The African green monkey kidney (Vero E6) cells were grown and maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat‐inactivated fetal bovine serum (FBS) (Gibco Invitrogen Corporation) at 37 °C. The SARS‐CoV‐WHU strain (Accession No. AY293 850) was maintained and propagated in Vero E6 cells as described 21.

SARS‐CoV total RNA was isolated and extracted from the SARS‐CoV‐WHU strain. The RNA was converted into cDNA by reverse transcription. The following primers were used to amplify a 264 bp fragment of the 5′UTR of the SARS genome by polymerase chain reaction (PCR): pUTR1, 5′AGAATTCTAGGTTTTTACCTACCCAGG‐ 3′ (EcoRI digestion site underlined); pUTR2, 5′AGGATCC CTTACCTTTCGGTCACACC‐3′ (BamHI digestion site underlined).

The conditions for PCR were 94 °C, 2 min; (94 °C, 40 s; 50 °C, 40 s; 72 °C, 2 min) for 30 cycles; followed by 72 °C, 10 min. The amplified product was analyzed by electrophoresis, cloned into pMD18‐T PCR cloning vector, then subcloned into the multiple cloning site (MCS) between EcoR1 and BamH1 in peGFP‐N1 vector (BD Biosciences Clonetech), and then transformed into Escherichia coli strain DH5a. The sequence integrity of the construct was verified by restriction mapping and DNA sequencing.

Based on Santoro and Joyce's report 8, we designed a RNA‐cleaving DNAzyme possessing the 10–23 motif. The target site was selected in the loop region of the predicted secondary structure of the 5′UTR. The designed DNAzyme, Dz‐104, consists of 9 + 9 arm sequences with a 10–23 catalytic core, and was synthesized using standard chemistry (Sangon, Shanghai, China). The details of the target sequences, the DNAzyme (Dz‐104) and the mutant control (Dz‐104 (mt)) are shown in Figure 2. As indicated, the A nucleotide of the target gene was left unpaired and cleavage is expected to take place after the A. The mutant DNA enzyme (Dz‐104 (mt)) possessed a single mutation in the catalytic motif (G to C).

To generate the substrate RNA for testing DNAzyme cleavage activity in vitro, the SARS‐CoV 5′UTR was produced by PCR from the recombinant plasmid p5′UTR‐eGFP with a forward primer containing a T7 promoter sequence (5′CTGTAATACGACTCACTATAGGGCCAAGTACGCCCC‐ CTATTGA and a reverse primer (5′GTAGTTGCCGTCGT‐ CCTTGA). The amplified fragment is 1000 bp in length and the predicted cleavage site is in the 5′UTR of SARS at the midpoint of the fragment.

The conditions for PCR were 94 °C, 2 min; (94 °C, 40 s; 50 °C, 40 s; 72 °C, 5 min) for 30 cycles; followed by 72 °C, 15 min. The amplified product was analyzed by electrophoresis and then recovered from the agarose gel. Then 1 µg purified DNA template was used for in vitro transcription with a T7 transcription kit (JINGMEI BIOTECH, Beijing, China) in a volume of 20 µl as recommended by the manufacturer and incubated for 2 h at 37 °C. Upon completion of the transcription, the DNA template was removed by digestion with DNase at 1 U DNase/1 µg template DNA (TaKaRa Biotechnology, Dalian, China) for 15 min at 37 °C. The transcript RNA was further purified with TE‐saturated phenol/chloroform and chloroform/isoamyl alcohol (24 : 1).

In cleavage reactions, the purified substrate RNA (100 nM) and the DNAzyme (various concentrations) were mixed in 10 µl of 50 mM Tris‐HCl, pH 7.5, in the presence of 10 mM MgCl₂ and 1 U RNasin to prevent non‐specific RNA degradation. Prior to mixing enzyme and target RNA, both solutions were denatured separately for 2 min at 85 °C. The cleavage reaction was incubated at 37 °C. For a single turnover kinetic experiment, 10‐fold excess of DNAzyme was used in the cleavage reaction (1000 nM). In a multiple kinetic reaction, 1/10 concentration of DNAzyme (10 nM) was employed and aliquots were taken after defined intervals during the first 10% of the reaction. The reaction was stopped by adding 83 mM EDTA and cooled on ice. 10× sample buffer (37% formaldehyde and 7% 5 × MOPS buffer in formamide) was added to the cleavage reactions prior to being loaded onto an agarose gel. The gel was stained with ethidium bromide and the intensities were quantified with the Quantity One software (Gene Snap). Data were further analyzed by fitting either linearly to obtain the initial velocity v _init for substrate excess experiments or with a single exponential decay function to obtain the observed cleavage rate for enzyme excess experiments using Origin (Microcal Software, Northampton, MA, USA). Values given are mean ± standard deviation (SD) of at least three independent experiments.

