Protocol for the generation and purification of high-molecular-weight covalent RNA-DNA hybrids with T4 RNA ligase

Summary

RNA-DNA covalent hybrids (RDHs) are widely employed in biology. Although RDHs can be manufactured, the synthesis of molecules longer than 120 nucleotides is challenging. Here, we present a protocol for the generation and purification of high-grade purified high-molecular-weight 5′-RNA-DNA-3′ hybrids. We describe steps for preparing oligos and buffers, ligation reaction, and high-performance liquid chromatography-based RDH purification. This protocol is executable in standard molecular biology laboratories.

Graphical abstract

Highlights

• •Protocol for the generation of large-sized 5′-RNA-DNA-3′ covalent hybrids
• •T4 RNA ligase I-mediated ligation for the generation of hybrid molecules
• •HPLC-based procedure for the purification of hybrid nucleic acidic molecules

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

RNA-DNA covalent hybrids (RDHs) are widely employed in biology. Although RDHs can be manufactured, the synthesis of molecules longer than 120 nucleotides is challenging. Here, we present a protocol for the generation and purification of high-grade purified high-molecular-weight 5′-RNA-DNA-3′ hybrids. We describe steps for preparing oligos and buffers, ligation reaction, and high-performance liquid chromatography-based RDH purification. This protocol is executable in standard molecular biology laboratories.

Before you begin

RDHs are currently employed in a wide variety of applications in biology such as stabilization of small interference RNA (siRNA) molecules,^1,2,3 generation of transcription factor targeting chimeras (TRAFTAC),⁴ design of ricin inhibitors⁵ and identification of single nucleotide polymorphisms (SNPs).⁶ Although RDHs are found in nature⁷ and can be obtained by means of chemical⁸ or enzymatic reactions,^9,10,11 currently, the majority of molecular biology companies are able to produce RDHs whose length is limited up to 120 nucleotides. Furthermore, a detailed protocol for the production and purification of RDHs, which allows circumventing the limitations due to the hybrid size, is yet to be available in the literature. Here, we provide the very first optimized protocol for the generation and purification of high-grade purified high molecular weight 5′-RNA-DNA-3′ covalent hybrids by using T4 RNA ligase I, followed by high-performance liquid chromatography (HPLC) purification. Our protocol provides a simple and rapid approach that is easily executable in standard molecular biology laboratories. More specifically, here we synthesized different RDHs starting from a single 81 nucleotide RNA molecule, namely the CRISPR-Cas9-binding RNA,⁴ and several DNA molecules of diverse lengths harboring different sequences. By following this protocol, we generated and purified RDHs covalent hybrids with a length ranging from 120 to 175 nucleotides.

Oligos design and preparation

Timing: 5–10 min

Note: The DNA sequences listed have been chosen in a random manner, with the aim of showing that the use of diverse DNA oligos differing in both sequence and size does not significantly affect either the RDH ligation reaction or the subsequent purification. However, the T4 RNA ligase manufacturer, i.e. New England Biolabs, recommends using oligos with a length of at least 10 nucleotides. The RNA oligo that has been employed in this study is the CrispR Cas9-binding RNA.¹² The choice of this stem-loop RNA for the optimization of RDH generation/purification has been dictated by the evidence that T4 RNA ligase efficiency is known to be affected by the secondary structure within an RNA substrate.¹³ Moreover, as described in the T4 RNA ligase manufacturer’s manual RNA secondary structure shielding the ends can inhibit ligation catalyzed by T4 RNA ligase. Thus, the use of a well-known RNA molecule showing a stable and complex secondary structure¹² for the protocol optimization has allowed to assess and address any potential influence of secondary structures on the ligation reaction and the subsequent RDH purification from the beginning and thus obtaining an optimized protocol for the generation/purification of RDHs even in the presence of non-linear conformations such as secondary structures.

Note: The oligonucleotides used are at an initial concentration of 100 μM.

• 1.
  Dilute the DNA and RNA oligonucleotides to a final concentration of 10 μM.

