A Pilot Study to Develop an Ultrasensitive Thio-NAD Cycling ELISA for Neisseria gonorrhoeae Detection

Abstract

Neisseria gonorrhoeae and its increasing antimicrobial resistance have raised global health concerns, creating an urgent need for more advanced diagnostics to contain the spread. We conducted a preclinical trial using an ultrasensitive enzyme-linked immunosorbent assay (ELISA) with Thio-nicotinamide-adenine dinucleotide (NAD) cycling for the detection of N. gonorrhoeae. Three independent researchers conducted the initial experiments to obtain calibration curves. The inter-assay reproducibility across the 3 investigators yielded coefficients of variation of 11% when measuring 500000 CFU/mL, confirming the consistent performance of the ultrasensitive ELISA. Specificity against common clinical microorganisms was also assessed. Performance was first evaluated in simulated urine, vaginal fluid, and saliva. A comparative evaluation of 90 clinical isolates was then performed using a conventional ELISA method. The ultrasensitive ELISA produced consistent calibration curves (R² = 0.99) with limits of detection of 1.1 × 10⁴ to 1.8 × 10⁴ CFU/mL. No cross-reactivity to common pathogens was observed. The assay results revealed linearity in simulated fluids, with a modestly reduced sensitivity in urine and vaginal fluid. At 50000 CFU/mL, the ultrasensitive ELISA generated signals >50-fold stronger than the p-nitrophenyl phosphate (pNPP) ELISA. For clinical isolates, the ultrasensitive ELISA identified 78.9% as positive compared with 30.0% by pNPP ELISA (p = 0.0005). The ultrasensitive ELISA detected N. gonorrhoea with high sensitivity and specificity across sample types, consistently outperforming the conventional ELISA, and demonstrating the potential of this approach as a cost-effective diagnostic alternative.

INTRODUCTION

Neisseria gonorrhoeae is the bacterial pathogen responsible for gonorrhea, one of the most common sexually transmitted infections worldwide. Gonorrhea is a major global health problem due to both its high incidence and clinical complications. Untreated infections can lead to pelvic inflammatory disease, chronic pelvic pain, infertility, and ectopic pregnancy.^1,2) Infected individuals also face a higher risk of acquiring or transmitting human immunodeficiency virus (HIV).³⁾ According to WHO, approximately 82.4 million new infections occurred in 2020 among individuals aged 15–49 years, with the highest number reported in Africa and the Western Pacific.⁴⁾ Transmission occurs through sexual contact involving the urogenital, rectal, and pharyngeal sites. Because a large proportion of cases are asymptomatic, infection control is difficult, contributing to continued community transmission.⁵⁾ Moreover, the emergence and spread of antimicrobial resistance due to a unique and ingenious mechanism of Neisseria spp. are major challenges to gonorrhea control,⁶⁾ with reduced susceptibility reported for nearly all antibiotic classes used in treatment.⁷⁾

Accurate diagnosis of N. gonorrhoeae infection is critical for initiating prompt, appropriate treatment, effective partner management, and monitoring antimicrobial resistance. Although conventional bacterial culture remains the reference method and allows for antimicrobial susceptibility testing,⁸⁾ it is technically demanding—requiring selective media, strict transport conditions, and skilled staff—and the sensitivity is reduced when specimens are not processed promptly. In clinical settings, recovery rates are strongly influenced by timely and appropriate specimen transport and handling, factors that are especially critical for extragenital specimens, such as rectal and pharyngeal swabs, where sensitivity is typically lower than that for urogenital sites, and further reduced by delays or suboptimal specimen handling.⁹⁾ Moreover, culture typically requires 24–48 h to yield results, which is impractical for providing timely treatment based on confirmed laboratory data in the outpatient settings.¹⁰⁾ These constraints contribute to empirical antibiotic prescribing practices, which may accelerate the emergence of antimicrobial resistance in N. gonorrhoeae and other pathogens. In addition, due to the reduced culture sensitivity of extragenital sites compared with urogenital sites, underdiagnosis occurs when culture alone is applied.¹¹⁾

Nucleic acid amplification tests (NAATs), such as PCR, are now primary diagnostic tools for N. gonorrhoeae, providing >90% sensitivity and >99% specificity. NAATs can be applied to noninvasive specimens such as first-catch urine, thereby increasing patient acceptability and facilitating screening of asymptomatic individuals.¹²⁾ NAATs have some disadvantages, however, including their susceptibility to bacterial genetic variation and the risk of false positive results due to cross-reaction with commensal Neisseria species if the assay target is not carefully designed.¹³⁾ In addition, NAATs are more expensive than culture, and their operation requires specialized equipment and personnel with advanced technical training. These factors limit the widespread use of NAATs in low-resource settings with a high disease burden. Moreover, NAATs detect nucleic acids rather than protein targets and therefore do not indicate bacterial viability, and residual DNA can remain detectable after effective treatment and complicate clinical interpretation.¹⁴⁾ In this context, ultrasensitive protein detection remains valuable because it reflects active infection and provides direct evidence of antigen presence, offering a biologically complementary approach to molecular diagnostics.

