Advertisement

Determination of the Role of the Human RNase H1 in the Pharmacology of DNA-like Antisense Drugs*

Open AccessPublished:February 11, 2004DOI:https://doi.org/10.1074/jbc.M311683200
      Although ribonuclease H activity has long been implicated as a molecular mechanism by which DNA-like oligonucleotides induce degradation of target RNAs, definitive proof that one or more RNase H is responsible is lacking. To date, two RNase H enzymes (H1 and H2) have been cloned and shown to be expressed in human cells and tissues. To determine the role of RNase H1 in the mechanism of action of DNA-like antisense drugs, we varied the levels of the enzyme in human cells and mouse liver and determined the correlation of those levels with the effects of a number of DNA-like antisense drugs. Our results demonstrate that in human cells RNase H1 is responsible for most of the activity of DNA-like antisense drugs. Further, we show that there are several additional previously undescribed RNases H in human cells that may participate in the effects of DNA-like antisense oligonucleotides.
      RNase H hydrolyzes RNA in RNA-DNA hybrids (
      • Stein H.
      • Hausen P.
      ). RNase H activity appears to be ubiquitous in eukaryotes and bacteria (
      • Itaya M.
      • Kondo K.
      ,
      • Itaya M.
      • McKelvin D.
      • Chatterjie S.K.
      • Crouch R.J.
      ,
      • Kanaya S.
      • Itaya M.
      ,
      • Busen W.
      ,
      • Rong Y.W.
      • Carl P.L.
      ,
      • Eder P.S.
      • Walder J.A.
      ). Although RNases H constitutes 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. Mg2+, Mn2+) to produce cleavage products with 5′ phosphate and 3′ hydroxyl termini (
      • Crouch R.J.
      • Dirksen M.L.
      ).
      Although a number of viral and bacterial polymerases and exonucleases have been shown to have RNase H activities (
      • Crouch R.J.
      • Dirksen M.L.
      ), in mammalian cells only two classes of RNase H enzymes have been identified (
      • Busen W.
      ,
      • Eder P.S.
      • Walder J.A.
      ,
      • Frank P.
      • Albert S.
      • Cazenave C.
      • Toulme J.J.
      ). These enzymes were shown to differ with respect to co-factor requirements and were shown to be inhibited by sulfhydryl reagents (
      • Frank P.
      • Albert S.
      • Cazenave C.
      • Toulme J.J.
      ,
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ). Although the biological 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 (
      • Busen W.
      • Peters J.H.
      • Hausen P.
      ,
      • Turchi J.J.
      • Huang L.
      • Murante R.S.
      • Kim Y.
      • Bambara R.A.
      ). Recently, an RNase H1 knock-out in the mouse was shown to be embryonically lethal, and it showed that RNase H1 is involved in mitochondrial DNA replication (
      • Cerritelli S.M.
      • Frolova E.G.
      • Feng C.
      • Grinberg A.
      • Love P.E.
      • Crouch R.J.
      ).
      Two human RNase H genes have been cloned and expressed (
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ,
      • Frank P.
      • Braunshofer-Reiter C.
      • Wintersberger U.
      • Grimm R.
      • Busen W.
      ,
      • Frank P.
      • Braunshofer-Reiter C.
      • Poeltl A.
      • Holzmann K.
      ,
      • Cerritelli S.M.
      • Crouch R.J.
      ). RNase H1 is a 286-amino acid protein and is expressed ubiquitously in human cells and tissues (
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ). The amino acid sequence of human RNase H1 displays strong homology with RNase H1 from yeast, chicken, Escherichia coli, and mouse (
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ). 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 (
      • Frank P.
      • Braunshofer-Reiter C.
      • Wintersberger U.
      • Grimm R.
      • Busen W.
      ). Human RNase H2 shares strong amino acid sequence homology with RNase H2 from Caenorhabditis elegans, yeast, and E. coli (
      • Frank P.
      • Braunshofer-Reiter C.
      • Wintersberger U.
      • Grimm R.
      • Busen W.
      ,
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ).
      The properties of the cloned and expressed human RNase H1 have recently been characterized; many of the properties observed for human RNase H1 are consistent with the E. coli RNase H1 isotype, (e.g. the cofactor requirements, substrate specificity, and binding specificity) (
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ,
      • Lima W.F.
      • Crooke S.T.
      ). 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 (
      • Kanaya S.
      • Katsuda-Nakai C.
      • Ikebara M.
      ,
      • Nakamura H.
      • Oda Y.
      • Iwai S.
      • Inoue H.
      • Ohtsuka E.
      • Kanaya S.
      • Kimura S.
      • Katsuda C.
      • Katayanagi K.
      • Morikawa K.
      • Miyashiro H.
      • Ikehara M.
      ,
      • Katayanagi K.
      • Miyagawa M.
      • Matsushima M.
      • Ishikawa M.
      • Kanaya S.
      • Ikehara M.
      • Matsuzaki T.
      • Morikawa K.
      ,
      • Yang W.
      • Hendrickson W.A.
      • Crouch R.J.
      • Satow Y.
      ). In addition, the lysine residues within the highly basic α-helical substrate-binding 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 (
      • Landt O.
      • Grunert H.
      • Hanh U.
      ). Recently a novel redox-dependent regulator element has been reported for the enzyme (
      • Lima W.
      • Wu H.
      • Nichols J.G.
      • Manalili S.M.
      • Drader J.J.
      • Hofstadler S.A.
      • Crooke S.T.
      ). In contrast to human RNase H1, nothing is known about the detailed enzymological properties of human RNase H2 because the cloned and expressed enzyme is inactive.
      Antisense oligonucleotides (ASOs)
      The abbreviations used are: ASO, antisense oligonucleotide; siRNA, small interference RNA; pfu, plaque-forming unit.
      1The abbreviations used are: ASO, antisense oligonucleotide; siRNA, small interference RNA; pfu, plaque-forming unit.
      have proven of value in determining gene functions and may be of value as a new therapeutic class (
      • Crooke S.T.
      ). Once ASOs bind via Watson-Crick hybridization to target RNAs, they may work through a variety of mechanisms of action (
      • Crooke S.T.
      ,
      • Crooke S.T.
      ,
      • Crooke S.T.