Vero E6 cells (2 × 10⁵) were grown to 60% confluence on a 24‐well plate and were then co‐transfected with a fixed amount of p5′UTR‐eGFP (1 µg) along with varying amounts of DNA enzymes (0.7, 3.5, 7 µg, respectively) for 12 h using Lipofectamine Plus (Invitrogen, USA) in a final volume of 500 µl. After 4 h of incubation in the presence of Lipofectamine Plus, the cells were washed with DMEM without serum and incubated for a further 12 h in the same medium with 10% FBS. Then, the cells expressing eGFP were visualized with fluorescence microscopy (Nikon, Tokyo, Japan).

The transfected cells were trypsinized and collected by centrifugation. The cells were further rinsed with phosphate‐buffered saline (PBS), re‐suspended in 100 µl 2% polyformaldehyde and kept at 4 °C; then 900 µl PBS were added to each sample. The cells were then analyzed on Beckman Coulter counter for the percentage of the cells expressing GFP.

At 24 h post‐transfection, the cells were collected for total RNA extraction using Trizol according to the manufacturer's instruction. Real‐time PCR reactions were carried out in ICycler iQ (BioRad). The primers used in the PCR were: GFP1, 5′CAAGCTGACC CTGAAGTTCA 3′ and GFP2, 5′ATGCGGTTCACCAGGGTGT3′, which gave rise to a 250 bp product of the GFP gene. The standard curve was generated using a serial of 10‐fold dilution of pEGFP‐N1 or p5′UTR‐EGFP (10¹⁰, 10⁹, 10⁸, 10⁷, 10⁶, 10⁵copies). The amplification condition was: room temperature (RT), 70 °C for 5 min; 1 × 94 °C for 2 min; 45 × 94 °C for 45 s, 60 °C for 45 s, and 72 °C for 30 s. The mean Ct values were collected and used for analysis.

SARS‐CoV belongs to the coronavirus family with a structure similar to the other coronaviruses, which contain S, M, E and N gene products. The virus features a discontinuous transcription of subgenomic mRNAs, regulated by the pre‐transcribed leader sequence and the specific TRSs in the 5′UTR of the genome. The 5′UTR sequence contributes to the regulation of the virus gene expression. Most strikingly, there is a high degree of sequence conservation in the 5′UTR region among all the SARS‐CoV strains identified thus far, which suggests the functional importance of the region. Targeting the 5′UTR would therefore not only present a feasible strategy, but also potentially avoid the possible emergency of escape mutant viruses.

A 10–23 DNAzyme was designed (Dz‐104) with a cleavage site at nucleotide 104. The arm length was chosen as a 9 + 9 format as previously reported, which gave rise to a much more efficient catalytic activity based on a balance of the hybridization strength (−ΔG) and kinetic turnover 14. The cleavage site was selected at an AU dinucleotide since a hierarchy of RY reactivity (where R is a purine and Y is a pyrimidine) to the 10–23 DNAzyme follows the general scheme AU = GU ≥ GC ≫ AC 15.

