Mutagenesis of the Novel Hericium erinaceus Ribonuclease, RNase He1, Reveals Critical Responsible Residues for Enzyme Stability and Activity

Abstract

Here, we determined the sequence of a cDNA encoding a guanylic acid-specific ribonuclease (RNase He1) from Hericium erinaceus that exhibits high sequence identity (59%) with RNase Po1, an enzyme with anti-cancer activity and which is found in Pleurotus ostreatus. RNase He1 and RNase Po1 have similar structures and heat stabilities; hence, RNase He1 may also have potential as an anti-cancer agent. Therefore, we initiated structure-function studies to further characterize the enzyme. Based on the RNase Po1 structure, RNase He1 is predicted to form 3 disulfide bonds involving Cys7–Cys98, Cys5–Cys83, and Cys47–Cys81 linkages. The Cys5Ala mutant exhibited no RNase activity, whereas the Cys81Ala mutant retained RNase activity, but had reduced heat stability. Therefore, the Cys5–Cys83 bond in RNase He1 is essential for the structure of the RNase active site region. Similarly, the Cys47–Cys81 bond helps maintain the conformational stability of the active site region, and may contribute to the greater heat stability of RNase He1.

Anticancer activity has been observed in frog-egg RNases (onconase,¹⁾ sialic acid-binding lectins [cSBL,²⁾ iSBL³⁾]), bovine seminal RNase,⁴⁾ and Momordica charantia ribonuclease (MC2).⁵⁾ These heat-stable enzymes belong to the vertebrate pyrimidine base-specific RNase subfamily. To date, only α-sarcin⁶⁾ from Aspergillus giganteus and RNase Po1⁷⁾ from Pleurotus ostreatus, which are guanylic acid-specific RNases of the T1 family, exhibit anti-tumor activity. Microbe-specific RNase T1 family RNases are potential anticancer agents owing to their possible resistance to mammalian proteases and/or enzymatic inhibitors. However, the anti-tumor activity of RNase Po1 is about 10-fold lower than that of onconase, which is similar to that of cSBL.⁷⁾ Thus, mutagenesis to improve the anti-tumor activity of RNase Po1 or its homologs is required. Here, we identified a new heat-stable RNase (RNase He1) from Hericium erinaceus that exhibited high identity with RNase Po1. Extracts from H. erinaceus induce apoptosis in human leukemia cells.⁸⁾ thus, RNase He1 could possess anticancer properties. We studied the RNase He1 cDNA sequence and to understand the underlying structure-function relationships, particularly heat stability, and enzyme activity we mutated critical cysteine residues that are predicted to participate in disulfide bonds and expressed the wild-type and mutant enzymes in Escherichia coli.

MATERIALS AND METHODS

Construction of Expression Vector

An expression vector for RNase He1 was constructed according to the method described by Huang et al.⁹⁾ The pelB signal sequence from pET22b was ligated with He1 gene (pelb-He1). The pelb-He1 was introduced into the pET11d expression vector. Flanking NcoI and BamHI sites were added upstream and downstream of the RNase He1 cDNA, respectively, by polymerase chain reaction (PCR).

Preparation of Two Cysteine Site-Directed Mutants of RNase He1

Mutants in which the Cys5 and Cys81 residues were either individually or simultaneously substituted to Ala were prepared by PCR.

Expression and Purification of RNase He1 and Site-Directed Mutants

RNase He1 and the site-directed mutants of RNase He1 were transfected into E. coli BL21 (DE3) pLysS (Promega, Madison, WI, U.S.A.) and incubated in Terrific Broth at 25°C containing 100 µg/mL ampicillin and a final concentration of 0.5 mM isopropyl β-D-1-thiogalactopyranoside (Wako Pure Chemical Industries, Ltd., Osaka, Japan) for 6 d. The culture medium was centrifuged at 12000×g for 30 min. The enzymes were precipitated with ammonium sulfate (90% saturation) and applied to a column of Sephadex G50 (3×180 cm; GE Healthcare, Uppsala, Sweden) and eluted with 10 mM acetate buffer (pH 6.0). Preparations were applied to a column of DEAE-Toyopearl (3×30 cm; Tosoh, Tokyo, Japan) and eluted with a linear gradient of NaCl (0–0.3 M) in 10 mM acetate buffer (pH 5.0). Samples were then passed through a column of Ultrogel AcA54 (3×180 cm; GE Healthcare) and eluted with 10 mM acetate buffer (pH 6.0). Elute was applied to SP-Toyopearl (1.5×40 cm; Tosoh) columns and eluted with a linear gradient of NaCl (0–0.3 M) in 10 mM Tris–HCl buffer (pH 6.0). Preparations were then added to a column of heparin-Sepharose (1.0×20 cm; GE Healthcare) and eluted with a linear gradient of NaCl (0–0.3 M) in 10 mM acetate buffer (pH 4.5). Following dialysis using a dialysis membrane (Spectra/Pro 7 with a 3.5-kD cut-off; Spectrum Labs, Dominquez, CA, U.S.A.) against de-ionized water, the enzyme was separated by HPLC (LC6A, Shimadzu; Kyoto, Japan) using a Shodex protein 802.5 column (0.8×80 cm; Showa Denko, Nagoya, Japan) and eluted with 5 mM trimethylamine acetate buffer (pH 6.0). The RNase active fractions were pooled as purified enzyme, the homogeneity of enzymes were confirmed by Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Tricine-SDS-PAGE).¹⁰⁾

