CRISPR/Cas12a‐loaded intelligent DNA hydrogel for universal and ultrasensitive exosome assay

Tumor‐derived exosomes are crucial for early non‐invasive and accurate tumor diagnosis in clinical diagnostics. The development of highly sensitive, simple, and intuitive exosome assays has sparked a research upsurge in clinical diagnostics. Here, we develop a bio‐responsive intelligent DNA hydrogel loaded with CRISPR/Cas12a for universal and ultrasensitive detection of the exosomes. The aptamer serves as the target response unit and switch, competitively disintegrating the region of the DNA linkers and then Cas12a/crRNA was released and activated, resulting in a high fluorescent intensity for exosome detection at the detection limit of 119 particles/μL. Moreover, a constructed colorimetric tube is made by loading a colorimetric filter membrane on the tube lid and intelligent DNA hydrogel on the tube bottom, which enables one‐pot portable colorimetric detection. Without the need for laboratory instruments and professionals, this strategy allows for naked eye detection with limit of detection as low as 104 particles/μL, and shows great applicability in distinguishing between healthy individuals, pretreatment patients, and post‐treatment patients after obtaining a testable analyte. In this study, an ultrasensitive detection platform for exosomes that enables one‐step sensing and dual signal output was introduced. The findings here suggest that this platform is a promising tool for the application of liquid biopsy based on exosomes in clinical diagnosis.


INTRODUCTION
Exosomes, which are nanoscale liquid biopsy biomarkers secreted by cells, are found in a variety of body fluids with high abundance and stability, and the minimal-invasive approach. 1Because of their roles in intercellular communication, angiogenesis, and cancer metastasis, 2,3 exosomes are very suitable target for early cancer screening, diagnosis, and disease surveillance. 4Recently, many studies have explored the screening and detection methods using tumor-derived exosomes. 5,6However, less specific, less sensitive, and less practical methods, such as nanoparticle tracking analysis (NTA), 7 flow cytometry (FCM), 8 and ultracentrifugation have resulted in unique clinical guidance for early warning, cancer treatment, and prognosis, especially when cancer cells secrete fewer exosomes, 9 thereby making it more difficult to detect the exosomes, efficiently and sensitively.Limitations to current methods indicate the need for a more straightforward and sensitive platform for exosome detection.
An intelligent DNA hydrogel is a prospective candidate for improved detection sensitivity because of its targetstimulated responsiveness and stable controllability of DNA sequences. 10,11As a functional material that perfectly integrates programmable skeletal structures with the biological functions of DNA, intelligent DNA hydrogels encapsulate functional nucleic acids (such as aptamers [APs] or DNAzyme) and signal probes in a polymer skeleton, [12][13][14] forming a three-dimensional (3D) network structure through cross-linking, which greatly improves the binding efficiency of analytes and their receptors.This colloidal structure contributes in part to high sensitivity, which is superior to homogeneous solution detection. 15ecently, various target-responsive DNA hydrogels have been established for the detection of biomolecules, chemicals, metal ions, etc. [16][17][18] ; however, some of these responses rely on indirect signal transduction such as an extra introduction of electroactive substances to have an impact on the result readout. 19As for other signaling methods, the signal molecules (e.g., fluorophores) are simply encapsulated in the hydrogel matrix, resulting in a low loading capacity and high signal background, which makes them less sensitive and unstable sensing systems. 20This research gaps inspire us to develop this intelligent DNA hydrogel with a one-step response, which is a promising candidate system for facilitating highly sensitive and bio-friendly detection of exosomes.
CRISPR-Cas systems (clustered regularly interspaced short palindromic repeat and CRISPR-associated protein) are a recent development in diagnostic biosensing because of their versatility and programmability. 21,22pecifically, Cas12a requires the activation of one CRISPR RNA (crRNA) by the target sequence to exert strong side-branch cleavage activity. 235][26] Presently, the detection strategies of various biomarkers (such as N-gene, 27 transcription factor, 28 nucleic acids 29 ) have been identified using intelligent hydrogel sensing platforms and Cas12a.In principle, these methods rely on introducing the Cas12a and reporter probes after the upstream recognition. 30,31owever, this is a two-step procedure, as the signal substance must be released first, before triggering the downstream response. 32Therefore, developing an intelligent program with "recognition-response" guided by Cas12a will simplify the analysis protocol and has potential clinical application value.
Thus motivated, an intelligent DNA hydrogel that integrates "Programming, Recognition, Response, and Dual Signal Output" has been explored for the ultrasensitive detection of exosomes.This assay included two stages.First, in the specific identification stage, the AP of human hepatocellular carcinoma (HCC) HepG2 cellderived exosome was functionalized as the molecular recognizer, which separated the cross-linked sequences in the intelligent hydrogel in the presence of exosomes. 33n the meantime, the encapsulated Cas12a/crRNA was released, and the branched ssDNAs (S1, S2) attached to the polyacrylamide backbone were exposed.Second, in the intelligent response of signal self-amplification stage, the released S1 and S2 served as activators of Cas12a/crRNA and the signal reporter, respectively.Guided by a single crRNA, Cas12a/crRNA recognized S1, and its indiscriminate ssDNA cleavage activity was activated to completely degrade S2, thereby enabling the release of the fluorescence signal. 34The more targets that were present in the system, the more the intelligent DNA hydrogel collapsed.In this case, Cas12a/crRNA was continuously released and activated, producing a highly amplified fluorescent signal.Except for the one-step fluorescence analysis, the different statuses of the intelligent DNA hydrogel textures before and after the reaction provided us with a colorimetric approach to detect tumor-derived exosomes with the naked eye or using a smartphone with the assistance of lyophilized gold nanoparticles. 35This intelligent DNA hydrogel does not require separate reporting probes or pre-amplification stage, instead only requiring a one-step assay from molecular recognition to signal response to the final visual readout, with minimum reliance on expensive instruments, thereby making it more promising for more sensitive bioanalysis.