  • a.
    Add in an autoclaved 1.5 mL Eppendorf tube 18 μL of Nuclease-free water and 2 μL of DNA oligonucleotide.
  • b.
    Repeat the same step for the dilution of the RNA oligonucleotide.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
T4 RNA ligase I kit	NEB	M0204S
Monarch RNA cleanup kit (10 μg)	NEB	T2030S
DMSO	Sigma	D2650
Nuclease-free water	Promega	P1193
TRIS-HCl pH 7.5	Alaos	TR085C
TRIS-HCl pH 8	Alaos	TR095C
NaCl	Alaos	CS131C
40% bis acrylamide/bis solution 19:1	Bio-Rad	1610144
Ammonium persulfate	Sigma	A3678-25G
TEMED	Bio-Rad	161-0800
TRIS-glycine buffer	Bio-Rad	1610734
Orange G	Sigma	O3756
Gelred	Biotium	41003
RNase AWAYy reagent	Invitrogen	10328-011
Ethanol	PanReac AppliChem	UN1170
Glycerol	Sigma	G6279-1L
Oligonucleotides
DNA 1 _oligo: CGACGGTTAGTCTG CCGACTATCGTTAGGGTTAGGGTT AGGGTTAGGGTTAGGGTTAGGGT TAGGGCGACGGTTAGTCTGCCGA CTATCG	Merck	N/A
DNA 2 _oligo: ATCCGCTCAACTGATCG TACGTCGCAGTCAGTACTAGTACTGTC AAGCACGCTGTACGCCTCAT	Merck	N/A
DNA 3 _oligo: GCTAGCGGGAATTTCCGG GGTTAGGGTTAGGGTTAGGGTTAGGGT TTCCGGGAATTTCCAGATCT	Merck	N/A
DNA 4 _oligo: AGCGTGTATGGAGCGGCG TCGACGGTTAGTCTGCCGACTATCGGGT TGTCTCTAACGTCATCTA	Merck	N/A
RNA oligo: CGUUUUAGAGCUAGAAAUAGC AAGUUAAAAUAAGGCUAGUCCGUUAUCAA CUUGAAAAAGUGGCACCGAGUCGGUGUUUUU	GenScript	N/A
Software and algorithms
Image Lab	Bio-Rad	https://www.bio-rad.com/es-es/product/image-lab-software?ID=KRE6P5E8Z
UNICORN 7.8	Cytiva	https://www.cytivalifesciences.com/en/us/shop/unicorn-5-31-p-01433
RNA CONTRAfold	CONTRAfold	http://contra.stanford.edu/contrafold/index.html
RNAstructure tool	RNAstructure, prof. David H. Mathews laboratory	https://rna.urmc.rochester.edu/RNAstructureWeb/Servers/Predict1/Example.php
FastPCR	PrimerDigital	https://primerdigital.com/fastpcr.html
Other
Low-binding tips 10 μL	Thermo Scientific	2139-05
Low-binding filter tips 20 μL	Thermo Scientific	2149P-05
Low-binding filter tips 200 μL	Thermo Scientific	2069-05
Low-binding filter tips 1,000 μL	Neptune	BT1000.96
Falcon 15 mL	Biofil	LC066
Thermomixer comfort	Eppendorf	20444601112
PCR thermocycler	MWG Biotech	Primus 96
Vortex minishaker	IKA	MS2
Centrifuge 1.5 mL tube	Eppendorf	5415 R
Superdex 200 increase 3.2/300	Cytiva	28990946
100 μL gastight syringe model 1710, small removable needle, 22s gauge, 2 in, point style 2	Hamilton	81030
ÄKTA pure micro	Cytiva	29302479
Mini-PROTEAN Tetra cell	Bio-Rad	1658000
Electrophoresis generator	Apelex	PS 608
Spectrafuge mini centrifuge	Labnet	c1301-p-230v
Scientific platform rocker	Stuart	STR6
200 μL (1–200 μL) MultiFlex Round	Sorenson	13810
Molecular Imager Gel Doc XR+ system with Image Lab software	Bio-Rad	1708195

Materials and equipment

Purification buffer
Reagent	Final concentration	Amount
NaCl (5 M)	50 mM	5 mL
TRIS-HCl pH 7.5 (1 M)	50 mM	25 mL
Nuclease-Free water	N/A	470 mL
Total	N/A	500 mL

Note: Store at 4°C. It can be stored at 4°C for not more than one week.

