Chenqi
Niu
a,
Xinhui
Xing
ab and
Chong
Zhang
*ab
aMOE Key Laboratory for Industrial Biocatalysis, Institute of Biochemical Engineering, Department of Chemical Engineering, Tsinghua University, Beijing, 100084, China. E-mail: chongzhang@mail.tsinghua.edu.cn
bCenter for Synthetic and Systems Biology, Tsinghua University, Beijing, 100084, China
First published on 21st November 2022
Profiling of aptamer dominated sites which have a huge impact on aptamer affinity is essential to characterize the relationship between an aptamer and target molecule and show the potential for boosting the sensitivity of aptamer-based biosensors. Here, we report a novel strategy to analyze aptamer dominated sites and detect aflatoxin B1 (AFB1) based on CRISPR–Cas12a. 28 different crRNAs were designed to hybridize with every region of the AFB1 47-nt aptamer, and sites 24–26 from the 5′ to 3′ end in the aptamer (“TGT”) were speculated to be the dominated sites according to the different responses based on the fluorescence intensity inhibition rate. Then, this was verified by replacing the three bases through the BLI assay. Three-dimensional (3D) modeling and molecular docking were also used to laterally validate the results. Moreover, the effects of special structures of DNA on the activated Cas12a trans-cleavage activity were evaluated to enrich our understanding of Cas12a. In addition, a strategy for the detection of AFB1 based on Cas12a was developed. The workflow is convenient, simple, and rapid (37 °C in 20 min) with a detection limit of 0.8 ng mL−1. The strategy presented good specificity over other mycotoxins and aflatoxin analogues and showed potential in complex sample analysis.
In recent years, several AFB1 aptamers with high binding affinity have been developed through systematic evolution of ligands by the exponential enrichment assay (SELEX).6–9 Many aptamer-based biosensors were reported accordingly to detect AFB1, including colorimetric assays, electrochemical assays, fluorescence assays, etc.10–14 Compared to the traditional detection methods, aptamer-based methods are easy to perform, have low cost and can improve the stability and sensitivity of AFB1 detection. Recently, Clustered Regularly Interspaced Short Palindromic Repeats associated protein 12a (CRISPR–Cas12a), an RNA guided endonuclease, has been applied for the development of aptamer-based small molecular biosensors.15,16 When the Cas12a–crRNA complex hybridizes with a region with aptamer/analyte dominated sites which have a huge impact on aptamer affinity, the structural change of the aptamer due to analyte binding has a great influence on the hybridization between the Cas12a–crRNA complex and aptamer, leading to a decrease of the trans-cleavage activity of Cas12a. The Cas12a-based competitive biosensor achieved high detection sensitivity without nucleic acid polymerases, and it has more potential for real samples with a complex matrix due to its simple reaction system. Nevertheless, there are still many challenges to be faced. In the competitive binding method with a fluorescence reporter system, crRNAs are designed to hybridize with aptamers and activate the Cas12a trans-cleavage activity, where target molecules act as an inhibitor to decrease the Cas12a trans-cleavage activity. In the presence of the target, the aptamers switching structure and no longer hybridizing with the crRNA–Cas12a complex are the key steps to the assay. However, considering that the hybridization region is 20 bp and the length of excellent aptamers usually ranges from 30–50 bp, the design of crRNAs is not an easy task. In addition, the lack of knowledge of aptamer dominated sites also poses great challenges for crRNA design.
Herein, we report a novel strategy for analyzing AFB1 aptamer dominated sites and detecting AFB1 based on CRISPR–Cas12. 28 different crRNAs were designed to hybridize with every region of the AFB1 47-nt aptamer. The dominated site was analyzed by comparing the repeated site of crRNAs with a high decrease of Cas12a trans-cleavage activity. Sites 24–26 from the 5′ to 3′ end in the aptamer (“TGT”) were speculated to be the dominated sites according to the different responses based on the fluorescence intensity inhibition rate. The results were verified by BLI assays and 3D modeling and molecular docking. In addition, according to the results of the analyzed aptamer dominated sites, a convenient, simple, and rapid method with high sensitivity for detection of AFB1 based on Cas12a was developed with crRNA 24. The strategy presented good specificity over other mycotoxins and aflatoxin analogues and showed potential in complex sample analysis.
