0, C-1; •œ, C-2; •£, C-3.

J. Biochem. CYTOSOL RIBONUCLEASE H FROM RAT LIVER 399

Fig. 5. Optimal pH and effect of ionic strength on the ribonuclease H reaction.- C-I and

C-2 were assayed at various pHs (A) or concentrations of ammonium sulphate (B) . Amounts of C-I and C-2 used were respectively 2.0 and 9.5 units in Fig . 5-A, and 9.2 and 3.4 units in Fig. 5-B. •ü, C-1; •œ, C-2.

Optimal Conditions for the Ribonuclease H Degradation of Double-Labelled Hybrid- Reaction-The three enzymes, C-1, C-2, and C-3, [14C]RNA-[3H]DNA hybrid was subjected to required absolutely divalent cations such as Mg2+ degradation by C-1 and C-2 enzymes. As shown in and Mn2+ for their activities. The optimal con Fig. 6, both enzymes specifically degrade the RNA centrations of Mg2+ for the activities of C-l, C-2, region of RNA-DNA hybrid. and C-3 were 6, 25, 3 mm, respectively (Fig. 4-A). Degradationsof Hybridand Heat-Denatured The optimal concentrations of Mn2+ for the three Hybridby C-3-C-3 degradedthe RNAhybridized enzymes all fell in the same range of about 0.6 mm to DNA as wellas the RNAmolecule dissociated (Fig. 4-B). C-I and C-2 were active at pH 7.5- fromthe heat-denatured hybrid, as shownin;Fig. 7. 8.0 and 8.5-9.0, respectively (Fig. 5-A). Optimal The ribonucleaseH and ribonucleaseactivities in concentrations of (NH4)2SO4 for C-I and C-2 were the C-3enzyme fraction could not be separatedby 6 and 50 trim, respectively (Fig. 5-B).

Fig. 6. Sensitivity of double-labelled hybrid to C-1 and Fig. 7. Degradation of hybrid and heat-denatured C-2.' The RNA-DNA hybrid containing 213 cpm of hybrid by C-3. •ü, Ribonuclease H activity; •œ, ribo [14C]RNA and 1,874 cpm of [3H]DNA was incubated nuclease activity against heat-denatured [14C]RNA-DNA with C-1 (0.9 units) and C-2 (0.5 units). •ü, 14C Radio hybrid' (100•Ž, 5 min). Amount of C-3 used was. 4.1 activity; •œ, 3H radioactivity. units.

Vol. 84, No. 2, 1978 400 F. TASHIRO and Y. UENO

CM-Sephadex and hydroxyapatite chromatography as shown in Fig. 8. These results strongly suggest (data not shown). that both enzymes are endonucleolytic. Substrate Specificity of C-1-As shown in As shown in Fig. 9, venom phosphodiesterase Table II, C-1 specifically degraded the RNA region completely converted the degradation products of of RNA-DNA hybrid, while heat-denatured hybrid, C-1 and C-2 into 5•L-UMP. We concluded that the single-stranded T7 DNA, poly(rU), poly(rU)- oligonucleotides produced by both enzymes termi poly(rA) and poly(rU)-poly(dA) were hardly nate in a 3•L-hydroxyl group (4). degraded. When double-stranded T7 DNA was Origin of C-2-The properties of C-2 described used as substrate, a little degradation was observed. above closely resembled those of nuclear ribo Mode of Action of C-1 and C-2-[14C]RNA- nuclease H from rat liver described in the preceding DNA hybrid was digested to various extents and paper (2). This suggested that C-2 might have the reaction products were chromatographed under been released from nuclei during the homogeniza the conditions described in the preceding paper tion of liver in buffer C. To confirm this, cytosol (2). At various stages of digestion by each enzyme, fraction was prepared from the liver homogenized the bulk of the product remained at the origin and with low ionic strength buffer containing only the amount of UMP was found to be very small, 250 mm sucrose and 10 mm MgCl2 and the cytosol

Fig. 8. Mode of action of ribonuclease H. [14C]RNA-DNA hybrid (3,653 cpm) with labelled UMP residues was incubated at 37•Ž for various lengths of time together with C-1 (15.7 units) or C-2 (32.9 units). The extent of degradation, determined by the standard assay, is indicated in parentheses. The reaction mixture was subjected to paper chromatography as described in the accompanying paper (2). One-centimeter strips were cut from the chromatogram and assayed for radioactivity. 0 epresents the origin, pU the position of 5•Lr-UMP and U the posi tion of uridine. a) Unreacted hybrid, b) digestion by C-1 for 15 min and C-2 for 7 min, c) digestion by C-I for 30 min and C-2 for 15 min, d) digestion by C-I for 60 min and C-2 for 30 min.