5X Orange G (100 mL)
Reagent	Final concentration	Amount
TRIS-HCl 7.5 (1 M)	10 mM	1 mL
Glycerol	60%	60 mL
EDTA (0.5 M)	60 mM	12 mL
Orange G	0.15%	150 mg
Nuclease-Free water	N/A	27 mL
Total	N/A	100 mL

Note: Prepare 1 mL aliquots of 5X Orange G in 1.5 mL tubes and store them at ‒20°C. The aliquots can be stored at ‒20°C for several months.

• •10% ammonium persulfate (APS) solution: add 0.1 g ammonium persulfate in 1 mL nuclease-free water.

Note: The aliquots can be stored at ‒20°C for several months.

• •1X GelRed solution: add 2 μL of 10.000X GelRed in 20 mL of nuclease-free water.

Note: Store at 20°C‒25°C in the dark for not more than one month.

• •1X TRIS-glycine buffer: add 100 mL 10X TRIS-glycine in 900 mL nuclease-free water.

Note: Store at 20°C‒25°C. It can be stored for months.

Step-by-step method details

Ligation reaction

Timing: 18–20 h

The protocol below describes the specific steps for the generation of 5′-RNA-DNA-3′ covalent hybrids by using T4 RNA ligase I (Figure 1A). 25 ligation reactions will yield elution fractions harboring ∼0.3–1 μM of >95% pure RDH following HPLC-based fractionation step.

Note: Before starting, clean the bench, pipettes and all the materials with Ethanol 70% and RNase-Free Away Reagent.

• 1.Prepare 25 autoclaved 200 μL tubes.
• 2.
  Thaw the following reagents.

  • a.
    T4 RNA ligase kit (keep T4 RNA ligase on ice).

    Note: T4 RNA ligase I is kept liquid at ‒20°C

  • b.
    DNA oligonucleotide.
  • c.
    RNA oligonucleotide.
• 3.Protocol reaction.

Reagent	Final concentration	Amount
DMSO	10%	2 μL
T4 RNA ligase buffer 10X	1X	2 μL
ATP (10 mM)	1 mM	2 μL
DNA (10 μM)	0.1 μM	0.2 μL
RNA (10 μM)	0.25 μM	0.5 μL
PEG (8000 units)	15%	3 μL
T4 RNA ligase (1000 U/mL)	10 U	2 μL
Nuclease-free H20	N/A	8.3 μL
Total	N/A	20 μL

Note: Keep the reaction mix on ice, and do not store it for more than 1 hour.

Note: Pulse-vortex each reagent before adding it to the mix in the 200 μL tubes.

CRITICAL: Due to the viscous nature of PEG, using standard pipette tips may lead to pipetting errors and handling of incorrect volumes of PEG, which might affect/interfere with the downstream purification steps. Cutting the final part of the pipette tip will avoid any obstruction caused by the viscous properties of PEG, thus allowing handling of the proper volume of PEG required for the ligation reaction (Figure 1B).

CRITICAL: A proper pipette cut is crucial for ensuring the correct collection of the desired volume of viscous reagents. We recommend to cut the tip from the narrower side in such a way as to reduce its length by 5–7 mm. A higher reduction in tip size (>7 mm) may result in the loss of pipetting precision with the risk to collect volumes that do not correspond to the actual intended volume needed. Smaller reductions in tip size (<5 mm) may not be sufficient to ensure the generation of an opening displaying the sufficient diameter that is required for properly handling of viscous liquids. Use scissors previously cleaned with ethanol and milli-Q pure water for cutting the pipette tip (Figure 1B).

Note: Before the addition of the ligase, briefly spin the samples in the Spectrafuge Mini Centrifuge for 5 seconds to spin down the reaction (Figures 1C and 1D).

