LUPP-LAMP: one-pot multiplatform universal LAMP assay for simultaneous detection and genotyping of DENV and ZIKV

Abstract

Dengue virus (DENV) and Zika virus (ZIKV) are mosquito-borne viruses that cause severe health problems upon infection, necessitating timely and accurate diagnostic testing for effective prevention and control of their transmission. In this study, a novel multiplex nucleic acids detection method was introduced based on ligation-based universal primer and probe loop-mediated isothermal amplification (LUPP-LAMP) for the simultaneous detection and genotyping of DENV and ZIKV. By employing ligation of universal primers and probes using splint DNA ligase, LUPP-LAMP simplifies primer design, eliminates primer dimer formation, and achieves high sensitivity and specificity at low temperatures, with a detection limit of 10 copies/reaction. The robustness of this strategy allows for precise differentiation even at low concentrations and in cases of co-infection, which is crucial for addressing early diagnostic challenges due to similar symptoms. LUPP-LAMP is further adapted for point-of-care testing (POCT) by integrating lateral flow test strips and fluorescence imaging, offering rapid and user-friendly results in resource-limited settings. This method is also applied to DENV genotyping (serotypes 1–4) using melting curve analysis (MCA) in cases of co-infection, aiding in epidemiological surveillance and intervention strategies. In summary, LUPP-LAMP represents a significant advancement in the detection and genotyping of DENV and ZIKV, offering a versatile tool for laboratory and POCT applications, thereby enhancing public health outcomes in the control of mosquito-borne diseases.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03581-x.

Introduction

Dengue and Zika fever are mosquito-borne infectious diseases caused by Dengue virus (DENV) and Zika virus (ZIKV). They are transmitted by Aedes aegypti mosquitoes and primarily prevalent in impoverished tropical and subtropical regions [1, 2]. In the early stages, both diseases exhibit similar symptoms such as fever, nausea, and joint pain, making diagnosis difficult. According to the World Health Organization, approximately 100–400 million people are infected with the DENV annually worldwide. In recent years, the incidence of this disease has increased alarmingly. Since 2015, the outbreak of the ZIKV has become a significant global health issue, raising serious alarms due to the potential risk associated with Guillain-Barré syndrome, particularly in newborn infants [3]. The concurrent outbreaks of DENV and ZIKV epidemics, coupled with the lack of vaccines and specific treatments, underscores the urgent need for early diagnostic methods to differentiate these two infections. Such capability is crucial for updating epidemiological data, delineating the geographical distribution of these diseases, and implementing more targeted prevention and control interventions [4–6].

Reverse transcription polymerase chain reaction (RT-PCR) is the gold standard method for diagnosing infections caused by DENV and ZIKV [7, 8]. However, its application in resource-limited settings is often hindered by the reliance for expensive equipment, rapid continuous temperature cycling, and the need for strict four-zone molecular laboratory requirements. Insufficient laboratory infrastructure frequently delays the timely acquisition of accurate diagnostic results, thereby causing delays in patient treatment. Therefore, developing rapid, accurate, and user-friendly nucleic acid detection technologies for simultaneous testing of DENV and ZIKV provides significant advantages in epidemic monitoring and prevention. It also offers robust protection against mosquito-borne infectious diseases. Latest advancements in isothermal nucleic acid amplification technology (INAT) provide rapid, efficient, and temperature-stable amplification, facilitating convenient on-site pathogen detection [9–11]. Notable techniques in this field encompass loop-mediated isothermal amplification (LAMP) [12], nucleic acid sequence-based amplification (NASBA) [13, 14], recombinase polymerase amplification (RPA) [15, 16], and other innovative platforms.

To further enhance the sensitivity and specificity of the method, cascade signal amplification techniques by coupling INAT with other methods such as DNAzyme [17, 18], rolling cycle amplification (RCA) [19, 20], clustered regularly interspaced short palindromic repeats/CRISPR-associated protein (CRISPR/Cas) based technologies [21, 22], exonuclease III-assisted cascade signal amplification [23, 24] and catalytic hairpin assembly [25, 26] have been widely developed. However, these technologies still require complex nucleic acid sequence design or stringent reaction conditions, which cannot be integrated with isothermal amplification reactions in a single tube, thereby limiting their widespread application. Additionally, the CRISPR/Cas system needs single guide RNA (sgRNA) with directional functionality, further increasing the method's cost and complexity [27]. Therefore, developing a new strategy in one pot for simultaneous detection of mosquito-borne infectious DENV and ZIKV remains challenging.

Among these techniques, LAMP stands out as the most extensively utilized method for rapid nucleic acid detection of pathogenic microorganisms. LAMP demonstrates high amplification efficiency and enhanced specificity through multiple DNA primers, making it an ideal option for point-of-care testing (POCT) platforms [28–30]. However, challenges faced by LAMP in complicated primer and probe design, along with its elevated reaction temperature (60–65 ℃), impede its full realization in POCT settings [31]. Additionally, LAMP assay necessitates the combined utilization of external and longer internal primer pairs, increasing the risk of nonspecific amplification issues such as primer dimers, which become more prominent across multiple LAMP reactions [32, 33]. Moreover, the strand replacement DNA polymerase utilized in LAMP lacks the probe hydrolysis capability seen with TaqMan probes in RT-PCR, further compromising specificity and leading to potential false-positive results. Therefore, developing new probe technology and addressing primer dimerization in multiplex LAMP assays are crucial to meet the growing demands of nucleic acid detection [34, 35].

