Fast microwave heating-based one-step synthesis of DNA and RNA modified gold nanoparticles

DNA/RNA-gold nanoparticle (DNA/RNA-AuNP) nanoprobes have been widely employed for nanobiotechnology applications. Here, we discover that both thiolated and non-thiolated DNA/RNA can be efficiently attached to AuNPs to achieve high-stable spherical nucleic acid (SNA) within minutes under a domestic microwave (MW)-assisted heating-dry circumstance. Further studies show that for non-thiolated DNA/RNA the conjugation is poly (T/U) tag dependent. Spectroscopy, test strip hybridization, and loading counting experiments indicate that low-affinity poly (T/U) tag mediates the formation of a standing-up conformation, which is distributed in the outer layer of SNA structure. In further application studies, CRISPR/Cas9-sgRNA (136 bp), SARS-CoV-2 RNA fragment (1278 bp), and rolling circle amplification (RCA) DNA products (over 1000 bp) can be successfully attached on AuNPs, which overcomes the routine methods in long-chain nucleic acid-AuNP conjugation, exhibiting great promise in biosensing and nucleic acids delivery applications. Current heating-dry strategy has improved traditional DNA/RNA-AuNP conjugation methods in simplicity, rapidity, cost, and universality. Simple methods for attaching polynucleotides to gold nanoparticles are of interest for simplifying conjugation in a range of applications. Here, the authors report a microwave heating-based method for the fast, one-step attachment of a range of thiolated or non-thiolated DNA and RNA to gold nanoparticles.

The construction of DNA-AuNP and RNA-AuNP conjugates is critical for these successful applications in biology, chemistry, medicine, and nanoscience. Conventionally, to attach negatively charged DNA and RNA to negatively charged AuNPs, a key is to screen the charge repulsion between AuNPs and nucleic acids. The most commonly used method is salt-aging method [17][18][19] , where salt was gradually added into the mixture of thiol (SH)-or poly (A)-tagged DNA and AuNPs to reduce charge repulsion. However, the salt-aging method required over two days to completion, making it a highly time-consuming process. A few new labeling methods were developed by adding acids 20,21 , surfactants 22 , polymers 23,24 , and organic solvent 25 , however, the use of extra reagents complicates the labeling process and may reduce the applicability of SNAs. Recently developed freeze-thaw method could attach thiol-modified oligonucleotides on AuNPs in a time-effective and reagentless form 26 . However, DNA/RNA thiolation increases the cost, especially for long oligonucleotides, whose chemical modification is very expensive and difficult 27,28 . Our group have successfully used a freeze-thaw method to attach thiol-free DNA/RNA on AuNPs, but this method still lacks universality due to sequence structure effect 29 . In summary, it seems that there is still a lack of an ideal DNA-AuNP and RNA-AuNP construction method that can simultaneously satisfy the simplicity, rapidity, low cost, and versatility. In particular, almost all currently reported DNA-AuNP and RNA-AuNP construction methods only demonstrated their feasibility for functionalization of short DNA and RNA sequences. Long DNA/RNA-conjugated AuNPs nanostructures, which may be developed as potential new biosensing, DNA/RNA vaccine 30,31 , and gene editing tool delivery strategies 32,33 , have been rarely developed.
Here, we report a microwave (MW)-assisted heating-dry method to construct SNAs, where a domestic microwave oven is used as a heater to drive the labeling process to completion in minutes without the need for extra reagents, and demonstrated its applicability to both thiolated and non-thiolated DNA/RNA. Further, for attaching non-thiolated DNA/RNA, we find that DNA/RNA-AuNP conjugates are poly (T/U) tag dependent under heating-dry circumstance. Spectroscopy, test strip hybridization, and loading counting experiments indicate that poly (T/U) tag is distributed in the outer layer of such a SNA structure. It is inferred that this essential low-affinity poly (T/U) tag is mainly responsible for the formation of a standing-up conformation in the AuNPs surface. Notably, our method addresses the deficiencies of traditional methods in long DNA/RNA labeling and allows selective functionalization of AuNPs with sgRNA (136 bp), RNA (1278 bp) fragment encoding nucleocapsid (N) protein (GU280_gp10 locus, termed N gene) of SARS-CoV-2 (accession: NC_045512.2), and rolling circle amplification (RCA) DNA products (over 1000 bp). In addition, the potential applications of this labeling strategy are demonstrated by developing new colorimetric identification of DNA single-base mutation or detecting viral DNA based on clustered regularly interspaced short palindromic repeats (CRISPR)based lateral flow strategy. The successful construction of long and structured DNA/RNA-conjugated AuNPs nanostructure may also indicate that a new functional nucleic acids delivery system based on AuNPs as a carrier can be developed. Overall, currently presented labeling strategy represents a fast, simple, cost-effective, and universal method that can be extended to the conjugation of almost all types of DNA and RNA sequences, which will expand the use of SNAs in biomedicine and nanotechnology in the future.