In the sensitivity test, the concentration of AFB1 ranged from 100 ng mL−1 to 2 μg mL−1.
In the specificity test, AFB1 was replaced with other mycotoxins and aflatoxin analogues with the same concentration.
In the complex matrix sample test, different concentrations of AFB1 were spiked into 5% red wine, 5% beer and 5% milk. The red wine, beer and milk were used directly without pretreatment. 50 μL aliquots of 10 μg mL−1 and 5 μg mL−1 AFB1 were mixed with 950 μL wine to obtain stock solutions of 500 ng mL−1 and 250 ng mL−1 AFB1 in 5% wine, respectively. And, AFB1 was also spiked into beer and milk in the same way as that for the wine to obtain 500 ng mL−1 and 250 ng mL−1 AFB1 in 5% beer and milk, respectively.
GraphPad Prism 7 graphics software was used to process data and draw images. In this study, the inhibition rate (%), IR (%) was defined to indicate the level of inhibition of trans-cleavage by small molecules. IR (%) = (A − B)/A × 100, where A is the fluorescence intensity in the control group, and B is the fluorescence intensity in the presence of AFB1.
The AFB1 binding energy and dominated sites of the ssDNA aptamers were estimated with the grid-based ligand docking program AutoDockTools (ADT) 1.5.6 package. The optimized structure of the aptamer was considered to be rigid receptors while AFB1 was considered to be a flexible ligand with only one rotatable bond. The docking simulations used the Lamarckian genetic algorithm and the best complex was selected according to the molecular docking results including binding energy, types of favorable interactions (mainly hydrogen bonds, and hydrophobic and electrostatic interactions) and dominated sites.
In the case of the Cas12a–crRNA complex hybridized with the region without an aptamer/analyte dominated site, the structural change of the aptamer due to analyte binding had little influence on the hybridization between the Cas12a–crRNA complex and aptamer. Cas12a was activated to show trans-cleavage activity. The ssDNA fluorescence-reporter probe labeled with a fluorophore and quencher pair at its two ends (denoted as the F-Q probe) was cleaved, resulting in separation of the fluorophore from the quencher, thereby producing an increase in fluorescence signal. And the decrease of trans-cleavage activity (decrease of fluorescence intensity) of Cas12a activated by the ssDNA aptamer was small compared to that of the control group (ssDNA aptamer activated Cas12a–crRNA to cleave the F-Q probe to show fluorescence), which resulted in a low inhibition rate (IR%).
In contrast, when the Cas12a–crRNA complex hybridized with the region with aptamer/analyte dominated sites, the structural change has a great influence on the hybridization between the Cas12a–crRNA complex and aptamer. The decrease of trans-cleavage activity of Cas12a activated by the ssDNA aptamer was sharp compared to that of the control group, which resulted in a high inhibition rate (IR%).
The dominated sites of the analyte and aptamer can be inferred by analyzing the results of different inhibition rates of different crRNAs.
This not only provides a supplement for our previously reported detection strategy for small molecules based on Cas12a, and further improves the integrity to ensure it can be applied to all kinds of aptamers, but also enriches the application fields of the Cas12a toolbox and fully exploits the potential of CRISPR technologies.
The AFB1 sample used in this study is dissolved in the organic solvent acetonitrile, which may have an effect on the Cas12a trans-cleavage activity. Before evaluating the IR% values of the different crRNAs, the effect of this organic solvent on the Cas12a trans-cleavage activity was evaluated (Fig. S2†). It is indicated that the organic solvent acetonitrile has no significant effect on the Cas12a trans-cleavage activity.