Enzymatic Assay

RNase activity was measured as described previously using 0.25 mg/mL yeast RNA (Wako Pure Chemical Industries, Ltd.) as a substrate at pH 4.5 and at 37°C.¹¹⁾

Additionally, the K_m and V_max of native RNase He1 and He1-C81A (0.03 nmol) were determined using RNA (0.5, 0.6, 1.0, 2.0, 3.0, 5.0 mg/mL) as a substrate and were calculated from a Lineweaver-plot.¹²⁾

Determination of Protein Concentration

The protein concentration of each of the final enzymes was determined spectrophotometrically, assuming an absorbance of 0.905 for a 0.1% solution at 280 nm. This value was based on the amino acid composition of RNase He1 (data not shown).

Effect of RNase He1 on the Proliferation of Human Leukemia Cells

The anti-proliferative activity of RNase Hel was measured with human leukemia cells (HL-60, Health Science Research Resources Bank, Osaka, Japan) according to the method described by Titani et al.²⁾ and Kobayashi et al.⁷⁾

Base Specificity

The base specificity of RNase He1 and He1-C81A was determined based on the rates of release of 3′-mononucleotides and 2′,3′-cyclic nucleotides during the hydrolysis of yeast RNA. The hydrolyzed nucleotides were identified, and their rates of release were determined by reversed-phase HPLC using a TSK gel carbon 500 column (Tosoh), as described previously.¹³⁾

Circular Dichroism (CD) Spectroscopy of RNase He1 and He1-C81A

To observe temperature-dependent changes in the secondary structure of the proteins, the CD spectra of RNase He1 and He1-C81A were measured between 200 and 250 nm with a spectropolarimeter (J600, JASCO, Tokyo, Japan) at 20–90°C. The sample cell had a 5-mm path length and the protein concentration was approximately 6.0 nmol/mL.

RESULTS AND DISCUSSION

Cloning of cDNA Encoding RNase He1 and Expression of RNase He1

We cloned the cDNA encoding RNase He1. The N-terminus of RNase He1 isolated from H. erinaceus (hereafter “raw RNase He1”) is most likely pyroglutamylated. Thus, both mature RNase He1 and RNase Po1 share glutamine as the first residue of the N-terminus. Furthermore, we found that the amino acid composition of raw RNase He1 and bacterially expressed RNase He1 were consistent (data not shown). Mature RNase He1 comprised 100 amino acid residues, with a predicted molecular mass of 10.69 kDa, and had the highest amino acid sequence identity with RNase Po1 (59%). We expressed wild-type RNase He1 and three mutants in E. coli. The expressed RNase He1 and the raw RNase He1 had similar optimum pH, base specificity, and specific activity (300 unit/mg); thus, we used E. coli-expressed RNase He1 in all subsequent experiments. The Cys81Ala mutant of RNase He1 (He1-C81A) was purified to homogeneity (Specific activity: 160 unit/mg, 60% of native RNase He1). The Cys5Ala mutant of RNase He1 (He1-C5A) and the double mutant of RNase He1 (He1-C5AC81A) did not exhibit RNase activity, as was confirmed via the enzymatic activity assays described above. This was not due to absence of expression, which we confirmed by performing Tricine-SDS-PAGE for proteins that were present in the culture medium.

Effect of RNase He1 on the Proliferation of Human Leukemia Cells

The effect of RNase He1 on the proliferation of HL-60 cells was studied. RNase Po1 showed remarkable proliferation inhibitory activity toward HL-60 cells (IC₅₀=0.2 µM), whereas RNase He1 had little effect.