Results and discussion
Fast construction of DNA-AuNP and RNA-AuNP conjugates based on MW-assisted heating-dry method. Both nucleic acids and AuNPs solutions exhibit negatively charged properties, in order to promote effective nucleic acid-AuNP conjugation, some well-proven strategies, such as salt-aging 17,18 and low-pH methods 20,21 , are to shield the charge to benefit the affinity interaction between nucleic acids and AuNPs. Other effective strategies, such as freeze-thaw 26 and solvent methods 25 , are to increase the local concentration by compressing the physical space, thereby forcing the chemical covalent reaction or physical cross-linking between AuNPs and nucleic acids in an ultralocalized reaction volume. Our previous study has shown that freeze-thaw method encounters the influence of secondary structure of oligonucleotides on AuNPs-based labeling reaction 29 , thus lacks sequence universality for construction of AuNPs-based bioprobes. Here, we invented a new DNA-AuNP and RNA-AuNP construction strategy through MW-assisted heating-dry method (Fig. 1a, b). Heating is well known to destroy the higherorder nucleic acid structure and stretch the nucleic acid strands 34 . The microwave oven, which is used in almost every kitchen, is a simple and cost-effective heating device 35,36 . In particular, the microwave radiation can be completely reflected on the metal surface and therefore has no effect on the metal particles, while the aqueous solution can be quickly heated and dried. We envisioned that this heating-dry strategy could be an ideal DNA/ RNA-AuNP labeling method due to its role in simultaneously stretch the nucleic acid structure and concentrate the reaction volume. Accordingly, we firstly mixed SH-DNA with AuNPs solution (Supplementary Fig. 1). Surprisingly, after a few minutes of MW heating and drying, AuNPs showed red clumps, while AuNPs showed purple-black clumps with thiol-free DNA or in the absence of DNA ( Supplementary Fig. 1). After addition of salt solution to the resulted red clumps, AuNPs immediately redispersed into a red colloidal state, indicating that AuNPs and SH-DNA may be covalently conjugated ( Supplementary Fig. 1). These phenomena signified that a simple DNA-AuNP labeling technique using microwave-assisted heating-dry method has been developed.
We further seek to develop a universal and thiol-free nucleic acid-AuNP construction method (Fig. 1a). Although thiol modification is very easy to obtain for short nucleic acid sequences, for long-chain DNA and RNA, the chemical modification is very expensive and difficult. Therefore, the conjugation of long-chain DNA and RNA to AuNPs is rarely achieved. Adenine (A) base has been well confirmed to have good affinity for Au 37,38 , and poly (A)-tagged nucleic acid sequences have also been successfully applied to the coupling of AuNPs based on freeze-thaw or salt-aging methods 18,29 . Therefore, we take it for granted that poly (A)-tagged nucleic acid sequences should be suitable for the heating-dry labeling strategy we developed. However, to our surprise, in our attempt we found that only the poly (T/U)-tagged nucleic acids showed a preferred labeling efficiency under heating-dry circumstance (Fig. 1a, b). In the absence of DNA/RNA sequences or in the presence of random DNA/RNA sequences without poly (T/U) tags, AuNPs solutions will aggregate irreversibly (Fig. 1b). Further, results showed that heating-dry strategy can also be applied to labeling of larger-sized AuNPs and SH-DNA-gold nanorods (AuNRs) conjugation ( Supplementary Fig. 2).
In order to further characterize this DNA-AuNP conjugate, we directly measured the UV-vis spectroscopy of the poly (T)-tagged DNA conjugated AuNPs solution. It was shown that the absorption peak of AuNPs with poly (T)-tagged DNA exhibited a 6 nm red-shift, suggesting that there are DNA attached on the AuNPs surface (Fig. 1c). To further demonstrate the poly (T)-tagged DNA attachment on AuNPs, the AuNPs solution was centrifuged and washed and then analyzed by dynamic light scattering (DLS). The hydrated particle size of DNA-AuNP conjugate is larger than the bare AuNPs ( Fig. 1d and Supplementary Table 1), further indicating that poly (T)-tagged DNA strands were attached. In addition, the electron microscope results confirmed that the MW-assisted heating treatment did not damage the morphology of AuNPs ( Supplementary Fig. 3).
Next, we wonder whether the DNA-AuNP conjugate is enough stable. Firstly, we tested its stability by resuspending it with different salt concentrations. Results showed that the DNA-AuNP conjugates were still monodispersed even challenge 1 M NaCl solution and its absorption peak did not show observable shift (Fig. 1e), while the bare AuNPs aggregated at a salt concentration Fl. intensity (x 10 a.u.) Wavelength (nm)  of 50 mM ( Supplementary Fig. 4). Such high salt stability indicated that there were many poly (T)-tagged DNA sequences attached to the AuNPs surface. We subsequently measured the number of FAM-labeled poly (T)-DNA probes assembled on each AuNP according to a β-mercaptoethanol (ME)-based displacement experiment ( Supplementary Fig. 5a-c) 39 . As a comparison, the number of FAM-poly (A)-DNA on each AuNP that labeled using two reported thiol-free methods, poly (A)-based salt-aging and low-pH methods, were also measured. Results showed that the number of poly (T)-tagged DNA probes attach to each AuNP by MW-assisted heating-dry method is 83, which is about 2 times higher than the salt-aging and low-pH method (Fig. 1f). Further test indicated that when SH-DNA was used, for each AuNP approximately 270 DNA strands can be loaded ( Supplementary  Fig. 5d). We believe that the high DNA density is not due to the multilayered adsorption caused by hydrogen bonding between DNA strands. This is confirmed in a subsequent thermal stability experiment, where FAM-labeled DNA-AuNP conjugates exhibited negligible fluorescence increase in the solution at different temperatures, indicating high thermal stability of the DNA-AuNP conjugates and non-specific adsorption was also excluded (Fig. 1g). However, a rapid increase in fluorescence intensity was observed when the replacement reagent ME was mixed with the DNA-AuNP conjugates ( Fig. 1g). At the same time, we also monitored the time stability based on the measurement of the fluorescence of DNA-AuNP conjugates. The results showed that only negligible fluorescence intensity increases even for up to 47-day storage, indicating that the conjugates are stable for longtime storage (Fig. 1h). Further, the DNA-AuNP conjugates stability at a wide pH range (pH 3~11) was also tested. It was showed that both thiolated and non-thiolated DNA/RNA-AuNP conjugates exhibit excellent pH tolerance (Fig. 1i).
One may worry about whether such a high-temperature labeling condition will damage DNA and affect subsequent use. We then tested the nucleic acid integrity and hybridization ability of the resulted DNA/RNA-AuNP conjugates. It was found that DNA/RNA is still intact and the conjugates work as well as the conventional SH-DNA-labeled AuNP probes based on a test strip hybridization experiment ( Fig. 1j and Supplementary Fig. 6). Additionally, we also evaluated the potential risk of two typical DNA damages, such as pyrimidine dimer and apurinic/ apyrimidinic (AP) sites, and the experimental results indicated that the MW-assisted heating-dry treatment did not generate potential pyrimidine dimers and AP sites on the DNA strands (Supplementary Figs. 7 and 8).