Before molecular methods became dominant, antigen detection methods based on enzyme-linked immunosorbent assays (ELISAs) were widely studied. Commercial kits such as the Abbott Gonozyme exhibit good sensitivity and specificity in symptomatic men and reasonable accuracy in cervical samples.¹⁵⁾ Because ELISA performance varies across populations and is less reliable in broader clinical use, ELISA use declined once NAATs became widely available. Traditional ELISA systems typically employ alkaline phosphatase with p-nitrophenyl phosphate (pNPP) as the substrate, producing a colorimetric signal at 405 nm. While this approach is inexpensive and compatible with basic laboratory equipment, its limited sensitivity restricts its application for N. gonorrhoeae detection. Alternative amplification chemistries, such as thio-NAD cycling, were recently developed to enhance alkaline phosphatase (ALP)-based assays.^16,17) One thio-NAD cycling ultrasensitive ELISA method, the TN-cyclon^TM, might be a practical tool for rapid and cost-effective detection of N. gonorrhoeae. In the present study, we applied the TN-cyclon^TM ultrasensitive ELISA for N. gonorrhoeae detection and evaluated its sensitivity, specificity, and performance across simulated body fluids, compared with the conventional pNPP ELISA. To assess the feasibility of this approach, we also conducted a pilot study using clinical specimens.

MATERIALS AND METHODS

Patient Specimens

A total of 90 clinical isolates of N. gonorrhoeae, originally obtained from genital swab specimens of male patients and confirmed as culture-positive, were used for comparative analyses. Colonies were obtained from primary culture, stored at −80°C, and subsequently recovered for the preparation of bacterial suspensions. To prepare bacterial suspensions, isolates were adjusted to a McFarland standard of 2 (approximately 1 × 10⁹ CFU/mL). Serial dilutions were then performed, and aliquots were plated and incubated overnight to obtain colony counts confirming the bacterial concentration. The resulting suspensions were further diluted to the appropriate concentrations for comparative assays.

Ethical Approval

This study was approved by the Institutional Review Board of Kaohsiung Medical University Chung-Ho Memorial Hospital (IRB No. KMUHIRB-SV(I)-20180072). The study was conducted in accordance with the principles of the Declaration of Helsinki and its later amendments. Written informed consent was obtained from all participants prior to specimen collection. Consent for publication was also obtained for cases involving identifying information.

Reagents and Chemicals

The Neisseria gonorrhoeae reference strain ATCC49926 was obtained from the American Type Culture Collection (ATCC, Manassas, VA, U.S.A.). The primary (capture) antibody (15803) and secondary (detection) antibody (15802) against N. gonorrhoeae were purchased from QED Bioscience (San Diego, CA, U.S.A.). Both antibodies were generated using a pool of UV-inactivated whole-cell N. gonorrhoeae strains (G-7, R-11, 7122 [W-I], 5766, 8038 [W-II], and 8660 [W-III]) as immunogens.^18,19) An ALP labeling kit (LK13) for secondary antibody conjugation was obtained from Dojindo Laboratories (Kumamoto, Japan). For the cycling reagents, NADH was purchased from Sigma-Aldrich (N1161-10VL; St. Louis, MO, U.S.A.), 3α-hydroxysteroid dehydrogenase (3α-HSD) was obtained from Asahi Kasei Pharma (T-58; Tokyo, Japan), and thio-NAD was purchased from Oriental Yeast (44104001; Tokyo, Japan). The substrate, 17β-methoxy-5β-androstan-3α-ol 3-phosphate (A3P), was synthesized and kindly gifted by Teruki Yoshimura. The pNPP was purchased from Merck (4876; Sigma-Aldrich).

Ultrasensitive ELISA with Thio-NAD Cycling

The ultrasensitive ELISA (TN-cyclon^TM) was conducted in a sandwich format.²⁰⁾ A capture antibody immobilized on the plate binds to the target antigen, and detection is performed using an ALP-conjugated secondary antibody. Signal amplification relies on an enzymatic cycling system involving ALP and 3α-HSD. ALP hydrolyzes a steroid phosphate substrate to generate a hydroxysteroid intermediate, which is then oxidized to a ketosteroid by 3α-HSD in the presence of thio-NAD as a coenzyme, producing thio-NADH. The reverse reaction reduces the ketosteroid back to the hydroxysteroid with NADH as a coenzyme. Continuous cycling between these 2 reactions leads to progressive accumulation of thio-NADH, which is measured spectrophotometrically at 405 nm. The triangular accumulation pattern of this amplification process allows detection of very low antigen concentrations within a short time.

To coat the plates, 100 μL of the capture antibody (100 μg/mL in 50 mM Na₂CO₃ buffer, pH 9.6) was added to each well, and the plates were incubated for 1 h at room temperature, followed by 3 washes with Tris-buffered saline (TBS) containing 0.05% Tween 20. Each well was then blocked with 300 μL of 1% bovine serum albumin (BSA) in TBS for 1 h at room temperature, and the plate was washed 3 times.

After blocking, 100 μL of bacterial suspension prepared from the N. gonorrhoeae reference strain was applied at concentrations of 500000, 100000, 50000, 10000, and 5000 CFU/mL. TBS containing 0.1% BSA served as the blank control. The plates were incubated for 1 h at room temperature or overnight at 4°C, and washed 9 times.

The ALP-conjugated detection antibody, diluted in TBS with 0.05% Tween 20 and 0.1% BSA, was added to each well at 100 μL, and the plates were washed 9 times. For signal amplification, 100 μL of thio-NAD cycling reagent containing 10 U/mL 3α-HSD, 0.4 mM A3P, 1.0 mM NADH, and 2.0 mM thio-NAD in 100 mM Tris–HCl was dispensed into each well. Absorbance at 405 nm was recorded using a microplate reader (SpectraMax ABS Plus, San Jose, CA, U.S.A.), with 660 nm used as the reference wavelength for background correction.