Apparatus
The fluorescence spectra were analyzed using a Thermo Scientific Varioskan Flash instrument.FCM analysis was performed using a BD FACSCanto II flow cytometer (BD Biosciences).Transmission electron microscopy (TEM) images were obtained using a JEM-1200EX instrument.Scanning electron microscopy (SEM) images were recorded using a ZEISS Sigma 300 instrument.NTA was performed using a Malvern NanoSight NS300 instrument.Rheological tests were performed using an Anton Paar MCR 302 rheometer.Confocal laser scanning microscopy imaging was performed using a ZEISS LSM780 (Zeiss).

Cell culture and exosome isolation
HepG2 cells were cultured in 1640 medium containing 1% (v/v) penicillin-streptomycin and 10% (v/v) FBS in a humid atmosphere with 5% CO 2 .The supernatant was removed and serum-free medium was added until enrichment of HepG2 cells reached 70%.After 48 h, the supernatant was collected to extract exosomes.The collected supernatant was centrifuged at 3000 × g for 15 min to remove cell debris and large particles.Then, the large granular vesicles were discarded after filtration (0.22 μm pore size) and ultracentrifugation at 11,000 × g for 30 min.Subsequently, the supernatant was ultracentrifuged at 120,000 × g for 2 h to obtain the exosome particles.Finally, the exosome particles were resuspended with PBS and stored at −80 • C prior to the experiment.The morphology and concentrations of the exosomes were measured using TEM and NTA, respectively.In addition" western blot analysis with a CD63 antibody was used to confirm the expression of surface proteins of exosomes, and the binding efficiencies of the CD63 AP and exosomes were analyzed using the NanoFCM platform.

Construction of functional DNA hydrogel
Before making the DNA hydrogel, the strand S1 was first hybridized with CD63 AP heated at 95 • C for 5 min and then cooled down to 4 • C at a rate of 4 • C/min to obtain a pre-hybridized solution (S1-AP).Simultaneously, 150 nM Cas12a was incubated with 180 nM crRNA in 1× reaction buffer (10 mM Tris-HCl and 5 mM MgCl 2 ) at 25 • C for 20 min.
The intelligent DNA hydrogel was constructed using the following steps.First, DNA-grafted polyacrylamide chains were synthesized.The pre-hybridized solution (S1-AP) and S2 were mixed separately with 4% acrylamide, 1× Tris-EDTA (TE) buffer (pH 8.0), 200 mM NaCl, and ultrapure water, and then degassed at 37 • C for 10 min in a vacuum oven to form S1-AP-acr and S2-acr.The polyacrylamide-DNA (PA-S1-AP and PA-S2) was obtained after polymerization by adding the freshly prepared APS (0.06%, m/v) and TEMED (0.6%, v/v) into the above solution and degassing for 6 min at 37 • C. Finally, the prepared PS1-AP (14 μM) and PS2 (14 μM) were mixed and incubated at 65 • C for 5 min, then the Cas12a/crRNA (150 nM) was added and incubated at 45 • C for 10 min to ensure homogeneity.The mixture was cooled slowly to 25 • C to form the intelligent DNA hydrogel.