Note: It is important that the T4 RNA Ligase is added as the last step and completely resuspended in the mix by pipetting.

• 4.Set the PCR thermocycler at 16°C.
• 5.Incubate the ligation samples in the PCR thermocycler for 16 h (Figure 1E).
• 6.
  Reaction inactivation with Monarch RNA Cleanup kit.

  • a.
    Mix all the ligation reactions together in a single Eppendorf tube (total volume ∼500 μL).
  • b.
    Add 2 volumes of RNA Cleanup Binding Buffer and transfer the resulting 1.5 mL mix into a 15 mL falcon tube.
  • c.
    Add 1 volume of ethanol (>95% pure) to your sample and mix well by pipetting up and down or flicking the 15 mL falcon tube. Do not vortex.
  • d.
    Insert the column into a collection tube, load the sample onto the column and close the cap. Spin for 1 min, then discard the flowthrough. Repeat if it is necessary.
  • e.
    Re-insert the column into the collection tube, add 500 μL of RNA Cleanup Wash Buffer, spin for 1 min, and discard the flow-through.
  • f.
    Repeat the wash: add 500 μL of RNA Cleanup Wash Buffer to the column, spin for 1 min and discard the flow-through.
  • g.
    Re-spin for 1 min to ensure traces of salt and ethanol are not carried over to the next step.
  • h.
    Transfer the column to an RNase-free 1.5 mL microfuge tube.
  • i.
    Add 50 μL of nuclease-free water to elute the sample.
  • j.
    Spin for 1 min.
  • k.
    Repeat steps I and j to a final 100 μL volume to get the sample.

Note: All centrifugation steps should be carried out at 20°C‒25°C at 16.000× g.

• 7.The samples are ready to be purified by HPLC.

Note: The volume used for the purification step is 100 μL.

Pause point: At this point the samples can be frozen and stored at ‒80°C. We do not recommend storing the samples for more than one week.

Note: Keep 10% of the ligation as a control input to run in the polyacrylamide gel.

Figure 1.

Schematic representation of ligation reaction as well as the corresponding equipment and settings employed for each step of this procedure

(A) Ligation reaction between RNA DNA oligos catalyzed by T4 RNA ligase I.

(B) Cut with clean scissors the p20 tip before pipetting PEG.

(C) A 200 μL PCR tube with some drops of ligation reaction in the wall.

(D) Spin the ligation reactions into a Spectrafuge Mini Centrifuge.

(E) The ligation reaction is incubated at 16°C for 16 h in the thermocycler.

(F) After the incubation at 16°C the ligation reaction is ready to be loaded in Superdex column (F) for the HPLC run.

(G) Before the HPLC run the Superdex column (shown in (F)) must be connected to the ÄKTA pure micro apparatus through the connectors.

(H) Settings for the washing step of the Superdex column employed in this protocol by using UNICORN version 7.8 software.

(I) Settings for the chromatography run employed in this protocol by using UNICORN version 7.8 software.

HPLC-based RDH purification

Timing: 7–8 h

For downstream applications of RDHs, it is highly recommendable to remove non-ligated RNA and DNA oligos as well as the reaction buffer, which might affect/interfere with the downstream uses. To this end, the ligated covalent RDHs were purified by HPLC gel filtration chromatography in order to obtain hybrid molecules that were as pure as possible and resuspended in a buffer suitable for most downstream applications, such as cell culture treatment or in vitro assays.

• 8.Mount the Superdex 200 increase 3.2/300 column in the ÄKTA pure micro apparatus (Figures 1F and 1G).
• 9.Equilibrate the Superdex 200 increase 3.2/300 column with purification buffer by using the settings shown in Figure 1H.
• 10.Introduce 100 μL of the sample by using the 100 μL Gastight Syringe in the Superdex200 increase 3.2/300 column coupled to the ÄKTA pure micro apparatus.

