Induction of the Interferon-Inducible RNA-Degrading Enzyme, RNase ...

[RNA Biology 1:1, 21-27; May/June 2004]; ©2004 Landes Bioscience

Induction of the Interferon-Inducible RNA-Degrading Enzyme, RNase L, by Stress-Inducing Agents in the Human Cervical Carcinoma Cells Research Paper

KEY WORDS

ABBREVIATIONS

ibu

Bio

2',5'-oligoadenylate polyinosine:polycytosine hydrogen peroxide calcium chloride tumor necrosis factor-alpha nuclear factor kappa B inhibitor kappa B-alpha

The type I interferons induce three well-studied antiviral pathways: the Mx proteins, the double stranded RNA (dsRNA)-dependent protein kinase (PKR) and the 2’,5’oligoadenylate (2-5A) pathway.1 The 2-5A-dependent ribonuclease L (RNase L) is the key player in the 2-5A Pathway.2 The enzyme is interferon-inducible, it gets activated by unique 2’,5’-linked adenosine cofactors and cleaves RNA of both viral and cellular origin in virus-infected cells, leading to their apoptosis. Induction of RNase L has been well-studied in response to picornaviruses like encephalomyocarditis virus (EMCV), mengo virus and dsRNA.3 Overexpression of human RNase L cDNA4 caused antiviral response and apoptosis of mammalian cells.5 RNase L knockout (RNase L-/-) mice showed impaired apoptosis in the thymus and spleen.6 Cellular RNA-degradation by RNase L needs both induction of RNase L gene, synthesis of 2-5A cofactor and activation of RNase L by 2-5A. However, regulation of RNase L gene has not been very well studied. We investigated if RNase L is induced by stress-inducing agents through cellular pathways other than antiviral response. This could link RNA-degradation and apoptosis by RNase L to stress-response. In the present study, we report that human RNase L is induced in response to various stress-inducing agents such as double-stranded RNA [poly(I:C)], a synthetic analogue of intermediates of replicating RNA viruses, calcium chloride, that induces membrane- and osmotic stress, hydrogen peroxide which causes oxidative stress, chemotherapeutic drugs and the proinflammatory cytokine, TNF that cause apoptosis of cancer cells. We propose that RNase L is a broad range stress-responsive gene and RNA-degradation by RNase L is involved in cellular stress-response.

nd

ACKNOWLEDGEMENTS

INTRODUCTION

es

2-5A poly(I:C) H2O2 CaCl2 TNF-α NF-κB IkB-α

sci

interferon, RNase L, RNA-degradation, dsRNA, chemotherapeutics, H2O2, CaCl2, TNF, apoptosis

No tD ist r

Previously published online as a RNA Biology E-publication: http://www.landesbioscience.com/journals/rnabiology/abstract.php?id=896

Do

Received 03/02/04; Accepted 04/05/04

ce.

*Correspondence to: Pramod C. Rath; Molecular Biology Laboratory; School of Life Sciences; Jawaharlal Nehru University; New Delhi-110067, India; Tel.: +91.11.26704525; Fax: +91.11.26717586; Email: [email protected]

RNA-degradation is one of the fundamental mechanisms of interferon (IFN)-inducible antiviral response in mammalian cells. This is primarily brought about by the IFNinducible 2’,5’-oligoadenylate (2-5A)-cofactor dependent ribonuclease L (RNase L). RNase L also functions as a tumor suppressor gene in case of prostate cancer due to its role in apoptosis. We report that RNase L is induced by stress-inducing agents such as double-stranded RNA [poly(I:C)], chemotherapeutic drugs, hydrogen peroxide (H2O2), calcium chloride (CaCl2) and tumor necrosis factor-α (TNF) in the human cervical carcinoma (HeLa) cells. The level of RNase L was not detected in the untreated cells. Induction of RNase L by such stress-inducing agents correlated with degradation of cellular RNA, fragmentation of chromatin-DNA and induction of apoptosis. We checked the stressinducible transcription factor, nuclear factor kappa B (NF-κB), which was persistently activated by cycloheximide but not by other agents after 24 hours indicating no role of NF-κB in the RNase L-induction. However, as expected, TNF-induced NF-κB activity was stimulated within 10-30 minutes through degradation of IκB-α. Our results strongly suggest that the IFN-inducible RNase L is induced by a broad range of stress-inducing signals such as double-stranded RNA (dsRNA) produced during viral infection, membrane- and osmotic shock caused by CaCl2 and oxidative stress induced by H2O2, inflammation stimulated by TNF-α and chemotherapy. Thus, in addition to its antiviral function, the IFN-inducible RNase L may play an important role during stress-response through RNA-degradation and apoptosis.

en

Molecular Biology Laboratory; School of Life Sciences; Jawaharlal Nehru University; New Delhi, India

te

ABSTRACT

Mitali Pandey Gagan Deep Bajaj Pramod C. Rath*

©

20

04

La

The kind gifts of human (pZC5) and murine (pZB1) RNase L cDNA clones, anti-human RNase L monoclonal antibody from Prof. R.H. Silverman, Cleveland Clinic Foundation, Cleveland, OH; Junior and Senior Research Fellowships to MP, GDB from the University Grants Commission (U.G.C.); financial support to the School of Life Sciences under the “Center of Advanced Study (CAS)”, “University of Potential for Excellence” programs of the U.G.C. and the FIST-program of the Department of Science and Technology (D.S.T.), Government of India are gratefully acknowledged.

