Sensitive RNA detection by combining three-way junction formation and primer generation-rolling circle amplification

Abstract

Recently, we developed a simple isothermal nucleic acid amplification reaction, primer generation-rolling circle amplification (PG-RCA), to detect specific DNA sequences with great sensitivity and large dynamic range. In this paper, we combined PG-RCA with a three-way junction (3WJ) formation, and detected specific RNA molecules with high sensitivity and specificity in a one-step and isothermal reaction format. In the presence of target RNA, 3WJ probes (primer and template) are designed to form a 3WJ structure, from which multiple signal primers for the following PG-RCA can be generated by repeating primer extension, nicking and signal primer dissociation. Although this signal primer generation is a linear amplification process, the PG-RCA exponentially can amplify these signal primers and thus even a very small amount of RNA specimen can be detected. After optimizing the structures of 3WJ probes, the detection limit of this assay was 15.9 zmol (9.55 × 10(3) molecules) of synthetic RNA or 143 zmol (8.6 × 10(4) molecules) of in vitro transcribed human CD4 mRNA. Further, the applicability of this assay to detect CD4 mRNA in a human mRNA sample was demonstrated.

Figure 1.

RNA detection mechanism by three-way junction probe and primer generation-rolling circle amplification. (A) Three-way junction (3WJ) probes (primer and template) are designed to form a 3WJ structure on target RNA, however they do not interact each other without target RNA because their complementary sequence is only 6–8 bases. (B) Addition of DNA polymerase and nicking enzyme initiates a reaction cycle of primer extension, nicking reaction and signal primer generation under an isothermal condition to generate signal primers. (C) The generated signal primers can be detected by primer generation-rolling circle amplification.

Figure 2.

Confirmation of 3WJ primer extension and signal primer generation products. 3WJ primer P1, 3WJ template T1b and RNA50 were mixed in different combinations [(A): P1/T1b/RNA50, (B): P1/T1b, (C): P1/RNA50 and (D): T1b/RNA50] and incubated to form a 3WJ structure. Signal generation reaction was conducted at 60°C for 0, 15 and 30 min by adding an enzyme mix [0.2 U Vent(exo-) DNA polymerase and 1 U Nb.BsmI] to the samples. The reaction products were analyzed on 15% denaturing polyacrylamide gel. P1* indicates a reaction product through primer extension of 3WJ primer and SP indicates signal primer generated from 3WJ structure. The numbers in parentheses indicate the length or expected length of each probe or reaction product. Lane M is 20–100 nt oligonucleotide marker.

Figure 3.

Detection of RNA50 by 3WJ probe and real-time PG-RCA. (A) Different concentrations of RNA50 were analyzed using 3WJ probe P1 and T1b at 37°C for 90 min following initial heat denaturation at 80°C for 5 min. PG-RCA was conducted at 60°C by adding a PG-RCA reaction mixture to the RNA samples and the fluorescent intensities of each reaction were monitored in real time. The RNA50 concentration in each reaction was prepared by 10-fold serial dilution from 250 amol to 25 zmol, and their signal amplification curves are indicated by colored lines (red, orange, green, light blue and blue, respectively). Negative controls are indicated by gray lines. (B) Threshold time T_T (the reaction time when fluorescent intensity of each reaction exceeds an arbitrary threshold) was plotted against the RNA50 concentration (S) of the reaction. The solid line indicates linear least squares fitting between 250 zmol and 250 amol RNA50 and its formulation is T_T = –25.2 log₁₀(S) + 110 (R²= 0.989). The perforated line indicates average T_T value of the negative controls. Detection limit is 76.5 zmol (4.59 × 10⁴ molecules) of RNA50 by calculation from the intersection of both lines.

Figure 4.

Suppression of background amplification using modified 3WJ templates. (A) Performance of modified 3WJ templates were compared by analyzing 50 amol RNA50 (solid lines) and negative control (perforated lines) with 3WJ probe and real-time PG-RCA. 3WJ templates used in this experiment are 3WJ template T1a (green), T1b (orange), T1c (red) and T1d (blue) in comparison with no 3WJ template (gray). (B) Threshold time T_T of the positive (open circle) and negative (cross) control experiments in (A) were analyzed and compared.

Figure 5.

RNA detection by modified 3WJ templates. (A and C) Different concentrations of RNA50 were analyzed by the modified 3WJ probes (in Table 1). The reaction condition is the same as in Figure 3 except using modified 3WJ template T1c (A and B) or T1d (C and D). The RNA50 concentration in each reaction was prepared by 10-fold serial dilution from 500 amol to 50 zmol, and their signal amplification curves are indicated by colored lines (red, orange, green, light blue and blue, respectively). Negative controls are indicated by gray lines. (B and D). Threshold time T_T was plotted against the RNA50 concentration (S) of the reaction. Solid lines indicate linear least squares fitting between 50 zmol and 500 amol RNA50 and their formulations are T_T = –23.2 log₁₀(S) + 109 (R²= 0.993) and T_T = –23.9 log₁₀(S) + 117 (R²= 0.991), respectively. Perforated lines indicate average T_T values of the negative controls. Detection limits are 15.9 zmol (9.55 × 10³ molecules) of RNA50 for T1c (B) and 18.2 zmol (1.09 × 10⁴ molecules) of RNA50 for T1d (D) by calculation from the intersections of both lines.

Figure 6.

RNA detection in the presence of excess amounts of unrelated mRNA. (A) 50 amol RNA50 (solid lines) and negative control (perforated lines) were analyzed by 3WJ probe P1 and T1c in the presence of 0–4000 pg rat spleen mRNA. (B) Threshold time T_T of the positive (open circle) and negative (cross) control experiments in (A) were analyzed and compared.

Figure 7.

Human CD4 mRNA detection. (A) Different concentrations of in vitro transcribed human CD4 mRNA were analyzed using 3WJ probe P1 and T1c. The CD4 mRNA concentration in each reaction was prepared by 10-fold serial dilution from 250 amol to 250 zmol, and their signal amplification curves in real-time PG-RCA are indicated by colored lines (orange, green, light blue and blue, respectively). Negative controls are indicated by gray lines. (B) Threshold time T_T was plotted against the CD4 mRNA concentration (S) of the reaction. The solid line indicates linear least squares fitting between 250 zmol and 250 amol CD4 mRNA and its formulation is T_T = –17.3 log₁₀(S) + 134 (R²= 0.953). The perforated line indicates average T_T value of the negative controls. Detection limit is 143 zmol (8.60 × 10⁴ molecules) of RNA50 by calculation from the intersection of both lines. (C) CD4 mRNA expressed in human mRNA was analyzed using 3WJ probe P1 and T1c. Human mRNA used in this experiment was 0 (gray), 10 (green), 100 (orange) and 1000 pg (red). (D) Threshold time T_T was plotted against the human mRNA concentration (S) of the reaction. The solid line indicates linear least squares fitting between 10 and 1000 pg human mRNA and its formulation is T_T = –3.67 log₁₀(S) + 146 (R²= 0.396). The perforated line indicates average T_T value of the negative controls.