The classic LAMP reaction typically operates within a temperature range of 60–65 ℃, which allows double-stranded DNA to remain in a dynamic and unstable state, promoting continuous denaturation and annealing [36]. During the initial stages, strand invasion serves as a rate-limiting step critical for the effective progression of the reaction. Once loop structures form, primers can efficiently bind to loop regions, facilitating extension and strand displacement reactions without the need for excessively high temperatures. To address the challenges and bottlenecks mentioned above, this work developed a ligation-based universal primer and probe (LUPP) mediated multiplex LAMP achieved with splint DNA ligase, enabling simultaneous detection and genotyping of DENV and ZIKV in a single tube at ambient temperature. The utilization of universal primers diminishes the need for multiple LAMP primer sets, overcoming the technical challenges linked with traditional multiplex LAMP, such as frequent primer dimerization and inconsistent amplification efficiency. Additionally, the developed technology can be employed in conjunction with lateral flow test strips, real-time fluorescence, or portable fluorescence imaging devices with potential clinical applications.

Experimental section

Materials and apparatus

Primers without modifications were PAGE purified and synthesized by Tsingke Biotechnology Co. (Haikou, China). Fluorescent, quencher, digoxin and biotin modified probes were HPLC-CE purified and purchased from Sangon Biotech (Shanghai, China). All primers and probes’ sequences are listed in Table S1-S4 and DEPC-treated water with individual packages (Bioligo, China) were used for all the experiments. The principle of enzyme selection is that the optimal reaction temperature of the enzyme ranges from 25 to 37℃, ensuring that our reactions occur at ambient temperature, facilitating the implementation at the point-of-need. The main reagents for LAMP were 1*isothermal amplification buffer (20 mM Tris–HCl, 10 mM (NH₄)₂SO₄, 50 mM KCl, 2 mM MgSO4, and 0.1% Tween 20; pH 8.8 at 25℃), dNTP/dUTP Mix (each dNTPs at a final concentration of 2 mM and dUTP of 4 mM), Thermolabile Uracil-DNA Glycosylase (UNG), Splint R DNA ligase and Ligase Reaction Buffer, Adenosine 5'-Triphosphate (ATP), Klenow DNA polymerase exo-, Bsu DNA polymerase large fragment, phi29 DNA polymerase, DNA Polymerase I, RNase H2, NEBuffer™ 2, phi29 DNA Polymerase Reaction Buffer, ThermoPol Reaction Buffer were obtained from NEB (Beijing, China). Fluorescent emission spectra were performed on F-4700 Fluorescence Spectrophotometer (Hitachi, Japan). Capillary electrophoresis (CE) and Sanger sequencing were performed on an ABI Prism 3730 Genetic Analyzer (Applied Biosystems, USA). Images were acquired using a ChemiDoc Touch MP imaging system (BioRad, USA) with its built-in 485 nm (λ ex) and 643 nm (λ ex) channels respectively. To ensure the reliability, all experiments were independently repeated three times.

Primer and probe design

To identify the most conserved regions within the genomes of DENV virus (DENV-1 (ACCESSION Number: OR029755, KM403578, NC_001477 and KX620455), DENV-2 (ACCESSION Number: NC_001474, OR125606, KU517847, MW512362), DENV-3 (ACCESSION Number: NC_001475, OR418423, OK605764, OP410995), DENV-4 (ACCESSION Number: KC762695, NC_002640, KY921909, MG601754) and ZIKV (NC_012532, NC_035889, MW680970, OM964565), genome sequences from China and neighboring countries, which were accessible during the design phase, were downloaded from the National Center for Biotechnology Information (NCBI) and utilized for assay design. Multiple sequence alignments encompassing all genomic sequences were generated using Clustal in SnapGene to obtain their conserved sequences. The target sequences for primer design were set at 28 nt in length. The linker on both left and right sides is designed to be 14 nt complement to target sequence and boost the activity of Splint DNA ligase.

For PCR, the linker on the left side is configured to be a 14 nt complement to the target sequence, along with a 20 nt PCR primer binding site. Conversely, on the right side, the linker is structured to be a 14 nt complement to the target sequence, paired with distinct 20 nt probe/primer binding sites. The primer binding sites consist of universal sequences adaptable for the detection of ZIKV and DENV. The 20 nt sequences for primer–probe binding sites were generated using Python code to derive random sequences (refer to support information). For LAMP, the two stem-loop primers on the left and right sides are employed to create the double stem-loop LAMP substrates. The structural domain of the stem-loop primer on the left side includes a 14 nt complement to the target sequence, along with F1C, F2C, the probe binding domain, and F1. Similarly, the structural domain of the stem-loop primer on the right side comprises a 14 nt complement to the target sequence, B1, B2, and B1C. Various domain lengths have been designed and optimized to achieve optimal performance. Melting temperatures (Tm) were confirmed using Primer Express v3.0 (Applied Biosystems, Life Technologies). Theoretical specificity of the system was assessed by performing a BLAST search against the NCBI nucleotide database.