Optimization of the detection
The key parameters of the DNA hydrogel construction were explored to optimize the performance of this assay.Specifically, for the cross-linking density of DNA hydrogel, the concentrations of the acrylamide (4, 6, 8, 10, and 12 wt%) and DNA probes (S1, S2, and AP) (6, 8, 10, 12, and 14 μM) were determined.In addition, the optimal DNA probe and acrylamide concentrations were determined by combining the two factors; the different reaction times for exosomes detection (30, 40, 50, 60, and 70 min) and temperatures (21 • C, 29 • C, 37 • C, 45 • C, and 53 • C) were carried out to gain the optimal results.In addition, as the amount of Cas12a/crRNA embedded in the DNA hydrogel can significantly affect the sensitivity of the analysis, therefore, the concentration of Cas12 (25, 50, 100, 150, and 200 nM) was also optimized.

The fluorescence detection of exosomes
Under the optimized conditions, the tumor-derived exosomes with certain concentrations (from 1 × 10 3 to 1 × 10 7 particles/μL) were added to 10× buffer (100 mM NaCl, 150 mM MgCl 2 , 100 mM Tris-HCl [pH 9.0], 0.5% Tween-20, and 10 mM Dithiothreitol (DTT)).The solution was loaded into the intelligent DNA hydrogel, the reaction was activated, and then incubated at 37 • C, 300 rpm for 1 h in the dark.The fluorescence intensity of the reaction mixture was measured at 520 nm using a Thermo Fisher Scientific Varioskan Flash.The fluorescence intensities of the different response groups were used to calculate the sensitivity of the intelligent DNA hydrogel.The limit of detection (LOD) was defined as the minimum detectable concentration with a fluorescence intensity greater than that of the negative control.
Polyacrylamide gel electrophoresis (PAGE) was performed to verify the binding of the different DNA strands.All samples were incubated at 25 • C, mixed with loading buffer (5:1 volume ratio), and then a 12% gel electrophoresis was performed at 125 V for 40 min in 1× Tris-Acetate-EDTA (TAE) buffer.After staining with Gel Red, gels were imaged using GenoSens (1860).
To assess the specificity of the hydrogel, the exosomes (1 × 10 5 particles/μL) and three interferents, including glucose, cytochrome C, and BSA, the concentration all are 6 μg/mL, were tested under the optimized fluorescence detection.Finally, six parallel tests for exosomes (1 × 10 5 particles/μL) were conducted using the proposed fluorescence assay.The relative standard deviation (RSD) was calculated to evaluate the repeatability.

The colorimetric detection of exosomes
Preparation of freeze-dried colorimetric filter paper: a workable filter paper was cut into a disc shape (5 mm in diameter) and 20 μL of NaCl (500 mM) solution was added dropwise.The NaCl-loaded colorimetric filter paper was obtained by freeze-drying.Then, another 5 mm filter paper disc was soaked in a nano-gold solution (9 × 10 11 particles/mL) and then freeze-dried to obtain a colorimetric filter paper loaded with gold nanoparticles.Two filter papers, one loaded with NaCl and one with gold nanoparticles, were fixed on the lid of the centrifuge tube.
Optimization of the conditions for color development: first, the reaction of AuNPs with different concentrations of NaCl (25, 50, 75, 100, and 200 mM) was explored in solution.Second, the DNA-loading capacity of different filter paper specifications (filter paper, Whatman paper, nonwoven fabrics, and glass fiber membranes) was tested to screen the most suitable filter paper substrate.
As mentioned in Sections 2.4 and 2.5, the intelligent DNA hydrogel was constructed in a 100 μL centrifuge tube.After adding target exosomes to the tube, the mixture was incubated at 37 • C for 60 min.During this process, the prepared colorimetric paper was fixed to the lid of a centrifuge tube in an upright state.However, when the centrifugal tube was inverted for 5 min at the end of the reaction, the colorimetric paper in the lid absorbed the sol-liquid in the tube, after which the tube was then rotated to the upright state with a visible color displayed on the lid.
Moreover, the image of the test tube lid was captured using a smartphone, and colorimetric detection was conducted by analyzing the RGB information of the region of interest on the colored lid using the Colorpicker App on a smartphone.The principle of smartphone color recognition in our study is based on the camera's ability to capture images using a sensor with photodiodes beneath RGB filters. 36The sensitivity of the proposed method was determined by investigating the relationship between the diluted exosomes and the relative color change.The visual LOD was defined as the lowest exosome concentration required to produce an observable pale purple color, observable with the naked eye, determined by random group of 30 observants.The R/B values were calculated using the following equation: where R x and B x represent the R and B values obtained from the image of colorimetric paper with different concentrations of exosomes input into the detection system.