www.landesbioscience.com

RNA Biology

21

INDUCTION OF RNASE L BY STRESS-INDUCING AGENTS

MATERIALS AND METHODS Cells, Treatments and Cell Extracts. All reagents were of tissue culture and molecular biology grade and were purchased from Sigma Chemical Co., USA and Life Technology, USA; γ(32P)ATP (3000 Ci/mmole, Amersham, USA), anti-IκB-α- and HRP-conjugated goat anti-rabbit IgG antibodies (Santa Cruz Biotech., USA) were used for the study. The human cervical adenocarcinoma HeLa S3 cells (A.T.C.C., USA) were grown in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% Fetal Calf Serum (FCS), Penicillin (100 U/ml), Streptomycin (100 mg/ml) at 37˚C and 5% CO2. Cells (0.5–1.0 x 106) were treated with Cycloheximide (50 µg/ml), poly(I:C) (25 µg/ml), clinical grade chemotherapeutic drugs (5 µM of cisplatin, doxorubicin, vinblastine and vincristine), 5 µM and 1 mM H2O2, 100 nM CaCl2 and 1 nM and 10 nM recombinant human TNF (5 x 107 U/mg, a kind gift from Genentech Inc., USA) in 6 well plates for 24 hours, cells were washed in Phosphate Buffered Saline (PBS) and lysed in 20 µl Lysis Buffer I (PBS with 1% Triton X-100, 1 mM DTT, 0.5 mM PMSF, 1 µg/ml each of Leupeptin and Aprotinin, 0.5 mg/ml Benzamidine and 1 mM Sodium Ortho-Vanadate) and centrifuged to prepare the whole cell extract.7 After parallel treatments, cells were lysed in 100 µl Lysis Buffer II (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, pH 8.0, 0.1 mM EGTA, pH 7.0, 1 mM DTT, 1 mM PMSF, 2 µg/ml Aprotinin and Leupeptin, 0.5 mg/ml Benzamidine), centrifuged to collect the supernatant (cytoplasmic extract), the nuclear pellet was extracted in the 25 µl Nuclear Extraction Buffer (20 mM HEPES, pH 7.9, 400 mM NaCl, 1 mM EDTA, pH 8.0, 1 mM EGTA, pH 7.0, 1 mM DTT, 0.5 mM PMSF, 2 µg/ml Aprotinin and Leupeptin, 0.5 mg/ml Benzamidine) to prepare the nuclear extract.8,9 Electrophoretic Mobility Shift Assay (EMSA). EMSA for NF-κB was carried out by using 16 fmole of a 32P labeled 45 bp HIV-LTR containing two NF-κB binding sites and 8 µg of the nuclear extract, the DNA-protein complex was resolved from the free-label in a 7.5% native polyacrylamide gel as described earlier.8,9 Western Blot Analysis for IκB-α and RNase L. Western blot analysis was carried out10 by using 50 µg of the whole cell extract and an antihuman RNase L monoclonal antibody11 for measuring RNase L-induction or 30 µg of the cytoplasmic extract and an anti-human IκB-α polyclonal antibody for measuring IκB-α levels. The secondary antibody was HRP-conjugated anti-mouse IgG for the RNase L-blot and HRP-conjugated anti-rabbit IgG for the IκB-α-blot. ECL reagent (Amersham) was used to develop the blots. Cellular RNA-degradation Assay. Total cellular RNA was purified from the cells by a LiCl-Urea extraction method12 after the treatments and the purified RNA was analyzed for degradation of the ribosomal RNAs by 1.2% native agarose gel electrophoresis. Chromatin-DNA-fragmentation Assay. Chromatin-DNA from the cells was purified by a method of gentle lysis followed by removal of RNAs and proteins after the treatments.13 Fragmentation of the chromatin-DNA was analysed by native 1.7% agarose gel electrophoresis. Morphology-based Apoptosis Assay. Cells were fixed and stained with Giemsa stain after the treatments.14 The cells were observed under a phasecontrast microscope and based on the morphological parameters clearly apoptotic cells were counted from 8 fields in each well. These included condensation of the nuclear chromatin-material as well as flattening and blebbing of the cytoplasmic-content.

RESULTS The human cervical carcinoma (HeLa) cells were used for the study because they are responsive to fibroblast interferon (IFN-β) and IFN-β is a strong inducer of RNase L. In addition, we investigated whether the stressinducing agents could cause induction of RNase L, RNA degradation and apoptosis in the cancer cell line. Induction of RNase L-expression. HeLa cells were treated with indicated concentrations of Cycloheximide, poly(I:C), Cycloheximide + poly (I:C), chemotherapeutic drugs (cisplatin, doxorubicin, vinblastine and vincristine), H2O2, CaCl2 and TNF for 24 hours. The cell extracts were prepared and

22

Figure 1. Induction of RNase L expression in the HeLa cells after treatment with various stress-inducing agents. HeLa cells were treated with various agents for 24 hours and 50 µg whole cell extract per lane was analysed for RNase L expression by Western blot analysis using a monoclonal antibody raised against the C-terminal ribonuclease domain of the human RNase L. Lane 1, untreated control (UC); lane 2, 50 µg/ml cycloheximide (CHX); lane 3, 25 µg/ml poly(I:C) (pI.C); lane 4, 50 µg/ml cycloheximide + 25 µg/ml poly(I:C); lane 5-8, 5 µM chemotherapeutics: lane 5, cisplatin (CP); lane 6, doxorubicin (DOX); lane 7, vinblastine (VB); lane 8, vincristine (VC); lane 9-10, hydrogen peroxide (H2O2) 5 µM and 1mM, respectively; lane 11, 100 nM calcium chloride (CaCl2); lane 12, 10 nM TNF; lane 13 and 14, extracts from E. coli cells (0.3 O.D.600 nm) expressing recombinant human RNase L from 1-741 a.a. (hRNase L) and recombinant mouse RNase L from 1-656 a.a. (mRNase L), respectively.