      ). DNA-like ASOs are thought to work by creating a substrate for cellular RNases H after binding to target RNAs (
      • Crooke S.T.
      ). Although a large body of inferential evidence supports this concept, direct proof of the mechanism is lacking, and to date compelling evidence has not been reported with regard to which of the RNases H may be responsible (
      • Crooke S.T.
      ,
      • Crooke S.T.
      ,
      • Giles R.V.
      • Spiller D.G.
      • Tidd D.M.
      ).
      The purpose of this study is to unequivocally demonstrate that DNA-like ASOs induce reduction of target RNAs by creating RNA-ASO duplexes that serve as substrates for mammalian cellular RNases H and to determine which of the two known mammalian RNases H plays a significant role in this process. To achieve these objectives, we have altered the level of expression and cellular activity of the RNase H1 and then determined the effects of these alterations on the ability of DNA-like ASOs to reduce various RNA targets. Because RNase H1 knock-outs have been reported to be lethal, we have employed overexpression and antisense or double-stranded siRNA (siRNA) techniques to alter levels of RNase H1. Our studies demonstrate that human RNase H1 is critically involved in the effects of DNA-like ASOs. Further, our data suggest that there may be other RNases H in mammalian cells that contribute to the activity of DNA-like ASOs.

      EXPERIMENTAL PROCEDURES

      Oligonucleotide Synthesis—Synthesis and purification of chimeric 2′-O-methoxyethyl (MOE)/deoxy phosphorothioate-modified oligonucleotides were as previously described (
      • McKay R.
      • Miraglia L.
      • Cummins L.
      • Owens S.
      • Sasmor H.
      • Dean N.M.
      ,
      • Baker B.F.
      • Lot S.S.
      • Condon T.P.
      • Cheng-Flournoy S.
      • Lesnik E.A.
      • Sasmor H.M.
      • Bennett C.F.
      ). Sequences of oligonucleotides and placement of 2′-O-MOE modifications are detailed below. Unmodified oligodeoxynucleotides were purchased from Invitrogen. The oligoribonucleotides were purchased from Dharmacon Research, Inc. (Boulder, CO). siRNA duplexes were formed in the solution containing 20 μm each oligoribonucleotide, 100 mm potassium acetate, 30 mm HEPES-KOH, pH 7.4, 2 mm magnesium acetates. Reactions were heated for 1 min at 90 °C and incubated for 1 h at 37 °C.
      The sequences of the ASOs used in the study are listed below (bold residues represent 2′-MOE-modified nucleotides; others are deoxynucleotides). All internucleotide links are phosphorothioate. ISIS-13650 (human c-Raf), TCCCGCCTGTGACATGCATT; ISIS101759 (human Jnk2), GCTCAGTGGACATGGATGAG; SIS116847 (human PTEN), CTGCTAGCCTCTGGATTTGA; ISIS104492 (mouse Jnk1), TGTTGTCACGTTTACTTCTG; ISIS22023 (mouse Fas), TCCAGCACTTTCTTTTCCGG; and ISIS194178 (human RNase H1), TGCAGGCTATTTTCCACACC.
      The siRNA sequence was: Si-H1, siRNA (human RNase H1), AAGUUUGCCACAGAGGAUGAG.
      The control ASO, ISIS 29848, is a random mixture of nucleotides at each position. The oligonucleotide sample was prepared by concurrent reaction of four amidites (A, G, C, and T) at each position of the oligonucleotides. The 2′-methoxyethyl amidites were employed at positions 1–5 and 16–20, whereas deoxyribose amidites were used at positions 6–15. The ratio of each amidite in the mixture was adjusted to ensure equivalent reactivities. The siRNA controls in transfection are either the single strand sense RNA or other RNA duplexes that were non-complementary to the target.
      Antibodies—Two human RNase H1 peptides, H-CRAQVDRFPAARFKKFATED-OH (amino acids 46–65) corresponding to the amino-terminal region, and H-CKTSAGKEVINKEDFVALER-OH, (amino acids 231–249) corresponding to the carboxyl terminus of the full protein (GenBank™ accession number AF039652), were conjugated to diphtheria toxin with maleimidocaproyl-N-hydroxysuccinimide and used to raise polyclonal antibodies in rabbits (
      • Harlow E.
      • Lane D.
      Antibodies: a Laboratory Manual.
      ). Polyclonal antibodies were also raised against the His-tagged partial human RNase H1 (amino acid 73–286) (
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ).
      Polyclonal antibodies were further purified with protein antigen using Aminolink immobilization kits (Pierce). 200 μg of purified H1 antibodies were then directly immobilized on agarose gel by using the Seize primary immunoprecipitation kit (Pierce) to create a permanent affinity support for immunoprecipitation without the need of protein A or G beads.
      Adenovirus Production—For overexpression of human RNases H1, the full-length RNase H1 cDNA coding region was amplified by PCR and cloned into the virus shuttle vector pACCMVplpA(–)LoxP-ssp. The insert fragments were confirmed by DNA sequencing, and then the adenovirus (H1) was generated by the Vector Core Laboratory of the University of Michigan. The control virus (LoxP) was also generated by using the same virus shuttle vector (without any insert). This virus shares all features with the RNase H1-containing virus except for the inserted genes. The viruses were prepared by either cell lysate (titration 3–7 × 109 pfu/ml) or CsCl purification (titration 1.38–1.61 × 1011 pfu/ml, 4 × 1012 viral particles/ml).
      Cell Culture and Treatment—HeLa, A549, and HepG2 cells (ATCC, Rockville, MD) were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FCS) in 6-well, 96-well, 10- or 15-cm culture dishes. MCF7 and T24 cell (ATCC) were cultured in McCoy's medium with 10% FCS. Mouse AML12 and HeLa cells were grown in DMEM with 0.005 mg/ml insulin, 0.005 mg/ml transferrin, 5 ng/ml selenium, 40 ng/ml dexamethasone, and 10% FCS. Media and all supplements were purchased from Invitrogen. For transfection of ASO or siRNA, cells were incubated with a mixture of 3 μg/ml lipofectin (Invitrogen)/1–200 nm ASO or siRNA in OptiMem medium (Invitrogen). After 4 h the transfection mixture was aspirated from the cells and replaced with fresh medium containing 10% FCS and the cells incubated at 37 °C, 5% CO2 until harvest or second transfection. For adenovirus infection, different amounts of virus (10–400 pfu/cell) were directly added into the cell culture.