The applicability for testing clinical samples
In addition, because this sensing platform is used for the detection of biological samples, real serum samples from 20 healthy individuals (H), 20 patients with HCC (P), and 20 post-treatment patients (PT), were randomly collected from volunteers at the Southwestern Hospital.Samples were tested to confirm their applicability in the clinical field.For the clinical sample application of the tumor exosome assay developed herein.The following inclusion criteria apply: 18-80 years old; patients with HCC diagnosed with primary liver cancer at least 6 months ago; treated HCC patients at least 1 month after undergoing surgery for HCC; healthy populations were selected from those who had qualified for a physical examination report in the last 6 months; and none of the above three groups were suffering from chronic diseases and other cancers.Serum samples were centrifuged at 5000 × g for 30 min.The upper plasma layer was collected and centrifuged at 500 × g for 10 min.The supernatant was collected and centrifuged at 15,000 × g for 20 min.Finally, the exosome solution was obtained by resuspension precipitation in PBS for subsequent detection analysis (referring to the aforementioned fluorescence and colorimetric detection schemes).The p-value represents the difference in the fluorescence intensity of the exosome responses between the different groups.The colorimetric results were analyzed using an algorithmic clustering analysis.This research proposal was approved by the Southwest Hospital.All the participants provided written informed consent to participate in this study.

Statistical analysis
All data are expressed as mean ± standard deviation of three experimental triplicates.Error bars were obtained by performing experiments in triplicate, except for special statements and the data were reported as mean.Comparisons between the P and PT groups were performed using independent sample t-tests.p-Values below 0.05 (*) or 0.001 (***) were considered to be statistically significant.Statistical analyses were performed using the Statistical Package for the SPSS (version 20.0).The LOD was estimated based on 3/s (where σ represents the standard deviation in the blank sample, or sample without exosomes, and s is the slope of linear calibration).Cluster analysis of the different groups (P, PT, H) was conducted using the K-means algorithm in Python.

Design of the dual-mode detection
Previously reported strategies for exosome detection often involved multiple steps and required specialized instruments for result interpretation.Therefore, we designed the fluorescence and colorimetric dual-mode intelligent DNA hydrogels to respond to CD63 expressed on exosomes of cancer cells to support the diagnosis and prognosis of the disease.Figure 1 shows that without stimulus exosomes, the functional DNA hydrogel wrapped with Cas12a is in a stable soft-solid state, and when stimulus exosomes are present, the surface CD63 protein dissociates the gel cross-linking sequences (AP), and the DNA hydrogel is transformed from a soft-solid texture to a more soluble, high fluorescence intensity mixture.This one-step response assay ensured that the fluorescence signal was trapped inside unless the target appeared, thereby avoiding false-positive results, amplifying the fluorescence intensity, and improving sensitivity.For colorimetric detection, freeze-dried AuNPs and NaCl were loaded onto a filter paper fixed on the lid of the tube.After the previous reaction, the tube was inverted until the liquid wetted the filter paper, and the color transition from blue to red could be observed by the naked eye with an increase in the concentration of the target exosome.Semiquantitative detection was established by analyzing RGB information using a smartphone.This rapid and portable colorimetric assay ensured the secondary confirmation of the test results without the need for additional laboratory instruments, providing a simple and portable way to interpret the results.