Preparation of the simulated DENV and ZIKV samples

Conserved sequences representing universal DENV and ZIKV were synthesized and subsequently cloned into the pUC57 plasmid. RNA targets were then generated via in vitro transcription using the HiScribe T7 High-Yield RNA Synthesis Kit. Briefly, RNA transcription templates containing T7 promoter were amplified from plasmid with specific primers using the PCR Master Mix according to manufacturer’s instruction, followed by purification using a gel extraction kit. In vitro transcription was performed in 20 μL reaction volume, according to a standard RNA synthesis protocol, at 37 ℃ for 16 h. Finally, the RNA transcripts were treated with DNase I to eliminate transcription templates and purified using an RNA cleanup kit. Varying concentrations of the above RNA were incubated with a fixed concentration of normal human serum to simulated ZIKV and DENV virus clinical samples. Viral RNA was extracted from 100 μL serum specimens by using commercial RNA Extraction Kits (QIAGEN, Shanghai) according to the manufacturer's instructions.

LUPP-PCR and one-pot LUPP-LAMP

The LUPP-PCR reaction mixture comprises with a 50 μL 1 × Taq master mix including Splint R Ligase (25U), L and R linkers (200 pM), universal primers and probes (200 nM). PCR were conducted with initial denaturation at 95 °C for 3 min, followed by 45 cycles of amplification consisting of 95 °C for 15 s (denaturation step), 60 °C for 30 s. The one pot LUPP-LAMP comprises with a 50 μL 1 × isothermal amplification buffer reaction system, including SLP-L (200 pM), SLP-R (200 pM), probe (100 nM), RNase H2 (0.5 mU/μL) and Splint R Ligase (25U), Bsu DNA polymerase large fragment (0.2 U/μL), dNTPs (2 mM), dUTPs (8 mM), UNG (50 mU/μL), ATP (1 mM), FIP-8/8 and BIP-8/8 (50 nM).

Procedures of LAMP only, two-step method and three-step method

In the ligation-LAMP method, a mixture containing DG-SLP-L (200 pM), DG-SLP-R (200 pM), DG-target, and Splint R Ligase (25U) was prepared in 1 × SplintR Ligase Reaction Buffer, with a total volume of 20 mL. This mixture was then incubated at 25 °C for 30 min, followed by heat inactivation at 65 °C for 10 min. Then, 2 mL of the ligation product was combined with a 50 mL 1 × isothermal amplification buffer reaction system, comprising Bsu DNA polymerase large fragment (0.2 U/mL), dNTPs (2 mM), FIP-8/8 and BIP-8/8 (50 nM), along with 1 mL of 20 × Eva Green. LAMP amplification was carried out at 37 °C for 45 min using the F-7000 Fluorescence Spectrophotometer device, with monitoring conducted at 1-min intervals. In the two-step method (Ligation LAMP + RNase H2), a 20 mL mixture comprising DG-SLP-L (200 pM), DG-SLP-R (200 pM), DG-target, and Splint R Ligase (25U), Bsu DNA polymerase large fragment (0.2 U/mL), dNTPs (2 mM), FIP-8/8, and BIP-8/8 (50 nM) was prepared. LAMP amplification was conducted at 37 °C for 45 min, followed by heat inactivation at 65 °C for 10 min. Subsequently, RNase H2 (0.5 mU/mL) and DG-probe2 (100 nM) were introduced into the system and incubated at 37 °C for 45 min using the F-7000 Fluorescence Spectrophotometer device, with monitoring conducted at 1-min intervals. In the three-step method, the ligation and LAMP steps were identical to those of the LAMP-only method. Post-LAMP, the enzyme was deactivated at 65 °C for 10 min, after which RNase H2 (0.5 mU/mL) and DG-probe2 (100 nM) were added.

Lateral flow assays and fluorescence imaging

Lateral flow nucleic acid test strips are composed of absorbent pad, interpretation area, and binding pad. Within the binding pad, Anti-biotin conjugated gold nanoparticles are incorporated, while primary antibody is embedded in the C line, and the Anti FAM and digoxin are situated in the T1 and T2 line within the interpretation area. In the lateral flow assay (LFA), universal DENV, ZIKV primers are tagged with biotin and FAM/digoxin at the 5' ends, respectively. The LAMP mixtures were diluted to a final volume of 100 μL and immediately immersed in the strips. After a 10-min reaction, the results were observed visually by assessing the color changes of the two test lines. Subsequently, the strip was photographed using a smartphone and the images were processed using Image J software to determine the optical density of the test bands.