      Western Blots—Whole cell lysates and non-nuclear or nuclear fractions from cells or mouse liver were prepared as described (
      • Dignam J.D.
      • Lebovitz R.M.
      • Roeder R.G.
      ); protein concentrations were measured by the method of Bradford (Bio-Rad Lab, Hercules, CA). The samples were boiled in SDS-sample buffer and then separated by SDS-PAGE using 4–20% Tris-glycine gels (Invitrogen) under reducing conditions. Molecular mass prestained markers were used to determine the protein sizes. The proteins were electrophoretically transferred to a polyvinylidene difluoride membrane and processed for immunoblotting using affinity-purified human RNases H antibodies at 0.5–1 μg/ml. The immunoreactive bands were visualized using the enhanced chemiluminescence method (Amersham Biosciences) and analyzed using PhosphorImager Storm 860 (Amersham Biosciences).
      Northern Blots—Total RNA was isolated from different human cell lines using RNAeasy kits (Qiagen, Valencia, CA). 5–10 μg of total RNA were separated on a 1.2% agarose/formaldehyde gel, transferred to Hybond-N+ (Amersham Biosciences), and fixed to the membrane using a UV cross-linker (Stratagene, La Jolla, CA). Hybridization was performed by using 32P-labeled human RNase H1, G3PHD, or c-Raf DNA probes in Quik-Hyb buffer (Stratagene) at 68 °C for 2 h. After hybridization, membranes were washed in a final stringency of 0.1×SSC/0.1% SDS at 60 °C for 30 min. Membranes were analyzed using PhosphorImager Storm 860.
      RNA Expression Analysis—At the indicated times following oligonucleotide treatment, total RNA was harvested from 96-well culture dishes using an RNAeasy 96 kit and a Bio Robot 3000 (Qiagen) according to the manufacturer's protocol. The RNA concentration was measured with the ribogreen RNA quantitation reagent (Molecular Probes, Eugene, OR). Gene expression was analyzed using quantitative RT/PCR as described elsewhere (
      • Winer J.
      • Kwang C.
      • Jung S.
      • Shackel I.
      • Williams P.M.
      ). Total RNA was analyzed in a final volume of 50 μl containing 200 nm gene-specific PCR primers, 0.2 mm of each dNTP, 75 nm fluorescently labeled oligonucleotide probe, 1× RT/PCR buffer, 5 mm MgCl2, 2 units of platinum TaqDNA polymerase (Invitrogen), and 8 units of ribonuclease inhibitor. Reverse transcription was performed for 30 min at 48 °C followed by PCR: 40 thermal cycles of 30 s at 94 °C and 1 min at 60 °C, using an ABI Prism 7700 sequence detector (Foster City, CA).
      The following primer/probe sets were used. Human c Raf kinase (GenBank™ accession number X03484): forward primer, AGCTTGGAAGACGATCAGCAA; reverse prime, AAACTGCTGAACTATTGTAGGAGAGATG; and probe, AGATGCCGTGTTTGATGGCTCCAGCX. Human PTEN phosphatase (GenBank™ accession number U92436): forward primer, AATGGCTAAGTGAAGATGACAATCAT; reverse primer, TGCACATATCATTACACCAGTTCGT; and probe, TTGCAGCAATTCACTGTAAAGCTGGAAAGGX. Human Jnk2 protein kinase (GenBank™ accession number U35003.1): forward primer, CGCTGGCCTCAGACACAGA; reverse primer, CTAACCTATCATCGACAGCCTTCA; and probe, AGCAGTCTTGATGCCTCGACGGGAX. Human RNase H1 (GenBank™ accession number AF039652): forward primer, GGTTTCCTGCTGCCAGATTTAA; reverse primer, GGCTTGCAGATTTCCTGACAA; and probe, TTTGCCACAGAGGATGAGGCCTGGX. Mouse Jnk1 protein kinase (GenBank™ accession number BU611812.1): forward primer, CAACGTCTGGTATGATCCTTCAGA; reverse primer, GTGCTCCCTCTCATCTAACTGCTT; and probe-AAGCCCCACCACCAAAGATCCCGX.
      Animal Experiments—Eight-week-old female Balb/c mice were purchased from Jackson Laboratory. Mice were treated with various doses of ASO (ISIS 22023, anti-mouse Fas) in saline (Invitrogen) or with saline alone in 200 μl by intraperitoneal injection before treatment with the RNase H1-containing adenovirus (6 × 109 pfu or 1.7 × 1011 viral particles in 200 μl of phosphate-buffered saline) or the control LoxP virus (6 × 109 pfu or 1.5 × 1011 viral particles in 200 μl of phosphate-buffered saline) by intravenous injection, according to the indicated schedules. Total RNA was extracted from mouse liver using RNAeasy kits (Qiagen). RNase protection assays (RPA) were performed according to the manufacturer's instructions (BD Biosciences). RPA template mApo-3 and a custom template (BD Biosciences) were used as probes. 20 μg of total RNA was analyzed on 6% denaturing polyacrylamide gels. Individual transcripts were then quantitated on a PhosphorImager. Fas mRNA expression levels were normalized to L32 or glyceraldehyde-3-phosphate dehydrogenase mRNA levels in each individual sample and presented as the percentage of saline (control) -treated animals.
      Statistics—To determine the IC50 values displayed in many figures, the curves were fit with a curve-fitting program in the linear dose range and the IC50 interpolated. To evaluate the statistical significance of treatment effects, pairwise comparisons of each point on each curve between treatment and control were performed using two-sided Student's t test. For every point in the RNase H1-treated samples, the p value compared with control was less than 10–3–10–41.

      RESULTS

      Overexpression of Human RNase H1—For overexpression of human RNase H1, a strain of adenovirus containing the insert shown in Fig. 1A was developed. Fig. 1B (right panel) shows that the full-length enzyme was overexpressed in HeLa and A549 cells. Peak expressions were observed 36–48 h after infection. In addition, the enzyme could be overexpressed in T24, MCF7, HepG2, and H293 cells (data not shown). Fig. 1B (left panel) shows a Western blot of the immunoprecipitated lysate from untreated HeLa cells; the right panel demonstrates that the full-length human RNase H1 virally encoded enzyme was overexpressed and comigrated with the enzyme from uninfected cells.