Characterization of the exosomes and functional DNA hydrogel
The polymerization process for the intelligent DNA hydrogel is shown in Figure 2A.Under the catalysis of APS and TEMED, S1-AP and S2 rapidly reacted with acrylamide to form polyacrylamide-DNA (PS1-AP, PS2), which was then co-incubated with Cas12a/crRNA to form an intelligent DNA hydrogel.The successful synthesis of the DNA hydrogels was verified by observing their morphology via SEM and by testing their rheological properties.As shown in Figure 2B, comparing the storage modulus (G′) and loss modulus (G′) of the optimized DNA hydrogels, G′′ was always smaller than G′, indicating the presence of gelation behavior in the DNA hydrogels. 37Additionally, DNA hydrogels were multi-cross-linked 3D reticular structures in SEM images (Figures 2C,D).
In this study, exosomes purified from the supernatants of HepG2 cell cultures were used as analytes.The concentration of exosomes analyzed by NTA was 2.2 × 10 10 particles/mL and the average hydrodynamic diameter was approximately 127 nm (Figure 2E).TEM images demonstrated that the small vesicles were wrapped in a bilayer of lipid molecules in the exosomes (Figure 2F), consistent with previous reports. 38In addition, significant CD63 and TSG101 bands were observed in western blot images (Figure 2G); however, the exosome-negative marker calnexin was not expressed, indicating that CD63 can serve as an effective biomarker for detecting exosomes.
Optimal formation of an intelligent DNA hydrogel is a prerequisite for this reaction.Factors during the synthesis of the intelligent DNA hydrogel were optimized, including the acrylamide concentration and the concentration of DNA probes (S1, S2, and AP), to stabilize the multidimensional extension of polymeric chains.As shown in Figure S1A, the fluorescence intensity continued to decrease with increasing acrylamide concentration, because high acrylamide concentration increased the stiffness of the DNA agarose gel, resulting in lowered efficiency of the gel-tosol transition.DNA concentration is not only a key factor in the cross-linking density of the DNA hydrogel, but also in the direct source of the fluorescence signal of the detection platform.Figure S1B shows that fluorescence intensity increased with increasing DNA concentration.Figure S1C shows the results of the interaction between the above two parameters, which yielded the same results as the singlefactor experiment.Therefore, the combination of 14 μM of DNA hydrogel and 4% wt of acrylamide were chosen for subsequent experiments.

Feasibility analysis of the working principle
Once the intelligent DNA hydrogels were established, they maintained a stable gel-solid state.One of the highlights of our design is that the AP sequences not only function as linkers of the DNA hydrogels, but also as specific recognition elements for exosomes, which means that the more exosomes present in the system, the more AP sequences will be combined, leading to less hydrogel formation.Meanwhile, the presence of exosomes leads to the dissociation of the 3D cross-linked structure, exposing the branching sequences and releasing Cas12a, which then shears the lateral branches.Theoretically, nearly no fluorescent signal can be detected before adding exosomes, whereas a strong fluorescent signal will be generated in the presence of exosomes.Therefore, a correlation between fluorescence changes and exosome concentration was observed in this one-step response system.
Furthermore, all switch elements of the entire reaction are exosomes, which can be competitively combined with the AP sequence.The cross-linked structure of the intelligent hydrogel was then dissociated to release Cas12a, activated by the exposed branching sequence, which sheared the lateral branches to generate a fluorescence signal, which realizes the one-step response of exosome detection (Figure 3A).Synthesis of intelligent DNA hydrogels and their successful dissociation in the presence of exosomes are critical for the success of the assay.First, PAGE electrophoresis was performed to verify the feasibility of the method.In Figure 3B, lanes 2, 3, and 4 represent the S1, S2, and AP bands, respectively.Lanes 5 and 6 display the bands of the S1-AP and S2-AP duplexes, which demonstrate that S1 and S2 could cross-link with AP.Lane 7 shows a triple-stranded complex of S1 and S2 hybridized with AP (AP-S1-S2).In the presence of exosomes, the bands were significantly reduced (lane 8), indicating the dissociation of the crosslinked structures.The ability of AP to bind to exosomes is a key step in determining whether the 3D network structure of intelligent DNA hydrogels can be destroyed.As shown in Figure 3C, the results of NanoFCM showed that approximately 99.6% of the exosomes were successfully recognized and bound by AP, indicating that the surface exosomes expressed abundant CD63 and were able to bind well to AP.The response triggered by exosomes was verified using fluorescence analysis.As shown in Figure 3D, almost no fluorescence was observed in the absence of exosomes, indicating that the disorderly cleavage activity of Cas12a/crRNA was not activated, thus leading to a low-fluorescence background of the intelligent DNA hydrogel.In contrast, in the presence of exosomes, the activated Cas12a/crRNA cleavage reaction yielded a significant fluorescent signal, which proved that exosomes could competitively dissociate AP sequences and expose the S1 and S2 sequences.This phenomenon was illustrated more clearly when the reaction system was exposed to ultraviolet irradiation (Figure 3E).As expected, the above results support that the intelligent DNA hydrogel collapsed after triggering the exosomes.Simultaneously, Cas12a/crRNA was successfully activated, and finally, the fluorescent signal was released, indicating that the assay can detect exosomes.

