Efficient Reverse Transcription Using Locked Nucleic Acid Nucleotides towards the Evolution of Nuclease Resistant RNA Aptamers

Background Modified nucleotides are increasingly being utilized in the de novo selection of aptamers for enhancing their drug-like character and abolishing the need for time consuming trial-and-error based post-selection modifications. Locked nucleic acid (LNA) is one of the most prominent and successful nucleic acid analogues because of its remarkable properties, and widely explored as building blocks in therapeutic oligonucleotides. Evolution of LNA-modified RNA aptamers requires an efficient reverse transcription method for PCR enrichment of the selected RNA aptamer candidates. Establishing this key step is a pre-requisite for performing LNA-modified RNA aptamer selection. Methodology In this study three different reverse transcriptases were investigated towards the enzymatic recognition of LNA nucleotides. Both incorporation as well as reading capabilities of the LNA nucleotides was investigated to fully understand the limitations of the enzymatic recognition. Conclusions We found that SuperScript® III Reverse Transcriptase is an efficient enzyme for the recognition of LNA nucleotides, making it a prime candidate to be used in de novo selection of LNA containing RNA aptamers


Introduction
Nucleic acid-based therapeutic approaches have gained significant interests in recent years towards the development of drugs against many diseases. Nucleic acid aptamers [1][2][3][4][5] are an emerging class of therapeutic molecules, and a chemicallymodified RNA aptamer MacugenH (Pegaptanib sodium), an inhibitor of vascular endothelial growth factor (VEGF) has been approved for the treatment of age related macular degeneration (AMD) [6,7]. To improve the pharmacokinetic properties of aptamers e.g., nuclease resistance, chemically-modified nucleotides have to be used. Aptamers are normally generated by an in vitro evolution process referred to as SELEX (Systematic Evolution of Ligands by EXponential enrichment) [8][9][10]. Application of modified nucleotides in SELEX processes is rather limited because of their poor substrate properties to various polymerases.
Locked Nucleic Acid (LNA) is one of the most prominent and successful nucleic acid analogues because of its remarkable properties, and it is widely explored as building blocks in therapeutic oligonucleotides [11][12][13][14]. LNA nucleotides are generally considered to be RNA mimicking molecules in which the ribose sugar moiety is locked by an oxymethylene linkage connecting the C29 and C49 carbon atoms, imposing conformational restriction to adopt C39-endo/N -type furanose conformation ( Figure 1) [15][16][17]. We have previously reported the enzymatic recognition capabilities of LNA nucleotides by DNA polymerases towards their applicability in the evolution of DNA aptamers [18][19][20][21][22]. Evolution of LNA-modified RNA aptamers requires an efficient reverse transcription method for PCR enrichment of the selected RNA aptamer candidates. Establishing this key step is a pre-requisite for performing LNA-modified RNA aptamer selection. Herein, for the first time, we report an efficient reverse transcription using LNA nucleotides.