      Figure thumbnail gr1
      Fig. 1Development of adenoviruses overexpressing human RNases H. A, human RNase H construct in adenovirus shuttle vectors. Full-length RNase H1 cDNAs were amplified by PCR and cloned into EcoRI and XhoI sites in the multiple cloning site (MCS) downstream from the cytomegalovirus promoter in the adenovirus shuttle vector, pACCCMVpLpA(-)Loxp-ssp (Core facility of University of Michigan). B, Western blot analysis of protein lysates from HeLa or A549 cells infected with full-length H1 virus (200 pfu/cell). The cells were harvested at different time points (0, 6, 12, 24, 36, 48 h) after virus infection (right panel). The protein concentrations of the cell lysates were measured. The lysates were subjected to 4–20% gradient SDS-PAGE (20 μg/lane) and Western blot analysis with anti-RNase H1 (against H1 carboxyl-terminal peptides, see “Experimental Procedures”). Immunoprecipitation was performed using uninfected HeLa cell lysate with purified H1 that was covalently immobilized to agarose beads (left panel). The eluted samples were subjected to Western blot analysis with H1 Ab. C, gel renaturation assay on uninfected HeLa cell lysate (5 μg) (panel 1); samples from immunoprecipitation with H1 Ab from uninfected HeLa cell nuclear and cytosolic extracts (see “Experimental Procedures”) (panel 2); samples from immunoprecipitation with H1 Ab from the lysates of HeLa cells infected with or without H1 or control virus (panel 3).
      To determine whether the overexpressed RNase H1 was active, we employed the gel renaturation assay. As previously reported, human RNase H1 can be renatured and was active in this assay (Fig. 1C). Human RNase H1 activity was present in both the cytosolic and nuclear fractions of uninfected HeLa cells. To confirm that the activity was indeed human RNase H1, the enzyme was immunoprecipitated from HeLa cells and then subjected to the gel renaturation assay (Fig. 1C, panel 2), again showing that the enzyme was present in both the cytosolic and nuclear fractions. Overexpression of the full-length human RNase H1 resulted in increased activity in the gel renaturation assay (Fig. 1C, panel 3).
      Overexpression of Human RNase H1 Increases the Potency of DNA-like ASOs—To evaluate the effect of overexpression of the RNases H on the potency of DNA-like ASOs, HeLa and A549 cells were infected with either the control (Lox P) or RNase H1 insert containing adenovirus and then the effects of several well characterized ASOs (
      • Monia B.P.
      ,
      • Geary R.S.
      • Leeds J.M.
      • Fitchett J.
      • Burckin T.
      • Truong L.
      • Spainhour C.
      • Creek M.
      • Levin A.A.
      ,
      • Bost F.
      • McKay R.
      • Dean N.
      • Mercola D.
      ,
      • Bost F.
      • McKay R.
      • Bost M.
      • Potapova O.
      • Dean N.
      • Mercola D.
      ,
      • Butler M.
      • Popoff I.J.
      • Gaarde W.A.
      • Witchell D.
      • Murray S.F.
      • Dean N.M.
      • Bhanot S.
      • Monia B.P.
      ) on the intracellular concentrations of target RNAs evaluated. Fig. 2, A–C, shows that the potencies of ASOs designed to bind to human c-Raf, PTEN, or c-Jun NH2-terminal kinase 2 were significantly increased by overexpression of human RNase H1 in HeLa cells. Fig. 2D shows that similar results were observed in A549 cells. Fig. 2E shows that similar results were observed whether RT-PCR or Northern blot analyses were performed. The IC50 values for each ASO under each condition are shown under the graphs.
      Figure thumbnail gr2
      Fig. 2Effects of RNase H1 overexpression on the potency of DNA-like ASOs. HeLa cells were split into 6000 cells/well in 96-well plates and then infected with H1 or control (LoxP) viruses (200 pfu/cell). 12 h later, the cells were transfected with the anti-cRaf (A) ASO (ISIS 13650) at different concentrations. The cells were harvested 24 h later. cRaf mRNA levels were measured with RT-PCR in which the reverse transcription and PCR amplification of cRaf mRNA were performed in 96-well format with the primer set described under “Experimental Procedures.” IC50 were calculated and presented under the graphs. The bars represent S.E. of the mean of 3–5 replicates of a representative experiment. B, a similar experiment with anti-PTEN ASO (ISIS116847). C, with anti-c-Jun NH2-terminal kinase 2 ASO (ISIS101759). D, a similar experiment in A549 cells. E, Northern blot analyses of the effects of RNase H1 on the potency of the cRaf ASO in HeLa cells. The cells were split into 106 cells/10-cm plate and incubated with control or H1 virus (200 pfu/cell) for 12 h before the cells were transfected with anti-cRaf ASO (ISIS13650) of different concentrations via lipofectin (see “Experimental Procedures”). The cells were harvested 24 h later, and the total RNA was prepared with RNAeasy kit (Qiagen). 5 μg of RNA/lane was subjected to 1.2% agarose/formaldehyde and further to Northern blot analysis with 32P-labeled human cRaf cDNA probe and housekeeping gene glycerol-3-phosphate dehydrogenase (G3) probe (for normalization). The experiment was performed in triplicate. Results were plotted with percentage normalized mRNA level versus ASO concentration. The bars represent S.E. of the mean of the triplicates. The experiment was repeated several times.
      Because DNA-like ASOs are frequently used in vivo and are being evaluated in multiple clinical trials in humans (
      • Crooke S.T.
      ), we extended our observations to include in vivo experiments. To do this we first demonstrated that human RNase H1 could be overexpressed in mouse cells (Fig. 3A). As was observed in human cells, overexpression of human RNase H1 increased the potency of a DNA-like ASO designed to bind to mouse c-Jun NH2-terminal kinase 1 RNA (Fig. 3B).
      Figure thumbnail gr3
      Fig. 3Overexpression of human RNase H1 enhances ASO activity in mouse cell lines. A, overexpression of human RNase H1 in mouse AML12 and HeLa cell lines. Adenoviral infection and Western blot analyses were performed as described in and under “Experimental Procedures.” B, overexpression of human RNase H1 increases anti-mouse c-Jun NH2-terminal kinase 1 ASO potency. Experimental procedures were as described for except for the increase in virus dosage (400 pfu/cell).
