Nuclease-induced stepwise photodropping (NISP) to precisely investigate single-stranded DNA degradation behaviors of exonucleases and endonucleases

Abstract

Here, we employed a fluorescence-based single molecule method called nuclease-induced stepwise photodropping (NISP) to measure in real time the DNA degradation mediated by mitochondrial genome maintenance exonuclease 1 (MGME1), a bidirectional single-stranded DNA (ssDNA)-specific exonuclease. The method detects a stepwise decrease in fluorescence signals from Cy3 fluorophores labeled on an immobilized DNA substrate. Using NISP, we successfully determined the DNA degradation rates of 6.3 ± 0.4 and 2.0 ± 0.1 nucleotides (nt) s-1 for MGME1 in the 5'-to-3' and 3'-to-5' directions, respectively. These results provide direct evidence of the stronger 5' directionality of MGME1, consistent with its established role in mitochondrial DNA maintenance. Importantly, when we employed NISP to investigate mung bean nuclease, an ss-specific endonuclease, we observed a markedly different NISP pattern, suggesting a distributive cleavage activity of the enzyme. Furthermore, we applied NISP to determine the ssDNA degradation behavior of the double-stranded-specific exonuclease, λ exonuclease. These findings underscore the capability of NISP to accurately and reliably measure the degradation of ssDNA by both exo- and endonucleases. Here, we demonstrate NISP as a powerful tool for investigating the ssDNA degradation behavior of nucleases at the single-molecule level.

Graphical Abstract

Figure 1.

Using NISP to investigate MGME1-mediated 5′-to-3′ ssDNA degradation. (A) Schematic of NISP investigating ssDNA degradation by exonuclease. A DNA substrate with a 5′-phosphorylated poly(dT)40 ssDNA overhang is anchored on a PEGylating slide. The ssDNA overhang region is labeled with two internal Cy3 (iCy3) fluorophores in various interval (Supplementary Table S1). S represents the number of nucleotides with scissile phosphodiester bonds between two Cy3 fluorophores. After injection of exonuclease, ssDNA degradation can be detected by the stepwise photodropping of Cy3 signals. Green asterisks represent iCy3 fluorophores and black circles represent MGME1 nucleases. (B) The typical NISP time traces illustrate the behaviors described in panel (A), obtained in the presence of 1 nM MGME1. The dwell times in the one-step NISP time trace (T₁) and two-step NISP time trace (T₂) are marked by double-arrowhead lines. Histograms of T₁ (C) and T₂ (D) of MGME1 in degrading the 5′-overhang DNA. For panels (C) and (D), N indicates the event number, and occurrence probabilities are enclosed by boxes. Solid lines in panel (D) indicate either the nonlinear polynomial fitting for T₂ distribution (in the absence of MGME1) or the nonlinear polynomial plus Gaussian fitting for T₂ distribution (in the presence of MGME1). (E) The plots of the population change in the one-step and two-step NISP methods at different concentrations of MGME1, compared to the control scenario without MGME1. (F) The plots of averaged T₂ acquired under different concentrations of MGME1 (solid squares; with no extra buffer wash). Open circles represent the experimental data acquired in the Mg²⁺-triggered condition (free MGME1 was removed by buffer wash before triggering the nuclease reaction). (G) The plot of averaged T₂, acquired for MGME1 acting on DNA substrates with varying spaces between di-iCy3 fluorophores (black squares) and di-eCy3 fluorophores (red squares), shows a linear fit (R²= 1.00).

Figure 2.

MGME1 cannot degrade blunt-end dsDNA. (A–D) Using NISP to investigate MEMG1’s activity on dsDNA. A DNA named 5′→3′ di-iCy3-18nt dsDNA was used, and the number of nucleotides with scissile phosphodiester bonds between two Cy3 fluorophores is 18 nt (Supplementary Table S1). NISP time traces in the absence (A) and presence of 5 nM MGME1 (C) are shown. T₁ and T₂ are marked with double arrowhead lines. Panels (B) and (D) show the histograms of T₁ and T₂ in the absence and presence of 5 nM MGME1, respectively. N indicates the event number. (E) Using FRET to investigate MGME1’s activity on dsDNA. Histograms of the FRET efficiency before (upper) and after adding 5.0 nM MGME1 (lower) are shown. N indicates the event number. Schematics of the DNA substrates used in the FRET assay are shown in the right panels. Green asterisks represent iCy3 fluorophores and red asterisks represent iCy5 fluorophores.

Figure 3.

