We obtained AuNPs-based LFB materials, including biotinylated bovine serum albumin (biotin-BSA; 4 mg/mL), a rabbit anti-fluorescein antibody (anti-FAM; 0.2 mg/mL), and a sheep anti-digoxigenin antibody (anti-Dig; 0.25 mg/mL) from Abcam Co., Ltd. (Shanghai, China), and crimson red dye streptavidin-coated AuNPs (SA-AuNPs; 40 ± 5 nm, 10 mg/mL, 100 mM borate [pH 8.5] with 0.1% BSA, 0.05% Tween 20, and 10 mM ethylenediaminetetraacetic acid) from Bangs Laboratories Inc. (Fishers, IN, USA). Four LFB components (sample, conjugate, absorption pads, and nitrocellulose membranes) were manufactured and laminated on plastic adhesive backing by HuiDeXin Biotech. Co., Ltd. (Tianjin, China) according to our design scheme (Figure 2). Nucleic acid releasers were obtained from BaiAoLaiBo Technique Ltd. (Beijing, China). Universal isothermal amplification kits and the colorimetric indicator, malachite green (MG), were obtained from HuiDeXin Biotech. Co., Ltd. (Tianjin, China). Commercial PCR diagnostic kits for C. trachomatis and N. gonorrhoeae were obtained from DaAn Gene Co., Ltd. (Guangzhou, China).

Full-length ompA sequences from 14 C. trachomatis serological variants (A-K and L1-L3; GenBank accession numbers JX548318.1, JX559518.1, JX559519.1, KP164991.1, JX559522.1, JX564244.1, JX564245.1, JX564246.1, JX564247.1, JX648604.1, JX564248.1, JX569832.1, KP120855.1, and JX569834.1) and the N. gonorrhoeae orf1 gene (M84113.1) were synthetically produced and cloned into the pUC57 vector. The basal concentration of all plasmids was 1×10⁸ copies. The synthesized plasmids were used as positive controls.

Participants at the Hangzhou Women’s Hospital provided 146 vaginal discharge samples for examination between August 2021 and March 2022 that were believed to include STIs caused by C. trachomatis and/or N. gonorrhoeae. Nucleic acid-releasing agents were used to extract crude genomic DNA according to the manufacturer’s guidelines. Briefly, a sample was incubated with 100 μl of a nucleic acid releasing agent for 5 min. The resulting supernatant was used as a template for LAMP amplification. The quantities of genomic DNA were calculated using the 260/280 nm absorbance ratio using a Nano-Drop ND 2000 instrument (Thermo Fisher Scientific; Waltham, MA, USA). Other microorganisms investigated in this study are listed in Table 1.

Figure 2 shows that the AuNPs-LFB, with dimensions of 60 mm × 4 mm, is divided into four sections: the sample pad, the conjugate pad, the detection region (nitrocellulose membrane), and the absorption pad. SA-AuNPs containing crimson red dye are deposited on top of the conjugate pad. Rabbit anti-FAM (0.2 mg/mL), sheep anti-Dig (0.25 mg/mL), and biotin-BSA (4 mg/mL) are immobilized onto the nitrocellulose membrane of test 1 (TL1) and 2 (TL2) and control (CL) lines, respectively, which were separated by 5 mm. Finally, the four separate parts are bonded together on a plastic card with an adhesive backing. The AuNPs-LFB was kept dry and at room temperature until used. The workflow of AuNPs-LFB assay as following, LAMP products and running buffer were added to the AuNPs-LFB sample pad concurrently. A capillary action carries the LAMP product-containing flowing buffer along the biosensor, rehydrating the SA-AuNPs immobilized in crimson red dye. Biotin-BSA was used to capture SA-AuNPs at the CL. Anti-FAM was used to capture FAM/biotin-labeled C. trachomatis-LAMP products at TL1. Anti-Dig was used to capture Dig/biotin-labeled N. gonorrhoeae-LAMP amplicons at the TL2. The purpose of control line (CL) is to monitor whether the AuNPs-LFB test is valid. The CL band must be shown in each test. When the CL brand is not appeared, indicating the test is invalid.