probed for RNase L-expression at protein level by western blot analysis using an anti-human RNase L monoclonal antibody raised against the C-terminal RNA binding and ribonuclease domain (amino acids 711–720) of the 741 amino acids human RNase L.11 As shown in Figure 1, in untreated HeLa cells (lane 1, Fig. 1) no RNase L-expression was detected, treatment with the protein synthesis inhibitor, Cycloheximide did not induce RNase L (lane 2, Fig. 1). On addition of poly(I:C), there was substantial induction of RNase L (lane 3, Fig. 1) which was significantly inhibited by Cycloheximide (lane 4, Fig. 1) indicating that the dsRNA-mediated induction of RNase L needed fresh protein synthesis. However, Cycloheximide was not sufficient to completely abrogate the poly(I:C)-mediated RNase L-induction (compare lanes 3 and 4, Fig. 1). Interestingly, treatment of the cells with different chemotherapeutic drugs produced varying levels of RNase L-induction (lanes 5–8, Fig. 1), cisplatin (lane 5, Fig. 1) and doxorubicin (lane 6, Fig. 1) strongly induced RNase L to high levels. Vinblastine (lane 7, Fig. 1) and vincristine (lane 8, Fig. 1) also induced RNase L but to low levels. Hydrogen peroxide (H2O2) at concentrations of 5 µM (lane 9, Fig. 1) and 1 mM (lane 10, Fig. 1) strongly induced RNase L-expression to very high levels. This induction was comparable to the induction of RNase L by 100 nM calcium chloride (CaCl2, lane 11, Fig. 1) or by 10 nM TNF (lane 12, Fig. 1). As a reference, extracts from the E. coli BL21(DE3) pLys E cells expressing the full length (741 amino acid) recombinant human RNase L protein, from the human RNase L cDNA (ZC5; see ref. 4) in a T7 RNA polymerase-driven pRSET A prokaryotic expression vector context, was loaded in lane 13 of Figure 1 which showed an approximately 97.6 kDa band along with its degradation products. A higher band at approximately 160 kDa (possibly RNase L-dimer) was also detected in lane 13. In contrast to this, extracts from the E. coli BL21(DE3) pLys S cells expressing the partial (656 amino acid truncated protein lacking the C-terminal substrate-binding and catalytic part) recombinant murine RNase L protein from the mouse RNase L cDNA (pZB1; see ref. 4) was also loaded in parallel (lane 14, Fig. 1), but no RNase L protein was detected in this case by the anti-RNase L-monoclonal antibody since it is specific for the human RNase L. This was expected as the corresponding 89 amino acids in the C-terminal region, against which the

RNA Biology

2004; Vol. 1 Issue 1

INDUCTION OF RNASE L BY STRESS-INDUCING AGENTS

A

B Figure 3. Degradation of ribosomal RNA in the HeLa cells treated with various stress-inducing agents. Total RNA was extracted from HeLa cells 24 hours after the treatments (same as Fig. 1 except for lane 11, 1nM TNF), analysed by 1.2% native agarose gel electrophoresis and ethidium bromide (0.5 µg/ml) staining. 28S and 18S ribosomal RNAs are marked. Degraded RNA indicates degradation of the cellular RNA.

C

Figure 2. NF-κB activity in the HeLa cells treated with the stress-inducing agents. (A) DNA binding activity of NF-κB in the nucleus was measured by EMSA, (B) IκB-α level in the cytoplasm was measured by western blot analysis after treatment of HeLa cells with 1 nM TNF for 5-60 min, (C) nuclear NF-κB DNA binding activity was measured in HeLa cells treated with the stress-inducing agents (same as in Fig. 1). Lane 1, free oligo (FO); lane 10, 5 µM H2O2. Arrows in (C) indicate the higher NF-κB complexes.

monoclonal antibody was raised, was absent in the mouse RNase L cDNA (pZB1). Although, in cells the active form of RNase L is a dimer, in HeLa cell extracts we detected monomeric form of RNase L. The dimeric form of RNase L is expected to be converted into the monomeric form of RNase L under SDS-PAGE followed by western blot conditions. This experiment showed that the HeLa cells used in the study contained undetectable level of RNase L and the cellular RNase L protein level was strongly induced by treatment with the stress-inducible agents. DNA Binding Activity of NF-κB. To study any possible correlation of the RNase L-induction with activation of the stress-inducible transcription