Cross-Reactivity Assay

Cross-reactivity was assessed using a panel of microorganisms frequently detected in clinical specimens. The panel included Proteus vulgaris ATCC 29905, Pseudomonas aeruginosa ATCC 27853, Klebsiella pneumoniae ATCC 700603, Staphylococcus aureus ATCC 25923, Acinetobacter baumannii ATCC 19606, Proteus mirabilis ATCC 7002, Streptococcus equi ssp. zooepidemicus ATCC 43079, and Candida albicans SC 5314. N. gonorrhoeae ATCC 49926 served as the positive control.

Each organism was cultured under standard laboratory conditions until the logarithmic growth phase. Suspensions were prepared in TBS containing 0.1% BSA, and the cell density was adjusted to 500000 CFU/mL. For each microorganism, 100 μL of the suspension was applied to antibody-coated wells, and the ultrasensitive ELISA with thio-NAD cycling assay was performed according to the protocol described above.

Absorbance at 405 nm was recorded with 660 nm as the reference wavelength. A blank control consisting of TBS with 0.1% BSA was included in each experiment. The signal-to-noise (S/N) ratio was calculated by dividing the absorbance of each sample by the mean absorbance of the blank. All assays were carried out in triplicate, and mean values were used for analysis.

Body Fluid Simulation Test

Simulated urine, vaginal fluid, and saliva were prepared according to established formulations, with compositions summarized in Table 1. A suspension of N. gonorrhoeae ATCC 49926 was adjusted to concentrations of 500000, 100000, 50000, 10000, and 5000 CFU/mL. Each concentration was diluted directly in the simulated fluids to generate test samples. Aliquots (100 μL) of the spiked samples were analyzed using the ultrasensitive ELISA as described above. A suspension of N. gonorrhoeae in TBS containing 0.1% BSA served as the standard reference. Absorbance at 405 nm was measured with 660 nm as the reference wavelength using a microplate reader. All assays were performed in triplicate, and mean values were used for subsequent analysis.

Table 1. Composition of Simulated Body Fluids for TN-cyclon^TM Performance Evaluation

Simulated fluid	Substance/Ingredient	Concentration (g/L)	pH
Urine	CaCl2·2H2O	0.651	—
	MgCl2·6H2O	0.651
	NaCl	4.6
	Na2SO4	2.3
	Na3C6H5O7	0.65
	Na2C2O4	0.023
	KH2PO4	2.8
	KCl	1.6
	NH4Cl	1.0
	CO(NH2)2 (urea)	25.0
	C4H9N3O2 (creatinine)	1.1
	Tryptic soy broth	10.0
Vaginal fluid	Sodium chloride	3.510	4.6
	Potassium hydroxide	1.400
	Calcium hydroxide	0.222
	Bovine serum albumin	0.018
	Lactic acid	2.000
	Acetic acid	1.000
	Glycerol	0.160
	Urea	0.400
	Glucose	5.000
Saliva	Disodium hydrogen phosphate	2.382	6.75
	Potassium dihydrogen phosphate	0.190
	Sodium chloride	8.000
	Distilled water	Up to 1 L
	Phosphoric acid	q.s. to pH 6.75

pNPP ELISA Assay

For comparison, a conventional ELISA using ALP with pNPP as the substrate was performed in parallel with the ultrasensitive thio-NAD cycling assay. To coat the plates, 100 μL of capture antibody (5 μg/mL in 50 mM Na₂CO₃ buffer, pH 9.6) was added to each well. The plates were incubated for 1 h at room temperature and then washed 3 times in TBS containing 0.05% Tween 20. Wells were then blocked with 300 μL of 1% BSA in TBS for 1 h at room temperature and the plates were washed 3 times.

Bacterial suspensions of N. gonorrhoeae ATCC 49926 were prepared and adjusted to 500000, 100000, 50000, 10000, and 5000 CFU/mL. A 100-μL aliquot from each dilution was applied to the wells. After incubation for 1 h at room temperature and washing, 100 μL of ALP-conjugated detection antibody diluted in TBS containing 0.05% Tween 20 and 0.1% BSA was added to each well, followed by additional washing.

For color development, 100 μL of the pNPP substrate solution (1 mg/mL in 1 M diethanolamine buffer, pH 9.8, containing 0.5 mM MgCl₂) was dispensed into each well, and the plates were incubated at room temperature. The absorbance at 405 nm was recorded with 660 nm as the reference wavelength. All experiments were performed in triplicate, and mean values were used for subsequent analysis.

Statistical Analysis

Background correction was performed by subtracting the mean absorbance of blank wells (0.1% BSA in TBS) from the values at each concentration and time point. The limit of detection (LOD) was defined as the mean blank absorbance plus 3 times the standard deviation (S.D.), and the limit of quantification (LOQ) was defined as the mean blank absorbance plus 10 times the S.D. Assay precision was evaluated by calculating the coefficient of variation (CV) for intra-assay and inter-assay reproducibility using N. gonorrhoeae ATCC 49926 measurements.

For cross-reactivity testing, signal-to-cutoff (S/CO) ratios were calculated, with the cutoff defined as the mean blank absorbance plus 3 times the S.D. Samples with S/CO ratios ≥1 were classified as N. gonorrhoeae-positive, while those below this threshold were considered negative. Statistical differences among microorganisms were analyzed using one-way ANOVA, followed by Dunnett’s multiple comparisons test to assess pairwise differences between N. gonorrhoeae and each non-target organism.

