Introduction
Madeline Hughes, Correa Lab
Supervisors/Lab Members: Madalee Wulf, Ivan Correa, Nan Dai Khalid Deojee
Oxford Nanopore Technologies (ONT) offers a new method for direct RNA sequencing. A membrane-embedded nanopore is applied with voltage to generate a current. The helicase motor protein feeds single-stranded RNA through the membrane (3’ to 5’), allowing each of its nucleotides to be identified by a unique change in current, ultimately providing a direct read-out of the RNA sequence (Figure 1). The technology is advantageous for its rapid, long-read sequencing, simple library preparation, and potential for RNA modification detection. Despite these benefits, the platform is unable to read the 5’ end due to the last ~15 nucleotides slipping through the pore rapidly without base identification as the helicase motor protein loses contact with the RNA.
To improve the accuracy of 5’ end sequencing, we previously altered the mRNA capping conditions to design a cap that can be adapted with additional 5’ sequence via click chemistry. First, we tested various GTP analogs in standard RNA 25mer capping reactions. These capped products can be clicked to an oligonucleotide sequencing adaptor, which allows for the 5’ cap structure to be read through the nanopore (previous work). Additionally, we generated oligo-GTPs through copper-mediated click reactions. We then used these oligo-GTPs as substrates in standard capping reactions in order to reduce the two-step process of capping into one. In an attempt to improve the capping reactions, we substituted different capping enzymes and polyamine additives into the protocol. After each reaction, we used intact mass spectrometry, an analytical tool that detects the components of a sample through their molecular weights, to evaluate the efficacy of the reaction modifications based on the presence and abundance of product. Direct capping of RNA with oligo-GTPs could allow modification of ONT library preparation to enable 5’ to 3’ sequencing.
Click Chemistry
RNA Capping Mechanism
Overview of Capping Conditions
Using Additives to Inhibit Dephosphorylation of Oligo-GTPs
Capping RNA with Various Enzymes
Standard Capping Protocol
RNA Oligo: 2 µM
10X Capping Buffer
GTP Analog: 0.5 mM
Figure 2. A) Cap 0 and Cap 1 structures of Eukaryotic mRNA caps. B) Enzymatic steps of mRNA capping. Step 1: RNA triphosphatase (Tpase) removes a phosphate to generate 5’ diphosphate. Step 2: RNA guanylyltranferase (Gtase) consumes a GTP molecule (2.1) and transfers a GMP to the 5’-diphosphate (2.2). Step 3: In the presence of S-adenosylmethionine (SAM), guanine-N7 methyltransferase (N7 Mtase), a methyl group is added to the N7 amine of the guanine cap, ultimately forming a Cap 0 structure. Step 4: m7G-specific 2’-O methyltransferase methylates the +1 ribonucleotide at the 2’-O position, ultimately creating the Cap 1 structure. C) Structure of Vaccina capping enzyme, one of the enzymes used to form the Cap 0 structure.
Copper-Mediated Click Protocol
10X TEAA Buffer
Oligo-N3: varies
DMSO: 25%
Figure 5. A) Structures of the GTP analogs tested in the standard capping reactions. B) After each reaction was completed, they were analyzed by intact mass spectrometry. The bar chart displays the average percent of desired product in each reacted sample. Based on the results, the most successful substrate was 3’-O-azido dGTP. The least successful substrates were the oligo-GTPs and the 3’-O-Propargyl GppCp. It is unclear why the 3’-O-Propargyl GppCp was ineffective since GppCp works fairly well as a substrate, suggesting that the sample we tested may need to be purified further. C) The mass spectrum indicates that the oligo-GTPs were all dephosphorylated, which inhibits the capping enzyme from recognizing the molecules as GTP, and prevents the capping reaction from proceeding.
Cu-Mediated Click of 3 Different N3-Oligos and 3’-O-Propargyl GTP
SAM (for methylation): 100 µM
Pyrophosphatase: 0.01 units/µL
Capping Enzyme: 1 unit/µL
42 °C, 1 hour
A) Structures of the GTP Analogs Tested
C)
B)
B) Protamine Sulfate Salt
A) Spermidine
C) 1:10 Dilution of Protamine
D) 1:100 Dilution of Protamine
Figure 6. Various polyamine additives are known to stabilize DNA, so we used them in RNA capping reactions discussed in Figure 5C to test their ability to prevent dephosphorylation of oligo-GTPs. A) Spermidine (250 mM) did not inhibit dephosphorylation of the 10mer-GTP. B) Protamine sulfate salt (50 mg in 2 mL of water) inhibited dephosphorylation of most of the 10mer-GTP starting material, however, it caused significant RNA degradation, and the capping reaction did not proceed. We attempted to mitigate the degradation by testing 1:10 dilutions (C) and 1:100 dilutions (D) of the protamine sulfate salt solution, however, these still caused degradation and did not prevent dephosphorylation.