Detection performance of the fluorescence assay
To ensure the best performance for this biosensing method, parameters such as reaction time, temperature, and concentration of Cas12a were optimized.As shown in Figure S2A, the number of APs competing with exosomes on the constructed functional DNA hydrogel increased with increasing reaction time, leading to the release of more Cas12a/crRNA.Subsequently, the cleavage function of the activated CRISPR-Cas12a system results in the restoration of fluorescence in a previously quenched solution.After 60 min, stable and remarkable fluorescence intensity was observed, indicating that the reaction reached saturation within 60 min.The appropriate temperature for biological activity detection is important.As shown in Figure S2B, low temperatures resulted in a weak fluorescent signal, which might be due to the poor biological activity of exosomes and the stable structure of the DNA hydrogel.However, a higher temperature (above 37 • C) generated a lower fluorescent signal, which was attributed to the inactivation of exosomes and the inactivation of Cas12a protein at high temperatures. 38 suggesting the possibility that the inactive Cas12a in the intelligent hydrogel binds to the substrate after reaching supersaturation, thereby protecting the substrate from cleavage by inactive Cas12a.The optimized conditions were used for subsequent tests.
The constructed intelligent DNA hydrogel was employed for exosome detection under optimized conditions, and its sensitivity was investigated to evaluate its analytical performance.As shown in Figure 4A, the fluorescence intensity of DNA hydrogels increased with the increase of the concentration of exosomes; a strong linear relationship (R 2 = 0.9705) was obtained between the logarithm of the target concentration ranging from 1 × 10 3 to 1 × 10 7 particles/μL and the fluorescence intensity (Figure 4B).The detection limit of exosome was 119 particles/μL, which was in accordance with the 3σ rule.

Specificity and reproducibility
To evaluate the specificity of the constructed method, possible coexisting substances in biological samples, such as cytochrome C, glucose, and BSA, were used as controls.The results shown in Figure 4D demonstrate that the proposed intelligent hydrogel has good specificity for exosomes when distinguishing between biological samples, as fluorescence intensity could only be determined when exosomes were tested in the intelligent hydrogel, suggesting the hydrogel was more specific than were the other three interferents.Then, the repeatability of the detection system was tested by constructing six parallel tests of exosomes at a concentration of 1 × 10 5 particles/μL (Figure 4E).The obtained 2.4% of RSD indicates a stable signal output across the six parallel tests, which supports the reliability and robustness of the established assay.