For body fluid simulation experiments, calibration curves were generated by linear regression of absorbance against bacterial concentration. Linearity was evaluated based on the coefficient of determination (R²) and the slope of the regression.

To compare the diagnostic performance between the ultrasensitive ELISA and pNPP ELISA, we constructed contingency tables. Sensitivity, specificity, and overall agreement were calculated, and group differences were analyzed using Fisher’s exact probability test.

All statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, CA, U.S.A.). A p value <0.05 was considered statistically significant.

RESULTS

Feasibility and Calibration Curve Analysis of the Ultrasensitive ELISA for Detecting Neisseria gonorrhoeae

Calibration curves for N. gonorrhoeae ATCC 49926 (5000 to 500000 CFU/mL) were independently established by 3 investigators. For all measurements, thio-NADH absorbance at 405 nm was recorded after 45 min of thio-NAD cycling. The first investigator obtained a linear calibration curve of y = 4 × 10⁻⁷x, with an R² value of 0.99. The calculated LOD was 10607 CFU/mL. The intra-assay CV was 6% at 500000 CFU/mL and 6% at 5000 CFU/mL. The second investigator reported a calibration curve of y = 4 × 10⁻⁷x, also with an R² value of 0.99. The corresponding LOD was 17677 CFU/mL. The intra-assay CV was 2% at 500000 CFU/mL and 2% at 5000 CFU/mL. The third investigator obtained a calibration curve of y = 3 × 10⁻⁷x, with an R² value of 0.99 and an LOD of 16329 CFU/mL. The intra-assay CV was 7% at 500000 CFU/mL and 2% at 5000 CFU/mL. Based on measurements at 500000 and 5000 CFU/mL, inter-assay reproducibility across the 3 investigators yielded CVs of 11 and 2%, respectively, confirming consistent performance across operators (Fig. 1).

Fig. 1. Calibration Curves for Detecting N. gonorrhoeae Using the TN-cyclon^TM Assay

Representative linear calibration curves obtained in 3 independent experiments (A–C). Serial dilutions of N. gonorrhoeae ATCC 49926 were analyzed with the TN-cyclon^TM ultrasensitive ELISA, and absorbance values at 405 nm were recorded after 45 min of thio-NAD cycling. Each panel shows the linear regression equation, R², and LOD. All data points represent the mean of triplicate measurements. ELISA: enzyme-linked immunosorbent assay; LOD: limit of detection.

Evaluation of Cross-Reactivity with Eight Pathogens

To evaluate the assay specificity, cross-reactivity was evaluated against a panel of microorganisms. At 50000 CFU/mL, N. gonorrhoeae exhibited an S/N ratio of 2.46 ± 0.24 (n = 3). In contrast, all non-target microorganisms yielded S/N ratios <1.0. The observed means were as follows: Proteus vulgaris (0.98 ± 0.00, n = 2), Pseudomonas aeruginosa (0.91 ± 0.05, n = 2), Klebsiella pneumoniae (0.94 ± 0.01, n = 2), Staphylococcus aureus (0.91 ± 0.02, n = 3), Acinetobacter baumannii (0.97 ± 0.01, n = 2), Proteus mirabilis (0.72 ± 0.01, n = 3), Streptococcus equi ssp. (0.73 ± 0.01, n = 3), and Candida albicans (0.75 ± 0.01, n = 3) (Fig. 2). One-way ANOVA demonstrated a statistically significant difference among groups (p < 0.0001). Post hoc analysis with Dunnett’s test confirmed that specificity was significantly greater for N. gonorrhoeae than for all other organisms tested (all adjusted p’s < 0.0001). These findings indicate that the ultrasensitive ELISA has high analytical specificity for N. gonorrhoeae.

Fig. 2. Cross-Reactivity Analysis of the TN-cyclon^TM Assay

The assay was evaluated against a panel of microorganisms commonly found in clinical specimens, including Proteus vulgaris, Pseudomonas aeruginosa, Klebsiella pneumoniae, Staphylococcus aureus, Acinetobacter baumannii, Proteus mirabilis, Streptococcus equi subsp., and Candida albicans. The N. gonorrhoeae ATCC 49926 strain served as the positive control. Bacterial and fungal suspensions were adjusted to 5 × 10⁵ CFU/mL, and the results are expressed as S/N ratios relative to the blank. All non-target organisms yielded S/N values <1, whereas N. gonorrhoeae exhibited significantly elevated signals (***p < 0.001, one-way ANOVA with Dunnett’s post-hoc test). Data represent mean ± S.D. from n = 2–3 replicates per organism. S/N: signal-to-noise.

Application of Simulated Body Fluid to Test the Performance of Ultrasensitive and Traditional ELISAs

To assess the influence of different biologic matrices on the assay performance, calibration curves for N. gonorrhoeae detection were established in simulated urine, vaginal fluid, saliva, and the standard buffer (0.1% BSA in TBS). Absorbance was measured after 40 min of cycling. The results are presented in Fig. 3. Each concentration point was measured in triplicate. All conditions yielded linear relationships with consistent slopes of 2 × 10⁻⁷. The highest linearity was observed in simulated saliva (R² = 0.976, LOD = 12247 CFU/mL), followed by vaginal fluid (R² = 0.961, LOD = 21213 CFU/mL) and the standard buffer (R² = 0.922, LOD = 7500 CFU/mL). Simulated urine showed the lowest linearity (R² = 0.845) with an LOD of 18708 CFU/mL. These findings indicate that performance of the ultrasensitive ELISA was reliable across different matrices, although sensitivity was slightly reduced in urine and vaginal fluid relative to the standard buffer.