A) H3C2 vs. VCE: Capping a ppp-A 25mer with GTP
Capping Enzyme | Percent Capped Product in Mass Spec |
H3C2 | 83.94% |
VCE | 95.63% |
B) H3C2 vs. Mutant H3C2: Capping a pp-G 25mer with Oligo GTPs
H3C2
Mutant H3C2
3784.8 g/mol
DNA 10mer Azido 3’ Oligo GDP
3864.7 g/mol
DNA 10mer Azido 3’ Oligo GTP
3784.8 g/mol
DNA 10mer Azido 3’ Oligo GDP
Figure 7. A) The table displays the percent of product in capping reaction samples that were catalyzed by H3C2 and VCE capping enzymes, detected through mass spectrometry. H3C2, an enzyme developed at NEB, consistently yields nearly the same amount of product as VCE. However H3C2 is much easier and less expensive to generate than VCE, making H3C2 a promising enzyme for standard capping reactions. B) We tested a mutant H3C2 capping enzyme that lacks Tpase activity to see if it could prevent dephosphorylation of oligo-GTPs. The mutant H3C2 capping enzyme was successful in this regard, while H3C2 was not, but the capping reaction still did not proceed to completion, suggesting that oligo-GTPs may not be effective substrates for the capping enzymes currently used.
16X Propargyl-Oligo Solution: varies
Cu(II)-TBTA Soln.: 500 µM
Ascorbic Acid Soln.: 500 µM
room temperature, overnight
GTP
dGTP
3’-O-Propargyl GTP
3’-azido dGTP
3’-O-azidomethyl dGTP
3’-O-Propargyl GppCp
MS-allyl-azido GTP
MS-allyl-biotin GTP
p-10mer GTP
P-DNA 10mer iSp18-N3 Oligo GTP (Ref. Figure )
DNA 10mer azido 3’ GTP (Ref. Figure )
5’ FAM DNA 10mer azido 3’ GTP (Ref. Figure )
GTP Analog
+
🡪
3’-O-Propargyl GTP
P-DNA 10mer iSp18-N3 Oligo
P-DNA 10mer iSp18-N3 Oligo GTP
+
🡪
3’-O-Propargyl GTP
DNA 10mer azido 3’
DNA 10mer azido 3’ GTP
3’-O-Propargyl GTP
+
🡪
5’ FAM DNA 10mer azido 3’
5’ FAM DNA 10mer azido 3’ GTP
Figure 3. Copper-mediated click reactions of 3’-O-propargyl GTP and three different oligos, which form unique oligo-GTP products. Click chemistry is a useful chemical tool due to its predictability and high yields. In this reaction, we utilized an azide-alkyne cycloaddition to generate the desired product. Copper-TBTA reacted with ascorbic acid functions as an effective catalyst for this reaction. Each click reaction consistently yields 80-100% percent product.
Figure 4. General outline of the order of reactions. First, the mRNA Is decapped (if it is eukaryotic mRNA). Then it is capped using 3’ pre-clicked oligo-GTP (ref. Figure 3) and VCE capping enzyme. VCE may be substituted with other effective alternatives such as H3C2 capping enzyme. The one-step capping process is favored over the two-step process as it mitigates the chance of RNA degradation. Once the new cap with the adaptor sequence (in green) is attached, sequencing from the 3’ to 5’ end may be more accurate, and 5’ to 3’ sequencing can be achieved, as it provides an ONT adaptor ligation site (ref. Figure 1).
Figure 1. A) The mechanism behind nanopore sequencing. Voltage is applied to the membrane to create a current. Single-stranded RNA is fed through the membrane via motor protein, and read by the unique change in current caused by each nucleotide. B) The device currently sequences from the 3’ end to the 5’ end, but Oxford Nanopore Technologies is interested in obtaining sequences from the 5’ end to the 3’ end, as another way to improve 5’ end sequencing coverage. By generating a cap with ligation sites for ONT adaptors, the motor protein can initiate 5’ to 3’ end sequencing.
Direct Capping of Oligonucleotides to RNA for Use in Oxford Nanopore Sequencing Technologies
MinION
Capping RNA with various GTP Analogs
Results
5’
3’
5’
3’
3’
5’
A)
B)
3’
5’
3’
5’
3’
5’
3’
5’
Helicase