Non-thiolated DNA/RNA-AuNP conjugation is poly (T/U) tag dependent. To further understand the sequence-based labeling effect, we designed four types of DNA with poly (A 10 ), poly (T 10 ), poly (C 10 ), and poly (G 10 ) tags at 5′-and 3′-terminal and mixed them with AuNPs, respectively. Moreover, four RNAs with poly (rA 10 ), poly (rU 10 ), poly (rC 10 ), and poly (rG 10 ) tag at 3′-terminal were also applied. After heating-dry and centrifugation-washing treatment, the DNA/RNA-AuNP conjugates were resuspended in 1 M NaCl solution. Results showed that only AuNPs with poly (T/U)-tagged DNA/RNA retained red color in 1 M NaCl solution, indicating AuNPs were protected by high-density DNA strands ( Fig. 2a(i), (ii), (iv)). Further assay showed that poly (A/T/C/G) tags at the middle position exhibited the similar phenomena ( Fig. 2a(iii)). Labeling using DNA sequences with poly (A/T/C/G) tags with varied lengths have also been tested. Under all situations only these poly (T)-tagged DNAs are efficient, thus the poly (A/ C/G) tags length effect is excluded ( Fig. 2b and Supplementary  Fig. 9). The results also showed that five T bases appeared to be enough for efficient labeling when the T bases are located at 5′-and 3′-terminal, and >10 T bases did not significantly improve the labeling efficiency (Fig. 2b). For T base located at the middle region of the DNA sequence, the labeling efficiency increased with the increase of T base number and T 30 base showed the highest labeling efficiency (Fig. 2b). All these observations indicated that poly (T) tag is essential for successful labeling and this labeling method exhibits flexibility in the position and length of poly (T) tag. In addition to the DNA sequence with terminal and middle tags, we also proved that the MW-assisted heating-dry method can achieve successful labeling for DNA sequence with poly (T) tag located at both ends or hairpin DNA sequence with shielded poly (T) tag in the stem region ( Supplementary Fig. 10a, b). The results indicated that heating-dry method can be applied to structured nucleic acid labeling, thus providing a more generic way than previously reported techniques, such as freeze-thaw method ( Supplementary Fig. 10c).
To more detailly evaluate the role of T base in labeling process, we replaced one or two T bases with A, C, or G base in poly (T 5 ) -tagged DNA sequences. Results showed that in most case the labeling efficiency dramatically decreased or totally lost upon this replacement (Fig. 2c), which indicated that five consecutive T bases are very critical. We also noticed that when replacing single T with G the labeling efficiency did not dramatically decreased. Previous observations and our colorimetric experiments indicated that T base has the lowest Au affinity ( Supplementary Fig. 11). Some pioneering works have concluded that the Au affinity order for all four bases is: A > C > G > T [40][41][42] . Dependence of lowaffinity T base during the heating-dry process prompted us to know how it is involved in this labeling method. To simplify the base effect evaluation, we designed four types of sequences consisting of two kinds of bases to observe this labeling response. Interestingly, we found that the T 10 A 1 sequence was effectively conjugated, while T 10 T 1 , T 10 C 1 , and T 10 G 1 did not (Fig. 2d). Further experiments showed that T 10 A 2 , T 10 C 2 , and T 10 G 2 can also be effectively conjugated while T 10 T 2 does not (Fig. 2d). However, when A 10 , C 10 and G 10 were used as the tag, all labeling reactions were unsuccessful (Fig. 2d). These experiments implied that T base is essential in the heating-dry labeling strategy but it is not directly involved in the DNA-AuNP conjugation.
Experimental observation of poly (T) tag orientation. In order to clarify the underlying labeling mechanism, we firstly used surface-enhanced Raman spectroscopy (SERS) to observe the poly (T) tag orientation in the SNA structure. Because the labeling mechanism based on Au-SH conjugates is well proved, we then tested the SERS signal based on the hybridization of a 5′-SH-DNA-AuNP conjugates and a complementary DNA (cDNA) probe with ROX labeled at 5′-and 3′-terminal, respectively. When 3′-ROX-cDNA-5′ and 3′-cDNA-ROX-5′ hybridized with the 5′-SH-DNA-AuNP conjugates, respectively, ROX will be close to or away from AuNPs surface. Expectantly, we observed significant SERS signal enhancement after 3′-ROX-cDNA-5′ probe hybridization ( Fig. 3a(i)). This indicated that SERS is suitable for observing the poly (T) tag orientation. Accordingly, we designed an "amphiphilic" DNA sequence with A 5 and T 5 tags located at 5′-and 3′-terminals, respectively, for AuNP conjugation. 3′-ROX-cDNA-5′ and 3′-cDNA-ROX-5′ probes were used for observing the SERS signals after hybridization. Experiments showed that 3′-ROX-cDNA-5′ probe mediated significant SERS signal enhancement ( Fig. 3a(ii)). When replaced the T 5 tag with T 10 tag, more significant SERS signal was observed ( Fig. 3a(iii)). These results supported that poly (T) tag is distributed in the outer layers of SNA structures.
We further executed a structured DNA labeling experiment to elucidate labeling mechanism. Two kinds of DNA sequences with well-confirmed G-quadruplex (G4) structures (G4-DNA1 and G4-DNA2) are employed. The circular dichroism (CD) spectroscopy was used to measure these two G4-contained sequences and random sequence, and the results showed G4-DNA1 and G4-DNA2 contained sequences present characteristic peaks at around 240 and 260 nm while random DNA sequence presents characteristic peaks at around 250 and 280 nm ( Fig. 3b(i), c(i)). Subsequently, we designed G4-DNA1 with T 10 tag at 5′-terminal and G4-DNA2 with T 10 tag at both 5′-and 3′-terminal for the AuNP conjugation. According to this design, it can be expected that if T 10 is involved in the AuNP conjugation, the G4 sequence will expose to the outer layer so that its structure is retained. Conversely, if the G4 sequence is involved in the AuNP conjugation, the G4 structure will be destroyed. CD spectra showed that G4 characteristic peaks disappeared after AuNP conjugation, verifying that G4 sequence is involved in the AuNP conjugation ( Fig. 3b(ii), c(ii)). Further, we replaced the T 10 tag at 3′-terminal with A 5 , A 10 , and A 15 tags, respectively. It is expected that high-affinity poly (A) tag will preferentially bind to the AuNPs surface, so the G4 sequence will be released to maintain its intact structure. CD spectra of these resulted DNA-AuNP conjugates are tested, which showed that G4 characteristic peaks are restored ( Fig. 3b(iii), c(iii)).