Two LAMP primer sets were created based on the ompA and orf1 target genes to detect C. trachomatis and N. gonorrhoeae, respectively. The ompA genes from 14 different C. trachomatis serological variations (serovars A, B, C, D, E, F, G, H, I, J, K, L1, L2, and L3) were aligned using the MEGA-X software (https://www.megasoftware.net/) (Supplementary Figure S1). Conserved sequences were used for C. trachomatis LAMP primer design with the Primer Explorer v.5 (http://primerexplorer.jp/e/) and Primer Premier v.5.0 software. Primer pair specificity was confirmed using the BLAST analysis tool. Table 2 lists the primer sequences and alterations, and Supplementary Figure S1 shows the primer locations. All primers were produced and purified via high-performance liquid chromatography by The TsingKe Biotechnology Company (Beijing, China).

The single LAMP reactions for C. trachomatis or N. gonorrhoeae were performed in 25 μL reaction volumes. Each reaction contained 1 μL of each standard plasmid template (5 μL of the clinical sample template); 0.8 μM of F3, B3, LF, and LB; 1.6 μM of FIP and BIP; 1 μL of Bst 2.0 DNA polymerase (8 U); 12.5 μL of 2× reaction buffer (2 M betaine, 16 mM magnesium sulfate, 40 mM potassium chloride, 20 mM ammonium sulfate, 40 mM Tris-hydrochloride [pH 8.8], and 0.2% Tween-20); 1 μL of the colorimetric indicator (MG); supplemented with double-distilled water (DW) to 25 μL.

The one-step mLAMP reaction was performed in a 25 μl reaction volume containing 1 μL of each standard plasmid template (5 μL of the clinical sample template); 0.3 μM of ompA-F3, ompA-B3, ompA-LF, and ompA-LB; 0.6 μM of ompA-FIP and ompA-BIP; 0.5 μM of orf1-F3, orf1-B3, orf1-LF, and orf1-LB; 1 μM each orf1-FIP, and orf1-BIP; 12.5 μL of 2×reaction buffer; 1 μL of Bst 2.0 DNA polymerase (8 U); 1 μL of the colorimetric indicator (MG); supplemented with DW to 25 μl in DW water. The reaction was performed in a heat-block or water bath at a constant temperature (conditions were optimized as outlined below).

Real-time turbidity LA-500 (Lumiprobe; Tokyo, Japan), visual detection reagents (MG), and the AuNPs-LFB were used to track the LAMP amplification products. A turbidity value of >0.1 indicated a successful outcome. Reaction mixtures that turned bright green indicated a successful MG visual detection; a colorless reaction indicated a failed detection. CL and TL1 simultaneously appearing on the AuNPs-LFB indicated a positive C. trachomatis-LAMP result. CL and TL2 simultaneously appearing on the AuNPs-LFB indicated a positive N. gonorrhoeae-LAMP result. CL only appearing on the AuNPs-LFB indicated a negative result.

The optimal mLAMP-AuNPs-LFB reaction temperature was determined by incubating the C. trachomatis-LAMP and N. gonorrhoeae-LAMP reactions at 62°C to 69°C (in 1°C increments) and assessing their amplification results using real-time turbidity. Then, the incubation time was varied from 15 to 45 min (in 10 min increments) at the optimal reaction temperature with the amplification results detected by AuNPs-LFB. Each assay was performed in triplicate.

Two conventional plasmids, the ompA and orf1 plasmids, were produced and then serially diluted by a factor of ten to provide copy numbers ranging from 5.0×10⁴ to 5.0×10^-1. LAMP reactions were performed under optimum conditions, and the data were examined using MG and AuNPs-LFB.