Base Specificity

Only 3′-GMP and 2′,3′-GMP were formed as a result of RNA hydrolysis by RNase He1 and He1-C81A. Therefore, RNase He1 was classified as a guanylic acid-specific RNase; the base specificity of He1-C81A was identical (data not shown).

Temperature and pH Optima, and Heat Stability of Wild-Type and C81A Mutant RNase He1

The optimal temperature for catalysis by both RNase Po1 and RNase He1 (65–70°C) was higher than that for Aspergillus oryzae RNase T1 (the optimum for this enzyme being 50°C; Fig. 1a). The optimal pH for RNase He1, measured using RNA as substrate, was 4.5; this is lower than that of RNase Po1 and RNase T1 (pH 7.5; Fig. 1b). The pH and heat stability of RNase He1 were stable over a wide range (Figs. 1c, d).

Fig. 1. Enzymatic Properties of RNase He1 and He1-C81A

(a) Effects of temperature on the enzymatic activity of RNase He1, RNase Po1, and RNase T1. Enzyme activity was determined as described in the text using RNA as the substrate at pH 4.5 (RNase He1) or pH 7.5 (RNase Po1, RNase T1), with incubation for 3 min at the indicated temperatures. The reactions were performed in 0.01 M buffer (acetate–NaOH buffer for pH 4.5, Tris–HCl buffer for pH 7.5) containing 1 mg/mL bovine serum albumin (BSA) and 0.2 M NaCl to help stabilize RNase. Activity is expressed as percentage of maximum activity. ●, RNase He1; ○, RNase Po1; ▲, RNase T1. (b) Effects of pH on the enzymatic activity of RNase He1, RNase Po1, and RNase T1. Enzyme activity was determined as described in the text using RNA as a substrate. The reactions were performed in 0.05 M buffer (acetate–NaOH buffer for pH 5.5–6.0, Tris–HCl buffer for pH 6.5–8.5). Activity is expressed as the percentage of maximum activity at 37°C. ●, RNase He1; ○, RNase Po1; ▲, RNase T1. (c) Heat stability of RNase He1 and RNase Po1. RNase He1 and RNase Po1 in 0.1 M acetate–NaOH buffer at pH 4.5, Tris–HCl buffer for pH 7.5 (containing 1 mg/mL of BSA and 0.2 M NaCl) were heated for 5 min at various temperatures and then quickly chilled in ice water. Enzymatic activity was determined with RNA as substrate at pH 4.5 or pH 7.5. Activity was expressed as percentage of maximum activity at 50°C. ●, RNase He1; ○, RNase Po1. (d) pH stability of RNase He1 and RNase Po1. RNase He1 and RNase Po1 in 0.01 M Tris–HCl buffer (containing 1 mg/mL of BSA and 0.2 M NaCl) were incubated individually for 1 h at various pH values at 37°C, and then 4/10 volume of 0.5 M acetate buffer (pH 4.5) or Tris–HCl buffer (pH 7.5) was added to the reaction mixture. Enzymatic activity was determined using RNA as a substrate as described in the text. Activity is expressed as percentage of maximum activity at pH 6.0 or 7.0. ●, RNase He1; ○, RNase Po1. (e) Effects of temperature on the enzymatic activity of RNase He1 and He1-C81A. The experimental details are described in the legend for Fig. 1a. ●, He1-C81A; ○, RNase He1. (f) Heat stability of RNase He1 and the He1-C81A mutant. RNase He1 and He1-C81A in 0.1 M acetate–NaOH buffer pH 4.5 (containing 1 mg/mL of BSA and 0.2 M NaCl to help stabilize RNase) were heated for 5 min at various temperatures and then quickly chilled in ice water. Enzymatic activity was determined with RNA as a substrate at pH 4.5. Activity is expressed as percentage of maximum activity at 50°C. ●, He1-C81A; ○, RNase He1.

The optimum pH of He1-C81A was pH 4.5, which is identical to that of wild-type RNase He1. The optimum temperature for He1-C81A activity was 50°C, which is 20°C lower than that of wild-type RNase He1 (Fig. 1e). The heat stability of He1-C81A decreased rapidly above 80°C as measured by its RNase activity, whereas wild-type RNase He1 retained more than 80% of its RNase activity at 95°C (Fig. 1f).

Kinetic Parameters of Wild-Type and C81A Mutant RNase He1

The V_max of wild-type and He1-C81A were almost the same (23.0 units⁻¹), however, K_m value of He1-C81A (0.91 mg mL⁻¹) was higher than that of wild-type RNase He1 (0.56 mg mL⁻¹).

