Colorimetric miRNA detection based on self-primer-initiated CRISPR-Cas12a-assisted amplification

Molecular metabolism changes significantly throughout pregnancy to support the development of the fetus. Different mechanisms carried out by biomolecules, such as miRNAs, are involved in such changes [1,2]. miRNAs are a kind of highly conserved endogenous RNA that are 19–25 nucleotides (nt) in length [3,4]. The crucial role of miRNAs in regulating embryo implantation, trophoblast differentiation and migration and maternal-fetal immune tolerance have made them promising biomarkers for evaluating pregnancy development [5–7]. Thus, the development of a sensitive and reliable method that can detect low abundance of miRNA is in high demand.

Due to the unique features of miRNA, such as low abundance, short length and high similarity among family members, sensitive and accurate quantitative determination of miRNA remains a huge challenge. Currently, the traditional miRNA quantification strategies [8–10] include reverse-transcription quantitative PCR (RT-PCR), northern blotting and microarrays. While these strategies have made remarkable contributions to miRNA detection and have been widely utilized in clinical applications, they are criticized for requiring cumbersome equipment, poor repeatability and complicated experimental procedures. The drawbacks of these conventional approaches have accelerated the establishment of novel miRNA detection approaches. In the past few years, various miRNA detection approaches have been established, such as strand displacement amplification and rolling circle amplification (RCA). For example, Zhao and colleagues proposed a sensitive miRNA detection method that integrates RCA with CRISPR-Cas12a-based signal amplification [11–14]. Consequently, the approach exhibited a low limit of detection for fM level. Recently established approaches can essentially be divided into electrochemistry and fluorescence, based on different signal transductors, however, they cannot be applied in resource-limited scenarios. Colorimetric methods can meet the requirement of direct readout of results, but the sensitivity of colorimetric approaches needs to be improved [15–17]. In addition, the detection range of colorimetric approaches is limited because subtle changes in color are hard to observe with the human eye [18]. Therefore, it is urgent to develop sensitive colorimetric miRNA detection approaches with a wider detection range.

Herein, a novel self-primer-initiated CRISPR-Cas12a-based signal amplification strategy for sensitive and colorimetric miRNA detection is described. The method was constructed on the basis of self-primer-initiated production of ssDNA sequences, CRISPR-Cas12a-system-based chain cleavage and functionalized gold nanoparticle (AuNP)-based color generation. Taking advantage of multiple signal amplification processes, the method exhibits high sensitivity. In addition, the cleavage of linker sequences by the CRISPR-Cas12a system blocked the aggregation of AuNPs to form color change, meaning the method has a wide detection range.

Materials & methods

Regents & materials

All essential nucleic acid sequences (Supplementary Table 1) were synthesized by Sangon Biotechnology Co. Ltd (Shanghai, China). The enzymes used in this research, including the T7 RNA polymerase, Klenow fragment (3′→5′exo-) and endonuclease, were bought from Takara Biotechnology Co. Ltd (Dalian, China). The RNase inhibitor, RNase-free water and dNTPs (2.5 mM) were purchased from Beyotime Biotechnology Co. Ltd (Shanghai, China). AuNPs were obtained from Shanghai ZZBIO Co., Ltd (Shanghai, China). The AuNPs were diluted with deionized water to 100 ml and stored at 4°C for following tests. Water used in all experiments was purified by a Milli-Q purification system. All chemicals used in this research were analytical reagents.

Assembly of dumbbell probe & feasibility characterization

The obtained synthesized dumbbell probe was diluted to 10 μM with DEPC water for subsequent experiments. For the assembly of the dumbbell probe, 10 μl of dumbbell probe mixture was heated to 90°C for 10 min and then cooled to room temperature. In the fluorescence assay, the FAM signal of the fluorescent dumbbell probe before and after assembly was measured by a Hitachi fluorospectro photometer F-7000 (Tokyo, Japan). The assembled 10-μl dumbbell probe (10 μM) was mixed with 2 μl of target miRNA (10 μM) and incubated for 10 min. The fluorescence signals when target miRNA existed or not were recorded by a Hitachi fluorospectro photometer F-7000 (Tokyo, Japan).