CRITICAL: Sample collection with a sterile syringe and the subsequent sample loading on the column should be performed extremely carefully with the aim of avoiding the introduction of air bubbles in the apparatus. Indeed, the loading of air bubbles into the ÄKTA pure micro system might cause HPLC apparatus damage/malfunctioning and loss of RDH sample. Therefore, particular attention should be given to the eventual presence of air bubbles in the sample before its loading on the column. Should air bubbles be present, they must be removed carefully from the sample before loading it on the column.

• 11.Start the chromatography run using the settings shown in Figure 1I.

Note: the settings chosen (Figures 1H and 1I) allow to minimize the interval volume of the system, therefore increasing the resolution of the size exclusion chromatography (SEC).

Note: Make sure the connection between the column and the ÄKTA pure micro apparatus is well-established.

• 12.Wait for the samples to run into the column and be eluted in different fractions on the basis of their size.
• 13.Collect the fractions corresponding to the peaks (Figure 2A), which potentially harbor the hybrid molecules.

Note: The first time, it is recommended to collect the fractions corresponding to all the peaks coming from the chromatogram so that you can easily identify the peak and the corresponding fractions containing the ligated RDH.

Note: Keep the fractions on ice.

• 14.Preparation of 15% polyacrylamide gel (2 gels).
  In a 15 mL falcon.

  • a.
    Add 2.7 mL of nuclease-free H₂0.
  • b.
    Add 4.8 mL of TRIS pH 8.0.
  • c.
    Add 4.5 mL of 40% Bis Acrylamide/Bis solution 19:1.
  • d.
    Add 200 μL APS 10%.
  • e.
    Add 10 μL TEMED.

    Note: Use 0.75 mm glasses and the corresponding 15-well comb for each gel.

• 15.Allow the gel to polymerize.
• 16.Pre-run the gel using TRIS-glycine 1X buffer with Electrophoresis power @100 V for 1 h@20°C-25°C.

Note: Remove potential gel impurities that may be settled in the bottom of the gel wells by washing the wells with millipore water.

• 17.Mix 1 μL of each sample (corresponding to ∼7% for each elution fraction) with 1 μL of Orange G 5X.

Note: Orange G volume can be increased up to 2 μL (2 volumes) to ensure sample precipitation in the gel wells.

• 18.Load the fractions in the gel and run it @150 V for 1.5 h (Figure 2B).

Note: Load the non-purified ligation, and the RNA and DNA oligonucleotides as a control on the same gel to distinguish the different bands on the basis of their molecular weight.

• 19.Dye the gel with Gel Red 1X for 10 min at 20°C‒25°C in agitation in rocking platform.
• 20.
  Position the gel in a gel imager (e.g., Molecular Imager Gel Doc XR) and follow the manufacturer’s instructions (Figure 2C).

  • a.
    Application→ Nucleic acids→Gel Red.
  • b.
    Adjust the position of the gel.
  • c.
    The Image Lab software automatically optimizes the exposure for intense bands.
  • d.
    Should be necessary, manually change the duration of UV exposure.

Figure 2.

Equipment and settings employed for purity evaluation and quantification of HPLC-purified RDH

(A) The fractions obtained after the HPLC-mediated purification are collected in a 96-well plate.

(B) The fractions harboring the RDH are then solved by electrophoresis to check their purity and concentration.

(C) Settings to observe the presence of the RDH in Gel Imaging with Image Lab program.

Expected outcomes

We determined that the range of 20–25 ligase reactions (yielding ∼0.35 μM of nucleic acids), corresponding to a total reaction volume of 0.4–0.5 mL, subsequently purified using the Monarch RNA cleanup kit and eluted in 100 μL for column loading, as being the optimal range for successful HPLC-mediated purification of the RDH. In our hands, loading the HPLC column with a number of reactions significantly below this range (<11) does not affect the degree of purification of the RDHs, but results in a significantly lower final yield of the purified RDH. With this protocol, the HPLC-based fractionation of ligation reactions results in a chromatogram characterized by three distinct peaks that can be more or less separated depending on the size of the RNA and DNA oligos employed in the reaction. In all the cases described in this report, the first (leftmost) peak corresponds to the RDH, whereas the remaining peaks are given by the free non-ligated RNA and free non-ligated DNA oligos (Figure 3). Importantly, regardless of the sequence/length of the DNA oligos used in this study (i.e., DNA1, DNA2, DNA3 and DNA4), the elution fractions of this first peak carry 95%-to-100% pure RDH with a final concentration ranging from ∼0.3 to ∼1 μM (Figure 3, see quantification section). The total amount of >95% pure RDH molecules that can be obtained with this protocol is ∼2–3 ng. The grade of separation of the first RDH-containing peak from the other peaks strongly affects the ability to isolate highly pure RDH fractions (Figure 3).