www.landesbioscience.com

factor NF-κB in the HeLa cells, we performed electrophoretic mobility shift assay (EMSA) using nuclear extracts from untreated cells and cells treated with the agents. A 32P-labeled 45 bp synthetic oligonucleotide representing the HIV-1 long terminal repeat (LTR) sequence with two NF-κB binding sites was used for the study as described earlier.10 As shown in Figure 2A, untreated cells did not show NF-κB activity (0 time, lane 1, Fig. 2A), TNF (1 nM) increasingly activated NF-κB at 10, 15, 30 min and it declined at 60 min (lanes 2–6, Fig. 2A). Consistent with the increase in DNA-binding activity of NF-κB in the nucleus, there was a corresponding decrease in the cytoplasmic level of IκB-α, the inhibitory protein for NF-κB (Fig. 2B). The IκB-α level remained more or less steady until 10 min, it decreased by 30 min and again increased at 60 min, the latter indicating synthesis of fresh IκB-α protein due to the activated-NF-κB. This increased level of IκB-α in the cytoplasm resulted in the decreased nuclear NF-κB activity at 60 min. As expected, NF-κB was inducible by TNF in the HeLa cells. NF-κB activity was measured after 24 hour-treatment of the cells with the RNase L-inducing agents shown in Figure 1. As shown in Figure 2C, untreated cells showed basal NF-κB activity (lane 2, Fig 2C) and there was a strong induction of NF-κB by Cycloheximide (lane 3, Fig 2C). The three NF-κB complexes detected here may be homodimers of p65/p65 and p50/p50 and p65/p50 heterodimer based on their relative mobility. Treatment with dsRNA [poly(I:C)] showed basal NF-κB activity (lane 4, Fig 2C). Interestingly, the Cycloheximideactivated NF-κB complex was abrogated by the Cycloheximide + poly(I:C) treatment (lane 5, Fig 2C). The cancer chemotherapeutic drugs such as cisplatin (lane 6, Fig 2C), doxorubicin (lane 7, Fig 2C) and vincristine (lane 9, Fig 2C) all induced low NF-κB activity which was even lower for vinblastine (lane 8, Fig 2C) and H2O2 (lane 10, Fig 2C). Thus, doxorubicin showed maximum induction of NF-κB which decreased in the following order : > cisplatin > vincristine > H2O2 > vinblastine. CaCl2-treated cells showed negligible NF-κB (lane 11, Fig 2C), so also TNF-treatment after 24 hours (lane 12, Fig 2C). It is possible that some of these agents might have activated NF-κB at an earlier time point but NF-κB activity did not correlate with RNase L-expression at 24 hours after treatment. RNA-degradation by RNase L-inducing Agents. The effect of the RNase L-induction was assayed in terms of degradation of cellular ribosomal RNAs in native agarose gel (Fig. 3) as described earlier.15 Untreated cells (lane 1,

RNA Biology

23

INDUCTION OF RNASE L BY STRESS-INDUCING AGENTS

Figure 4. Chromatin-DNA fragmentation in the HeLa cells treated with various stress-inducing agents. HeLa cells were treated with various agents [same as Fig. 1 except for lane 10, 50 nM CaCl2 and lane 12 (positive control), 1 nM TNF + 50 µg/ml CHX]. Genomic DNA was isolated from 1 × 106 cells and analysed by 1.8% agarose gel electrophoresis and ethidium bromide (0.5 µg/ml) staining. Fragmented genomic DNA was observed. Lane 1, negative control.

Fig. 3) showed distinct bands of 28S and 18S rRNAs after 24 hours. In sharp contrast, there were no distinct rRNA bands after 24 hour-treatment with the stress-inducing agents (lanes 2–11, Fig. 3). In most lanes, the 28S rRNA band was more or less absent due to RNA-degradation but the 18S

A

B

rRNA band showed incomplete degradation. In cells treated with Cycloheximide (lane 2, Fig. 3), poly(I:C) (lane 3, Fig. 3) and Cycloheximide + poly(I:C) (lane 4, Fig. 3), the 28S rRNA was completely degraded. However, weak 18S rRNA band was still visible. Cisplatin-treated cells (lane 5, Fig. 3) showed neither 28S nor 18S rRNA band, so also vinblastine-treated cells (lane 7, Fig. 3). Cells treated with doxorubicin (lane 6, Fig. 3), vincristine (lane 8, Fig. 3), H2O2 (lane 9, Fig. 3) and CaCl2 (lane 10, Fig. 3) showed complete degradation of 28S rRNA and residual 18S rRNA. Treatment of cells with 1 nM TNF-α (lane 11, Fig. 3) showed degradation of both the 28S and 18S rRNAs. Thus the ribosomal RNA-degradation correlated with the RNase L-induction in the cells treated with the stressinducing agents. Chromatin-DNA Fragmentation by RNase L-inducing Agents. Chromatin-DNA fragmentation was studied as a parameter of apoptosis in the cells treated with the stress-inducing agents for 24 hours. Untreated cells (lane 1, Fig. 4) as well as Cycloheximide-treated cells (lane 2, Fig. 4) showed no DNA-laddering or appearance of any low molecular weight DNA as compared to the TNF + Cycloheximide treatment (lane 12, positive control for DNA fragmentation during apoptosis, Fig. 4). Little DNA-laddering was observed in poly (I:C)-treated cells (lane 3, Fig. 4). Cells treated with Cycloheximide+ poly(I:C) (lane 4, Fig. 4), cisplatin (lane 5, Fig. 4), doxorubicin (lane 6, Fig. 4), vinblastine (lane 7, Fig. 4) and vincristine (lane 8, Fig. 4), 50 nM- and 100 nM CaCl2 (lanes 10 and 11, respectively, Fig. 4) showed distinct ladders of the fragmented chromatin-DNA. H2O2-treated cells showed almost no DNA-laddering. Undegraded high molecular weight genomic DNA was observed in untreated cells as well as in cells after all the treatments indicating the fraction of intact chromatin-DNA. Apoptosis by RNase L-inducing Agents. Apoptosis was assessed by judging the microscopic morphological parameters of the cells i.e., condensation of nuclear chromatin-material, cytoplasmic-flattening and blebbing after 24-hour treatment with the stress-inducing agents. Cells showed morphological features characteristic of apoptosis such as cytoplasmic shrinkage, nuclear condensation, blebbing and deeply stained rounded cells as described earlier.16 Cells with apoptotic morphology were counted from eight fields from each well. Relative levels of apoptosis was expressed as fold increase in the number of apoptotic cells over the untreated control, this showed varying extents of apoptosis by the agents. Figure 5 shows number of apoptotic cells in 8 random fields per well after 24-hour treatment (Fig. 5A), fold increase in apoptosis (Fig. 5B) and