      To determine whether overexpression of human RNase H1 in mouse liver increased the potency of DNA-like ASOs, we evaluated the effects of a well characterized mouse Fas ASO (
      • Zhang Z.
      • Cook J.
      • Nickel J.
      • Yu R.
      • Stecker K.
      • Myers K.
      • Dean N.M.
      ). Groups of mice were treated with the control and human RNase H1-containing adenovirus as described under “Experimental Procedures.” Fig. 4A shows that human RNase H1 was significantly overexpressed in the liver of the animals that were infected with the adenoviruses containing the insert. The human RNase H1 expressed in mouse liver was active in the gel renaturation assay. Moreover, the degree of overexpression was reasonably consistent. Fig. 4B shows that the ASO caused the selective reduction of Fas RNA in mouse liver. To evaluate the effects of overexpression of RNase H1 on the potency of the Fas ASO, the bands were quantitated and normalized to glycerol-3-phosphate dehydrogenase. Those data were then used to construct the dose response curves shown in Fig. 4C. The effects of overexpression of human RNase H1 were further confirmed by immunostaining of Fas protein with a Fas antibody (data not shown). Under the conditions employed, there was no evidence of significant liver toxicity and no histological differences among the saline, the LoxP control, and human RNase H1 virus-treated livers (data not shown). Furthermore, the concentrations of Fas ASO in the liver of mice treated with different doses of ASO were similar in mice treated with control virus or RNase H1-containing virus (Group 20 mg/kg: control, 29.01 ± 1.9; RNase H1, 31.1 ± 2.3. Group 30 mg/kg: control, 37.5 ± 15; RNase H1, 40.9 ± 4.1. Group 40 mg/kg: control, 50.1 ± 7.8; RNase H1, 48.5 ± 3.0 μg of parent ASO/g liver). Thus, the effects observed were neither secondary to hepato-toxicity nor due to changes in uptake of the Fas ASO into liver as a result of adenovirus infection.
      Figure thumbnail gr4
      Fig. 4Effects of overexpression of human RNase H1 on Fas ASO potency in mouse liver. A, analysis of expression of human RNase H1 in mouse liver. Mice were treated with different amounts of Fas ASO (ISIS 22023) and then the H1 or control viruses 4 h later as indicated under “Experimental Procedures.” After another 24 h, the animals were sacrificed and the livers harvested. Liver tissue lysate was prepared with SDS RIPA lysis buffer (see “Experimental Procedures”). 20 μg of protein were used in the gel renaturation assay (GRN) in the presence of 10 mm Mg2+ and Western blot (WB) with anti-human RNase H1 Ab. Each lane represents a sample from an individual animal (n = 4 for each group). B, RNA protection assay. Total RNA was extracted from the livers of the same mice as in panel A. The expression of Fas mRNA in liver was determined by an RNA protection assay. Each lane in the gel represents a sample from an individual animal. The figure shows only two lanes for each group (n = 4). Fas and other RNAs are labeled to the left of the figure. C, effects of different doses of Fas ASO on Fas mRNA levels were compared with the saline control group after normalization to glyceraldehyde-3-phosphate dehydrogenase mRNA expression, respectively. The bars represent the S.E. of the mean of four animals in each group. This experiment was repeated three times with equivalent results.
      Reduction of Human RNase H1 Reduces the Potency of DNA-like ASOs—To complement the overexpression experiments, we have reduced RNase H1 levels and evaluated the effects of reduced enzyme levels on the potencies of DNA-like ASOs. Potent selective DNA-like ASOs and siRNA were identified for both enzymes by cellular screening as previously described (
      • Crooke S.T.
      ,
      • Vickers T.A.
      • Koo S.
      • Bennett C.F.
      • Crooke S.T.
      • Dean N.M.
      • Baker B.F.
      ).
      To identify the most potent ASOs and siRNAs to inhibit human RNase H1, we screened candidates designed to bind to multiple sites in the RNAs. The most potent ASO (ISIS 194178) in inhibiting human RNase H1 was located at nucleotides 1026–1006 in the 3′-untranslated region of the RNA (GenBank™ accession number AF039652). The most potent siRNA for RNase H1 (si-H1) was located at nucleotides 259–279 in the coding region of the RNA.
      The effects of various concentrations of each of the optimized inhibitors were then evaluated (Fig. 5). Both the ASO and siRNA inhibitors resulted in potent dose-dependent selective loss of RNase H1 RNA in both HeLa and A549 cells (Fig. 5A). Both the 1.3- and the 5-kb bands thought to be preprocessed human RNase H1 RNA were reduced. Further, the RNase H1 activity in both cell lines was reduced as shown in the gel renaturation assay (Fig. 5B). There were no effects on RNase H2 levels (data not shown). The duration of effect for both the ASO and siRNAs was greater than 48 h (data not shown).
      Figure thumbnail gr5
      Fig. 5ASO or siRNA reduction of RNase H1 in HeLa and A549 cell lines. A and B, ASO ISIS194178 or si-H1 reduces RNase H1 mRNA levels and enzyme activity. Cells were treated with different amounts of ASO or siRNA for 24 h. Total RNA and cell lysates were prepared. As described earlier, the RNA was subjected to 1.2% agarose/formaldehyde gel (5 μg of total RNA/lane) and Northern blot analysis with 32P-labeled human RNase H1 or a glycerol-3-phosphate dehydrogenase (G3PDH) cDNA probe. 20 μgof proteins of cell lysate were used for gel renaturation assay to test RNase H1 activity.
      Fig. 6 shows that reduction in the levels of human RNase H1 reduced the potency of the ASO targeting c-Raf RNA in HeLa cells. Increasing concentrations of the siRNA to human RNase H1 resulted in a comparable reduction in the potency of the c-Raf ASO. Further, there is a clear correlation between the reduction of c-Raf RNA by the ASO and the cellular level of human RNase H1 (R2 = 0.91 or 0.69; p <0.01) (Fig. 6, A and B). Note that an extrapolation of the dose response curves for the c-Raf ASO would not demonstrate zero antisense activity when there was no human RNase H1 mRNA.