Results and discussion

Working principle

Splint DNA ligase is a novel enzyme capable of catalyzing the ligation of complementary DNA strands in an RNA-templated manner, making it highly promising for RNA detection [37, 38]. The reaction principle of the multiplex LAMP mediated by universal primers and probes based on splint DNA ligase is illustrated in Scheme 1A. In the presence of single-stranded DENV or ZIKV genomic RNA, Splint DNA ligase catalyzes ligation of the left and right connectors, forming a double-loop substrate. This formation process bypasses the initial rate-limiting step in traditional LAMP, where the double-loop primers need to be generated through displacement of external primer. This innovative strategic design enables the reaction to occur at a lower temperature. Subsequently, continuous polymerization and chain displacement reactions take place between the universal primers FIP and BIP, producing a large amount of cauliflower-like LAMP amplification products, thereby achieving effective amplification of the target gene. It is worth emphasizing that for real-time signal monitoring, we adopted RNase H2-mediated signal output technology (Scheme 1B). The designed dual-labeled DNA probes containing RNA bases can fully complement and pair with the target gene sequence. Utilizing the unique property of RNase H2, which can specifically hydrolyze RNA in DNA-RNA hybrids but has no digestion function for phosphodiester bonds in single-stranded or double-stranded DNA or RNA, allows continuous real-time monitoring of the amplification process and assists in detecting the endpoints for single-tube detection of ZIKV and DENV. Compared with Cas enzyme-based systems for signal amplification, the RNase H2 mediated approach simplifies the reaction setup by eliminating the need for additional components such as Cas enzymes and crRNAs, thereby reducing system complexity and overall cost. Furthermore, the combination with nucleic acid paper strip technology has significantly enhanced the rapid detection in a POCT manner (Scheme 1C).

Scheme 1.

Scheme illustration of the LUPP-LAMP process at ambient temperature. A Overview of the LUPP-LAMP reaction. B Fluorescence signal generated by RNase H2-mediated cleavage of the RNA-modified dual-labeled probe. C Detection of the RNase H2 cleavage product using a lateral flow strip

Construction and analytical performance of LUPP-PCR

PCR is the most common and robust detection system, thus we first chose to use the PCR system to develop and evaluate the feasibility and detection performance of LUPP. Figure 1A describes the operational concept of LUPP-PCR in detecting single-strand viral RNA target, with primers and probe connection points located at both ends of the linker. During the entire PCR amplification process, the 5'-3' exonuclease activity of Taq polymerase guides the cleavage of the TaqMan probe, leading to the continuous accumulation of fluorescence signals. Sequence conservation analysis was conducted primarily using Clustal Omega for multiple sequence alignment, obtaining conserved sequences of approximately 28 nucleotides for DENV and ZIKV (Figure S1 and S2). To confirm the ligation of the left and right connectors by Splint DNA ligase in the presence of the target RNA single strand, we performed Sanger sequencing, capillary electrophoresis (CE), and polyacrylamide gel electrophoresis (PAGE) on the ligated products. For DENV RNA target as example, results of Sanger sequencing demonstrated effective ligation of the linker sequences on both ends (Fig. 1B). The sequencing results for the ZIKV target are shown in Figure S3. Its sequence was completely consistent with the theoretic spectra of DENV-left linker and DENV-right linker. CE (Fig. 1C, S4) and PAGE (Fig. 1D) results showed a ligation product of approximately 89 bp, further verifying the successful connection of the left and right linkers.

Fig. 1.

A Working principle of splint DNA ligase connecting universal primers and probes. B Sanger sequencing result of DENV target. Shown primer, probe, DENV target binding domains. C CE results of LUPP-PCR product for DENV RNA target. D 12% native PAGE of DENV target (1), ZIKV target (2), mixed DENV and ZIKV (3), and negative control (4). E Amplification curves of LUPP-PCR for DENV and ZIKV. The concentrations are range from 1E1 to 1E6 copies/reaction. F Standard curve of the Ct values at varying DENV and ZIKV concentrations. G LOD confirmation for DENV and ZIKV at 10 copies/reaction with a 95% confidence interval. H Ct values of simulated 24 clinical samples in total

The experimental parameters, including connector concentration, ligation time, and ligase concentration, were optimized to achieve best Ct value for LUPP-PCR. After experimental optimization, the optimal reaction parameters were determined to be 200 pM for either the left or right connector, a 10-min ligation time and 20 U/reaction ligase concentration (Figure S5). Next, the established LUPP-PCR system was used to evaluate the analytical performance for universal DENV (FAM channel) and ZIKV (ROX channel) virus. The universal primers used in LUPP-PCR suppress the formation of primer-dimers, allowing the simultaneous detection of multiple targets in a single tube. The amplification results for DENV/ZIKV in two fluorescent channels shown in Fig. 1E indicate that as the target concentration increases from 1E1 to 1E6 copies/reaction, the required Ct value decreases accordingly. Moreover, the Ct values for DENV and ZIKV are very close, indicating similar amplification efficiencies for both viruses. This similarity can be attributed to the design of universal primers, which helps reduce amplification imbalance caused by inconsistent primer amplification efficiency. Figure 1F highlights a strong linear correlation (R² = 0.99) between the logarithmic concentration of DENV/ZIKV targets and the Ct values. As shown in Fig. 1G, the reliable detection limit is calculated at 10 copies/reaction with 95% confidence interval, comparable to individual reaction system (Figure S6). This excellent limit of detection (LOD) performance is attributed to the use of universal primers to amplify multiple targets, thereby eliminating amplification imbalance. Co-infection of DENV and ZIKV is common in clinical samples. Using dual infection of DENV and ZIKV as an example, we investigated the ability of the constructed LUPP-PCR system to differentiate between the two viruses when they are simultaneously present in the same reaction. By varying their relative proportions, the results showed that even when the virus ratio in mixed infection was 1000:1, the system could still detect viral loads as low as 0.1% (Figure S7). These data indicate that our LUPP-PCR method can detect mixed infections over a wide range. We further validated the method’s potential application in clinical samples. Figure 1H summarizes the Ct values for the amplification of 24 clinical samples with DENV, ZIKV, or mixed infections (amplification curves are shown in Figure S8). All results were consistent with the preset simulated sample results, further confirming the superior analytical performance of the constructed LUPP-PCR system.