Figure 5. Apoptosis in the HeLa cells treated with various stress-inducing agents. HeLa cells were treated with the agents (same as Fig. 1 except for lanes 11 and 12, 1 hour and 24 hours of 1nM TNF-treatment, respectively) and stained with Geimsa stain. Apoptotic cells were scored under a phase-contrast microscope on the basis of their morphological features. (A) Total number of apoptotic cells from 8 fields, (B) Fold increase in the number of apoptotic cells in the treated vs. untreated control cells and (C) Representative morphological features characteristic of apoptotic cells at 1250 x magnification as described in results. UC, untreated control; CHX, cycloheximide; pI.C, poly(I.C); CHX + pI.C, cycloheximide and poly(I.C); CP, cisplatin; DOX, doxorubicin; VB, vinblastin; VC, vincristine; H2O2, hydrogen peroxide; CaCl2, calcium chloride and TNF, recombinant human tumor necrosis factor-α.

C

24

RNA Biology

2004; Vol. 1 Issue 1

INDUCTION OF RNASE L BY STRESS-INDUCING AGENTS

⊥

Figure 6. A schematic representation of RNase L-induction by various stress-inducing agents. Double stranded RNA (poly I:C), chemotherapeutic drugs (cisplatin, doxorubicin, vinblastin and vincristine), membrane- and osmotic stress by CaCl2, oxidative stress by H2O2 and the inflammatory cytokine (TNF) all induced RNase L protein which correlated with RNA degradation, DNA fragmentation and apoptosis. +, activation/stimulation; , suppression; dotted lines indicate information already reported; solid lines indicate results from this study.

morphological features of the apoptotic cells (Fig. 5C). As expected, untreated cells exhibited very low level of apoptosis, whereas, Cycloheximide, poly(I:C), Cycloheximide + poly(I:C) treatments showed less number (100–200) of apoptotic cells with intense nuclear condensation and it caused a 5-10x fold increase in apoptosis. The chemotherapeutic drugs clearly demonstrated greater extents of apoptosis, e.g., cisplatin (10x fold), doxorubicin (25x fold), vinblastine (40x fold), vincristine (35x fold), respectively. The cellular features observed were deeply-stained condensed nuclear material, rounded morphology, blebbing and release of apoptotic granules for vinblastine and vincristine; flattened cytoplasm with condensed nucleus for doxorubicin and cytoplasmic vacuoles for cisplatin. H2O2 and CaCl2 showed a 10–15x fold increase in apoptosis and cytoplasmic vacuoles and bulging nuclear material, respectively. TNF treatment caused 15x (1 hour) and 30x (24 hour) fold increase in apoptosis, respectively. By 1 hour, it showed larger nuclear area with cytoplasmic vacuoles and nuclear condensation at the later time point (24 hour). The morphological changes and the extent of apoptosis after the treatments (Fig. 5) correlated with the extent of DNA-fragmentation (Fig. 4), e.g., vinblastine, vincristine, CaCl2, TNF + Cycloheximide treatments showed increased apoptosis as well as DNA-laddering. Thus, in the HeLa cells, induction of RNase L-expression (Fig. 1) correlated with RNA-degradation (Fig. 3), chromatin-DNA fragmentation (Fig. 4) and apoptosis (Fig. 5) but not with NF-κB activity (Fig. 2C) after 24 hour treatment with the stress-inducing agents. Interestingly at 24 hours after the treatments, the double-stranded RNA strongly induced expression of the RNase L-protein which was significantly inhibited by Cycloheximide and Cycloheximide strongly induced the nuclear DNA binding activity of NF-κB which was completely inhibited by the double-stranded RNA in the HeLa cells.

DISCUSSION The present study assigns a novel role to RNase L, i.e., RNase L is a broad range stress-responsive factor. Induction of RNase L expression by the stress-inducing agents correlated with RNAdegradation, DNA-fragmentation and apoptosis. As shown in the www.landesbioscience.com