To test the cleavage activity of Dz‐104, we conducted in vitro cleavage analyses using a 1 kb RNA transcript that was transcribed from a T7‐containing primer‐based PCR fragment of the 5′UTR. The cleavage point was in the middle of the transcript; thus the two bands of the cleavage product would appear at the same position in a gel. In a single turnover assay where 10‐fold access of Dz‐104 was used, the substrate RNA was efficiently cleaved with the expected size of the cleavage products (Figure 3A) and the observed rate was 0.064 min⁻¹ (Figure 3B). Under a multiple turnover condition, the substrate was cleaved with an initial velocity at 0.4 nM min⁻¹ (Figure 3C). When the mutant DNAzyme (Dz‐104 (mt)) was subjected to the cleavage assay under the same conditions, it did not show any cleavage (Figure 3A, lower panel). These results demonstrated that the designed DNAzyme was chemically active in cleaving its substrate RNA with a capacity to turnover under the in vitro conditions. It also suggested that the local RNA structures where the Dz‐104 targeted are accessible under the simulated physiological conditions.

To examine the efficacy of the designed DNAzymes, it is crucial to have a sensitive and meaningful system to perform the assay. We chose to use a fusion transcript of the 5′UTR and GFP gene as a model system. In this system, the DNAzyme target was linked to the 5′ end of the GFP gene, which provided a versatile assay to monitor DNAzyme activity in cells (Figure 4). When the DNAzyme cleavage occurred, it was anticipated that the truncated fusion RNA would be subjected to the nuclease degradation; therefore, the level of the GFP expression would be reduced, which could be monitored by fluorescence microscopy.

When the p5′UTR‐eGFP construct was transfected into Vero E6 cells, the expression of the eGFP gene could be easily observed, but appeared to be a slightly weaker than the parental vector peGFP‐N1 (Figure 5). This minor change could be due to the effect of the extra 5′UTR sequences on the promoter activity or transcription efficiency of the GFP gene. However, the system should be sensitive enough to be used for assays of the DNAzyme activity.

The advantage of using GFP is that it facilities the direct observation of the target gene expression by fluorescence microscopy. To demonstrate that the designed DNAzyme could target the mRNA with the SARS 5′UTR and suppress the eGFP expression, we co‐transfected Vero E6 cells with p5′UTR‐eGFP and Dz‐104 at different ratios. The data showed that co‐transfection of p5′UTR‐EGFP/Dz‐104, compared to the Dz‐104 (mt) control, significantly decreased the fluorescent level, while it had no effect on the GFP in the cells co‐transfected with pEGFP‐N1 that contained no 5′UTR sequence (Figure 5). The effect of Dz‐104 on eGFP expression was shown to be dose‐dependent and the transfection procedure had no effect on cell growth (data not shown).

To confirm the DNAzyme effect on GFP expression, the percentage of the GFP‐expressing cells was measured using flow cytometry. The results showed that there was a dose‐dependent reduction in the number of GFP+ cells that were transfected with the p5′UTR‐eGFP and Dz‐104 (Table 1). This reduction was not seen in the cells transfected with either the parental vector pEGFP‐N1 or mutant DNAzyme Dz‐104 (mut).

To determine if the effect of Dz‐104 on the GFP expression was due to a cleavage of the 5′UTR‐GFP mRNA, real‐time PCR was performed to measure the level of the target RNA. It was shown that Dz‐104 treatment of the p5′UTR‐eGFP‐transfected cells led to a marked reduction in the target mRNA level, as evidenced in Table 2. This demonstrates that DNAzyme elicits the degradation of the 5′UTR‐EGFP mRNA sequence in a sequence‐specific manner, resulting in suppression of the eGFP gene expression.

In the present study, the DNAzyme was co‐transfected with the target vector into Vero cells. Although this co‐transfection did not fully mimic the cells infected with the virus, it did show that the DNAzyme could enter the cells and find the target RNA. A previous study showed that the DNAzyme transfected into mammalian cells remained stable and biologically active up to 24 h 22, which supports our results, that sufficient DNAzymes had been transfected into the cells and functioned as a target‐specific gene suppressor.

Taken together, our study shows that the DNAzyme can efficiently cleave the 5′UTR of SARS‐CoV both in vitro and in cells, implying that the 5′UTR may well be a good target for molecular targeting. The study also presents a convenient system for screening other nucleic‐acid‐based agents for inhibition of SARS replication. Since DNAzymes have inherent physicochemical advantages over other nucleic‐acid‐based agents, such as higher target flexibility, minimal chemical modification, relative resistance to nuclease degradation and low cost of synthesis, they may be a clinically appealing new class of anti‐SARS agent.