Detection procedures

The analytical reaction was performed in a 50-μl mixture containing 5 μl dumbbell probe (1 μM), 5 μl of different concentrations of target miRNA, 0.2 U DNA polymerase (2 μl), 0.4 U endonuclease (2 μl), 2 μl dNTPs, RNase inhibitor (2 μl) and 32 μl PBS buffer solution. The mixture was incubated at 45°C for 60 min. Afterward, the mixture was heated to 70°C for 15 min to inactive the DNA polymerase and endonuclease. A total of 5 μl template DNA sequence (1 μM), 5 μl linker sequence (1 μM), 2 μl NTP mix (5 mM), 2 μl T7 RNA polymerase (1 U μl^-1) and 2 μl RNase inhibitor (1 U μl^-1) were then added to the mixture, and the mixture was incubated at 30°C for 30 min. A total of 5 μl AuNPs was then added to the mixture, and the RGB value of the AuNP solution was detected using a smartphone (Vivo S7) with a 64-megapixel camera.

Results & discussion

Working mechanism of the established approach for colorimetric miRNA detection

The working principle of the established approach is illustrated in Figure 1. In this method, a dumbbell probe is designed to specifically bind with target miRNA and induce subsequent signal amplifications. In the presence of target miRNA, with miRNA-21, which is closely associated with pregnancy development, as an example, it can bind with one of the loop parts in the dumbbell probe by hybridizing with the c-section and gradually unfolding it. As a result, the a’ section is exposed to hybridization with a section, forming a self-primer. With the section as primer, the chain is extended and miRNA that was originally hybridized with the c-section is displaced. The released miRNA binds with the c-section in the next dumbbell probe and induces signal amplification, constituting the first signal recycle. The transcribed sequences that are complementary to the b-section can be recognized and digested by endonuclease. Under the cooperation of both DNA polymerase and endonuclease, a large number of c' sequences and d' sequences are produced. The c' sequences bind with the c-section in the dumbbell probe to induce signal amplifications and the d' sequence hybridizes with template DNA and functions as a primer to induce T7 RNA polymerase-based generation of RNA sequences. The transcribed sequences can fold to sgRNA and assemble with Cas12a to the CRISPR-Cas12a system. The CRISPR-Cas12a system can specifically identify linker sequences and digest them to block the aggregation of AuNPs, leading to color change.

Feasibility of the established approach

The designed dumbbell probe is crucial in recognizing target miRNA and inducing subsequent signal amplification. To test the assembly of the dumbbell probe and its feasibility in identifying target miRNA, a fluorescence assay was conducted. In detail, a fluorescent moiety (FAM) and the corresponding quenching moiety were labeled on the two terminals of the probe. When the linear dumbbell probe sequences were assembled to dumbbell structure, the FAM moiety was closed with BHQ in distance, leading to the quenching of the FAM signal. In the presence of target miRNA, miRNA can bind with c sequences and unfold the loop section in the dumbbell probe, leading to the recovery of the FAM signal (Figure 2A). The result in Figure 2B shows a remarkable decrement in the FAM signal when the probe was assembled to dumbbell structure. In the presence of target miRNA, the FAM signal recovered, indicating that the dumbbell probe can specifically bind with target miRNA. The cleavage activity of the Cas12a enzyme was then tested. In this experiment, the T7 RNA polymerase transcribed products were first mixed with the Cas12a enzyme to assemble the CRISPR-Cas12a system, and the CRISPR-Cas12a system was utilized in cutting FAM- and BHQ-labeled linker sequences. The result in Figure 2C shows a greatly increased fluorescence signal when T7 RNA polymerase transcribed products were mixed with Cas12a enzyme, while no significant enhancements of fluorescence signals were observed, suggesting robust cleavage activity of CRISPR-Cas12a. The linker was designed to mediate the conjugation of ssDNA-functionalized AuNPs, thus mediating the color change. Therefore, the linker-based aggregation of AuNPs was investigated. The result in Figure 2D shows that the solution color changed from red to purple, while the purple color changed to red upon the addition of the CRISPR-Cas12a system, indicating the linker sequences mediated the aggregation of AuNPs.