In order to strengthen the understanding of labeling mechanism, and also supply guide for future applications, we employed test strip hybridization assay to evaluate the hybridization iii. iv. properties of the resulted DNA-AuNP conjugates. DNA sequences with varied poly (T) tag lengths were designed for AuNP conjugation and a cDNA can hybridize with these DNA sequences was pre-coated on the test strip. It was found that with the increase of poly (T) tag length, the hybridization efficiency decreased, whether 13 nm or 30 nm AuNP bioprobes (Fig. 4a(i)). These results indicated that increasing the poly (T) length leads to an increase in steric hindrance and a decrease in hybridization efficiency. Subsequently, the same DNA sequences were tagged with a fixed poly (T 7 ) at 5′-terminal and varied poly (A) lengths at 3′-terminal. The resulted DNA-AuNP conjugates were applied to the test strip hybridization and presented almost equal hybridization efficiency when increasing the poly (A) length from A 2 to A 15 (Fig. 4a(ii)). It was also noticed that further increase of poly (A) length to 25 has a reverse effect on hybridization (Fig. 4a(ii)). It is possible that the longer poly (A) tag leads to a decrease in the number of probes conjugated to the AuNPs surface, thus affecting the hybridization. At last, we pre-coated a poly (A 7 ) probe on the test strip to test the above-mentioned DNA-AuNP conjugates. Robust and equal hybridization efficiency can be observed for DNA-AuNP conjugates with varied poly (A) lengths ( Fig. 4a(ii)). These experiments provide strong evidence that poly (T) is not involved in conjugation with AuNPs and is distributed in the outer layer of the SNA structure. This can also be further Intensity (x 10 a.u.) 3 Raman shift (cm -1 ) a i. ii. ii. ii. iii. iii. iii. i. i.
Keep the poly (T 10 ) tag fixed, while increasing the length of poly (C) and poly (A) tags, attached DNA is gradually reduced (Fig. 4b(ii), (iii)).
Mechanism description. It is fascinating that poly (T/U)-tagged DNA/RNA can be loaded on the AuNPs surface at high density by simply heating-dry treatment. Previous study has shown that heating could drive DNA to be adsorbed onto AuNPs or graphene 43,44 . To understand whether this labeling is only induced by heating, some random DNA sequences with poly (A 10 /T 10 /C 10 /G 10 ) tags at 5′-terminal were mixed with AuNPs under heating ( Fig. 5a and Supplementary Fig. 15). As observed, under both 90 and 150°C, the AuNPs aggregated after the centrifugation-resuspension treatment, which show that heating process cannot mediate efficient DNA-AuNP conjugation (Fig. 5a, top panel and Supplementary Fig. 15). Conversely, all poly (T)-tagged DNA sequences could be efficiently attached to AuNPs using heating-dry method (Fig. 5a, below panel). In another test, when AuNPs solution were mixed with six kinds of DNAs and dried at lower temperatures (30 and 90°C), feasible labeling was achieved although with nonuniform labeling efficiency ( Supplementary Fig. 15). The nonuniform labeling efficiency may be due to the structural influence of some DNA sequences, which in turn affects the stability, density, and activity  T10  T15 T20  T25  T30   13 nm AuNPs   T5  T10  T15  T5  T10  T15   13 nm AuNPs  30 AuNP   T30  T25  T20  T10 T15  T5 T6 T7 T8 T9   0 Fig. 15). It is well known that heat can destroy the higher-level structure of nucleic acid strand, thereby stretching them into an uncoiled state. Therefore, in the present work, we tried to use heating to stretch the nucleic acid strands, which can make them to be well bound with AuNPs surface, especially the long-chain/structural nucleic acid strands. In addition, the MW-assisted heating condition can promote the rapid water evaporation in the solution of AuNPs and nucleic acid strands to reach a drying state, which compresses their physical space and increases their local concentrations to achieve rapid labeling, thereby greatly improving the labeling efficiency and density. Further, we found that poly (T 10 )-tagged DNA at the concentration of 1.25 µM is enough for heating-dry labeling ( Supplementary Fig. 16a). Based on the data of DNA loading per AuNP (83 poly (T 10 )-tagged DNA strands each AuNP) and the measured AuNPs concentration after labeling completion, we were surprised that >80% of the DNA strands were utilized in the labeling process. Such a high DNA utilization ratio should be ascribed to the concentration effect under the heating-dry condition. For SH-DNA, a critical DNA concentration of 5 µM is required for successful conjugation ( Supplementary Fig. 16b), which is consistent with the results of about 270 SH-DNA strands attached for each AuNP (Supplementary Fig. 3d). These observations indicated that both evaporation-mediated DNA condensing and heating-mediated DNA stretch cooperatively promote the high-density DNA loading on AuNPs surface (Fig. 5b). Based on the above observations and poly (T) tag orientation experiments, we concluded that the heating-dry method presents the following mechanism: (i) for random DNA/RNA sequence, both 5′-and 3′-terminal or middle region have medium and undifferentiated Au affinity. Under rapid heating-dry condition, flat and cross-linked DNA-AuNP adsorption behavior is dominant. The resulted DNA-AuNP complex is thus unstable against heating and high salt concentration due to the low DNA adsorption density (Fig. 5c); (ii) poly (T/U)-tagged DNA/RNA exhibits an "amphiphilic"-like characteristic, where poly (T/U) tag has low Au affinity and the other end has medium Au affinity (random base composition) or high Au affinity (A/C-rich nucleic acid sequences). Under rapid heating-dry condition, nucleic acid sequences are clustered and medium-and high-affinity regions are preferentially bound to AuNPs, thus the low-affinity poly (T/ U)-contained region is distributed in the outer layer due to the repulsion effect. Such a "standing-up" conformation allows highdensity nucleic acid strands attached thus mediating the formation of high-stable SNA structure (Fig. 5c). It is important to point out that for efficient conjugation the low-affinity tag is essential from all our observations, which is consistent with the previous report 45 . In a test, labeling failure occurred when the Au affinity of poly (T) was enhanced by phosphorothioate (PS) modification. However, addition of modification-free poly (T) tag at the other end, the successful conjugation was restored ( Supplementary Fig. 17). In addition, to investigate the relationship between the nucleic acid strand length and the number of T bases for successful adsorption, we performed molecular dynamics simulation and binding free energy calculation and the results showed that with the same affinity sequence, the poly (T 10 )-DNA strands is more inclined to be adsorbed on the AuNPs surface than that of poly (T 5 )-DNA strands, and more T bases may provide higher driving force to attach on the AuNPs surface ( Supplementary Fig. 18), which can also be seen by naked-eye observation of these two labeling results (Fig. 2b).