      Figure thumbnail gr6
      Fig. 6Effects of RNase H1 siRNA on the potency of cRaf ASO (ISIS 13650) in HeLa cells. Cells were first transfected with various concentrations of RNase H1 siRNA as indicated for 10 h before the cells were split into 96-well format cell culture plates (6000 cells/well) and incubated for 10–14 h. The cells were transfected with various concentrations of ISIS 13650 for 24 h before harvest. Total RNAs were prepared, and the cellular cRaf and RNase H1 mRNA levels were determined with RT-PCR in which the reverse transcription and PCR amplification of cRaf and RNase H1 mRNAs were performed in the 96-well format with the primer sets described under “Experimental Procedures.” The vertical bars represent S.E. of the mean of 3–6 replicates of a representative experiment. A, reduction of cellular RNase H1 by H1 siRNA. B, effects of RNase H1 siRNA treatment on the potency of cRaf ASO (ISIS 13650). IC50 were calculated and presented under the graph. C, correlation of cellular RNase H1 mRNA levels with the potency of ISIS 13650. Cellular RNase H mRNA levels were determined by RT-PCR as described. The RNase H1 mRNA levels in arbitrary results for untreated cells were divided by the level of the RNase H1 mRNA from treated cells to obtain the relative level of RNase H1 RNA. Percent reduction of cRaf RNA was calculated as previously described.
      Fig. 7 shows that an ASO versus human RNase H1 reduced the potency of the c-Raf ASO in HeLa and A549 cells. These results were entirely comparable with the effects of siRNAs to RNase H1.
      Figure thumbnail gr7
      Fig. 7Effects of ASO reduction of RNase H1 (ISIS 194178) on cRaf ASO (ISIS 13650) potency in HeLa and A549 cells. Reduction of cRaf mRNA in HeLa cells (A) and A549 cells (B) pretreated with H1 ASOs. Experimental procedures as in . Each RNase H ASO was transfected at 150 nm concentration. The vertical bars represent S.E. of the mean of six replicates of a representative experiment.
      There Are Multiple RNase H-like Activities in Human Cells— Two related questions are prompted by the RNase H1 inhibition experiments. First, if RNase H1 is required for the activity of DNA-like ASO, why was the c-Raf ASO active when cellular RNase H1 was reduced by more than 90%? Second, if only RNase H1 is involved in the activities of DNA-like ASOs, why do the RNase H1 inhibition experiments generate curves that do not extrapolate to zero activity at zero RNase H1?
      We observed several additional higher molecular mass (50–70-kD bands) as well as several lower molecular mass bands from cell homogenates in the gel renaturation assay. These bands were observed in a variety of cells when the renaturation assay was performed under standard conditions (10 mm Mg2+). The level of activity and the number of extra bands varied from cell type to cell type and from cell preparation to cell preparation (Fig. 8A). When the assay was performed in the presence of 0.5 mm Mn2+, the extra RNase H activity bands were more apparent and at least one higher molecular mass activity band was observed in all cell lines studied (Fig. 8B).
      Figure thumbnail gr8
      Fig. 8Several RNases H are present in human cells. Cell lysates were prepared in RIPA lysis buffer from human HeLa, A549, T24, MCF7, and HepG2 cells as described under “Experimental Procedures.” 20 μg of protein from each lysate were used in gel renaturation assay (see “Experimental Procedures”). Lanes 1–2, HeLa cell lysates; lanes 3–4, A549 lysates; lanes 5-6, T24; lane 7, MCF7; lane 8, HepG2 lysate. The lysates from lanes 2, 4, and 6 were prepared with the lysis buffer without phosphatase inhibitors. A, gel renaturation assay in the presence of Mg2+. B, gel renaturation in the presence of Mn2+. This is a representative experiment that has been repeated more than five times.
      Fig. 9, A and B, shows that reduction of RNase H1 with either an ASO or siRNA in both HeLa and A549 cells reduced the RNase H1 band of activity and had no effect on the higher molecular mass bands of RNase H activity. In Fig. 9C, we performed quantitative immunoprecipitation of RNase H1 in cell lysates. The supernatant after immunoprecipitation of RNase H1 contained no detectable RNase H1. Further, in the gel renaturation assay (Fig. 9D) there was no RNase H1 activity in the immunoprecipitation supernatant. Nevertheless, in the immunoprecipitation supernatant, several of the novel RNase H activity bands remained (Fig. 9D). The same results are produced from similar experiments with anti-H2 antibody immunoprecipitation (data not shown).
      Figure thumbnail gr9
      Fig. 9Several RNases H are present in human cells, and they are not RNase H1. A, gel renaturation assay in the presence of Mg2+ of HeLa cell lysates prepared as described under “Experimental Procedures.” Prior to preparation of the lysates, the HeLa cells were treated with either a control ASO or the RNase H1 ASO (ISIS194178) at the concentration indicated. This is a representative experiment repeated more than three times. B, gel renaturation assay in the presence of Mn2+ of A549 cell lysates. Cells were treated with a control or the siRNA for RNase H1 as indicated. C, Western blot analysis of RNase H1 from HeLa cell lysates. Cell lysates were prepared as previously described. These were subjected to immunoprecipitation with the purified polyclonal antibodies to human RNase H1. The supernatant was separated from the protein A beads by centrifugation. All samples were then subjected to SDS-PAGE and probed with the purified human RNase H1 antibody. D, gel renaturation assay of the HeLa cell lysates after immunoprecipitation. Experimental procedures are as described earlier.
      These results show that there are several previously unidentified RNase H activities in human cells that are not RNase H1 or H2 yet are active in a gel renaturation assay. Neither inhibition at the RNA level with ASOs or siRNAs nor precipitation with RNase H1 or H2 antibodies affected the level of activities of the unidentified RNases H.

      DISCUSSION

      Although it has been assumed that DNA-like ASOs cause target RNA reduction by binding to the target RNA and creating a DNA-RNA duplex that serves as a substrate for RNase H, definitive proof that this mechanism is responsible for the observed effects in mammalian cells and animals is lacking (
      • Crooke S.T.
      ,
      • Crooke S.T.
      ). In cell-free systems, the addition of E. coli RNase H or human RNase H1 to DNA-RNA duplexes results in degradation of the target RNA (
      • Wu H.