Fig. 3. Calibration Curves of the TN-cyclon^TM in Simulated Body Fluids

The N. gonorrhoeae ATCC 49926 was serially diluted in simulated urine (red), vaginal fluid (blue), saliva (purple), or standard buffer (green), and analyzed by the TN-cyclon^TM. Linear regression analysis was performed, and the corresponding regression equation, R², and LOD are shown for each condition. All assays were performed in triplicate, and mean values are presented. LOD: limit of detection.

Comparison of the pNPP ELISA and Ultrasensitive ELISA for Clinical Isolates

To directly compare the performance of the ultrasensitive ELISA with that of the conventional pNPP-based ELISA, both assays were first evaluated using N. gonorrhoeae ATCC 49926 at 50000 CFU/mL. The ultrasensitive ELISA assay produced a steadily increasing absorbance signal over time, whereas the pNPP ELISA showed negligible change (Fig. 4). At 60 min, the ultrasensitive ELISA reached an absorbance of 0.377 ± 0.021, compared with 0.007 ± 0.004 (p < 0.0001) for the pNPP ELISA. The LOD of ultrasensitive ELISA ranged from 10607 to 17677 CFU/mL (Fig. 1), whereas the conventional pNPP ELISA reached an LOD of approximately 565000 CFU/mL (Supplementary Fig. S1), corresponding to a >30-fold improvement in detection sensitivity. This finding underscores the strong signal amplification capability of the thio-NAD cycling system.

Fig. 4. Comparison of Signal Development between the TN-cyclon^TM and Conventional pNPP ELISA

N. gonorrhoeae ATCC 49926 (50000 CFU/mL) was tested in parallel using TN-cyclon^TM (red) and pNPP ELISA (blue). Absorbance at 405 nm was recorded every 5 min for 120 min. The TN-cyclon^TM produced a steadily increasing signal, reaching 0.377 ± 0.021 at 60 min, whereas the pNPP assay yielded only 0.007 ± 0.004 (p < 0.0001). Data are presented as mean ± S.D. (n = 3). ELISA: enzyme-linked immunosorbent assay.

We next extended this comparison to 90 clinical isolates of N. gonorrhoeae, all confirmed to be culture-positive prior to testing. The ultrasensitive ELISA identified 71 of 90 isolates as positive (78.9%) and 19 as negative (21.1%), whereas the pNPP ELISA detected only 27 as positive (30.0%) and 63 as negative (70.0%) (Fig. 5), indicating a higher detection rate for the ultrasensitive ELISA than for the pNPP ELISA. Fisher’s exact probability test confirmed that the difference in detection outcomes between the 2 assays was significant (p = 0.0005, two-sided). Together, these findings demonstrate that the thio-NAD cycling method provides markedly improved sensitivity compared with the conventional pNPP ELISA when applied to both reference strains and clinical isolates of N. gonorrhoeae.

Fig. 5. Comparison of the TN-cyclon^TM and Conventional pNPP ELISA in Detecting N. gonorrhoeae

A total of 90 N. gonorrhoeae ATCC 49926 cultures were tested in parallel using the TN-cyclon^TM and conventional pNPP ELISA. The TN-cyclon^TM identified 71 positive and 19 negative samples, whereas the pNPP ELISA detected 27 positive and 63 negative samples. Statistical comparison by Fisher’s exact probability test showed a significant difference between the 2 assays (p = 0.0005, ***p < 0.001). ELISA: enzyme-linked immunosorbent assay.

DISCUSSION

In the present study, we evaluated the ability of the TN-cyclon^TM ultrasensitive ELISA to detect N. gonorrhoeae and demonstrated several key findings. The assay produced reproducible calibration curves, showed no cross-reactivity with a panel of common pathogens, maintained performance in simulated body fluids, and achieved higher detection rates than the conventional pNPP-based ELISA when tested on frozen clinical isolates obtained from genital swabs. Earlier enzyme immunoassays, including the Abbott Gonozyme, demonstrated good diagnostic performance in symptomatic male urethral specimens. The sensitivity and specificity of these earlier immunoassays, however, were more variable in cervical samples.²¹⁾ In a family-planning clinic study, the cervical sensitivity and specificity of the Gonozyme were 87.2 and 89.1% respectively, with a positive predictive value of only 37.2% at a prevalence of 6.9%.²²⁾ The low positive predictive value highlights a key drawback of earlier antigen assays. In populations with a lower prevalence, diagnostic methods with only moderate sensitivity and specificity could lead to a higher rate of false-positive results, substantially reducing their clinical utility and application.

Most earlier ELISA investigations rarely included extragenital sites or asymptomatic infections, and their performance data were derived mainly from freshly collected urethral specimens of symptomatic men. The overall sensitivity of 78.9% observed in the present study was lower than that reported in earlier ELISA-based studies, such as those using the Abbott Gonozyme assay.^21,23) The Gonozyme and other commercial enzyme immunoassays have been discontinued, however, precluding direct comparison of their performance with that of our method. Their withdrawal from the market likely reflects limited reproducibility and variable clinical performance due to the antigen/phase variation of N. gonorrhoeae,²⁴⁾ which contributed to their eventual replacement by NAATs. Because the immunogens used to generate the antibodies in this study were a pool of UV-inactivated N. gonorrhoeae cells, further development using antibodies against a highly conserved and specific N. gonorrhea antigen may improve the sensitivity. Jen et al. reported that the N. gonorrhea surface protein methionine sulfoxide reductase (MsrA/B) is immunogenic,²⁵⁾ suggesting that it may be a useful target for further development of ELISA-related techniques.