MW-assisted heating-dry method enables colorimetric detection of poly (T)-interspaced RCA products. After establishing a fast, low-cost and facile DNA-AuNP construction method, we wondered whether this method could be used for long-stranded DNA-AuNP conjugation. RCA, an isothermal amplification reaction, is performed using circular DNA template and DNA primer to produce long single-stranded DNA (ssDNA) which contains many tandem repeating sequences complementary to circular DNA template 46 . Here, we designed a padlock probe containing a A 40 sequence at non-recognition region and a 20-bp sequence for specific target DNA recognition. After performing ligation and RCA reaction, the resulting long ssDNA products will contain interspaced tandem repeated T 40 sequence, which is potential suitable for AuNP-based labeling using heating-dry method (Fig. 6a). The gel analysis showed that the length of RCA products is over 1000 bp (Fig. 6b). With MW-assisted heating-dry method, we observed that the poly (T)-interspaced RCA products could be effectively attached thus AuNPs presented in a red monodispersed solution, while AuNPs aggregated when mixed with RCA products with random repeated sequence (Fig. 6c), which indicated a selective AuNPs-based attachment with T 40 sequence in RCA products. Based on this observation, we further developed a simple colorimetric single-base mutation diagnostic platform. As shown in Fig. 6d, the padlock probe will be ligated in the present of perfectly matched target sequence thus the RCA reaction could generate long ssDNA strands containing T 40 sequence, while no ligation and RCA amplification occur with single-base mismatched sequence (Fig. 6e). Subsequently, heating-dry method was utilized to colorimetric detection of the RCA products. The solution retained red when the sample with perfectly matched target, while it was colorless when the sample with single-base mismatched sequence or without target due to AuNPs aggregation (Fig. 6f). These results indicated that our developed diagnostic platform based on padlock-dependent ligation, RCA reaction, and MW-assisted heating-dry method enables a new colorimetric gene mutation detection of DNA sequence.
MW-assisted heating-dry method enables construction of long and structured ssRNA-AuNP nanostructure. Conjugation of long and structured RNA on AuNPs is of significance due to a wide range of potential applications in biomedicine and nanobiotechnology 47,48 , which, however, remains a challenge. This is because chemical thiol-modification of long-chain RNA is very expensive and difficult 25,26 . Currently, high-fidelity chemical synthesis of RNA with more than 200 bases is almost impossible to achieve. Here, we challenged to conjugate AuNPs with long and structured RNA using the heating-dry method. To demonstrate this possibility, poly (U)-tagged sgRNA (136 bp) and N gene RNA (N-RNA, 1278 bp) of SARS-CoV-2 were obtained using in vitro T7 transcription based on a poly (A)-tagged DNA template (Fig. 6g). The sequence, structure, and labeling details of sgRNA and N-RNA are shown in Fig. 6h, i, respectively. sgRNA is a guide RNA in Cas9 gene editing tool and N gene RNA is an important functional gene of SARS-CoV-2. Conjugation of these RNAs to AuNPs may give birth to new medical applications, such as nucleic acid drugs and vaccines 30,31 , delivery of gene editor tool 32,33 , due to the excellent cell entry capabilities of AuNPs 12,49 .
As shown in Fig. 6j, after mixture of RNA and AuNPs and a few minutes of heating-dry process, both the absorption spectra and photographs indicated that poly (U)-tagged sgRNA and N-RNA can be successfully functionalized on AuNPs, while it fails to efficiently load sgRNA and N-RNA without poly (U) tag. The measured hydrodynamic diameters of sgRNA-AuNP and N-RNA-AuNP conjugates are about 141 and 255 nm ( Fig. 6k and Supplementary Table 1), respectively, revealing larger RNA strands attached. Next, RNAs were displaced from the AuNPs surface by ME and were subjected to the gel electrophoresis, showing the same map when compared to the original in vitro T7 RNA transcription products (Fig. 6l), indicating that RNA sequence was not damaged during the heating-dry process.