Construction and analytical performance of warm-start LUPP-LAMP

In this study, we designed a novel LAMP system capable of operating at room temperature, aiming to improve the efficiency and versatility of LAMP, thereby expanding its potential applications across various fields. Figure 2A illustrates the ligation-based LAMP process at ambient temperature, focusing on the formation of the dumbbell-shaped stem-loop primers (SLP). The SLP consists of five distinct regions: the left loop (blue, LL), the left stem (red, LS), the central region (gray, C), the right stem (yellow, RS), and the right loop (green, RL). The secondary structures of DG-L, DG-R linker and SLP predicted using the NUPACK online software are shown in Figure S9. The strand displacement activity of Bsu DNA polymerase is utilized to initiate DNA synthesis. This process is facilitated by specially designed universal primers which bind to SLP loop domains. These bindings trigger the amplification of repeated cycles by extending the loop and promoting subsequent annealing of additional primers. As a result, long DNA products are formed at ambient temperature, containing repetitive sequences of the SLP sequence, connected through single-stranded loop regions in long tandem arrays. However, the lack of 5'-3' exonuclease activity in the Bsu polymerases used in this work prevents the utilization of hydrolysis fluorescent probes. Therefore, we considered using the previously established RNase H2-based probe for monitoring, which releases significantly enhanced fluorescence signals upon hybridization and cleavage.

Fig. 2.

A Structure of SLP and simplified steps of LAMP. Optimization of reaction conditions including (B) TTR values according to the different combinations of DNA polymerase and reaction buffers. 1, 2, 3, and 4 indicate Klenow DNA polymerase exo-, Bsu DNA polymerase large fragment, DNA Polymerase I Klenow Fragment, phi29 DNA polymerase, respectively; while A, B, C and D indicate isothermal amplification buffer, NEBuffer™ 2, phi29 DNA Polymerase Reaction Buffer, ThermoPol Reaction Buffer. The LAMP reactions were carried out at 37 °C. C TTR values according to the different concentrations of Mg²⁺ and dNTPs. TTR values according to the different react temperature (D) and different concentrations of Tween 20 (E)

Methodical exploration of diverse pivotal reaction parameters was optimized. First, we evaluated the performance of different DNA polymerases and reaction buffers in promoting low-temperature LAMP. Specifically, we assessed the amplification activity of DNA polymerases possessing strand displacement activity in different reaction buffers, including Klenow DNA polymerase exo-, Bsu DNA polymerase large fragment, DNA Polymerase I Klenow Fragment and phi29 DNA polymerase (Fig. 2B). The amplification process proved ineffective when utilizing phi29 DNA polymerase, despite the integration of all four buffers. Meanwhile, changes in the TTR values clearly indicated that Bsu DNA polymerase large fragment exhibited the highest amplification efficiency, especially when used in the isothermal amplification buffer. Therefore, Bsu DNA polymerase in the isothermal amplification buffer will be used for subsequent LAMP experiments at ambient temperature. Next, we explored the effects of magnesiumion (Mg²⁺) and dNTP concentrations (Fig. 2C). To ensure high-specificity isothermal amplification, we ultimately selected 0.2 mM dNTPs and 2.5 mM Mg²⁺ for the LAMP assay. The developed method also exhibited excellent analytical performance within the temperature range of 20 to 40℃, demonstrating broad temperature robustness (Fig. 2D). This makes it more suitable for POCT that requires minimal temperature control. This advantage allows the proposed strategy to operate effectively in both indoor and outdoor environments. Tween 20 plays several key roles, including enhancing nucleic acid solubility, preventing non-specific binding, improving reaction kinetics, and stabilizing enzymes [39, 40]. In the developed LAMP detection system, 1% Tween 20 is considered the optimal concentration as it facilitates the fastest amplification rate (Fig. 2E). Subsequently, the optimal concentration of FIP and BIP, as well as probe were set to 100 nM. The lengths of FIP-8/8, BIP-8/8 and the 15 nt probe in length were established as the optimal conditions (Figure S10).