Figure 3.

Generation and HPLC-mediated purification of different RDH molecules

(A-D) CrispR gRNA molecule was covalently ligated to single stranded (A) DNA1, (B) DNA2, (C) DNA3 and (D) DNA4 oligonucleotides, each of which differing in their sequence and length (see key resources table).The selected HPLC-separated fractions (up), indicated by the blue brackets and corresponding arrows, were run on a denaturing PAGE (bottom) and solved on the basis of their size. Non-ligated RNA and DNA oligonucleotides were run as a control (on the left of each gel). The concentration and the percentage of purity of each RDH fraction are indicated in each gel. DNA = DNA oligos, RNA = RNA oligo, LIG = ligation reaction, ssRNA = single stranded RNA, ssDNA = single stranded DNA. Note that the slight shift in the run of the nucleic acids in the reaction ligation (LIG) might be ascribable to the presence of PEG.

Quantification and statistical analysis

The concentration of the HPLC-purified RDH fractions (Figure 3) was determined by using Image Lab tool and employing the non-ligated DNA and RNA oligos at known concentrations as standards. More specifically, 0.1 μM of DNA oligo and 0.25 μM of RNA oligo (i.e., 0.2 μL and 0.5 μL of respectively DNA and RNA oligos from an initial concentration of 10 μM) were diluted in 2 μL of Orange G dye (required for visualizing nucleic acids on polyacrylamide gels) and run in the gel. The level of purity of the HPLC-purified RDH fractions was measured by determining the relative amount of ligated hybrid molecule with respect to total nucleic acids (RNA-DNA hybrid + free non-ligated DNA + free non-ligated RNA) for each HPLC-purified fraction (Figures 3 and 4).

Figure 4.

Optimization of ligation reaction for the improvement of the RDH yield

Test with different concentrations of DNA, RNA and ligase. Changes in ligase concentration do not affect the efficiency of the reaction. The stoichiometric ratio 2.5:1 RNA-DNA (lane 5) displays the best efficiency in promoting the ligation reaction. The numbers below the bands indicate the relative abundance (in percentage) of each band compared to the total nucleic acids in the corresponding lane.

Limitations

Potential influence of oligo size on the final yield of RDHs: here we used several DNA oligos of different sizes, ranging from 64 to 90 nucleotides in length, and a single RNA oligo of 81 nucleotides. Importantly, regardless of the oligo length, we obtained similar yields (0.4–1 μM) for all the corresponding HPLC-purified hybrids we tested. Thus, the protocol presented here is suitable for the generation of RDHs ranging in length from 145‒171 nucleotides and yields comparable amounts of different RDHs in an oligo-size independent manner. However, it is essential to note that the manufacturer of the T4 RNA ligase, New England Biolabs, recommends using oligos of at least 10 nucleotides in length. Therefore, it cannot be excluded that the use of oligos outside the above tested range may affect the final yield of RDH. Similarly, we cannot rule out the possibility that the final amount of purified RDH molecules, falling outside the previously specified length range, might be affected and require further optimization. Although this protocol allows the generation of at least ∼2 ng of highly purified (>95% pure) RDHs regardless of their sequence and length, the final yield of the purest RDH fractions suitable for downstream applications may be further improved. A possible strategy to increase the amount of RDHs might be scaling up the number of initial ligation reactions. However, it would also be valuable to explore alternative methods that can further concentrate the reaction without compromising the integrity of the RDHs. Such improvements would allow working with higher concentrations of the desired hybrids.