Finally, we demonstrated an application of the resulted RNA-AuNP conjugates by developing a CRISPR/Cas9-based nucleic acid test strip using sgRNA-conjugated AuNPs. Previously, we have developed a test strip strategy by employing Cas9/sgRNA as the recognition element to specifically recognize double-stranded DNA (dsDNA), which is superior to the traditional test strip in terms of simplicity, specificity, and portability 50,51 . This conventional Cas9/sgRNA-based test strip employed a DNA-AuNP probe to hybridize with the loop region of sgRNA for lateral flow detection of virus and pathogenic bacteria. Benefiting from the new developed RNA-AuNP conjugation method, now we can further simplify the previous test strip procedure by direct construction of the Cas9/sgRNA-AuNP probes. We next tested if the resulted Cas9/sgRNA-AuNP probes still possess the recognition and cleavage ability. Results showed that both of Cas9/  sgRNA and Cas9/sgRNA-AuNP probes could maintain the DNA recognition and cleavage ability (Fig. 6m, n and Supplementary  Fig. 19a). It has now been widely accepted that the binding free energy of bases to the AuNPs surface is greater than the pairing between bases 38 . Thus, it is confusing how specific parts of longer DNA/RNA are able to recognize and hybridize to other structures if the entire sequence has some affinity for the AuNPs. We therefore hypothesized that these SNA structures must contain some recognition sequences that exhibit fully erect and free conformational characteristics. The SNA obtained by MWassisted heating-drying method may follow the site-specific adsorption mechanism: during the initial water evaporation process, nucleic acid strands may be randomly attached to the AuNPs surface. Most of the sites on the AuNPs may be occupied by such random adsorption strands, but there are still some narrow gaps. As the water is completely evaporated, the concentration of nucleic acid strands on the surface of the AuNPs increases dramatically, resulting in saturated adsorption. It can be inferred that only a few bases at the end of the nucleic acid strands may be able to participate in saturation adsorption, so that the resulting SNA contains some nucleic acid strands that exhibit fully erect and free conformational characteristics. These free-conformation nucleic acid strands are easily used for biorecognition reaction (Supplementary Fig. 20).
To demonstrate the detection application of this Cas9/sgRNA-AuNP probe, we amplified the B646L gene encoding the viral protein 72 (VP72) of African swine fever virus (ASFV, accession: MN715134.1) with biotinylated primers and designed the corresponding Cas9/sgRNA-AuNP probe to recognize the B646L (VP72) gene amplicons (Supplementary Fig. 19b). Subsequent test strip results showed that the positive sample presents a band with high intensity at test line, while the negative sample does not (Fig. 6o), indicating the successful application of the novel Cas9/sgRNA-AuNPs test strip.
In summary, we invented a fast, general, and cost-effective method to functionalize AuNPs with DNA and RNA using a MW-assisted heating-dry method. The whole labeling process could be completed in minutes due to the rapid water evaporation by only employing a domestic microwave oven (Supplementary Movie). Rapid water evaporation by MW-assisted heating-dry treatment is a convenient and environmentally friendly way to force AuNPs to bind nucleic acid strands efficiently. Because both AuNPs and nucleic acid strands are negatively charged, previously, the binding of AuNPs to nucleic acid strands was regulated by adjusting pH 20,21 , adding salts 17,18 , or organic solvents 25 . Compared to these state-of-the-art labeling methods, MW-assisted heating-dry method exhibits maximum labeling efficiency, excellent sequence generality, and can produce the most stable DNA-AuNP conjugates, representing a new and advantageous construction method for AuNP conjugates (Supplementary Fig. 21). In addition to its confirmed hybridization performance, AuNP-based bioprobes obtained from MW-assisted heating-dry method also exhibit promising new application potential, such as colorimetric DNA single-base mutation analysis and virus DNA detection. Moreover, the heating-dry method can stretch any nucleic acid strand with a complex secondary structure, making the labeling process unaffected by structureresulted affinity sequence shielding, thus this method is universal to almost any sequence. We have successfully demonstrated its ability for attaching the long-chain and structured nucleic acids on AuNPs surface, which is difficult to achieve by routine methods. Therefore, it may stimulate academic interest in branching out into some relatively unexplored areas based on the applications of long-chain DNA/RNA-AuNP conjugates, such as the delivery of gene editing tools, RNA vaccine, and other functional DNA/RNAs.

Methods
Preparation of 13 nm AuNPs. The 13 nm AuNPs were synthesized using citrate reduction method as previously reported 52,53 . In brief, 100 mL of 1 mM HAuCl 4 solution (catalog number: 484385) was added into glass flask and heated to boiling. Then quickly stirred HAuCl 4 solution and added 10 mL of 38.8 mM sodium citrate solution. Keep the above solution boiling and stirred for 15-20 min, finally cooled to room temperature with stirring. The prepared AuNPs were stored at 4°C in the dark.
Preparation of 30 nm and 40 nm AuNPs. The 30 nm and 40 nm AuNPs were synthesized using citrate reduction method as previously reported 54 . In brief, 100 mL of 1 mM HAuCl 4 solution was added into glass flask and heated to boiling. Then quickly stirred HAuCl 4 solution and added 6 mL (5 mL) of 38.8 mM sodium citrate solution for 30 nm (40 nm) AuNPs preparation. Keep the above solution boiling and stirred for 15-20 min, finally cooled to room temperature with stirring. The prepared AuNPs were stored at 4°C in the dark.
Preparation of AuNRs. The AuNRs were synthesized as previously reported 55 . Firstly, 220 µL of 50 mM HAuCl 4 solution was added to 10 mL of 0.2 M cetyltrimethylammonium bromide (CTAB, catalog number: C-3960-53) solution, followed by a slight vortex mixing. When the floccule completely dissolved, 21 µL of 0.1 M AgNO 3 (catalog number: 204390) solution and 212 µL of 0.1 M ascorbic acid were added to the mixture, followed by a slight vortex mixing. Subsequently, 2.5 µL of 1 mM NaBH 4 was added to above mixture and incubated at room temperature for 12 h. The prepared AuNRs were washed two times with water and centrifugated (3100×g) at 15°C for 10 min. Finally, the pellet was resuspended in water. The prepared AuNRs were stored at 4°C in the dark. Modification of AuNPs using low-pH method. For low-pH method 21 , 400 µL of AuNPs solution was mixed with 20 μL of 100 μM poly (A/T/C/G)-tagged DNA probes. Then, 8 μL of 500 mM citrate buffer (pH 3) was added to lower the pH of the solution. After brief vortex mixing, the mixture was incubated at room temperature for 3 min. Next, 12 µL of 1 M HEPES buffer (pH 7.6) was added to the mixture to adjust the pH back to a neutral state. Subsequently, the particles were washed three times with 0.01 M phosphate buffer (0.1 M NaCl, pH 7.4) through centrifugation (12,400×g) at 4°C for 20 min. Finally, the resulted conjugates were resuspended in buffers with different concentrations of NaCl.