schematic representation in Figure 6, all arrows in the scheme represent information derived from this study, while all dotted lines represent information reported earlier. This scheme highlights RNA-degradation during stress-response and apoptosis. It also opens the possibility of induction of RNase L gene for apoptosis of cancer cells. RNase L is a tumor suppressor. Recently, RNase L gene has been identified as the human prostate cancer 1 (HPC1) locus that is involved in susceptibility to prostate cancer. Germ line mutations in the RNase L gene in families showing linkage with human prostate cancer gene (HPC1) region has been reported.17 Recently, implications for RNase L in prostate cancer has been reviewed.18 Mutations in the RNase L gene leading to decreased RNase-activity may be linked to decreased apoptosis and development of prostate cancer. In Ashkenazi Jewish men, the RNase L 471 delAAAG allele and prostate cancer have been linked.19 In hereditary prostate cancer, RNase L mutations have been reported.20 This indicates expression of RNase L in cells should be involved with control of cell growth regulation as well as tumor suppression. Further, the cellular stress response kinase, the c-jun N-terminal kinase (JNK) has been linked to RNase L.21 By linking RNase L-expression with cellular stress-response induced by a number of different agents, our study suggests that RNase L is a broad range stress-inducible factor. In the present study, we investigated if expression of RNase L can be induced in human cancer cells by stress-inducing agents. The human cervical cancer (HeLa) cells are responsive to IFN-β which is a strong inducer of RNase L. Interferons are antiproliferative and proapoptotic cytokines. A number of stress-inducing agents were selected for the study, they were double-stranded RNA, chemotherapeutic drugs, hydrogen peroxide, calcium chloride, and tumor necrosis factor-α. Expression and activation of RNase L in mammalian cells has been implicated with antiviral response. Induction of RNase L expression is usually a signal for RNA degradation during apoptosis in mammalian cells following virus infection. Our study shows that induction of RNase L expression is also involved in stress-response. The induction of RNase L expression was correlated with RNA degradation, therefore, it indicated that the induced level of RNase L was active in the cells. Poly(I:C) is a synthetic dsRNA that mimics replicating intermediates of viral RNA-genome. It strongly induced RNase L-expression, RNA-degradation and DNA-fragmentation but caused lower level of apoptosis in the HeLa cells. Earlier reports with NIH 3T3 and L929 cells transfected with the human RNase L cDNA showed increased RNase L-dependent apoptosis upon addition of exogenous 2-5A but not poly(I:C).5 We observed that induction of RNase L by poly(I:C) was abrogated by Cycloheximide suggesting the requirement for de novo protein synthesis for the induction. This suggests that the induced level of RNase L expression may require synthesis of novel protein factor(s) inducible by dsRNA. This may link dsRNA-activated factors (DRAFs) 1, 222 to induction of RNase L. The mechanism by which RNase L gene is induced at transcriptional level is unknown. Since poly(I:C) induces PKR-pathway which also activates NF-κB, we checked if there is any possible correlation between RNase L-induction and NF-κB activity after the cells were treated with poly(I:C) for 24 hours. Except for Cycloheximide, other agents did not show strong NF-κB activity at 24 hours after the treatments. However, it is possible that NF-κB might have been activated at an earlier time point. The persistent induction of DNA binding activity of NF-κB by Cycloheximide corroborated with earlier reports that protein synthesis inhibitors such as Cycloheximide and Anisomycin are potent inducers of NF-κB and IL-6.23 Interestingly, we found

RNA Biology

25

INDUCTION OF RNASE L BY STRESS-INDUCING AGENTS

that poly(I:C) strongly inhibited this persistent NF-κB induction by Cycloheximide. In turn, Cycloheximide significantly inhibited the strong induction of RNase L expression by poly(I:C). This reciprocal inhibitory loop of poly(I:C) and Cycloheximide in terms of induction of NF-κB and RNase L needs further investigation. We used two groups of clinical grade chemotherapeutic agents that are widely used in the treatment of various types of human cancers. Cisplatin and doxorubicin are genotoxic drugs that interfere with DNA replication and cause damage to genomic DNA and RNA, while vinblastine and vincristine interfere with the mitotic spindle apparatus. Interestingly, these chemotherapeutic agents showed a general pattern of drug-mediated RNase L-induction. Cells treated with these drugs caused RNase L-induction, rRNA-degradation and apoptosis. This is the first report of induction of RNase L expression by chemotherapeutic drugs. Reports of the combinatorial therapy of IFN-β and tamoxifen for cancer treatment24 may involve a similar drug-induced mechanism of RNase L-induction, IFN-β being a strong inducer of RNase L. Thus, RNase L-mediated RNA-degradation induced by drugs and stress-inducing agents may lead to apoptosis. Hydrogen peroxide, a potent inducer of oxidative stress also showed substantial induction of RNase L and rRNA-degradation. This is comparable to an earlier report of induction of RNase L after microwave radiation.25 Doxorubicin, is a potent inducer of hydrogen peroxide, which ultimately leads to induction of eNOS in endothelial cells.26 In our experiments, doxorubicin strongly induced RNase L. Calcium chloride is another potent inducer of apoptosis following membrane- and osmotic stress which involves various Ca++ and Mg++ dependent endonucleases that are mobilized by cytosolic Ca++. Although, recombinant RNase L does not directly require Ca++ for either 2-5A binding or ribonuclease function,27 our study is the fist report of RNase L induction by calcium chloride. It will be interesting to find out how calcium-induced stress leads to induction of RNase L. Analysis of microarray-gene expression pattern during cisplatinmediated nephrotoxicity has suggested occurrence of apoptosis and perturbation of intracellular calcium homeostasis.28 Our data showed that treatment of HeLa cells with CaCl2 resulted in substantial induction of RNase L and it also caused apoptosis. However, the exact mechanism of RNase L-induction by calcium chloride needs further investigation. Interestingly, the inflammatory cytokine, TNF also substantially induced RNase L expression after 24 hour-treatment but NF-κB activity was not detected in the nucleus at this time point, although, TNF (1 nM) showed degradation of IκB-α and activation of NF-κB within 10–30 min in these cells. This is the first report of RNase L-induction by TNF. Activation of RNase L by a combination of IFN-γ and TNF has been reported earlier.29 As reported, HeLa cells lack the p75-TNF receptor-2 (TNFR2),30 therefore, induction of RNase L by TNF should involve the p55-TNFR1-associated (apoptotic) pathway.31 RNase L-knockout (RNase L-/-) mice showed decreased apoptosis in the spleen and thymus tissues in vivo.6 It has been shown that the apoptotic function of RNase L is accompanied by release of Cytochrome C from the mitochondria leading to activation of caspase 3.32 It would be interesting to study the mechanism of TNF-mediated signaling leading to RNase L-induction. Apoptosis caused by various stress-inducing agents, such as serum starvation, ceramide, arsenite and peroxide treatment, have been shown to be mediated by the antiviral, dsRNA-dependent kinase, PKR.33 These functions invoke activators other than dsRNA and include RAX, the ubiquitously expressed PKR-associated protein in mouse,34 PACT, the protein activator of PKR,35 and p67, a cellular 26