Detection performance of the colorimetric assay
To improve the applicability and portability of the detection method, a one-pot visual test strategy was proposed.First, colorimetric filter paper was loaded on the lid of the detection tube.After the experimental procedure described in Section 2.7, as illustrated in Figure 4A, the color of the tube lid changed from blue to red with an increase in the exosome concentration.The main theoretical support for this inference is that more exosomes dissociate the cross-linking sequence, which makes the tensile strength and supporting force of DNA hydrogel decrease significantly, lose its mechanical rigidity, and change into a more fluid state. 39Then, the DNA single strands in the hydrogel in the fluid state will protect the aggregation of gold nanoparticles induced by NaCl, thereby achieving significantly different colors.
To verify the above scheme, a colorimetric filter paper was prepared and loaded onto a tube lid (Figure S3A). Figure S3B shows the principle and results of the confirmatory experiment.When the target was present, the intelligent DNA hydrogel underwent a gel-to-sol transition.The sol contained a large amount of single-stranded DNA that protected the AuNPs from aggregation, resulting in a red color on the colorimetric paper.Conversely, in the absence of the target, the double-stranded DNA encapsulated in the gel remains intact with less liquid.When the test tube was inverted, almost no single-stranded DNA in the liquid dissolved in the colorimetric filter membrane, leading to the aggregation of AuNPs and blue coloration.This phenomenon demonstrates the successful design of the colorimetric assay.To achieve maximum color intensity of the colorimetric filter paper, different concentrations of NaCl solution were checked to further determine the aggregation degree of gold nanoparticles; a remarkable blue color was observed at a concentration of 100 mM NaCl solution, while more NaCl resulted in less chroma (Figure S3C).Moreover, the adsorption capacity of the colorimetric filter paper for DNA and AuNPs is crucial to the results.Therefore, four types of colorimetric filter papers were chosen, and their microscopic structures were observed under an electron microscope (Figure S4).The DNA adsorption efficiency of each filter paper was studied using LSCM, and the results showed that the non-woven fabric membrane exhibited satisfactory DNA adsorption efficiency (Figure S5A).
Under optimized conditions, the developed assay can be used to detect exosomes with the naked eye.As shown in Figures 4A and S5B, the colorimetric filter paper exhibited a blue-red transition as the concentration of exosomes increased from 1 × 10 3 to 1 × 10 7 particles/μL.Additionally, fewer exosomes drove little color change (Figure S5C), which might be attributed to the fact that the DNA hydrogel barely dissolves at excessively low exosome concentrations.The color change was recognized by the naked eye until the exosome concentration reached 1 × 10 3 particles/μL, indicating that the colorimetric detection performance was comparable to that of the fluorescence assay.To eliminate the influence of different individual color sensitivities, we used smartphones to identify color information and generate a standard curve.This allowed for semiquantitative detection of the test samples, making the results more objective and traceable.The R and B values were identified using a smartphone and a strong linear relationship (R 2 = 0.976) was calculated in the range of 1 × 10 3 to 1 × 10 7 particles/μL (Figure 4C), and the LOD was set as 1 × 10 4 particles/μL based on the visual recognition results of 30 random observants (Figure S6).

The applicability for testing clinical samples
To further evaluate the feasibility of the proposed intelligent DNA hydrogel in a real clinical diagnostic scenario (Figure 5A), 60 serum samples from healthy individuals, patients, and post-intervention patients with HCC (20 samples per group) were randomly selected and analyzed using the proposed dual-mode detection method (Figure S7).As shown in Figure 5B, the fluorescence intensities in the buffer and serum were almost identical, indicating that the intelligent DNA hydrogel is applicable to complex matrices.In the fluorescence assay, as shown in Figure 5C, the fluorescent signal of healthy human samples was significantly lower than that of the patient group, which is consistent with previous reports. 39In the patient group, there was a significant difference (p-value < 0.0001) in fluorescence signaling between preoperative and postinterventional patient samples.In the colorimetric assay, after application of the K-means algorithm, 60 samples were scattered into three clusters: the patient group, postintervention patients, and healthy individuals (Figure 5D), demonstrating the considerable capability of the colorimetric assay.
A comparison of the fluorescence and colorimetric methods for exosome detection is summarized in Table S2.Compared to other fluorescence methods, the LOD obtained by this intelligent DNA hydrogel "one-step method" is equivalent to other single colorimetric methods and better than other fluorescence methods.Compared to other colorimetric methods, the operation process of our prefabricated colorimetric tube is simple, and naked eye detection can be realized without large instruments.In addition, the results of the dual-mode detection in a reaction tube can be mutually verified, which ensures the reliability of the results.Overall, with the advantages of high sensitivity, easy operation, rapid readout, and applicability to real serum samples, this dual-mode strategy could help identify cancer at different stages and may be promising for tumor outcome monitoring and prognostic assessment.