Synthetic nucleic acid sequences and bacterial, viral, and fungal nucleic acid templates were detected at ≥1.0×10⁴ copies to assess the mLAMP-AuNPs-LFB assay’s specificity (Table 1). All assays were performed at least thrice, with DW serving as the negative control.

Genital secretion samples were obtained from 146 women at the Hangzhou Women’s Hospital who were believed to have STIs caused by C. trachomatis and/or N. gonorrhoeae. Genomic DNA templates were obtained rapidly using the Nucleic Acid Releasers Kit (BaiAoLaiBo Technique Ltd.) according to the manufacturer’s guidelines. We compared the mLAMP-AuNPs-LFB assay to commercially available real-time TaqMan PCR Kits for C. trachomatis and N. gonorrhoeae (DaAn Gene Co., Ltd.; Guangzhou, China) on an Applied Biosystems™ 7500 Real-Time PCR System (Life Technologies; Singapore). The basis for determining the copy number of the samples according to the corresponding standard curve. C. trachomatis and N. gonorrhoeae concentrations of >500 copies were considered positive based on the manufacturer’s recommendations. The Human Ethics Committee of Hangzhou Women’s Hospital approved the lawful collection and analysis of these genomic DNA templates (approval no. [2021]-K[2]-8). The identification investigations were performed at biosafety level 2 based on the WHO Biosafety Manual (3^rd Edition). The mLAMP-AuNPs-LFB data were compared with quantitative real-time PCR (qPCR) data.

A representative schematic and workflow of the mLAMP-AuNPs-LFB assay are shown in Figures 1, 2, respectively. Briefly, genomic DNA templates were rapidly extracted using nucleic acid releasers (Figure 1, step 1). The ompA-FIP* and ompA-LF* primers were 5’-labeled with fluorescein (FAM) and biotin, and orf1-FIP* and orf1-LF* with digoxigenin (Dig) and biotin. The ompA-LAMP products were labeled with FAM and biotin and the orf1-LAMP products with Dig and biotin after a 35 min reaction at 67°C (Figure 1, step 2). Finally, the mLAMP products were visually monitored with an AuNPs-LFB within 2 min (Figure 1, step 3).

LAMP products (0.5 µL) and running buffer (100 µL; 100 mM phosphate-buffered saline with 1% Tween 20 [pH 7.4]) were added to the AuNPs-LFB sample pad concurrently (Figure 2A). A capillary action carries the LAMP product-containing flowing buffer along the biosensor, rehydrating the SA-AuNPs immobilized in crimson red dye (Figure 2B). Biotin-BSA was used to capture SA-AuNPs at the CL. Anti-FAM was used to capture FAM/biotin-labeled ompA-LAMP products at TL1. Anti-Dig was used to capture Dig/biotin-labeled orf1-LAMP amplicons at the TL2. The result is negative when only biotin-BSA captures the SA-AuNPs at the CL (Figure 2C). The interpretation of the mLAMP-AuNPs-LFB assay is outlined in Figure 2D.

The ompA-, orf1-, or mLAMP-amplification mixes were incubated at a fixed temperature (65°C) for 1 h to confirm the reaction system with the two LAMP primer sets. Then, a visual detection reagent (MG) and the AuNPs-LFB were used to assess the ompA/orf1 and mLAMP amplification results, respectively. DW was used as the negative control. While positive ompA-, orf1-, and mLAMP reactions changed from colorlessness to bright green, the negative control remained colorless (Figure 3A). The co-appearance of crimson-red TL1 and CL bands with AuNPs-LFB detection indicated a positive ompA-LAMP result. The co-appearance of crimson-red TL2 and CL bands indicated a positive orf1-LAMP result. A positive mLAMP result was indicated by the simultaneous appearance of TL1, TL2, and CL bands in the AuNPs-LFB. In contrast, the negative control showed only a crimson red CL band (Figure 3B). In conclusion, our findings supported using both LAMP primer sets in the mLAMP reaction.