Results and Discussion
First, we investigated LNA-TTP (LNA-T-59-triphosphate) as a substrate of three different reverse transcriptases; SuperScriptH III, M-MuLV (Moloney Murine Leukemia Virus) and AMV (Avian Myo-blastosis Virus). The designed RNA template T1 (43 nt, Figure 2A) has three sites of incorporations for LNA-T nucleotides. In all experiments, we included positive and negative control reactions, in parallel to the one involving LNA-TTP in place of dTTP. The positive control reaction contained all four natural dNTPs, resulting in full-length cDNA products. The negative control reaction mixture lacked the nucleoside triphosphate similar to the LNA nucleotide to be tested for incorporation, and thus the extension was expected to stop at the nucleotide prior to the first site of LNA incorporation. The resulting cDNA products were resolved by denaturing polyacrylamide gel electrophoresis for which the primer DNA P1 (19 nt, Figure 2A) The experiments demonstrated that SuperScriptH III Reverse Transcriptase efficiently incorporated LNA-T nucleotides, encod-ed by RNA nucleotides of the template strand, and are capable of producing full-length cDNA products in good yield ( Figure 2B, lane 4 & C). To check the fidelity and accuracy of the reaction products, we performed polymerase chain reaction (PCR) using the cDNA product as the template followed by cloning and sequencing. We have previously reported the PCR involving LNA-nucleotides [20]. The sequencing chromatogram clearly matched with the expected PCR product sequences confirming the accuracy of the reverse transcription by SuperScriptH III ( Figure S1, supporting information). The yield of full-length cDNA containing LNA nucleotides was greater at 45 minutes of incubation ( Figure 2C). M-MuLV Reverse Transcriptase was also able to produce the full-length cDNA, however, it was not as efficient as SuperScriptH III Reverse Transcriptase ( Figure S2B lane 4, supporting information). AMV Reverse Transcriptase afforded the full-length cDNA product in only very low yield ( Figure S2C lane 4, supporting information).
Next, we performed an experiment using both LNA-TTP and LNA-ATP in one reaction by replacing dTTP and dATP. The designed RNA template sequence T1 directs incorporation of first three LNA-T nucleotides followed by three LNA-A nucleotides in the resulting cDNA product. The results show that SuperScriptH III Reverse Transcriptase efficiently incorporated both LNA-T and LNA-A nucleotides and furnished the full-length cDNA product in excellent yield ( Figure 3B, lane 4). In this experiment,  both AMV and M-MuLV Reverse Transcriptases failed to afford the expected full-length cDNA product.
The first enzymatic step involved in RNA aptamer selection by SELEX is to convert the selected RNA aptamer sequences to a complementary DNA for PCR amplification. Encouraged by the initial results, we then performed reverse transcription experiments using an LNA-modified RNA template T2. The designed RNA template has four LNA nucleotides with two LNA-A nucleotides first followed by two LNA-T nucleotides opposite to which the desired DNA nucleotides have to be incorporated. We have also conducted one reaction to check the incorporation of LNA-T nucleotides opposite to LNA-A nucleotides of the template sequence by replacing dTTP with LNA-TTP in the dNTP mixture. These experiments revealed that SuperScriptH III Reverse Transcriptase efficiently transcribed the LNA nucleotides of the template RNA strand by incorporating DNA nucleotides and afforded the cDNA product in very good yield ( Figure 3C, lane 2). PCR amplification followed by cloning and subsequent sequencing analysis was performed to verify the cDNA product. The resulting sequence chromatogram clearly matched with the expected PCR product sequences confirming the accuracy of the reverse transcription using LNA-modified RNA template ( Figure 4C, lane 2 & D). Furthermore, SuperScriptH III Reverse Transcriptase was able to incorporate LNA-T nucleotides opposite to LNA-A nucleotides of the template strand and furnished an LNA-containing full-length cDNA product in reasonably good yield ( Figure 3C, lane 4). AMV and M-MuLV Reverse Transcriptases were less efficient to afford the full-length cDNA products compared to the Super Script III Reverse Transcriptase.
In summary, we have demonstrated that SuperScriptH III Reverse Transcriptase is an efficient enzyme to recognize LNA nucleotides and to synthesize a complementary DNA copy from an LNA-modified RNA template. These findings are vital and pave the way towards applying LNA nucleotides in RNA aptamer selection by in vitro evolution processes.

Materials
DNA sequences were purchased from Integrated DNA Technologies (Coralville, Iowa). SuperScriptH III Reverse Transcriptase, RNase OUT TM (Ribonuclease inhibitor) and RNase H were purchased from Invitrogen Australia (Mulgrave, Victoria). M-MuLV and AMV Reverse Transcriptases were purchased from New England biolabs (supplied by Genesearch Australia). Phusion DNA polymerase was purchased from Finnzymes (supplied by Genesearch Australia).

Reverse transcription
The 59-end labelled primer DNA was annealed with the templates by mixing primer and template in a molar ratio of 1:2 and heating to 80uC for 3 min followed by slow cooling to 37uC.