NAATs exhibit exceptional analytical performance, with limits of detection often below 1000 CFU/mL or even down to a few genome copies per reaction, and remain the current diagnostic standard.^26,27) Clinical evaluations consistently report sensitivity in the range of 95–100% and specificity exceeding 98% across genital and many extragenital specimens.^28,29) The US Center for Disease Control and Prevention recommend NAATs as the preferred method for routine detection of N. gonorrhoeae, emphasizing their superior sensitivity compared with culture and antigen detection.¹⁰⁾ Nevertheless, NAATs are costly and require specialized instrumentation, strict quality control, and trained personnel, which limits their widespread application in resource-constrained settings.¹²⁾ In the present study, the ultrasensitive ELISA achieved limits of detection ranging from 10600 to 17700 CFU/mL across operators. While this threshold is higher than that of NAATs, it is sufficient for the bacterial loads typically found in clinical specimens, with a recent study quantifying mean gonococcal loads of approximately 32000 CFU/mL in urine and 20000 CFU/mL in vaginal swabs.³⁰⁾ These findings suggest that the ultrasensitive ELISA can reliably detect most urogenital infections. Its relatively simple workflow and minimal equipment requirements further support its potential as a cost-effective screening tool in low-resource environments where the use of NAATs is not feasible.

The cross-reactivity results further highlight the specificity of ultrasensitive ELISA. Although cross-reaction with commensal Neisseria was not tested, the absence of reactivity with other clinically relevant bacteria and fungi supports its analytical reliability. Previous studies reported that several non-gonococcal Neisseria species, including N. meningitidis, N. lactamica, N. subflava, N. cinerea, and N. flavescens, can yield false-positive results in some NAAT platforms, especially when testing pharyngeal specimens.^10,12,13) Clinical studies have also reported false positives from commercial assays, emphasizing the importance of evaluating commensal cross-reactivity in diagnostic development.³¹⁾ Expanding the cross-reactivity panel to include these related species in future studies will be essential for establishing the ultrasensitive ELISA as a diagnostic alternative that complements NAATs.

Evaluation of simulated body fluids suggests that the ultrasensitive ELISA tolerates diverse sample environments. Previous reports demonstrated that body fluids can interfere with immunoassay performance. For example, urine often exhibits variable recovery and nonlinearity due to matrix effects.³²⁾ Endogenous peroxidases in sputum may catalyze horseradish peroxidase substrates and generate false-positive signals in ELISAs.³³⁾ In addition, cervicovaginal secretions can contain mucus and mucins that bind analytes or alter antigen–antibody interactions.³⁴⁾ In our simulated matrices, the urine formulation contained urea and high ionic strength, which can weaken antibody–antigen interactions, whereas the simulated vaginal fluid included mucin components capable of increasing background or partially masking target antigens. These matrix characteristics likely contributed to the higher detection thresholds observed for urine and vaginal fluid compared with saliva. In the present study, whereas urine and vaginal fluid showed modestly higher detection thresholds, saliva and buffer conditions preserved strong linearity. The consistent performance of the ultrasensitive ELISA across multiple matrices suggests that the assay can be applied to a broad range of clinical specimen types. This flexibility is particularly important for extragenital testing, which contributes substantially to case detection but continues to present diagnostic challenges.³⁵⁾

The present pilot study has several limitations that should be considered when interpreting the findings. One limitation is that all 90 clinical isolates were previously cultured and stored frozen, rather than fresh specimens, which may not fully reflect real-time clinical conditions. Additionally, the study included only male urogenital samples and did not assess female or extragenital specimens, which are clinically important yet diagnostically challenging sites. The absence of a direct comparison with NAATs further limits conclusions regarding the assay’s relative performance within routine diagnostic workflows. Finally, only a single pair of monoclonal antibodies targeting whole-cell antigens was evaluated; alternative antibody combinations may yield different analytical characteristics. These limitations indicate that further validation is required before clinical application.

Future work should expand upon the current findings in several important directions. First, larger prospective studies including female and extragenital specimens will be essential to confirm the broad applicability of the ultrasensitive ELISA. A good clinical trial is also mandatory. Second, optimizing assay conditions to minimize borderline results could further improve robustness in clinical practice. In parallel, improving antibody performance remains an important avenue for enhancing sensitivity. The current assay relied on commercially available monoclonal antibodies raised against whole-cell N. gonorrhoeae, which may not target the most abundant or diagnostically stable epitopes. Ongoing work in our laboratory focuses on developing monoclonal antibodies directed against N. gonorrhoeae methionine sulfoxide reductase AB (MsrAB) proteins, which may offer stronger binding affinity and more efficient signal generation. Third, direct comparison with NAATs across diverse patient populations would better clarify the complementary role of the ultrasensitive ELISA within diagnostic algorithms. Together, these next steps will help establish whether the TN-cyclon^TM can serve not only as a reliable alternative to conventional ELISA but also as a practical adjunct to NAATs, particularly in resource-limited settings.³⁶⁾

CONCLUSION

Application of an ultrasensitive ELISA based on thio-NAD cycling (TN-cyclon^TM) for N. gonorrhoeae detection yielded a lower detection limit and better sensitivity than the conventional pNPP ELISA when testing isolates. The method has good specificity, and testing simulated fluids yielded promising results. These findings suggest the high potential of this ultrasensitive ELISA as a rapid, cost-effective, and scalable diagnostic platform, particularly for use in settings where molecular testing is not widely accessible.