Troubleshooting

Problem 1

Potential influence of secondary structures of the oligos on ligation efficiency: one of the main causes for oligo aggregation is secondary structure formation, oligonucleotides can fold into secondary structures, such as hairpins or dimers, which can lead to self-association and aggregation. This is particularly relevant when designing longer oligos or sequences with repeated motifs. T4 RNA ligase efficiency is known to be affected by the secondary structure within an RNA substrate,¹³ and according to T4 RNA ligase manufacturer’s manual, RNA secondary structure shielding the ends can inhibit ligation catalyzed by T4 RNA ligase. Furthermore, according to T4 RNA ligase manufacturer, i.e., NEB, T4 RNA ligase will work in the presence of DNA dimers as long as the overhangs are long enough. However, if the overhangs are too short, e.g., 1–3 nucleotides long, the dimers might affect the ligase reaction. (Related to oligos design and preparation section).

Potential solution

We always recommend to carry out preliminary in silico tests for the DNA/RNA oligo sequences that will be employed for the RDH generation before performing the ligation reaction. Such in silico tests, which can be run by using either on-line prediction tools or open source softwares (see below), would be useful to predict the ability of the DNA/RNA oligos to give rise to dimers/secondary structures. To this end, we recommend RNA CONTRAfold for the RNA secondary structure prediction. For the prediction of DNA and RNA secondary structures we also suggest RNAstructure tool. Finally, for the prediction of oligo dimer formation we recommend the free-downloadable software FastPCR. For all the cases, we suggest, as a starting point, to use default settings for the prediction. Importantly, should the presence of secondary structures in the oligos significantly affect the final yield of the RDH, as indicated in T4 RNA ligase manufacturer’s manual, the addition of DMSO 10% (v/v) can increase ligation efficiency. It is important to emphasize that we already included in our protocol the use of 10% DMSO in the ligation reaction as default parameter. Indeed, for all the above-mentioned reasons, we chose to test for the initial optimization of our protocol a well-known RNA molecule, i.e., CrispR Cas9-binding RNA, that is known to show a stable and complex secondary structure.¹² In this manner, we reasoned that we could assess and address any potential influence of secondary structures on the ligation reaction from the beginning and thus obtaining a good yield of purified RDHs even in the presence of non-linear conformations such as secondary structures/oligos dimers/aggregates. Thus, in principle the protocol presented here has been optimized taking into account the potential limitations due to the presence of secondary structures in the oligos and yielded high-grade purified RDHs harboring complex conformations such as stem-loop structures. It is also recommendable for potential users to consider using an RNA molecule like the one tested here as a positive control of the procedure.

Problem 2

For a successful covalent ligation of the DNA and RNA oligos in the proper orientation, it is essential that the donor molecule, i.e., the DNA oligo, possesses a 5′-PO4 group and it is blocked by phosphorylation at the 3′ end (Figure 5). This configuration allows the T4 RNA ligase to catalyze the ligation between the 3′-OH end of the RNA molecule and the DNA 5′-PO4 group, thus allowing the formation of 5′-RNA-DNA-3′ hybrid molecules. (Related to oligos design and preparation section).

Figure 5.

Schematic representation showing the blocked 3′termini of the DNA oligonucleotides described in this study

LCAA (long chain alkylamine); CPG (controlled pore glass).

Potential solution

When using this protocol, be sure to use 5′Phosphorylated DNA oligos that are blocked at their 3′end (Figure 5). DNA oligo blocking by 3′ phosphorylation is performed by standard molecular biology companies, therefore blocked DNA oligos can be ordered with the block already pre-installed, and no additional step is required to be performed in the lab.

Problem 3

In step 3, choosing the right concentration of polyethylene glycol (PEG) is a critical point. Indeed, PEG is a viscous liquid and solidifies at 4°C at concentrations higher than 15%. Since the working temperature of the ÄKTA pure micro column is 4°C, high concentrations of PEG in the ligation reaction will lead to column clogging and inefficient RDH purification. (Related to step 3).