For SuperScriptH III Reverse Transcriptase
The reaction mixtures were prepared in a total volume of 20 mL by adding 4 mL of 56 First Strand buffer, 4 mL of MgCl 2 (25 mM), 1.5 mL of dithiothreitol (DTT, 100 mM), 1.5 mL of dNTPs/LNA-NTPs (2.5 mM of dNTP and 5 mM for LNA-NTP) mixture, 1.5 mL of a solution containing 59-FAM-labelled primertemplate complex (5:10 pmol ratio), 5 mL of two times distilled water, 1 mL RNase OUT TM (40 U/mL) and 1 mL of SuperScriptH III Reverse Transcriptase (250 U/mL). The reaction mixture was gently vortexed and incubated for 45 min at 50uC. Positive and negative control reactions were stopped after 3 minutes of incubation at 50uC as prolonged incubation resulted in product degradation. Then, the reaction mixtures were heated at 80uC for 5 minutes to inactivate the enzyme. To the resulting mixture, 1 mL of RANase H (5 U/mL) was added and incubated for 1 h at 37uC. The reactions were then quenched by the addition of half volume of loading buffer. Analysis of the products was by performing 13% 7 M urea denaturing polyacrylamide gel electrophoresis in the presence of TBE buffer (100 mM Tris, 90 mM boric acid, 1 mM EDTA).

For M-MulV and AMV Reverse Transcriptase
In this case, the reaction mixtures were prepared in a total volume of 20 mL by adding 2 mL of 106 RT buffer, 4 mL of dNTPs/LNA-NTPs mixture (prepared the same way as mentioned above), 2 mL of a solution containing 59-FAM-labelled primer-template complex (5:10 pmol ratio), 10 mL of two times distilled water, 1 mL RNase OUT TM (40 U/mL) and 1 mL of M-MuLV (200 U/mL) or AMV (10 U/mL) Reverse Transcriptase. The reaction mixtures were gently vortexed and incubated for 1 h at 42uC. Positive and negative control reaction was stopped after 3 minutes of incubation at 42uC. Then the reaction mixtures were heated at 90uC for 10 min to inactivate the enzyme. To the resulting mixture, 1 mL of RNase H (5 U/mL) was added and incubated for 1 h at 37uC. The reactions were then quenched by the addition of half volume of loading buffer and analysed as discussed above.

General procedure for polymerase chain reactions
The PCR reaction mixture was prepared in a total volume of 50 mL by adding 10 mL 56 Phusion HF buffer (included in the Phusion DNA polymerase kit), 4 mL of dNTPs (400 mM), 28 mL of two times distilled water, 1.5 mL of forward primer (50 mM), 1.5 mL of reverse primer (50 mM), 5 mL of template (0,7 mM) stemming from the reverse transcription and 0,5 mL of Phusion DNA polymerase (250 U/mL). The reaction mixtures were gently vortexed and then amplified using a thermal cycler (S1000 TM Thermal cycler, Bio-Rad). A 25-cycle PCR consisted of denaturation at 98uC for 10 second, annealing at 55uC for 15 seconds and extension at 72uC for 25 seconds. After the polymerase reactions, gel-loading buffer (included in Ultra Low DNA size marker kit from Fermentas, supplied by VWR Australia) was added (1.5 mL) and the products were analysed by 4% agarose gel electrophoresis followed by UV-photography.

Cloning and Sequencing
Purified PCR products (50 ng) were cloned into pCR-Blunt (Invitrogen) according to the manufacturer's instructions. Plasmid DNA was extracted using QIAprep (Qiagen) and sequenced by the Australian Genome Research Facility (AGRF, Brisbane, Australia) to validate sequence changes. Figure S1 Sequence analysis of PCR product obtained from a cDNA template containing LNA-T by cloning and sequencing. Position of LNA-T in the expected PCR product is matched with a double headed arrow as identified in the sequencing chromatogram. (TIF)