Attachment of a 32 P- phosphate to the 3′ Terminus of a DNA Oligonucleotide

[Abstract] Biochemical investigations into DNA-binding and DNA-cutting proteins often benefit from the specific attachment of a radioactive label to one of the two DNA termini. In many cases, it is essential to perform two versions of the same experiment: one with th e 5′ DNA end labeled and one with the 3′ DNA end labeled. While homogeneous 5′ -radiolabeling can be accomplished using a single kinase-catalyzed phosphorylation step, existing procedures for 3′ -radiolabeling often result in probe heterogeneity, prohibiting precise DNA fragment identification in downstream experiments. We present here a new protocol to efficiently attach a 32 P- phosphate to the 3′ end of a DNA oligonucleotide of arbitrary sequence, relying on inexpensive DNA oligonucleotide modifications (2′ -O-methylribonucleotide and ribonucleotide sugar substitutions), two enzymes (T4 polynucleotide kinase and T4 RNA ligase 2), and the differential susceptibility of DNA and RNA to hydroxide treatment. Radioactive probe molecules produced by this protocol are homogeneous and oxidant-compatible, and they can be used for precise cleavage-site mapping in the context of both DNase enzyme characterization and DNA footprinting assays.

aims to determine the mechanism of a novel DNase enzyme or discern protection parameters from DNA footprinting assays. In the classic version of this experiment, the investigator first uses T4 polynucleotide kinase (PNK) to transfer a 32 P-phosphate from [γ-32 P]-adenosine triphosphate (ATP) to the 5′-OH of a DNA oligonucleotide of interest, yielding a 5′-radiolabeled DNA "probe". After subjecting the probe to a cleavage process, the investigator then resolves the cleavage fragments by polyacrylamide gel electrophoresis (PAGE), visualizes them by autoradiography, and determines cleavage position and frequency from band location and intensity, respectively. However, if a DNA probe molecule is cleaved more than once, the described experiment only allows detection of the cleavage event closest to the 5′ end. To detect additional cleavage events with equivalent positional resolution, an analogous experiment must be performed with the 3′ DNA end radiolabeled.
Because there is currently no known kinase activity that can specifically transfer a 32 P-phosphate to the 3′-OH of a DNA oligonucleotide, standard 3′-radiolabeling procedures instead employ either a polymerase or a terminal transferase to catalyze primer extension at the 3′ DNA end, using [α-32 P]nucleotide triphosphates (NTPs) as a reagent (in different strategies, the sugar moiety of the NTPs could be 2′-deoxyribose, ribose, or 3′-deoxyribose) (Wu et al., 1976). While the resulting probes are suitable for certain experiments, they cannot always be used for unambiguous DNA cleavage-site mapping (i.e., determining at exactly which phosphodiester a DNA molecule was cut, as measured from the 3′ DNA end). For example, the polymerase-or transferase-based methods suffer from heterogeneity in probe length and/or incorporation of radionuclides at internal backbone positions, both of which can destroy the unique band-to-fragment correspondence that is required for unambiguous cleavage-site mapping. When using probes generated by one of these methods, a single band on a PAGE autoradiograph could be attributed to any of several distinct radioactive cleavage fragments that migrate with indistinguishable electrophoretic mobility. Note that these problems arise even when the 32 P-phosphate lies between the final two nucleotides, as certain nuclease enzymes and chemical processes (such as those associated with permanganate footprinting) can remove the final nucleoside without its phosphate. In one protocol that cleverly addresses this problem, the authors expose an internal radionuclide by sequentially treating a ribonucleotide-tailed radioprobe with alkali, phosphatase, and periodate. However, while this procedure does yield the desired outcome-a DNA probe homogeneously labeled with a terminal 3′-32 Pphosphate-the initial tailing step requires high concentrations of expensive radioactive ribonucleotides to promote efficient labeling (Jay et al., 1974).
More recently, many synthesis companies have begun to offer custom DNA oligonucleotides with a fluorophore chemically conjugated to the 3′ end. If a laboratory has the optical instrumentation to directly image fluorophore-tagged DNA in a polyacrylamide gel, these reagents offer a viable alternative to 3′radiolabeled DNA probes. Still, this detection technique is far less sensitive than autoradiography and will miss minor products in complex cleavage patterns. Additionally, common fluorophores may be degraded under the harsh chemical treatments associated with certain experiments, such as permanganate footprinting.
Considering the previously mentioned shortcomings, the ideal 3′-labeling strategy would: (1) take advantage of the high sensitivity and chemical compatibility of 32 P-phosphate-based detection; (2) 3 www.bio-protocol.org/e3787 produce a homogeneous population of DNA probe with a single 3′-terminal radionuclide; and (3) employ catalytically efficient combinations of common enzymes and substrates to make the procedure inexpensive and fast. We developed a protocol to meet these needs in the context of our studies of DNA binding and cleavage by the RNA-guided DNase CRISPR-Cas12a (Cofsky et al., 2020). In our protocol, we begin with a DNA oligonucleotide in which the penultimate and final deoxyribonucleotides have been changed to a 2′-O-methylribonucleotide and a ribonucleotide, respectively, which are standard and cheap modifications at oligonucleotide synthesis companies ( Figure 1A). To introduce the 32 P-phosphate to the system, we first perform a standard 5′-radiolabeling procedure on a "phosphate shuttle" RNA oligonucleotide ( Figure 1B). We then form a double-stranded junction by annealing the phosphate shuttle RNA oligonucleotide, the DNA oligonucleotide, and a bridging splint RNA oligonucleotide ( Figure   1C). With the sugar modifications at the DNA oligonucleotide's 3′ end, this bridge structure can be efficiently ligated by T4 RNA ligase 2, attaching the 32 P-phosphate to the 3′-OH of the DNA oligonucleotide (Nandakumar and Shuman, 2004; Figure 1D). After ligation, we treat the probe precursor with hot hydroxide, leaving the DNA intact and dissociating the RNA into mononucleotides that can be removed with a spin column ( Figures 1E, 1F). The final result is a DNA oligonucleotide with a penultimate 2′-O-methylribonucleotide and a 3′-terminal ribonucleotide bearing a 2′,3′-cyclic-32 P-phosphate ( Figure   2).  produce the required probes, as compared to polymerase-/transferase-based strategies that would also require expensive [α-32 P]-NTPs. In our experiments, we used these probes to detect Cas12a-generated DNA cleavage products (including minor and transient products), leading to the unforeseen finding that Cas12a cleaves its DNA substrate multiple times to form a five-nucleotide gap. We also used these probes in permanganate footprinting experiments that yielded highly position-specific information about the conformation of Cas12a-bound DNA (Cofsky et al., 2020). Additionally, by using a 32 P-phosphate label in these experiments, we avoided the problems associated with the use of fluorophores in the presence of strong oxidants. Finally, when using this protocol to generate many variant DNA probes, reagent costs were ~90% the cost of the cheapest commercially synthesized fluorophore-tagged DNA oligonucleotides, according to list prices at the time of writing. Therefore, this method may be used to prepare inexpensive DNA probes for any cleavage-site mapping experiment that requires high precision, sensitivity, and chemical compatibility.  B. Order or transcribe the phosphate shuttle RNA oligonucleotide and the splint RNA oligonucleotide 1. The phosphate shuttle RNA oligonucleotide can be ordered from an oligonucleotide synthesis company or produced by in vitro transcription. If producing by in vitro transcription (Notes 3 and 4), use a method whose final product contains a 5′-OH (either by treating a T7 RNA polymerase product with phosphatase or including a self-cleaving hammerhead ribozyme in the transcript).