Modification of AuNPs using freeze-thaw method. For freeze-thaw method 26,29 , 400 µL of AuNPs solution was mixed with 20 μL of 100 μM poly (A/T/C/G)-tagged DNA probes. The above mixture was frozen at -20°C for 1 h and then thawed at room temperature. After that, the particles were washed three times with 0.01 M phosphate buffer (0.1 M NaCl, pH 7.4) through centrifugation (12,400×g) at 4°C for 20 min. Finally, the resulted conjugates were resuspended in buffers with different concentrations of NaCl.
Modification of AuNPs using butanol extraction method. For butanol extraction method 25 , 100 µL of AuNPs solution was mixed with 5 μL of 100 μM poly (A/T/C/ G)-tagged DNA probes. And then, 900 μL of n-butanol was added to the mixture, followed by a quick vortex mixing. Subsequently, 200 μL of 0.5× TBE (44.5 mM Tris, 1 mM EDTA, 44.5 mM boric acid, pH 8.0) buffer was added to the above solution followed by quick vortex mixing. The mixture was centrifugated at 2000×g for several seconds to facilitate a liquid phase separation. The liquid phase with particles was pipetted and washed three times with 0.01 M phosphate buffer (0.1 M NaCl, pH 7.4) through centrifugation (12,400×g) at 4°C for 20 min. Finally, the particles were resuspended in buffers with different concentrations of NaCl.
Modification of AuNPs with DNA/RNA using MW-assisted heating-dry method. In total, 400 µL of AuNPs solution was mixed with 20 μL of 100 μM SH-DNA or poly (T)-DNA or poly (U)-RNA probes in 5 mL glass bottle. Then, put the glass bottle into a domestic microwave oven (the microwave input power is 1150 W and the microwave output power is 700 W) and heated at middle-high model for 3 min, followed by resuspension with ddH 2 O or 0.01 M phosphate buffer (0.1 M NaCl, pH 7.4). After that, the particles were washed with 0.01 M phosphate buffer (0.3 M NaCl, pH 7.4) and centrifugated (12,400×g) at 15°C for 20 min. Finally, the pellet was resuspended in buffers with different NaCl concentrations, and the resulted nucleic acid-AuNP conjugates were stored at 4°C in the dark or used for subsequent experiment. Fluorescence quantification of the surface density of DNA-AuNP conjugates. Poly (T)-tagged DNA used in these studies were labeled with FAM at 3′-terminal (Supplementary Table 2). FAM-labeled DNA were first absorbed to AuNPs surface following the protocol as above described. The FAM-DNA-AuNP conjugates were resuspended in 0.01 M phosphate buffer (0.3 M NaCl, pH 7.4) or 10 mM HEPES buffer. The surface density of the DNA-AuNP conjugate was quantitated according to the published protocol 56 . First, β-mercaptoethanol (ME, catalog number: 63689) was added (20 mM final concentration) to the FAM-labeled DNA-AuNPs solution and incubated overnight with shaking at room temperature. Then, the mixture was centrifugated at 15°C at 12,400×g for 20 min and the supernatant was collected. The released FAM-DNA probes were in the supernatant and the fluorescence was measured by SpectraMax iD5 Multimode Microplate Reader (Molecular Devices, CA, USA). The fluorescence intensity was converted to the molar concentration of DNA probes by interpolation from a standard linear calibration curve that was prepared with known concentrations of same FAM-DNA probes at same buffer condition and ME concentration. AuNPs concentration was determined via the absorption of UV-vis spectra and finally the number of DNA on each AuNP could be calculated.
Surface-enhanced Raman spectroscopy (SERS) analysis. The DNA-AuNP conjugates were obtained using MW-based heating-drying method as above described and resuspended in 0.01 M phosphate buffer (0.3 M NaCl, pH 7.4). 10 μL of 10 μM ROX-labeled DNA probes complementary to DNA-AuNP conjugates were firstly added to 200 μL DNA-AuNP solution. After 2 h incubation at 37°C for efficient hybridization, the particles were washed three time with 0.01 M phosphate buffer (0.3 M NaCl, pH 7.4) and centrifugated at 15°C at 12,400×g for 20 min. Then the supernatant was discarded and the pellet (20 μL) was collected for SERS measurements. For SERS analysis, the 20 μL pellet was dropped on the Si substrate and the SERS measurement was performed using Renishaw inVia microspectrometer (Derbyshire, England). A He-Ne laser operating at λ = 633 nm was used as the excitation source with a laser power of approximately 10 mW. All the Raman spectra reported in this work were collected for exposure times of 10 s in the range from 1000 to 1800 cm -1 .
Circular dichroism measurements. The G4-DNA-AuNP conjugates were obtained using MW-based heating-drying method as above described. The G4-DNA was diluted in 50 mM Tris-HCl (100 mM KCl, pH 8.5). The G4-DNA-AuNP conjugates were washed two times with 50 mM Tris-HCl (100 mM KCl, pH 8.5) and centrifugated at 15°C at 12,400×g for 20 min, followed by resuspending in 50 mM Tris-HCl (100 mM KCl, pH 8.5). The CD spectra of G4-DNA and G4-DNA-AuNP conjugates were collected by a JASCOJ-810 spectropolarimeter (Tokyo, Japan), of which the lamp was always kept under dry purified nitrogen condition during measurements. Three times of scanning from 220 to 310 nm with the scan rate of 100 nm min -1 was performed and averaged. The background signal of the 50 mM Tris-HCl (100 mM KCl, pH 8.5) buffer was subtracted from the CD data.