J. Biochem. CYTOSOL RIBONUCLEASE H FROM RAT LIVER 401

TABLE 11. Substrate specificity of C-1. The follow- ing amounts of substrates were assayed under standard assay conditions: 1,335 cpm of native or 1,691 cpm of denatured [14C]RNA-DNA hybrid (1,156 cpm/ƒÊg RNA); 669 cpm of native or 722 cpm of denatured [3H]T7 DNA

(3,478 cpm/g.g DNA); 276 cpm of [3H]poly(rU) (7.76 Ci/mol); 975 cpm of [3H]poly(rU)-poly(dA); 352 cpm of

[3H]poly(rU)-poly(rA). For denaturation of RNA- DNA hybrid and T7 DNA, the samples were heated at 100•Ž for 5 min and quickly chilled in ice. Amount of enzyme used was 9.2 units.

obtained was applied to DEAE-Sephadex and Sephadex G-200 chromatographies. Figure 10 Fig. 9. Terminal group of the digestion products of shows the elution profile of the cytosol ribo ribonuclease H. Digestion products by C-1 and C-2 nuclease H prepared with 250 mm sucrose-l0 mm as shown in Fig. 8-d were treated with snake venom MgCl2 on Sephadex G-200 chromatography. C-2 phosphodiesterase (10 ƒÊg) for 30 min at 37•Ž. The mixture was then subjected to paper chromatography as activity was greatly reduced compared with that described in the legend of Fig. 7. pU and U represent of the cytosol ribonuclease H prepared with buffer 5•L-UMP and uridine, respectively. (a) Digestion by C (see Fig. 2-A). C-1 as shown in Fig. 8-d followed by treatment with snake venom phosphodiesterase. (b) Digestion by DISCUSSION C-2 as shown in Fig. 8-d followed by treatment with the phosphodiesterase. We have detected in rat liver cytosol three forms of enzyme cleaving the RNA moiety of RNA-DNA hybrid. In molecular weight, optimal pH, optimal described in the preceding paper (2), the nuclei concentration of Mg2+ and Mn2+, mode of action isolated from rat liver with 2.3 Msucrose and 10 mm and substrate specificity, C-1 is very similar to MgCl2 contain only the ribonuclease H corre ribonuclease H reported by Roewekamp and sponding to C-2, and not that corresponding to Sekeris (1) and C-2 closely resembles the nuclear C-l. It seems likely that the nuclear ribonuclease enzyme reported in the preceding paper (2). H, which is presumed to be the same entity as C-2, When the cytosol was prepared with a solution functions in the DNA replication or transcription containing 250 mm sucrose and 10 MM MgCl2, the process in the nucleus. activity of C-2 was significantly reduced in com According to Busen et al. (3), parison with that prepared with buffer C (Fig. 9). HI and HIIb of bovine lymphocytes differ from This result implies that C-2 is loosely bound to each other and serve different physiological func the nucleus and reading released. Moreover, as tions. The possibility that C-1 and C-2 also serve

Vol. 84, No. 2, 1978 402 F. TASHIRO and Y. UENO

Fig. 10. Sephadex G-200 chromatography of the cytosol ribonuclease H pre

pared with 0.25 M sucrose-l0 mm MgC12. The cytosol of rat liver was prepared with 0.25 M sucrose-10 mm MgC12 instead of buffer C and applied to DEAE- Sephadex column as described in the legend of Fig. 1. After the elution, the enzyme activity corresponding to peak B in Fig. I was precipitated by the addition of solid (NH,)2SO, and subjected to Sephadex G-200 column as described in the legend of Fig. 2. ----, Absorbance at 280 nm; •ü, ribonuclease H activity; •œ, ribonuclease activity.

different physiological functions in different cellular location cannot be discounted. REFERENCES C-3 acts on the RNA strand released by heat denaturation of RNA-DNA-hybrid as well as on 1. Roewekamp, W. & Sekeris, C.E. (1974) Eur. J. the RNA hybridized to DNA. This enzyme could Biochem. 43, 405-413 2. Tashiro, T. & Ueno, Y. (1978) J. Biochem. 84, not be separated into the ribonuclease H and the 385-393 ribonuclease activities. It is not still clear whether 3. Busen, W., Peter, J.H., & Hausen, P. (1977) Eur. J. C-3 consists of one or several enzymes. Further Biochem. 74, 203-208 purification and characterization of this enzyme 4. Harberkern, R.C. & Cantoni, G.L. (1973) Biochemis are currently being pursued in our laboratory. try 12, 2389-2395 5. Lowry, O.H., Rosebrough, N.J., Farr, A.L., & Ran We wish to thank Dr. T. Higashinakagawa of Mitsu dall, R.J. (1951) J. Biol. Chem. 193, 265-275 bishi-Kasei Institute of Life Sciences and Dr. T. Mita of National Cancer Center Research Institute for helpful discussions.

J. Biochein.