glycoprotein36 and possibly others. In this context, it is appropriate to imagine RNase L as more than an antiviral molecule, linked to stress-response and apoptosis. Our data provide evidence that RNase L is a stress-responsive gene in mammalian cells. However, RNase L differs from PKR in its requirement for 2-5A cofactor for dimerization and enzymatic activation. Thus, signaling cascades leading to induction of RNase L should logically be linked to production of 2-5A also. Therefore, caspase-mediated apoptosis, induction of RNase L and RNA degradation may be inter-linked. The association of RNA metabolism with the process of apoptosis is still not clearly understood. It is possible that PKR is involved in translational inhibition, whereas RNase L is involved in degrading the transcripts to ensure complete annihilation of translation. It will be interesting to study the TNF-signaling events which cause induction of NF-κB within 10–30 min but induction of RNase L by 24 hours, possibly a late response to the cytokine. For example, TNF-inducible NF-κB is essential for transcription of IFN-β gene and IFN-β is a potent inducer of RNase L. Recently, we have shown that expression of the recombinant human RNase L caused RNA degradation and inhibition of cell growth in E. coli.37 This indicated that the human RNase L protein was biochemically active in the prokaryotic cells. RNase L is one of the many proteins involved in the process of apoptosis but it is unique in terms of inducible RNA degradation. In the present study, we observed a correlation between RNase L-induction and RNA degradation during apoptosis after treatment with various stress-inducing agents in the HeLa cells. Such RNase L-induction by the stress-inducing agents may occur at transcriptional or posttranscriptional level. To establish RNase L as a direct cause for apoptosis under such conditions, further investigation is necessary. Studying apoptosis by such stress-inducing agents in the cells/tissues from RNase L knock out mice may be informative in this aspect. References 1. Stark GR, Kerr IM, Williams BRG, Silverman RH, Schreiber RD. How cells respond to interferons. Annu Rev Biochem 1998; 67:227-64. 2. Silverman RH, Krause D. In: Clemens MJ, Morris AG, Gearing AJH, eds. “Lymphokines and Interferons: A Practical Approach.” Oxford, England: IRL Press, 1997:149-93. 3. Silverman RH, Cirino NM. RNA decay by the interferon-regulated 2-5A system as a host defense against viruses. mRNA Metabolism and Post-Transcriptional Gene Regulation. Wiley-Liss, Inc., 1997:295-309. 4. Zhou A., Hassel B, Silverman RH. Expression cloning of the 2-5A-dependent RNase: A uniquely regulated mediator of interferon action. Cell 1993; 72:753-65. 5. Castelli JC, Hassel BA, Wood KA, Li XL, Amemiya K, Dalakas MC, et al. A study of the interferon antiviral mechanism: Apoptosis activation by the 2-5A system. J Exp Med 1997; 186:967-72. 6. Zhou A., Paranjape JM., Brown TL, Nie H, Naik S, Dong B, et al. Interferon action and apoptosis are defective in mice devoid of 2’,5’-oligoadenylate-dependent RNase L. EMBO J 1997; 16:6355-63. 7. Walczak H, Bouchon A, Stahl H, Krammer PH. Tumor necrosis factor-related apoptosisinducing ligand retains its apoptosis-inducing capacity on Bcl-2- or Bcl-xL-overexpressing chemotherapy-resistant tumor cells. Cancer Res 2000; 60:3051-7. 8. Schreiber E, Matthias P, Muller MM, Schaffner W. Rapid detection of octamer binding proteins with ‘mini-extracts’, prepared from a small number of cells. Nucl. Acids Res 1989; 17:6419. 9. Chaturvedi M.M, Lapushin R,. Aggarwal BB, Tumor necrosis factor and lymphotoxin. Qualitative and quantitative differences in the mediation of early and late cellular response. J Biol Chem 1994; 269:14575-83. 10. Rath PC, Aggarwal BB. Antiproliferative effects of IFN-α correlate with the downregulation of nuclear factor-κ B in human burkitt lymphoma daudi cells. J Interferon Cytokine Res 2001; 21:523-8. 11. Dong B, Silverman RH. 2-5A-dependent RNase molecules dimerize during activation by 2-5A. J Biol Chem 1995; 270:4133-7. 12. Auffray C ,Rougeon F. Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur J Biochem 1980; 107:303-14. 13. Loweth AC ,Morgan NG. Methods for the study of NO-induced apoptosis in cultured cells. Methods Mol Biol 1998; 100:311-20. 14. Freshney RI. In: Alan R, ed. Culture of Animal cells: A Manual of Basic Technique. 2nd ed. Liss, New York: 1987.