Acknowledgments

The authors thank Teruki Yoshimura for providing the A3P.

DECLARATIONS

Funding

This study was supported by the National Science and Technology Council (114-2327-B-037-003 (PLL) and 114-2320-B-715 -005 (TYY)), Kaohsiung Medical University Hospital (KMUH110-0R26 (PLL)), Kaohsiung Medical University Research Center Grant, Center for Liquid Biopsy and Cohort Research (KMU-TC112B04 (PLL)), SPRING from the Japan Science and Technology Agency (JPMJSP2128 (PKC)), and a collaborative research grant from BioPhenoMA Inc. (EI). The funders had no role in the study design; data collection, analysis, or interpretation; manuscript preparation; or the decision to publish the study.

Author Contributions

Conceptualization, TYY, PLL, and EI; Methodology, PKC, TYY, and EI; Formal analysis, PKC and TYY; Investigation, PKC, TYY, YWC, and YCC; Resources, JWL and HJT; Writing—original draft, PKC and TYY; Writing—review and editing, JWL, HJT, YWC, YCC, PLL, and EI; Funding acquisition, PKC, TYY, PLL, and EI.

Conflict of Interest

E.I. has received an honorarium from BioPhenoMA Inc. All other authors declare no conflict of interest.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1)  Darville T. Pelvic inflammatory disease due to Neisseria gonorrhoeae and Chlamydia trachomatis: immune evasion mechanisms and pathogenic disease pathways. J. Infect. Dis., 224 (Suppl. 2), S39–S46 (2021).
• 2)  Wei J, Huang H, Fan L. Global burden of female infertility attributable to sexually transmitted infections and maternal sepsis: 1990–2021 and projections to 2050. Sci. Rep., 15, 15189 (2025).
• 3)  WHO. “Gonorrhoea (Neisseria gonorrhoeae infection).”: <https://www.who.int/news-room/fact-sheets/detail/gonorrhoea-(neisseria-gonorrhoeae-infection)>, accessed 11 November, 2025.
• 4)  WHO. “Global progress report on HIV, viral hepatitis and sexually transmitted infections, 2021.”: <https://www.who.int/publications/i/item/9789240027077>, accessed 11 November, 2025.
• 5)  Barbee LA. Preparing for an era of untreatable gonorrhea. Curr. Opin. Infect. Dis., 27, 282–287 (2014).
• 6)  Wajima T, Tanaka E, Uchiya K. Unique and ingenious mechanisms underlying antimicrobial resistance and spread of Haemophilus influenzae. Biol. Pharm. Bull., 48, 205–212 (2025).
• 7)  Unemo M, Shafer WM. Antimicrobial resistance in Neisseria gonorrhoeae in the 21st century: past, evolution, and future. Clin. Microbiol. Rev., 27, 587–613 (2014).
• 8)  Unemo M, Dillon JA. Review and international recommendation of methods for typing Neisseria gonorrhoeae isolates and their implications for improved knowledge of gonococcal epidemiology, treatment, and biology. Clin. Microbiol. Rev., 24, 447–458 (2011).
• 9)  Nash EE, Pham CD, Raphael B, Learner ER, Mauk K, Weiner J, Mettenbrink C, Thibault CS, Fukuda A, Dobre-Buonya O, Black JM, Johnson K, Sellers K, Schlanger K. Impact of anatomic site, specimen collection timing, and patient symptom status on Neisseria gonorrhoeae culture recovery. Sex. Transm. Dis., 48 (Suppl. 2), S151–S156 (2021).
• 10)  Centers for Disease Control and Prevention. Recommendations for the laboratory-based detection of Chlamydia trachomatis and Neisseria gonorrhoeae—2014. MMWR Recomm. Rep., 63 (RR-02), 1–19 (2014).
• 11)  Meyer T, Buder S. The laboratory diagnosis of Neisseria gonorrhoeae: current testing and future demands. Pathogens, 9, 91 (2020).
• 12)  Whiley DM, Tapsall JW, Sloots TP. Nucleic acid amplification testing for Neisseria gonorrhoeae: an ongoing challenge. J. Mol. Diagn., 8, 3–15 (2006).
• 13)  Tabrizi SN, Unemo M, Limnios AE, Hogan TR, Hjelmevoll SO, Garland SM, Tapsall J. Evaluation of six commercial nucleic acid amplification tests for detection of Neisseria gonorrhoeae and other Neisseria species. J. Clin. Microbiol., 49, 3610–3615 (2011).
• 14)  Nakaishi K, Watabe S, Kitagawa T, Ito E. Immunochromatographic detection of MPB64 secreted from active BCG by heating: toward same-day diagnosis of tuberculosis. Biotechniques, 66, 240–242 (2019).
• 15)  Chen PK, Lu PL, Ito E, Yang TY. Enzyme-linked immunosorbent STI assays: development, current status and future perspective. J. Microbiol. Immunol. Infect., (2025), in press.
• 16)  Kobayashi Y, Kyosei Y, Ogawa R, Okita K, Yoshimura T, Ito E. Ultrasensitive protein-level detection for respiratory infectious viruses. Front. Immunol., 15, 1445771 (2024).