      • Lima W.F.
      • Crooke S.T.
      ,
      • Crooke S.T.
      • Lemonidis K.
      • Neilson L.
      • Griffey R.
      • Lesnik E.A.
      • Monia B.P.
      ,
      • Lima W.F.
      • Mohan V.
      • Crooke S.
      ). The ability of DNA-like ASOs to cause a reduction in target RNAs in cells has been demonstrated many times as well (
      • Crooke S.T.
      ). Moreover, changes in the structure of a DNA-like ASO that resulted in loss of the ability of the duplex to serve as a substrate for RNase H in cell-free systems have been reported to result in a loss of target RNA reduction in cells treated with the modified ASO (
      • Crooke S.T.
      ,
      • Chiang M.-Y.
      • Chan H.
      • Zounes M.A.
      • Freier S.M.
      • Lima W.F.
      • Bennett C.F.
      ). Additionally, Giles et al. (
      • Giles R.V.
      • Spiller D.G.
      • Tidd D.M.
      ) used reverse ligation PCR to identify cleavage products from bcr-abl mRNA in cells treated with a DNA-like ASO. Nevertheless, none of these studies directly demonstrates that DNA-like ASOs reduce target RNA by activating RNase H nor has the specific RNase H that is responsible been identified. Because knock-outs of human RNase H1 are lethal (
      • Busen W.
      ,
      • Cerritelli S.M.
      • Frolova E.G.
      • Feng C.
      • Grinberg A.
      • Love P.E.
      • Crouch R.J.
      ), we have chosen to overexpress human RNase H1 and to employ ASO and siRNA reduction of RNase H1 as complementary approaches to determine whether RNase H is required for target RNA reduction by DNA-like ASOs.
      DNA-like ASOs Cause RNA Reduction by Serving as Substrates for Human RNase H1—For the first time, in this report we provide direct evidence demonstrating that DNA-like ASOs work via an RNase H mechanism. We show that overexpression of human RNase H1 increases the potencies of several ASOs against several target RNAs in several human cell lines (Fig. 2). We also demonstrate that overexpression of human RNase H1 in mouse cells and mouse liver increases the potencies of DNA-like ASOs (Figs. 3 and 4). More compellingly, reduction of RNase H1 results in a loss of potency for ASO to RNA target in several human cell lines (Figs. 6 and 7). Thus, both overexpression and reduction of RNase H1 demonstrate that RNase H1 is involved in the effects of DNA-like ASOs. This is the first direct demonstration of the role of an RNase H in the activity of a DNA-like ASO. The conclusions differ from a previous publication: ten Asbroek et al. (
      • ten Asbroek A.L.
      • van Groenigen M.
      • Mooij M.
      • Baas F.
      ) concluded that RNase H2 plays an essential role in the activity of DNA-like ASOs based on indirect evidence. In essence, they inferred that RNase H2 must be involved because it is more prevalent and differs in subcellular localization from RNase H1. Obviously, direct evidence based on selective alteration of the levels of both enzymes provides a more compelling answer.
      Other RNases H May Play a Role in the Activity of DNA-like ASOs—The failure of siRNA or ASO inhibition of RNase H1 to eradicate ASO activity suggests that although RNase H1 plays a dominant role other RNases H may be involved, particularly if RNase H1 is reduced. We have identified several RNases H that are not immunoprecipitated by antibodies to RNase H1 or H2 (Figs. 8 and 9). These enzymes are not reduced by ASOs or siRNAs to RNase H1. They are active in the gel renaturation assay, suggesting that they are active as monomers and can be renatured. Some of them are more active in the presence of Mn2+ than Mg2+, and their activities are quite variable from cell line to cell line and within cell lines. These data suggest that these enzymes differ from the cloned mammalian RNase H and that they are regulated differently from RNase H1, because the levels of these enzymes vary far less than the novel activities we have observed. They may even be chimeric proteins with polymerase or other enzymatic activities fused to a peptide with RNase H activity. We are in the process of isolating and characterizing the enzymes.
      Implications for Antisense Therapeutics—The demonstration that human RNase H1 plays a dominant role in the activities of DNA-like ASOs suggests that additional studies that explore the substrate preferences, enzymology, and regulatory processes for RNase H1 should support improved design of antisense agents. The demonstration that increases in RNase H1 activity correlated with increases in potency suggests that recruitment of RNase H1 to the ASO-RNA duplex and/or cleavage of the RNA by the enzyme is limiting for ASO activity. Any strategy that would improve these processes should improve ASO potency. Medicinal chemical approaches that optimize the structures of chimeric ASOs, the placement of DNA-like domains relative to preferred sites of cleavage, or enhance recruitment of the enzyme are in progress. It will also be important to better understand the novel RNases H and begin to design ASOs optimized for them.

      Acknowledgments

      We thank Loren Miraglia, Tim Vickers, Chenguang Zhao, and Wayne Lam for technical assistance, Frank Bennett and Dave Ecker for helpful discussions, and Bella Aguilar for manuscript preparation.

      References

        • Stein H.
        • Hausen P.
        Science. 1969; 166: 393-395
        • Itaya M.
        • Kondo K.
        Nucleic Acids Res. 1991; 19: 4443-4449
        • Itaya M.
        • McKelvin D.
        • Chatterjie S.K.
        • Crouch R.J.
        Mol. Gen. Genet. 1991; 227: 438-445
        • Kanaya S.
        • Itaya M.
        J. Biol. Chem. 1992; 267: 10184-10192
        • Busen W.
        J. Biol. Chem. 1980; 255: 9434-9443
        • Rong Y.W.
        • Carl P.L.
        Biochemistry. 1990; 29: 383-389
        • Eder P.S.
        • Walder J.A.
        Biochimie (Paris). 1993; 75: 6472-6479
        • Crouch R.J.
        • Dirksen M.L.
        Linn S.M. Roberts R.J. Cold Spring Harbor. Cold Spring Harbor Press, Plainview, New York1982: 211-254
        • Eder P.S.
        • Walder J.A.
        J. Biol. Chem. 1991; 266: 6472-6479
        • Frank P.
        • Albert S.
        • Cazenave C.
        • Toulme J.J.
        Nucleic Acids Res. 1994; 22: 5247-5254
        • Wu H.