RNA Biology

2004; Vol. 1 Issue 1

INDUCTION OF RNASE L BY STRESS-INDUCING AGENTS

15. Silverman RH, Skehel JJ, James TC, Wreschner DH ,Kerr IM. rRNA cleavage as an index of (A2’p)nA activity in interferon-treated encephalomyocarditis virus-infected cells. J Virol 1983; 46:1051-5. 16. Rath PC, Aggarwal BB. TNF-induced signaling in apoptosis. J Clin Immunol 1999; 19:350-64. 17. Carpten J, et al. Germline mutations in the ribonuclease L gene in families showing linkage with HPC1. Nat Genet 2002; 30:181-4. 18. Silverman RH. Implications for RNase L in prostate cancer biology. Biochemistry 2003; 42:1805-12. 19. Kotar K, Hamel N, Thiffault I, Foulkes WD. The RNASEL 471delAAAG allele and prostate cancer in Ashkenazi Jewish men. J Med Genet 2003; 40:e22. 20. Nupponen NN, Wallen MJ, Ponciano D, Robbins CM, Tammela TL, Vessella RL, et al. Mutational analysis of susceptibility genes RNASEL/HPC1, ELAC2/HPC2, and MSR1 in sporadic prostate cancer. Genes Chromosomes Cancer 2004; 39:119-25. 21. Li G, Xiang Y, Sabapathy K, Silverman RH. An apoptotic signalling pathway in the interferon antiviral response mediated by RNase L and c-Jun NH2-terminal kinase. J Biol Chem 2004; 279:1123-31. 22. Weaver BK, Ando O, Kumar KP Reich NC. Apoptosis is promoted by the dsRNA-activated factor (DRAF1) during viral infection independent of the action of interferon or p53. FASEB J 2001; 15:501-15. 23. Faggioli L, Costanzo C, Merola M, Furia A Palmieri M. Protein synthesis inhibitors cycloheximide and anisomycin induce interleukin-6 gene expression and activate transcription factor NF-κ B. Biochem Biophys Res Commun 1997; 233:507-13. 24. Lindner DJ, Hofmann ER, Karra S, Kalvakolanu DV. The interferon-β and tamoxifen combination induces apoptosis using thioredoxin reductase. Biochim. Biophys Acta 2000; 1496:196-206. 25. Krause D, Mullins JM, Penafiel LM, Meister R, Nardone RM. Microwave exposure alters the expression of 2-5A-dependent RNase. Radiat Res 1991; 127:164-70. 26. Kalivendi SV, Kotamraju S, Zhao H, Joseph J, Kalyanaraman B. Doxorubicin-induced apoptosis is associated with increased transcription of endothelial nitric-oxide synthase. Effect of antiapoptotic antioxidants and calcium. J Biol Chem 2001; 276:47266-76. 27. Dong B, Xu L, Zhou A, Hassel B A, Lee X, Torrence PF, et al. Intrinsic molecular activities of the interferon-induced 2-5A-dependent RNase. J Biol Chem 1994; 269:14153-8. 28. Huang Q, Dunn 2nd RT, Jayadev S, Disorbo O, Pack FD, Farr SB, et al. Assessment of cisplatin-induced nephrotoxicity by microarray technology. Toxicol Sci 2001; 63:196-207. 29. Chapekar, MS, Glazer RI. The synergistic cytocidal effect produced by immune interferon and tumor necrosis factor in HT-29 cells is associated with inhibition of rRNA processing and (2’,5’) oligo (A) activation of RNase L. Biochem. Biophys Res Commun 1988; 151:1180-7. 30. Haridas V, Darnay BG, Natarajan K, Heller R, Aggarwal BB. Overexpression of the p80 TNF receptor leads to TNF-dependent apoptosis, nuclear factor-κ B activation, and c-Jun kinase activation. J Immunol 1998; 160:3152-62. 31. Chen G, Goeddel DV. TNF-R1 signaling : A beautiful pathway. Science 2002; 296:1634-5. 32. Rusch L, Zhou A, Silverman RH. Caspase-dependent apoptosis by 2’,5’-oligoadenylate activation of RNase L is enhanced by IFN-β. J Interferon Cytokine Res 2000; 20:1091-100. 33. Gil J, Esteban M. Induction of apoptosis by the dsRNA-dependent protein kinase (PKR): Mechanism of action. Apoptosis 2000; 5:107-14. 34. Ito T, Yang M, May WS. RAX, a cellular activator for double-stranded RNA-dependent protein kinase during stress signaling. J Biol Chem 1999; 274:15427-32. 35. Patel CV, Handy I, Goldsmith T, Patel RC. PACT, a stress-modulated cellular activator of interferon-induced double-stranded RNA-activated protein kinase, PKR. J Biol Chem 2000; 275:37993-8. 36. Gil J, Esteban M, Roth D. In vivo regulation of the dsRNA-dependent protein kinase PKR by the cellular glycoprotein p67. Biochemistry 2000; 39:16016-25. 37. Pandey M, Rath PC. Expression of interferon-inducible recombinant human RNase L causes RNA degradation and inhibition of cell growth in Escherichia coli. Biochem Biophys Res Commun 2004; 586-97.

www.landesbioscience.com

RNA Biology

27