• 17)  Okita K, Arita H, Sudo K, Yoshimura T, Ito E. Establishment of an assay with ultrahigh sensitivity for detecting sEV-derived PD-L1 as a serum biomarker for lung cancer—a pilot study using TN-cyclon^TM. Curr. Issues Mol. Biol., 47, 564 (2025).
• 18)  Bioscience Q. “Anti-Neisseria gonorrhoeae Antibody (15802).”: <https://www.qedbio.com/product/neisseria-gonorrhoeae-2/>, accessed 11 November, 2025.
• 19)  Bioscience Q. “Anti-Neisseria gonorrhoeae Antibody (15803).”: <https://www.qedbio.com/product/neisseria-gonorrhoeae-3/>, accessed 11 November, 2025.
• 20)  Chen PK, Chang JH, Ke LY, Kao JK, Chen CH, Yang RC, Yoshimura T, Ito E, Tsai JJ. Advanced detection method for dengue NS1 protein using ultrasensitive ELISA with thio-NAD cycling. Viruses, 15, 1894 (2023).
• 21)  Jephcott AE. Microbiological diagnosis of gonorrhoea. Genitourin. Med., 73, 245–252 (1997).
• 22)  Nachamkin I, Sondheimer SJ, Barbagallo S, Barth S. Detection of Neisseria gonorrhoeae in cervical swabs using the Gonozyme^TM enzyme immunoassay. Clinical evaluation in a university family planning clinic. Am. J. Clin. Pathol., 82, 461–465 (1984).
• 23)  Schachter J, McCormack WM, Smith RF, Parks RM, Bailey R, Ohlin AC. Enzyme immunoassay for diagnosis of gonorrhea. J. Clin. Microbiol., 19, 57–59 (1984).
• 24)  Anderson MT, Seifert HS. Phase variation leads to the misidentification of a Neisseria gonorrhoeae virulence gene. PLoS One, 8, e72183 (2013).
• 25)  Jen FE, Semchenko EA, Day CJ, Seib KL, Jennings MP. The Neisseria gonorrhoeae methionine sulfoxide reductase (MsrA/B) is a surface exposed, immunogenic, vaccine candidate. Front. Immunol., 10, 137 (2019).
• 26)  Golparian D, Hellmark B, Unemo M. Analytical specificity and sensitivity of the novel dual-target GeneProof Neisseria gonorrhoeae PCR kit for detection of N. gonorrhoeae. Acta Pathol. Microbiol. Scand. Suppl., 123, 955–958 (2015).
• 27)  van Niekerk JM, van der Veer B, Hoebe C, van de Bovenkamp J, van Herk C, van Loo IHM, van Alphen LB, Wolffs PFG. Despite excellent test characteristics of the cobas 4800 CT/NG assay, detection of oropharyngeal Chlamydia trachomatis and Neisseria gonorrhoeae remains challenging. J. Clin. Microbiol., 59, e02137–20 (2021).
• 28)  Adamson PC, Klausner JD. Diagnostic tests for detecting Chlamydia trachomatis and Neisseria gonorrhoeae in rectal and pharyngeal specimens. J. Clin. Microbiol., 60, e0021121 (2022).
• 29)  Bromhead C, Miller A, Jones M, Whiley D. Comparison of the cobas 4800 CT/NG test with culture for detecting Neisseria gonorrhoeae in genital and nongenital specimens in a low-prevalence population in New Zealand. J. Clin. Microbiol., 51, 1505–1509 (2013).
• 30)  van der Veer B, Hoebe C, Dukers-Muijrers N, van Alphen LB, Wolffs PFG. Men and women have similar Neisseria gonorrhoeae bacterial loads: a comparison of three anatomical sites. J. Clin. Microbiol., 58, e01171–20 (2020).
• 31)  Upton A, Bromhead C, Whiley DM. Neisseria gonorrhoeae false-positive result obtained from a pharyngeal swab by using the Roche cobas 4800 CT/NG assay in New Zealand in 2012. J. Clin. Microbiol., 51, 1609–1610 (2013).
• 32)  Chatziharalambous D, Lygirou V, Latosinska A, Stravodimos K, Vlahou A, Jankowski V, Zoidakis J. Analytical performance of ELISA assays in urine: one more bottleneck towards biomarker validation and clinical implementation. PLOS ONE, 11, e0149471 (2016).
• 33)  Kim LH, Plaza K, Thomas SR, Draijer C, Radford K, Peters-Golden M, Mukherjee M, Nair P. Endogenous peroxidases in sputum interfere with horse-radish peroxidase-based ELISAs. J. Immunol. Methods, 454, 76–79 (2018).
• 34)  Fichorova RN, Richardson-Harman N, Alfano M, et al. Biological and technical variables affecting immunoassay recovery of cytokines from human serum and simulated vaginal fluid: a multicenter study. Anal. Chem., 80, 4741–4751 (2008).
• 35)  Friedman DS, O’Byrne P. Extragenital testing increases case detection of gonorrhea and chlamydia: The impact of implementing nucleic acid amplification testing. Can. Commun. Dis. Rep., 46, 285–291 (2020).
• 36)  Chen PK, Lin KH, Ke GM, Yang SC, Lin CJ, Lee CH, Yang RC, Hsu CS, Yang TY, Hasunuma M, Sudo K, Ito E, Kao JK. The role of P30 adhesin detected by ultrasensitive ELISA in the clinical diagnosis and management of Mycoplasma pneumoniae. Diagn. Microbiol. Infect. Dis., 114, 117192 (2026).