Optimizing the incubation temperature for LAMP reactions is essential. Therefore, we used a standard plasmid copy number (1.0×10⁴) to evaluate temperatures from 62°C to 69°C in 1°C increments. The LAMP amplification results were tracked in real-time using turbidity measurements. The results indicated that robust ompA-LAMP amplification occurred at 66°C to 69°C and orf1-LAMP amplification at 67°C to 69°C (Figure 4). Therefore, 67°C was the most suitable amplification temperature for our assay.

Serial dilutions of nucleic acid templates (ompA- and orf1-plasmid constructs) with 5.0×10⁴ to 5.0×10^-1 copies were used to determine the LoD of our assay. mLAMP reactions were performed as described above, and the results were visualized using MG and the AuNPs-LFB. The sensitivity of our assay was 50 copies per test for both the ompA (Figures 5A, B) and orf1 (Figures 5C, D) plasmids. The two target genes were simultaneously detected and identified in a single reaction (Figures 5E, F). The mLAMP results using the AuNPs-LFB were consistent with the visual detection reagent (MG; Figure 5). However, it should be noted that the visual detection method (MG) cannot discriminate target genes in a multiplex assay. The sensitivity of mLAMP-AuNPs-LFB was consistent with single ompA and orf1 mLAMP-AuNPs-LFB assays (Figure 5).

Four assay reaction times (15, 25, 35, and 45 min) were evaluated at 67°C to determine the optimal reaction time for the LAMP amplification stage of the protocol. The AuNPs-LFB was used to determine the outcomes. A 35 min LAMP reaction was found to be optimal at the genomic DNA template LoD (50 copies; Figure 6). Therefore, our assay’s entire detection process can be completed within 45 min: rapid genomic DNA preparation (5 min), mLAMP amplification (35 min), and visual outcome (<2 min).

Our assay’s specificity was evaluated using C. trachomatis ompA plasmids (serovars A, B, C, D, E, F, G, H, I, G, K, L1, L2, and L3), an N. gonorrhoeae orf1 plasmid, positive C. trachomatis and N. gonorrhoeae clinical samples (confirmed by qPCR), and 15 other pathogens (Table 1). Both C. trachomatis and N. gonorrhoeae strains’ genomic DNA extractions were successful. Positive C. trachomatis and N. gonorrhoeae results were indicated by the presence of three red bands (TL1, TL2, and CL) on the AuNPs-LFB. C. trachomatis isolates caused TL1 and CL bands on the AuNPs-LFB, indicating success. N. gonorrhoeae strains caused TL2 and CL bands on the AuNPs-LFB, indicating success. All other pathogens and control groups examined returned negative results (Figure 7). Therefore, our assay could distinguish between C. trachomatis and N. gonorrhoeae strains.

We further evaluated the clinical feasibility of our mLAMP-AuNPs-LFB assay for accurately identifying C. trachomatis and N. gonorrhoeae with 146 suspected C. trachomatis and/or N. gonorrhoeae genital secretion samples examined using qPCR and our assay. Forty-eight of the 146 samples were qPCR-confirmed (>500 copies) as C. trachomatis-positive, 29 as N. gonorrhoeae-positive, and six as C. trachomatis and N. gonorrhoeae co-infections. All C. trachomatis and N. gonorrhoeae positive samples were confirmed by our mLAMP-AuNPs-LFB assay. In addition, three C. trachomatis-negative and one N. gonorrhoeae-negative samples (genomic DNA concentrations of 100–500 copies) were positive with our mLAMP-AuNPs-LFB assay (Table 3 and Supplementary Table S1). Our assay’s positive results with these four discordant samples were confirmed by PCR amplification and DNA sequencing (Supplementary Table S1). Therefore, these data indicate that our assay can function as an advanced clinical diagnostic tool for C. trachomatis and N. gonorrhoeae.