To directly assess the impact of dual activities of DNA ligase, polymerase and RNase H2 on the tri-enzyme reaction network, we compared DENV RNA detection under the same reaction conditions using: Two-step tandem ligation-assisted LAMP (Fig. 3B); Two-step tandem ligation-LAMP and RNase H2 hydrolysis (Fig. 3C); Three-step ligation, LAMP and RNase H2 hydrolysis (Fig. 3D); All-in -one-pot assays (Fig. 3E). The conventional assay, composed of two sequential reactions of ligation and LAMP affords a high LOD of 1*10E3 copies/mL for DENV. This could be attributed to the potential generation of non-specific amplification products by the LAMP reaction and the non-specific binding affinity of Eva Green (Fig. 3B). The tandem combination of ligation and LAMP amplification with RNase H2-based signal amplification significantly improved the sensitivity for detecting DENV below 1*10E3 copies/mL, contributed to continuous RNase H2 hydrolysis (Fig. 3D). Based on this three-step tandem assay, we tested a simplified version that combining ligation and LAMP in one pre-amplification reaction followed by an RNase H2-powered fluorogenic readout. The two-step LAMP-RNase H2 approach, while simplifying the workflow, slightly improved detection sensitivity and speed (Fig. 3C). This may be due to the reaction buffer being suboptimal for others enzymes, affecting the overall system. In contrast, our one-pot assay overcame this challenge, enabling sensitive detection as low as 1E2 copies under optimized conditions, notably outperforming the two- or three-step assays (Fig. 3E). The one-pot LUPP-LAMP assay was estimated to yield 1.62*10E8 ~ 1.1*10E9 fold amplification with 100 copies and 1000 copies of DENV or ZIKV input (Figure S11).

Fig. 3.

A Scheme illustration of RNase H2 mediated probe hybridization and hydrolysis. Comparison of the kinetics and detection sensitivity of (B) Two-step tandem ligation-assisted LAMP; (C) Two-step tandem ligation-LAMP and RNase H2 hydrolysis; (D) Three-step ligation, LAMP and RNase H2 hydrolysis; (E) All-in -one-pot assays

Analytical performance of RNase H2 probe assistant LUPP-LAMP

To determine the analytical specificity of the RNase H2 probe-assisted LUPP-LAMP assay for detecting DENV and ZIKV prior to its application, the assay was meticulously validated against a panel of interference viruses, including H1N1, COVID, parainfluenza virus and yellow fever virus. The results demonstrated that the LUPP-LAMP assay specifically detected DENV and ZIKV without cross-reactivity or amplification of non-target viral species (Fig. S12). These findings confirm that the LUPP-LAMP assay is highly specific and efficient for the accurate detection of DENV and ZIKV in serum samples. The excellent specificity of the assay can be attributed to the high-fidelity Splint R ligase-mediated ligation step. Efficient ligation occurs only when both SLP-L and SLP-R probes perfectly hybridize to the target RNA, leading to the formation of the dumbbell-shaped product necessary for subsequent LAMP amplification. In the absence of the target or in the case of mismatched hybridization, the ligation reaction is significantly suppressed, thereby preventing nonspecific amplification. Next, a range of target DENV and ZIKV nucleic acid concentrations (NTC, 1E1, 1E2, 1E3, 1E4, 1E5 and 1E6 copies/reaction) was tested to evaluate the performance of the one-pot RNase H2 probe-assisted real time LUPP-LAMP. Figure 4A and D show that positive results were obtained at concentrations of 1E6, 1E5, 1E4, 1E3, 1E2, and 1E1 copies/reaction. Additionally, the fluorescence emission spectra of FAM- and ROX-labeled probes were tested and shown in Fig. 4B and E. The fluorescence intensities of FAM and ROX increased with higher concentrations of DENV and ZIKV. At concentrations ranging from 1E1 to 1E6 copies/mL, Figure C and F show a positive linear correlation between the log of DENV and ZIKV concentrations and RFU. For DENV, the linear mathematical mode is RFU = 17.51*lgC-8.82 (R² = 0.99), with a minimum detectable concentration of 7.2 copies/reaction (S/N = 3). For ZIKV, the linear mathematical mode is RFU = 18.09*lgC-8.89 (R² = 0.98), with a minimum detectable concentration of 6.5 copies/reaction (S/N = 3). Subsequently, we assessed the feasibility of employing direct fluorescence imaging for the above RNase H2 probe-assisted real time LUPP-LAMP reactions (Fig. 4G). For DENV samples, the resulting amplicon hybridized with FAM-conjugated probes, and RNase H2 cleavage elicited green fluorescence emission. Similarly, ZIKV samples produced amplicons that hybridized with ROX-conjugated probes, resulting in RNase-mediated hydrolysis and red fluorescence emission. When both DENV and ZIKV are co-detected, green and red fluorescence signals are produced concurrently, merging to form composite yellow fluorescence. The LAMP amplification products were confirmed by native PAGE, revealing a distinct stepwise distribution of varying sizes (Fig. 4H). Finally, we performed fluorescence imaging (Fig. 4I) and relative quantified fluorescence intensity (Fig. 4J, S13-15) on FAM and ROX channel of 24 clinical specimens, respectively. Fluorescence imaging and intensity measurements strongly correlated with previous PCR assays. The aforementioned results validate that the LUPP-LAMP method developed in this study exhibits robust platform compatibility.

Fig. 4.