Using NISP to investigate MGME1-mediated 3′-to-5′ ssDNA degradation. (A) Schematic of the 3′-poly (dT)₄₀ overhang DNA substrate used in NISP. The ssDNA overhang region is labeled with two internal Cy3 (iCy3) fluorophores in various intervals (Supplementary Table S1). S represents the number of nucleotides with scissile phosphodiester bonds between two Cy3 fluorophores. Green asterisks represent iCy3 fluorophores and black circles represent MGME1 nucleases. (B) The typical NISP time traces obtained from the 3′-overhang DNA substrate in the presence of 1 nM MGME1 as shown in panel (A). T₁ and T₂ are marked with double-arrowhead lines. Histograms of T₁ (C) and T₂ (D) of MGME1 in degrading the 3′-overhang substrates. N indicates the event number, and occurrence probabilities are enclosed by boxes. Solid lines in penal (D) indicate either the nonlinear polynomial fitting for T₂ distribution (in the absence of MGME1) or the nonlinear polynomial plus Gaussian fitting for T₂ distribution (in the presence of MGME1). (E) The plots of the population changes in the one-step and two-step NISP methods at different concentrations of MGME1, compared to the control scenario without MGME1. (F) The plots of averaged T₂ acquired under different concentrations of MGME1 (solid squares; with no extra buffer wash). Open circles represent experimental data acquired in the Mg²⁺-triggered condition (free MGME1 was removed by buffer wash before triggering the nuclease reaction). (G) The plot of averaged T₂, acquired for MGME1 acting on DNA substrates with varying spaces between di-iCy3 fluorophores (black squares) and di-eCy3 fluorophores (red squares), shows a linear fit (R²= 1.00).

Figure 4.

Using NISP to investigate MBN-mediated ssDNA degradation. (A) Schematic of the assay with DNA substrate named 5′→3′ di-eCy3-19nt (Supplementary Table S1). The number of nucleotides with scissile phosphodiester bonds between two eCy3 fluorophores is 19 nt. Green asterisks represent eCy3 fluorophores and black circles represent MBN. (B) The typical NISP time traces of the used substrate in the presence of 5 nM MBN. T₁ and T₂ are marked with double-arrowhead lines. Histograms of T₂ of MBN in degrading the (C) 5′-overhang substrates and (D) 3′-overhang substrates. N indicates the event number, and occurrence probabilities are enclosed by boxes. The solid lines in panels (C) and (D) indicate either the nonlinear polynomial fitting to T₂ distribution (in the absence of MBN) or the nonlinear polynomial plus Gaussian fitting to T₂ distribution (in the presence of MBN). (E) The plots of the population changes in the one-step and two-step NISP methods at different concentrations of MBN along 5′- and 3′-ssDNA overhangs, compared to the control scenario without MBN. (F) The plots of averaged T₂ acquired under different concentrations of MBN along 5′- and 3′-ssDNA overhangs.

Figure 5.

The investigation of λ exonuclease-mediated dsDNA unwinding and ssDNA degradation processes. (A–C) Using FRET to investigate the activity of λ exonuclease. (A)FRET histograms of the dsDNA and 3′-overhang DNA used in the assay in the enzyme-free condition. (B) The representative FRET efficiency time trace and the schematic to illustrate the λ exonuclease-mediated dsDNA unwinding process. (C) The histograms of unwinding times at different concentrations of the exonuclease. (D–H) Using NISP to investigate the activity of λ exonuclease. (D) Schematic and (E) the typical NISP time traces of the used substrates, respectively. T₁ and T₂ are marked with double-arrowhead lines. (F) The population changes in the one-step and two-step NISP methods at different concentrations of λ exonuclease compared to the control scenario without λ exonuclease. (G) Histograms of T₂ of λ exonuclease in degrading the eCy3-labeled substrates. The solid lines indicate the nonlinear polynomial plus Gaussian fitting to T₂ distribution obtained from DNA substrates labeled with di-eCy3 spaced differently. (H) The plot of averaged T₂, acquired for λ exonuclease acting on DNA substrates with varying spaces between di-eCy3 fluorophores (black suqares), shows a linear fit (R²= 1.00).

Figure 6.

The kinetic model of the studied nucleases established by NISP in this work. (A) A kinetics model of MGME1-mediated ssDNA degradation. MGME1 can degrade ssDNA in either the 5′-to-3′ direction (right panel) or the 3′-to-5′ direction (left panel) in a processive manner. In this process, MGME1 exhibits a faster degradation rate when processing DNA from the 5′ end. (B) A kinetics model for MBN-mediated ssDNA degradation. MBN degrades ssDNA in a distributive manner. (C) A kinetics model for λ exonuclease-mediated DNA degradation. λ exonuclease can unwind dsDNA (left panel) and degrade one of the ssDNA (right panel) simultaneously in the 5′-to-3′direction.