Materials and Reagents
It is essential that the phosphate shuttle RNA contain a 5′-OH for the 5′-radiolabeling step.
a. Phosphate shuttle RNA oligonucleotide sequence (Note 5): We include below the sequence of an in vitro transcription DNA template for T7 RNA polymerase to generate the phosphate shuttle RNA oligonucleotide (Note 6). Transcripts from this template contain a 5′ hammerhead ribozyme that cleaves to yield a 5′-OH on the final phosphate shuttle oligonucleotide.  2. Run a sample of the labeled DNA oligonucleotide on denaturing PAGE to verify that the sample contains a pure radiolabeled DNA fragment of the correct size (Notes 10 and 11, Figure 3).
3. Store the radiolabeled DNA at -20 °C. The limiting factor for the expiration of the radiolabeled oligonucleotide is the radioactivity, as the chemical stability of the product (> years) long outlasts the radioactive decay lifetime of 32 P (half-life 14 days). The product can continue to be used as long as it retains enough radioactivity for the researcher's particular experimental purposes.

Notes
1. For the purposes of mapping a DNA cleavage site with single-nucleotide resolution, the baseline purification option from oligonucleotide synthesis companies is often insufficient (the product contains significant concentrations of truncated synthesis products that may confound experimentally important cleavage products). To achieve adequate DNA oligonucleotide purity before 3′-radiolabeling, either perform a PAGE-purification in house or opt for more extensive purification in your commercial order.
2. Standard precautions for working with RNA should be followed. Keep gloves and surfaces clean, and when possible, use filter-containing pipette tips. We do not detect any RNase activity when using MilliQ water in reactions containing RNA, but the experimentalist can purchase commercially certified RNase-free water if there are suspicions of RNase contamination. RNA samples should be stored at -80 °C (ideal) or -20 °C when not in use, but they can be briefly handled on a room-temperature bench between the steps described in this protocol.
3. Consult any standard protocol for the optional steps involving in vitro transcription, PAGE purification, analytical PAGE, and autoradiography.
4. If RNA oligonucleotides are produced by in vitro transcription, the resulting transcript should then be PAGE-purified for optimal results. 5. The phosphate shuttle RNA oligonucleotide sequence is arbitrary, and we simply included here a sequence that worked well in our experiments. This sequence could presumably be changed provided that compensatory substitutions are also made in the splint RNA oligonucleotide. 8. In the absence of a thermocycler, an alternative annealing procedure is as follows: incubate samples at 95 °C for 2 min on a heat block, turn off the heat block, and allow the system to gradually equilibrate to room temperature.
9. The ~75% yield is after accounting for the stoichiometric excess of phosphate shuttle RNA oligonucleotide in the annealing reaction, which results in a 17% subpopulation whose 5′-32 Pphosphate is unavailable for transfer. The absolute fraction of recovered radioactivity is ~60%.
10. The most common contaminants are depurination products and n+1 products (some fraction of the DNA molecules contain two ribonucleotides on the 3′ end instead of one, as the n+1 phosphodiester was not cleaved in the RNA degradation reaction). In our preparations, these products were minor or not detected at all (there was variation between different probe preparations), and they did not interfere with our experiments. If the contaminant concentration is too high for your particular oligonucleotide preparation, try quenching the RNA degradation reaction with HCl at a variety of temperatures or time points. Identify a combination of temperature and quenching time that yields an acceptable balance of depurination products (which accumulate over time) and n+1 products (which disappear over time due to cleavage and release of the terminal ribonucleotide). Alternatively, PAGE-purify the 3′-radiolabeled DNA oligonucleotide for maximal purity.
11. The identity of the final species can also be confirmed by treating with T4 PNK and checking by PAGE/autoradiography that all radioactivity has moved from the DNA oligonucleotide to an inorganic phosphate (due to the phosphatase activity of the T4 PNK enzyme that removes 2′,3′cyclic phosphates).