A RNase H2 probe-assisted real-time LUPP-LAMP of DENV in the FAM channel. B Fluorescence emission spectrum of DENV in FAM the channel. C Calibration plot of FAM fluorescence intensity versus the logarithm of DENV concentrations. D RNase H2 probe-assisted real-time LUPP-LAMP for ZIKV in the ROX channel. E Fluorescence emission spectrum of ZIKV in the ROX channel. F Calibration plot of ROX fluorescence intensity versus the logarithm of ZIKV concentrations. The concentrations of DENV and ZIKV are (from bottom to top) 1E1, 1E2, 1E3 1E4, 1E5 and 1E6 copies/ reaction. G Fluorescence imaging of NTC, co-infected, DENV and ZIKV (from left to right) in FAM, ROX and merged channels. H 12% native PAGE of LUPP-LAMP products of co-infected, DENV and ZIKV. Lane 1 to 5: 25 bp marker, DENV target, ZIKV target, mixed target, and NTC. I Merged fluorescence imaging of simulated clinical samples No. 1 to 24. J Fluorescence emission heatmap of simulated clinical samples No. 1 to 24 clinical samples in the FAM and ROX channels

Lateral flow LUPP-LAMP assay for point-of-care (POC) testing of DENV and ZIKV

Subsequently, we developed a rapid nucleic acid test strip method to further evaluate the potential application of our constructed assay in POCT. As shown in Fig. 5A, the extracted RNA samples were first subjected to a low-temperature LAMP reaction at 30 °C. After the reaction, the products were promptly analyzed using the test strips. Notably, the LAMP primers used in this assay were labeled with FAM/biotin or TAMRA/Digoxin (Figure S16). Results from typical samples are presented in Fig. 5B. A color band at T1 indicated a positive result for DENV, while a band at T2 signified a positive result for ZIKV. When both lines appeared, it indicated a mixed infection. Serial dilutions of the RNA target of DENV, ZIKV and coinfected virus, ranging from 5 to 100 copies/reaction, were tested to determine the detection limit of the developed lateral flow LUPP-LAMP assay. As shown in Fig. 5C, the T1 test line for individual DENV detection showed positive red colorations at 10, 50, and 100 copies/reaction, but no color change at 5 copies/reaction. Similar results were observed for individual ZIKV detection. For simultaneous detection of DENV and ZIKV, the LOD was consistently 10 copies/reaction. This consistency is attributed to the design of our universal primers, which minimize amplification imbalance during LAMP. Since the test strips require handling of the amplified products, we evaluated the anti-contamination performance by incorporating dUTP and UNG (Fig. 5D). Our findings demonstrated that UNG successfully deactivated carryover contaminants from previous LUPP-LAMP reactions with mixed DENV and ZIKV infections, preventing their utilization as templates in subsequent cycles. The analysis performance of the proposed LUPP-LAMP is comparable to or better than that of published methods (Table S5). Further, we used lateral flow test strips to assess clinical samples No.1 to 24. Figure S17 shows images of the test strip outcomes, and Fig. 5E presents a heatmap of the relative optical density (ROD) values for the T1, T2, and C lines. The results obtained from LUPP-PCR, RNase H2 probe-assisted real-time LUPP-LAMP, fluorescence imaging, and lateral flow test strips were in complete agreement for detecting 24 clinical samples. This uniformity validates the LUPP-LAMP platform for detecting DENV and ZIKV, supporting its application in PCR, LAMP, as well as in visualization and fluorescence-based detection methods.

Fig. 5.

A Scheme illustration of lateral flow LUPP-LAMP assay. B Lateral flow strip and corresponding optical density results of NTC, DENV, ZIKV, and co-infected samples. C LOD confirmation for DENV, ZIKV, and co-infected samples. Sample concentrations are 100, 50, 10, and 5 copies/reaction. D Anti-contamination capacity of the LUPP-LAMP lateral flow strip system. The DENV and ZIKV mixed LUPP-LAMP amplification product was used as a simulated contaminant. and the DENV, ZIKV and mixed samples were used as the templates, respectively. Group 1 to 4 had the UNG enzyme added, while group 5 to 8 did not. E ROD of the control, Test 1, and Test 2 lines on lateral flow strips for 24 simulated clinical samples. F Specificity analysis of DENV and ZIKV. ROC curve of lateral flow test (green line) and qPCR (red line) results for DENV and ZIKV (G). The AUC values were 0.960 for DENV and 0.953 for ZIKV at a 95% confidence interval

Selectivity was tested using five different pathogens: Influenza virus (H1N1), COVID-19, Parainfluenza virus (PIV), yellow fever virus (YFV) and NTC. As depicted in Fig. 5F, significant alterations in the ROD values were observed exclusively in the presence of DENV and ZIKV. However, for other interfering viruses and blank samples, the strips show no color change. These results indicate that the proposed lateral flow biosensor is specific for detecting DENV and ZIKV. The receiver operating characteristic (ROC) curve analysis demonstrated an area under curve (AUC) value of 0.960 for the DENV target, indicating high accuracy with a sensitivity of 86% and a specificity of 90% (Fig. 5G). For the ZIKV target, ROC analysis resulted in a slightly lower AUC value of 0.953, while maintaining high sensitivity (93%) and specificity (90%). These results show that the one-pot LUPP-LAMP LFA is highly sensitive and specific for the detection of DENV and ZIKV in real samples.

