Efficient DNA ligation in DNA–RNA hybrid helices by Chlorella virus DNA ligase

Abstract

Single-stranded DNA molecules (ssDNA) annealed to an RNA splint are notoriously poor substrates for DNA ligases. Herein we report the unexpectedly efficient ligation of RNA-splinted DNA by Chlorella virus DNA ligase (PBCV-1 DNA ligase). PBCV-1 DNA ligase ligated ssDNA splinted by RNA with kcat ≈ 8 x 10−3 s−1 and KM < 1 nM at 25°C under conditions where T4 DNA ligase produced only 5′-adenylylated DNA with a 20-fold lower kcat and a KM ≈ 300 nM. The rate of ligation increased with addition of Mn2+, but was strongly inhibited by concentrations of NaCl >100 mM. Abortive adenylylation was suppressed at low ATP concentrations (<100 µM) and pH >8, leading to increased product yields. The ligation reaction was rapid for a broad range of substrate sequences, but was relatively slower for substrates with a 5′-phosphorylated dC or dG residue on the 3′ side of the ligation junction. Nevertheless, PBCV-1 DNA ligase ligated all sequences tested with 10-fold less enzyme and 15-fold shorter incubation times than required when using T4 DNA ligase. Furthermore, this ligase was used in a ligation-based detection assay system to show increased sensitivity over T4 DNA ligase in the specific detection of a target mRNA.

Assessing long-distance RNA sequence connectivity via RNA-templated DNA–DNA ligation

“…Moreover, Rnl2 could not ligate DNA oligos hybridized to the DNA template, eliminating the possibility of contaminating genomic DNA confounding SeqZip ( Figure 2A ). We note that Chlorella virus DNA ligase was recently commercialized for the purpose of RNA-templated DNA–DNA ligation (SplintR ligase; NEB) ( Lohman et al, 2014 ). We found, however, that SplintR ligase produces more non-templated DNA–DNA ligation events than Rnl2 ( Figure 2—figure supplement 1 ).…”

A high-throughput assay for the comprehensive profiling of DNA ligase fidelity

“…Additionally, the ligation product resulting from a T:G upstream mismatch ligating to a correct pA:T base pair on the downstream side (the T/pA product) can be observed. Additional peaks can be observed eluting near the starting materials; based on previous work ( 47 , 48 ), we identified these as 5′-adenylylated (App) probe molecules resulting from aborted ligation reactions. The 5′-adenylylated probes AppG and AppT were resolved cleanly from the starting pN peaks, while AppA coelutes with pG and AppC coelutes with pT.…”

Efficient Generation of Transcriptomic Profiles by Random Composite Measurements

“…To extend to unequal weights, the relative abundance of probes targeting each gene in a set is adjusted proportionally, if the weights a i are all positive (we extend below for negative weights). One implementation would create a pair of a single-stranded DNA probes, L i and R i , that hybridize to adjacent locations on the mRNA of gene i , so that they may be ligated with an enzyme that selects for RNA-DNA hybrids (such as in (Lohman et al, 2014)). The left and right probes L i and R i contain a left and right barcode sequence, respectively, which are each the same for all the genes measured.…”

“…For the final sets of experiment, 10ng of total RNA were used for each reaction, and a final concentration of 25pM or 0.25nM probe mix were added to the RNA sample to a final reaction volume of 20uL along with SplintR ligase buffer (New England Biolabs) (Lohman et al, 2014) and RNase Inhibitor. To hybridize the probe library to the RNA sample, a slow ramping protocol was applied by incubating the mixture first at 75°C for 5 minutes for denaturing of any possible RNA/DNA secondary structure, then slowly ramping down to 37°C at a ramping rate of 0.1°C per second with one minute incubation for every degree drop in temperature in a cycling manner over ~ 4 hours in Mastercycler Pro thermocycler (Eppendorf).…”