Potential solution

Concentrations of PEG ≤15% are ideal to prevent the solidification of the ligation reaction and therefore will allow an efficient HPLC-based RDH purification. In addition, the inactivation with the Monarch RNA Cleanup Kit will remove most of the excess PEG in the solution.

Problem 4

In step 3, the stoichiometric ratio of DNA and RNA may affect the final yield of RDH, thus potentially hindering the use of RDHs in downstream applications. Therefore, in order to increase the final yield of RDH, it is advisable to find the optimal stoichiometric ratio of DNA and RNA oligonucleotides that will be used in the ligation reaction. For this purpose, different concentrations of DNA and RNA oligonucleotides should be tested to find the optimal stoichiometric ratio, as shown in Figure 4. (Related to step 3).

Potential solution

An optimal ratio is crucial to increase the final yield of RDHs. In relation to the oligonucleotides described in this protocol, a stoichiometric ratio of 1:2.5 (DNA:RNA) was found to be the optimal condition (Figure 4). Indeed, such a ratio allows the ligase to convert its substrates, i.e., the DNA and RNA oligos, into the RDH with the highest efficiency. At this ratio, the yield of RDHs is ∼45% of the nucleic acids in the ligation reaction, thus minimizing the presence of non-ligated DNA or RNA.

Problem 5

RNA oligonucleotides are highly unstable, and improper handling may lead to their degradation. (Related to step 3).

Potential solution

RNA degradation can be avoided by storing RNA oligonucleotides and RDHs @‒80°C, always working in RNase-free conditions and handling them on ice after thawing.

Problem 6

Selecting fractions in the case of low-resolution HPLC peaks. (Related to step 13).

Potential solution

Under the conditions used, HPLC size exclusion chromatography is a highly sensitive purification system that can effectively separate oligonucleotides of similar molecular weight. Even in instances of partial peak overlapping (as depicted in Figure 3A), precise selection of elution fractions can prevent potential undesired contamination. Specifically, if the separation between RDH and free RNA/DNA fractions is unclear, it is advisable to selectively collect only the earliest elution fractions of the RDH peak. This precaution is taken because the middle/late RDH fractions might be susceptible to contamination from free, non-ligated RNA/DNA oligos associated with overlapping RNA/DNA peaks. To facilitate this selection, we propose drawing an imaginary line from the top of the adjacent RNA/DNA peaks to extrapolate the position corresponding to the initial fractions of peaks containing unligated RNA and DNA oligos (refer to Figure 6). Once the earliest fractions of the non-ligated RNA/DNA peaks are determined, it is recommended to choose only the fractions of the RDH peak preceding the commencement of elution of the non-ligated RNA/DNA-containing fractions (as illustrated in Figure 6). Moreover, should peak overlap be too high, resolution can be also improved by injecting a smaller volume of sample and performing multiple runs. Finally, it is also important to point out that another valuable option that might be considered for improving the peak resolution is to modify some parameters of the chromatography run such as increasing the working temperature of the HPLC-run (i.e., at 20°C‒25°C) and/or decreasing the mobility phase (i.e., flow rate = 0.01 mL/min).

Figure 6.

RDH fraction selection when the HPLC-peaks are not well resolved

In the presence of low-resolution HPLC-peaks, fitting a Gaussian curve to the experimental curves of DNA and RNA non-ligated oligos will allow to extrapolate the complete shape of the DNA/RNA peaks and consequently to select the RDH fractions supposedly showing the minimal free DNA/RNA oligo contamination. In this example, the black and red dashed lines represent the predicted curves of the DNA and RNA oligo peaks, respectively. To minimize potential RNA/DNA contamination of the hybrid it is suggested to take the RDH fractions comprised between the fraction 2.B.10 and fraction 2.B.8, indicated by the blue rectangle.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Maria A. Blasco (mblasco@cnio.es).

Technical contact

Questions about the technical specifics of performing the protocol should be directed to and will be answered by the technical contact, Maria A. Blasco (mblasco@cnio.es).

Materials availability

This study did not generate new unique reagents.

Data and code availability

This study did not generate/analyze either datasets or codes.