LUPP-LAMP MCA assay for DENV genotyping

Dengue virus serotyping is of great significance in improving diagnosis and treatment, controlling epidemic spread, studying virus evolution, and developing vaccines. Thus, we developed the LUPP-LAMP melting curve analysis (MCA) assay to genotype DENV strains 1, 2, 3, and 4. Initially, the genomic sequences of the common strains were downloaded from the NCBI database and performed sequence alignment to identify strain-specific sequences (Figure S18-21). Figure 6A illustrates the principle of the LUPP-LAMP MCA detection process. This reaction used four sets of target-specific ligation oligos targeting DENV 1, 2, 3, and 4. Each oligo was equipped with a distinct Tm-tag sequence. These sequences allowed differentiation of DENV subtypes through specific hybridization with a universal dual-labeled probe, varying in melting temperature. We refined our assay by developing a dual-labeled probe of 20 nucleotides. Systematic evaluation showed that variations in Tm tag length had negligible impact on the probe’s melting temperature (Fig. 6B). However, incorporating mismatches at various Tm sequence locations effectively altered the probe’s Tm (Fig. 6C). Utilizing this discovery, we crafted a stem-loop structured dual-labeled probe optimized for subsequent experimental procedures (Fig. 6D).

Fig. 6.

A Scheme illustration of LUPP-LAMP mediated MCA for DENV genotyping. B of 15, 17, 19, and 21 nucleotides (nt). Linear dual labeled (C) and hairpin dual labeled probes (D) for melting tags with mismatch base pairs in different positions. E DENV-1 MCA results at 65.4 ℃. F DENV-2 MCA results at 60.7 ℃. G DENV-3 MCA results at 55.1 ℃. H DENV-4 MCA results at 44.3 ℃

Co-infection with multiple DENV subtype also occurs clinically. Rapid detection and differentiation of DENV are crucial for timely clinical treatment, etiologic investigation, and disease control. Using dual infection of DENV 1, 2, 3, and 4 as examples, we studied the ability to distinguish the four types when they are partially or completely present in the same reaction. MCA revealed distinctive thermal profiles for each DENV serotype configuration. Specifically, the presence of serotype 1 alone resulted in a unimodal peak at 65.4 °C (Fig. 6E). Coexistence of serotypes 1 and 2 induced a bimodal profile with peaks at 60.7 °C and 65.4 °C, respectively (Fig. 6F). Inclusion of serotype 1, 2 and 3 introduced a trimodal pattern with peaks at 55.1 °C, 60.7 °C, and 65.4 °C (Fig. 6G). Simultaneous presence of all four DENV serotypes (1, 2, 3, and 4) yielded a quadruple-peaked curve with distinct melting points at 44.1 °C, 55.1 °C, 60.7 °C, and 65.4 °C (Fig. 6H), enabling differentiation through MC. In the next phase of our study, we employed both probe-based and melting curve analysis to examine 24 clinical specimens containing mixed DENV subtypes (Figure S22, 23). The concordance between the two methods was remarkable, underscoring the robustness and reliability of our assay. This strong agreement between the probe and MCA techniques suggests that our methodology holds considerable promise for clinical diagnostics, offering a valuable tool for accurate detection and differentiation of mixed DENV infections.

Conclusions

This study introduces a groundbreaking multiplex LAMP technique, known as LUPP-LAMP, for the simultaneous genotyping of DENV and ZIKV. LUPP-LAMP harnesses the power of splint DNA ligase to connect universal primers and probes, overcoming the complexities of primer design and eliminating primer dimer formation commonly encountered in traditional LAMP assays. This innovative approach yields a highly sensitive and specific assay with a detection limit of 10 copies/reaction for both DENV and ZIKV, setting a new standard for nucleic acid detection. LUPP-LAMP’s adaptability for POCT is exemplified by its seamless integration with lateral flow test strips and fluorescence imaging. This integration provides rapid and user-friendly alternatives for resource-limited settings, expanding the accessibility of accurate diagnostics. Furthermore, LUPP-LAMP extends its capabilities to DENV genotyping using MCA. This enables the differentiation of DENV serotypes 1 to 4 in mixed infections, offering crucial insights for epidemiological surveillance and targeted interventions. While the current study could not include real clinical sample validation due to practical limitations, future work will prioritize validating the LUPP-LAMP system with clinical specimens, including samples with varying concentrations of DENV serotypes and ZIKV. This will be essential for further assessing its diagnostic accuracy and applicability in real-world settings. In conclusion, the LUPP-LAMP method represents a significant leap forward in the detection and genotyping of DENV and ZIKV. Its simplicity, high sensitivity, specificity, and adaptability make it a promising tool for both laboratory and POCT applications, contributing to improved diagnosis, treatment, and control of these mosquito-borne diseases, ultimately enhancing public health outcomes.

Author contributions

All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Funding

This work is financially supported by State Key Research Development Program of China (2022YFC2305003, 2022YFC2305000). National Natural Science Foundation of China (82460414). Sichuan Science and Technology Program (2023NSFSC1478).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors agreed to submit this manuscript.

Competing interests

The authors declare no competing interests.