Optimization of experimental conditions

All essential experimental parameters in the research, such as incubation time, concentration of linker sequences and concentration of DNA polymerase, endonuclease and Cas12a enzyme were tested and optimized. To directly compare the detection performance of the established method, the method was utilized in detecting 10 nM target miRNA, and the red (R) values of the color were compared. As shown in Figure 3A, the R-value gradually reduced with incubation times ranging from 0–75 min, and no more decrements were observed. Therefore, 75 min was applied for the following experiments. The concentration of linker sequences plays a crucial role in mediating the color generation. As shown in Figure 3B, the R-value decreased when the concentrations of linker sequences ranged from 0 to 50 nM. When the system was added with more concentrations of linker sequences, no more decrements were observed. Eventually, a 50-nM linker sequence was used in this research. In addition, the optimized concentrations of DNA polymerase, endonuclease and Cas12a enzyme were determined to be 0.6 U/l, 0.4 U/l and 0.8 U/l, respectively (Figure 3C).

Analytical performance of the established colorimetric approaches

Under the optimized experimental conditions, the effect of the concentrations of target miRNA on the R-value of the established colorimetric approach was evaluated. As shown in Figure 4A, the R-value elevated as the concentration of target miRNA increased from 100 fM to 10 nM. A good linear relationship between the R-value and the concentration of target miRNA in the range of 100 fM to 1 nM was observed. The linear correlation equation was determined to be Y = 5.147*lgC+40.99, where C refers to the concentration of target miRNA and the correlation coefficient was determined to be 0.9897 (Figure 4B). The detection limit of the approach was 54 fM, which was calculated using the equation DL ¼ 3s/slope (s denotes the standard deviation of the blank).

To verify the specificity of the established approach, the method was utilized in analyzing target miRNA and interferent miRNAs that have 1, 2 and 3 base pairs with the target. Notably, the concentrations of target miRNA and interferent miRNAs were the same. As shown in Figure 4C, the R-values of the method when detecting interferent miRNAs were almost the same as those of the blank group. On the contrary, the R-value of method when detecting target miRNA was high, implying the high selectivity of the established approach. Considering the crucial role of miRNA in regulating pathophysiological process, such as regulating embryo implantation, a variety of methods have been proposed for miRNA detection based on different signal transducer modes, such as fluorescent assay and colorimetric assay. Compared with the fluorescent approaches, this method does not require cumbersome equipment to readout signals (e.g., fluorescence spectrophotometer), making it possible to read the result directly by the naked eye. Compared with the colorimetric assays, the method possesses a much higher sensitivity in fM level (Supplementary Table 2) [19].

Application of the approach in quantifying miRNA

To test whether the approach could be utilized in noninvasive disease diagnosis, the detection performance of the established approach was tested by detecting miRNA-21 in urine samples. In detail, different concentrations of miRNA-21 were mixed with urine samples and the amounts of miRNA in samples were calculated by the established approach. The result in Figure 5 shows that the calculated miRNA-21 concentration showed a good linear relationship with the added amounts of miRNA-21, implying that the method could be potentially applied in clinical practice. To further investigate the selectivity of the approach in clinical application, the proposed approach was exploited to detect different miRNAs in artificial serum samples that were constructed by diluting miRNAs by commercial serum. The result in Supplementary Figure 1 shows that the proposed approach also showed a high selectivity to miRNA-21 in serum samples, indicating that the method could also be utilized in analyzing clinical samples.

Conclusion

In this paper, we prosed a sensitive and reliable miRNA detection colorimetric approach by integrating self-assembly-initiated signal recycles and CRISPR-Cas12a-based color generation. In this method, a dumbbell probe was designed integrating the functions of specific target miRNA recognition and signal amplifications. The proposed approach possesses several advantages, including multiple signal recycles integrated with CRISPR-Cas12a-based cleavage of linker sequences, endowing the method with high sensitivity and a low limit of detection of 254 aM. The result can also be observed directly by the naked eye. In addition, the method was utilized in detecting miRNA-21 from urine samples, indicating its promising potential in noninvasive disease diagnosis.

Future perspective

Sensitive and reliable detection of miRNA remains a huge challenge due to its low abundance and high similarity between homogenous family members. Many efforts have been made in the past decades to improve detection sensitivity by exploiting different signal amplification strategies. In this method, self-assembly-initiated signal recycles and transcleavage activity of CRISPR-Cas12a were integrated in the color reaction, endowing the method with high sensitivity. owing to the advantages of high sensitivity, good selectivity and being free of instruments, the method shows promising prospects in diseases diagnosis and on-site evaluation of fetal development in the future. Besides miRNA detection, the proposed colorimetric approach could also be potentially extended to protein biomarker analysis simply by replacing the c-section with aptamer sequences.