Temperature Dependence of the CD Spectra of Wild-Type and C81A Mutant RNase He1

The CD spectrum of native RNase He1 remained essentially unchanged up to 60°C, and then declined very rapidly with increasing temperature. In contrast, the spectrum of He1-C81A declined rapidly above 40°C (Figs. 2a, b). RNase He1 and He1-C81A were heated to 90°C, then cooled to 40°C or 25°C and their CD spectra were again measured. The CD spectrum of wild-type RNase He1 cooled to 40°C was essentially identical to that of unheated RNase He1, whereas the original spectrum of He1-C81A was not regained even after cooling to 25°C (Figs. 2c, d).

Fig. 2. Temperature Dependence of the CD Spectrum of RNase He1 and He1-C81A

(a), (b) CD spectra of RNase He1 and He1-C81A at 25–90°C. (a), RNase He1; (b), He1-C81A. , 25°C; ······, 40°C; –·–, 60°C; – – –, 70°C; —, 90°C. (c), (d) CD spectra of RNase He1 and He1-C81A cooled from 90°C to 40°C and 25°C. (c), RNase He1; (d), He1-C81A. , 25°C; —, 90°C; – –·, cooled at 40°C; ···, cooled at 25°C.

RNase He1 contains six cysteine residues, which directly correspond to those found in RNase Po1, whereas RNase T1 has only four cysteines¹⁴⁾ (Fig. 3). Thus, we predicted that RNase He1 would exhibit heat stability similar to that of RNase Po1. We previously reported the X-ray crystallographic structure of RNase Po1 and investigated the location of its three disulfide bonds (Cys7–Cys84, Cys9–Cys99, and Cys48–Cys82).¹⁵⁾ We therefore attempted to superimpose the disulfide bonds of RNase Po1 onto a homology model of RNase He1 (Fig. 4). The Cys7–Cys98 bond of RNase He1 (the Cys9–Cys99 in RNase Po1) corresponded to those found in other RNase T1 family members from fungi. The Cys5–Cys83 bond (the Cys7–Cys84 in RNase Po1) connects the β1-strand to the β6-strand. Because He1-C5A exhibited no RNase activity, this bond likely confers rigidity to the RNase He1 structure and is thus required for the active conformation of RNase. The Cys47–Cys81 bond of RNase He1 (the Cys48–Cys82 in RNase Po1) connects the loop between the β3- and β4-strands to the β6-strand; consequently, on this loop, there were base recognized amino acid residues. We overlaid the structure of C81A-mutant with that of wild type of RNase Po1, then, the end of β6-strand moved to outside (Fig. 4C). Therefore, by losing the Cys47–Cys81 bond, the affinity of He1-C81A for RNA may reduce (higher K_m value) because of the β6-strand may be slid; this may have lowered the specific activity of He1-C81A.

Fig. 3. Comparison of the Amino Acid Sequences of RNase He1 with RNase Po1, RNase T1

He1, RNase He1 from Hericium erinaceus; Po1, RNase Po1¹⁷⁾ from Pleurotus ostreatus; T1, RNase T1¹⁴⁾ from Aspergillus oryzae. Residues in common are enclosed in boxes. Numbers above the alignment show RNase He1 numbering. * Residues that contribute to the catalytic site; ^# Residues that contribute to base recognition. □; pyroglutamyl residue. - - -, the disulfide bonds of RNase T1; , the disulfide bonds of RNase Po1.

Fig. 4. The Location of Cysteine Residues of RNase He1 over the Tertiary Structure of RNase Po1

The figure was drawn with PyMOL (http://pymol.sourceforge.net). The location of the cysteine residues in RNase He1 over the structure of RNase Po1 are shown in yellow. The catalytic residues are shown in red. (A) Tertiary structure of RNase Po1. α-Helices and β-strands are marked α1 and β1–7, respectively. RNase Po1 is shown in purple (PDB ID: 3WHO). The amino acid numbers of RNase Po1 are shown followed by those of RNase He1 in parentheses. (B) The location of cysteine residues of RNase He1 over the tertiary structure of RNase Po1. The location of C81A was constructed by employing the web server SWISS-MODEL¹⁶⁾ with the crystal structure of RNase Po1 as template and shows in pink. (C) Structural overlay of C81A-mutant with that of wild type of RNase Po1. C81A-mutant colored green and wild type of RNase Po1 colored purple.

Similar to RNase T1 that has no anti-tumor activity, He1-C81A retains RNase activity but has decreased heat stability. Thus, this Cys47–Cys81 bond may be dispensable with respect to RNase activity, but potentially facilitates the renaturation observed after heat denaturation. Therefore, the Cys47–Cys81 bond confers greater heat stability to RNase He1 and may be involved in the overall stabilization of the enzyme, which potentially enhances its efficacy as an anti-tumor agent.