        • Lima W.F.
        • Crooke S.T.
        Antisense Nucleic Acid Drug Dev. 1998; 8: 53-61
        • Busen W.
        • Peters J.H.
        • Hausen P.
        Eur. J. Biochem. 1977; 74: 203-208
        • Turchi J.J.
        • Huang L.
        • Murante R.S.
        • Kim Y.
        • Bambara R.A.
        Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 9803-9807
        • Cerritelli S.M.
        • Frolova E.G.
        • Feng C.
        • Grinberg A.
        • Love P.E.
        • Crouch R.J.
        Mol. Cell. 2003; 11: 807-815
        • Frank P.
        • Braunshofer-Reiter C.
        • Wintersberger U.
        • Grimm R.
        • Busen W.
        Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 12872-12877
        • Frank P.
        • Braunshofer-Reiter C.
        • Poeltl A.
        • Holzmann K.
        Biol. Chem. 1998; 379: 1407-1412
        • Cerritelli S.M.
        • Crouch R.J.
        Genomics. 1998; 53: 300-307
        • Wu H.
        • Lima W.F.
        • Crooke S.T.
        J. Biol. Chem. 1999; 274: 28270-28278
        • Lima W.F.
        • Crooke S.T.
        Biochemistry. 1997; 36: 390-398
        • Kanaya S.
        • Katsuda-Nakai C.
        • Ikebara M.
        J. Biol. Chem. 1991; 266: 11621-11627
        • Nakamura H.
        • Oda Y.
        • Iwai S.
        • Inoue H.
        • Ohtsuka E.
        • Kanaya S.
        • Kimura S.
        • Katsuda C.
        • Katayanagi K.
        • Morikawa K.
        • Miyashiro H.
        • Ikehara M.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 11535-11539
        • Katayanagi K.
        • Miyagawa M.
        • Matsushima M.
        • Ishikawa M.
        • Kanaya S.
        • Ikehara M.
        • Matsuzaki T.
        • Morikawa K.
        Nature. 1990; 347: 306-309
        • Yang W.
        • Hendrickson W.A.
        • Crouch R.J.
        • Satow Y.
        Science. 1990; 249: 1398-1405
        • Landt O.
        • Grunert H.
        • Hanh U.
        Gene. 1990; 96: 125-128
        • Lima W.
        • Wu H.
        • Nichols J.G.
        • Manalili S.M.
        • Drader J.J.
        • Hofstadler S.A.
        • Crooke S.T.
        J. Biol. Chem. 2003; 278: 14906-14912
        • Crooke S.T.
        Abraham D.J. Burger's Medicinal Chemistry. 6th Ed. Vol. 5. John Wiley and Sons, Inc., New York2003: 115-166
        • Crooke S.T.
        Biochim. Biophys. Acta. 1999; 1489: 30-42
        • Crooke S.T.
        Crooke S.T. Antisense Technology: Principles, Strategies, and Applications. Marcel Dekker, Inc., New York2001: 1-28
        • Giles R.V.
        • Spiller D.G.
        • Tidd D.M.
        Antisense Res. Dev. 1995; 5: 23-31
        • McKay R.
        • Miraglia L.
        • Cummins L.
        • Owens S.
        • Sasmor H.
        • Dean N.M.
        J. Biol. Chem. 1999; 274: 1715-1722
        • Baker B.F.
        • Lot S.S.
        • Condon T.P.
        • Cheng-Flournoy S.
        • Lesnik E.A.
        • Sasmor H.M.
        • Bennett C.F.
        J. Biol. Chem. 1997; 272: 11994-12000
        • Harlow E.
        • Lane D.
        Antibodies: a Laboratory Manual.
        Cold Spring Harbor Laboratories, Cold Spring Harbor, New York1988: 53-138
        • Wu H.
        • Lima W.F.
        • Crooke S.T.
        J. Biol. Chem. 2001; 276: 23547-23553
        • Dignam J.D.
        • Lebovitz R.M.
        • Roeder R.G.
        Nucleic Acids Res. 1983; 11: 1475-1489
        • Winer J.
        • Kwang C.
        • Jung S.
        • Shackel I.
        • Williams P.M.
        Anal. Biochem. 1999; 270: 41-49
        • Monia B.P.
        Anti-cancer Drug Design. 1997; 12: 327-339
        • Geary R.S.
        • Leeds J.M.
        • Fitchett J.
        • Burckin T.
        • Truong L.
        • Spainhour C.
        • Creek M.
        • Levin A.A.
        Drug Metab. Dispos. 1997; 25: 1272-1281
        • Bost F.
        • McKay R.
        • Dean N.
        • Mercola D.
        J. Biol. Chem. 1997; 272: 33422-33429
        • Bost F.
        • McKay R.
        • Bost M.
        • Potapova O.
        • Dean N.
        • Mercola D.
        Mol. Cell. Biol. 1999; 19: 1938-1949
        • Butler M.
        • Popoff I.J.
        • Gaarde W.A.
        • Witchell D.
        • Murray S.F.
        • Dean N.M.
        • Bhanot S.
        • Monia B.P.
        Diabetes. 2002; 51: 1028-1034
        • Zhang Z.
        • Cook J.
        • Nickel J.
        • Yu R.
        • Stecker K.
        • Myers K.
        • Dean N.M.
        Nature Biotechnol. 2000; 18: 862-867
        • Vickers T.A.
        • Koo S.
        • Bennett C.F.
        • Crooke S.T.
        • Dean N.M.
        • Baker B.F.
        J. Biol. Chem. 2003; 278: 7108-7118
        • Crooke S.T.
        • Lemonidis K.
        • Neilson L.
        • Griffey R.
        • Lesnik E.A.
        • Monia B.P.
        Biochem. J. 1995; 312: 599-609
        • Lima W.F.
        • Mohan V.
        • Crooke S.
        J. Biol. Chem. 1997; 272: 18191-18199
        • Chiang M.-Y.
        • Chan H.
        • Zounes M.A.
        • Freier S.M.
        • Lima W.F.
        • Bennett C.F.
        J. Biol. Chem. 1991; 266: 18162-18171
        • ten Asbroek A.L.
        • van Groenigen M.
        • Mooij M.
        • Baas F.
        Eur. J. Biochem. 2002; 269: 583-592