To evaluate the effect of different crRNAs on the inhibition rate of Cas12a trans-cleavage activity in the presence of AFB1, 28 different crRNAs were tested in the Cas12a fluorescence reporter system compared to the respective control groups (Fig. S3†). As shown in Fig. S3,† there were two kinds of IR% vs. time curves. The IR% values of several curves did not change drastically with time, such as crRNA 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15, 18, 20, 22, 23, 24, 25, 26, 27, and 28. Meanwhile, the IR% values of crRNA 14, 16, 17, 19, and 21 were relatively stable at first, and then dropped suddenly. The second curve was caused by the gradual saturation of the fluorescence intensity of the control group, and the dropped IR% values could not reflect accurate results. Thus, the appropriate data were collected to evaluate the IR% values of different crRNAs (crRNA19 and 21 at 5 min and the others at 10 min) (Fig. 2). According to the schematic, crRNA with high IR% indicated that the dominated site was within the hybridization region. The mean IR% value of crRNA 7–24 was significantly higher than the mean value of the others (One-way ANOVA analysis). This suggested that the repeated site in crRNA 7–24 (sites 24–26 from the 5′ end to the 3′ end in the AFB1 ssDNA aptamer) was speculated to be the dominated site (Fig. S4†). Few reported studies have specifically analyzed the dominated sites between AFB1 and its 47 nt ssDNA aptamer. However, several studies have been reported to develop AFB1 biosensors from different experimental strategies. For example, the fluorescence anisotropy-based AFB1 detection approach was developed by labeling fluorophores on different bases.17 However, the quenching effect of the G base made it hard to be labeled with a fluorophore, greatly limiting the application in aptamer sequences with the G base. Among these studies, we could analyze that the sites causing the maximum change of signal were sites 24 and 26,18 sites 25–27 (ref. 11) and sites 20, 24, 26, and 28.17 Although the results are different, they showed that the dominated sites are near the 3′ end of the loop structure and our results are consistent with the reported results in this study.
Fig. 3 BLI assay to analyze the interaction between AFB1 and aptamers. A) Aptamer-WT. B) Aptamer-MT. 0 to 120 s is the association process and 120 to 240 s is the dissociation process. |
Molecular docking was used to predict the dominated sites and binding energy of interactions between the 47-nt ssDNA aptamer and AFB1. The docking energy is −5.83 kcal mol−1 and the binding residues of the aptamer are C17, G25, T26 and C27 (Fig. 4C). The results showed that these four sites interacted with the furan and coumarin moieties of AFB1 through hydrogen bonding which are consistent with reported results.11,17 This is not completely consistent with our experimental results (sites 24–26), which illustrates that computational tools for 3D modeling still have limitations in predicting accurate molecular interactions under complex conditions, especially in the case of lacking computational tools for 3D modeling of ssDNA molecules. However, it is still good enough as an auxiliary tool and it will achieve excellent performance in analyzing dominated sites of the aptamer together with our experimental strategy.
The trans-cleavage activity of Cas12a would be activated in the case of the crRNA/Cas12a complex hybridized with complementary ssDNA or complementary dsDNA with the PAM sequence (TTTN).20,21 However, whether a special DNA structure like a hairpin can activate the trans-cleavage activity of Cas12a has never been reported. Therefore, six DNA structures were designed to evaluate the effects of different special DNA structures on the activated trans-cleavage activity: ssDNA, hairpin, dsDNA-non PAM, triple-complex-non PAM, dsDNA-PAM and triple-complex-PAM (Fig. S6†). In addition, to illustrate that this is not a single case, the crRNA in this design (crRNA-special structure DNA in Table S1†) is different from crRNA 1–28. The annealed six DNA structures were hybridized with the same crRNA sequence and the activated trans-cleavage activities were analyzed through the fluorescence intensity. As shown in Fig. 5, ssDNA and dsDNA-PAM could activate trans-cleavage activity and dsDNA-non PAM could not activate trans-cleavage activity, which were consistent with the reported study mentioned above. Triple-complex-non PAM is a structure between ssDNA and dsDNA-non PAM and crRNA hybridized with the middle ssDNA region. This demonstrated that the process of crRNA hybridization was mainly controlled by the ssDNA structure yet the partial dsDNA at both ends near the middle ssDNA structure affects the trans-cleavage activity. The trans-cleavage activity activated by triple-complex-PAM was weaker than that activated by triple-complex-non PAM. This result supposed that the hybridization efficiency affected due to the PAM sequence is one more step to be verified in the hybridization process. In addition, the trans-cleavage activity activated by hairpin was similar to that activated by triple-complex-non PAM. Partial dsDNA at both ends and a 20-nt middle ssDNA to be hybridized were common characteristics for these two structures. Therefore, the recognition of the hairpin structure could be equivalent to the process shown in Fig. S7.† And the trans-cleavage activity activated by these two structures was weaker than that activated by ssDNA, which could be an explanation for the result of crRNA 10 activated weak trans-cleavage activity.