Synthesis of sgRNA. sgRNA and poly (U)-tagged sgRNA have been transcribed in vitro. The DNA template with or without poly (T) sequence was obtained as previous report by a fill-in PCR method 57 . 500 μL transcription reactions were performed with 0.5 mM NTPs, 250 U T7 RNA polymerase (catalog number: M0251L), 50 U recombinant RNase inhibitor (RRI, catalog number: 2313A), and 500 ng DNA template. The mixture was incubated for 12 h at 37°C. Then, DNase I (catalog number: M0303S) was added to digest the DNA template and the RNA transcripts were purified by RNA Clean & Concentrator Kits. The concentration of RNA products was determined by Nanodrop 2000.
Synthesis of long RNA of N gene of SARS-CoV-2. Poly (U)-tagged long RNA (1278 bp) of N gene of SARS-CoV-2 has been transcribed in vitro. The DNA template with or without poly (T) sequence was obtained by a RT-PCR of SARS-CoV-2 RNA standards. The primers were designed according to the sequence of N gene (Supplementary Table 3). The RT-PCR thermal cycling program: reverser transcription at 50°C for 45 min, pre-denaturation of cDNA at 94°C for 2 min, followed by 40 cycles of amplification at 94°C for 30 s, 55°C for 30 s, 72°C for 2 min, finally incubated at 72°C for 10 min. The PCR products containing T7 promoter was verified by gel electrophoresis and the template was collected using gel extraction kit. The purified DNA templates were then used for in vitro transcription system as above described. RNA transcripts were purified and concentrated by RNA Clean & Concentrator Kits (catalog number: R1017) and the resulted RNA concentration was determined by Nanodrop 2000.
Preparation of the test strip device. The test trip device consists of sample pad, conjugate pad, nitrocellulose (NC) filter membrane, test line, control line, absorbent pad, and adhesive pad. The sample pad was firstly immersed into a 50 mM Tris-HCl buffer (0.15 M NaCl, 0.25% Triton X-100, pH 8.0) and then dried at 37°C for 2 h. Streptavidin was embedded in the test line and streptavidin-biotinylated DNA probes was embedded in the control line. The intervals between two adjacent lines were 6 mm. Then the NC membrane was dried at 37°C for 1 h. Subsequently, the sample pad, conjugate pad, and absorbent pad were attached to the adhesive pad with 2 mm overlap. Finally, the prepared test strip was cut with a width of 4 mm and stored at 4°C.
Test strip assay to evaluate the hybridization efficiency of DNA-AuNP conjugates. The DNA-AuNP conjugates were obtained using MW-assisted heating-dry method. The 200 μL DNA-AuNP solution was washed twice with 0.01 M phosphate buffer (0.1 M NaCl, pH 7.4) by centrifugation at 15°C at 12,400×g for 20 min. Subsequently, the pellet was resuspended in 200 μL resuspension buffer (20 mM Na 3 PO 4 , 5% BSA, 0.25% Tween-20, and 10% sucrose). In all, 10 μL resuspended DNA-AuNP conjugates were mixed with 10 μL running buffer (4× SSC, 0.05% Tween-20 (v/v), 1× PBS, and 1% BSA), and then the above mixture was added to the sample pad of the test trip device. Finally, 100 μL running buffer was added on the sample pad to wash the test strip, and the test strip was photographed after 5 min.
Padlock-based ligation and RCA reaction. Ligation of the linear padlock probe was performed in a 10 μL reaction solution containing 10 mM Tris-HCl buffer (50 mM KCl, 1.5 mM MgCl 2 , pH 8.9), 500 nM padlock probe, 500 nM matched target DNA, 10 U T4 ligase. The above solution was incubated at 37°C for 30 min. Subsequently, a total of 20 μL RCA reaction system containing 10 mM Tris-HCl buffer (50 mM KCl, 1.5 mM MgCl 2 , pH 8.9), 1.25 mM dNTP mix, 3.2 U Bst 2.0 WarmStart DNA polymerase, and 10 μL ligation products, was perform at 60°C for 60 min. The RCA products were analyzed by gel electrophoresis.
Loading of AuNPs with RCA products using MW-assisted heating-dry method. 400 µL of AuNPs solution was mixed with 20 μL RCA products in 5 mL glass bottle. Then, put the glass bottle into a microwave oven and heated at middlehigh model for 3 min, followed by resuspension with ddH 2 O. After that, the solution was washed with 2 mM Tris-HCl buffer (10 mM KCl, 0.3 mM MgCl 2 , pH 8.9) and then washed with water by centrifugation at 15°C at 12,400×g for 20 min. Finally, the pellet was resuspended in 1 M NaCl solution.
Modification of AuNPs with poly (U)-tagged sgRNA and poly (U)-tagged long RNA of N gene (1278 bp) of SARS-CoV-2. In all, 400 µL of AuNPs solution was mixed with 10 μL of 50 μM poly (U)-tagged sgRNA or 320 μg poly (U)-tagged long RNA in 5 mL glass bottle. Then, put the glass bottle into microwave oven and heated at middle-high model for 3 min, followed by resuspension with ddH 2 O. After that, the solution was firstly washed with 0.02 M phosphate buffer (0.3 M NaCl, pH 7.4) and then washed with water by centrifugation at 15°C at 12,400×g for 20 min. Finally, the pellet was resuspended in 1 M NaCl solution.
Amplification of B646L (VP72) gene of ASFV by PCR. The PCR reaction of the B646L (VP72) gene was performed using a Premix Ex Taq™ II. PCR primers were designed according to the B646L (VP72) conserved region of African swine fever virus (ASFV) (Supplementary Table 3). A total of 40 μL amplification system contained 500 nM forward and reverse primer, the genome samples and 1× PCR reagent. The PCR thermal cycling program was: 95°C for 30 s, followed by 40 cycles of amplification at 95°C for 5 s, 59°C for 30 s, a final extension at 72°C for 10 min and a 10°C hold. The PCR products were analyzed by gel electrophoresis.