The lower optimal pH of RNase He1 may be in part explain why RNase He1 did not inhibit the proliferation of HL-60 cells. In the future, we will design the RNase He1 mutants that exhibit anti-tumor activities to confirm that hypothesis.

Acknowledgment

We thank Mr. K. Kawasumi for his technical assistance. The RNase He1-encoding cDNA has been deposited as DDBJ accession number AB429363.

REFERENCES

• 1) Ardelt W, Mikulski SM, Shogen K. Amino acid sequence of an anti-tumor protein from Rana pipiens oocytes and early embryos. Homology to pancreatic ribonucleases. J. Biol. Chem., 266, 245–251 (1991).
• 2) Titani K, Takio K, Kuwada M, Nitta K, Sakakibara K, Kawauchi H, Takayanagi G, Hakomori S. Amino acid sequence of sialic acid binding lectin from frog (Rana catesbeiana) eggs. Biochemistry, 26, 2189–2194 (1987).
• 3) Kamiya Y, Oyama F, Oyama R, Sakakibara F, Nitta K, Kawauchi H, Takayanagi Y, Titani K. Amino Acid Sequence of a Lectin from Japanese Frog (Rana japonica) Eggs. J. Biochem., 108, 139–143 (1990).
• 4) Sacco G, Drickamer K, Wool G. The primary structure of the cytotoxin alpha-sarcin. J. Biol. Chem., 258, 5811–5818 (1983).
• 5) Fang EF, Zhang CZ, Fong WP, Ng TB. RNase MC2: a new Momordica charantia ribonuclease that induces apoptosis in breast cancer cells associated with activation of MAPKs and induction of caspase pathways. Apoptosis, 17, 377–387 (2012).
• 6) Endo Y, Huber PW, Wool IG. The ribonuclease activity of the cytotoxin alpha-sarcin. The characteristics of the enzymatic activity of alpha-sarcin with ribosomes and ribonucleic acids as substrates. J. Biol. Chem., 258, 2662–2667 (1983).
• 7) Kobayashi H, Motoyoshi N, Itagaki T, Tabata K, Suzuki T, Inokuchi N. The Inhibition of Human Tumor Cell Proliferation by RNase Po1, a Member of the RNase T1 Family, from Pleurotus ostreatus. Biosci. Biotechnol. Biochem., 77, 1486–1491 (2013).
• 8) Kim SP, Kang MY, Choi YH, Kim JH, Nam SH, Friedman M. Mechanism of Hericium erinaceus (Yamabushitake) mushroom-induced apoptosis of U937 human monocytic leukemia cells. Food Funct., 2, 348–356 (2011).
• 9) Huang HC, Wang SC, Leu YJ, Lu SC, Liao YD. The Rana catesbeiana rcr gene encoding a cytotoxic ribonuclease. Tissue distribution, cloning, purification, cytotoxicity, and active residues for RNase activity. J. Biol. Chem., 273, 6395–6401 (1998).
• 10) Schägger H, von Jagow G. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem., 166, 368–379 (1987).
• 11) Irie M. Isolation and Properties of a Ribonuclease from Aspergillus saitoi. J. Biochem., 62, 509–518 (1967).
• 12) Lineweaver H, Burk D. The determination of enzyme dissociation constants. J. Am. Chem. Soc., 56, 658–666 (1934).
• 13) Ohgi K, Horiuchi H, Watanabe H, Iwama M, Takagi M, Irie M. Role of Asp51 and Glu105 in the Enzymatic Activity of a Ribonuclease from Rhizopus niveus. J. Biochem., 113, 219–224 (1993).
• 14) Takahashi K. The amino acid sequence of ribonuclease T-1. J. Biol. Chem., 240, 4117–4119 (1965).
• 15) Kobayashi H, Katsutani T, Hara Y, Motoyoshi N, Itagaki T, Akita F, Higashiura A, Yamada Y, Inokuchi N, Suzuki M. X-Ray Crystallographic Structure of RNase Po1 That Exhibits Anti-tumor Activity. Biol. Pharm. Bull., 37, 968–978 (2014).
• 16) Arnold K, Bordoli L, Kopp J, Schwede T. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics, 22, 195–201 (2006).
• 17) Nomura H, Inokuchi N, Kobayashi H, Koyama T, Iwama M, Ohgi K, Irie M. Purification and primary structure of a new guanylic acid specific ribonuclease from Pleurotus ostreatus. J. Biochem., 116, 26–33 (1994).