CONCLUSION
In summary, a CRISPR/Cas12a signal amplification system was embedded in a spatial confinement based on a DNA hydrogel, creating a novel and user-friendly intelligent DNA hydrogel exosome detection platform.On this detection platform, the output results included fluorescent and colorimetric signals, which enabled both accurate quantification and rapid visual detection of exosomes.By utilizing target recognition based on APs, the target signal was converted to the cutting efficiency of Cas12a/crRNA on the hydrogel skeleton chain.It is worth noting that the Cas12a/crRNA system was pre-encapsulated in the DNA hydrogel, enabling one-step completion of the target input-signal output.The 3D mesh structure of the DNA hydrogel increases the chances of AP-target binding and provides a rich surface area, thereby enhancing the sensitivity of detection.A detection limit of 119 particles/μL was achieved, and good specificity was maintained even in the presence of other interfering substances.Additionally, this reaction system enables rapid and convenient visual detection with an LOD of 10 4 particles/μL.In the detection of clinical samples, both fluorescence and colorimetric detection have demonstrated good ability to identify differences in the expression levels of exosomes between healthy individuals and patients.However, further research is required to assess its clinical applicability, as the sample pool in this study was not large enough.Regardless, our approach still provides a novel system for exosome detection, and the detection strategy provided by this intelligent DNA hydrogel can be extended to detect other biological targets by changing the adapter sequence.The results of this study set a foundation for the combined detection of disease biomarkers and improve the accuracy of disease diagnosis.

F I G U R E 1
Overview of the dual-mode detection of exosomes based on the intelligent DNA hydrogel loaded with CRISPR/Cas system.F I G U R E 2 Characterizations of the intelligent DNA hydrogel and exosomes.(A) The process of synthesizing the intelligent DNA hydrogel loaded with CRISPR/Cas system.(B) The rheological parameters of the synthesized the intelligent DNA hydrogel.(C and D) Establishment of the 3D network structure of the intelligent DNA hydrogel under the scanning electron microscopy (SEM) microscope.Scale bar: 5 and 10 μm.(E) Results of the average size of exosomes tested by nanoparticle tracking analysis (NTA).(F) Transmission electron microscopy (TEM) images of exosomes.Scale bar: 200 nm.(G) Western blot analysis of CD63 and TSG101 expression on exosomes.

F I G U R E 3
Feasibility of the working principle.(A) Schematic illustration of exosomes triggering dissociation, activation and cleavage of the intelligent DNA hydrogel.(B) 12% polyacrylamide gel electrophoresis (PAGE) analysis of functional DNA synthesis process.Lane 1: 500 DNA ladder; lane 2: S1; lane 3: S2; lane 4: aptamer (AP); lane 5: S1 and AP; lane 6: S2 and AP; lane 7: S1, S2, and AP; lane 8: exosomes, S1, S2, and AP.(C) The banding efficiency of AP to exosomes analyzed by nano flow cytometry.(D) Fluorescence spectra with response to the presence of target and absence of target from 560 to 650 nm.(E) Results of differential fluorescence of target (-) and target (+) under ultraviolet (UV) light.
Figure S2C depicts that the fluorescence intensity increased continuously until the concentration of Cas12a reached 150 nM, F I G U R E 4 The performance of the intelligent DNA hydrogel.(A) Illustration of the schematic diagram of dual-mode (fluorescence and colorimetry) of signals.Inset: fluorescence spectra and images of color filter paper with response to different concentrations of exosomes from 505 to 580 nm.(a-f) Blank (without exosomes), 1 × 10 3 particles/μL, 1 × 10 4 particles/μL, 1 × 10 5 particles/μL, 1 × 10 6 particles/μL, 1 × 10 7 particles/μL.(B) The linear relationship between fluorescence intensities and the logarithmic values of different exosomes concentrations.(C) The linear relationship between R/B values and the logarithmic values of different exosomes concentrations.(D) Result of the specificity.(E) Repeatability of the intelligent DNA hydrogel.

F I G U R E 5
Applicability of the detection of exosomes in real samples by dual-mode detection.(A) The extraction of exosomes from serum samples.(B) Fluorescence response in different environments without (1) and with (2) exosomes, data are presented as mean ± standard deviation (SD), and significance is determined using independent samples t-tests.*p < 0.05.(C) Scatter plot corresponding to fluorescence intensity of clinical samples (healthy individuals (H), patients (P), and post-treatment patients with hepatocellular carcinoma (HCC) (PT), with 20 people in each group) analyzed by the proposed method.Significance is determined using analysis of variance (ANOVA) with post hoc multiple comparisons (Lest Significant Difference (LSD)).****p < 0.0001.(D) Cluster analysis diagram.Dots with different colors represent colorimetric values of different samples.