Fig. 5 Effects of different special DNA structures on activated trans-cleavage activity. Error bars in the figures present the standard deviation (SD) based on at least three individual tests. See the sequence of symbols in Fig. S6.† |
Quantitative analysis was performed by monitoring the inhibition rate % (IR%) 20 min after the addition of different concentrations of AFB1 (Fig. 6A and S9†). For biosensor-crRNA 24, the resulting plot of IR% vs. AFB1 concentration was linear over the range of 100 ng mL−1 to 2 μg mL−1. The limit of detection (LOD) was calculated to be 0.8 ng mL−1, equivalent to 2.6 nM based on 3*SD/slope, where SD is the standard deviation of nine NC samples (Fig. S9A†). For biosensor-crRNA 11, the linear range was 200 ng mL−1 to 2 μg mL−1 and the LOD was calculated to be 4.2 ng mL−1, equivalent to 13.3 nM (Fig. S9B†). For biosensor-crRNA 13, the linear range was 500 ng mL−1 to 2 μg mL−1 and the LOD was calculated to be 4.5 ng mL−1, equivalent to 14.3 nM (Fig. S9C†). For biosensor-crRNA 27, the linear range was 750 ng mL−1 to 2 μg mL−1 and the LOD was calculated to be 4.7 ng mL−1, equivalent to 15.0 nM (Fig. S9D†). crRNA 10 has no response to AFB1 (Fig. S9E†). The results illustrated that the performance of the different biosensors varied with the different crRNAs, and biosensor-crRNA 24 performed best, reaching a LOD of 2.6 nM within 20 min. The sensitivity of our assay was comparable to those of fluorescence biosensors reported for AFB1 detection (Table S2†).
Among the five crRNAs studied in this section, the dominated sites 24–26 (“TGT”) were covered in the hybridization regions of crRNA 24, 11, 13 and 10. Apart from crRNA 10 with a special hybridization region, the biosensors based on crRNA 24, 11, and 13 all had better performance than that based on crRNA 27. This illustrated that our strategy not only analyzed the dominated sites between AFB1 and the aptamer but also developed an AFB1 biosensor with high sensitivity.
The performance of our biosensor in the complex sample matrix was evaluated by spiking with different concentrations of AFB1 in 5% red wine, beer and milk samples.
The estimated recovery of the strategy ranged from 91.6% to 94.47% in red wine, 104.8% to 110% in beer and 87.87% to 90.2% in milk (Table S3†). Further, we tested the AFB1 samples in the complex matrix using the ELISA detection kit method, and the results indicated the accuracy of the method used in the work (Fig. S10†). The above results illustrated the feasibility of our biosensor for detecting AFB1 in a complex sample matrix, and it has the potential to be applied for real sample analysis.
Footnote |
† Electronic supplementary information (ESI) available: The secondary structure (a simple stem-loop structure) of the 47-nt AFB1 aptamer from oligo analysis website Mfold; effect of organic solvent (acetonitrile) on the trans-cleavage activity of Cas12a; kinetic fluorescence results IR% of 28 RNAs; repeat sites in crRNA 7–24 (sites 24–26 from the 5′ end to the 3′ end in the AFB1 ssDNA aptamer) were speculated to be the dominated sites; RNA 10 hybridization with the whole loop sequence from sites 10–29 of the aptamer; sequences and symbols of six structures; the predicted process of recognition of hairpin structure; sensitivity tests for 5 different crRNAs. See DOI: https://doi.org/10.1039/d2sd00152g |
This journal is © The Royal Society of Chemistry 2023 |