A Novel Point-of-Care Approach for Cytokine Detection in Latent Tuberculosis Using Enzymatic Metallization: Cost-Effective and High-Throughput Method for Resource-Limited Settings

Introduction: Tuberculosis (TB), caused by Mycobacterium tuberculosis, affects nearly 2 billion people, of which 10 million are active cases. There are methods in place to screen active cases, but 90-95% of worldwide cases are latent (LTB) – meaning that the patients are asymptomatic1.

Currently, the two most widely used methods for LTB diagnosis are the tuberculin skin test (TST) and the interferon-gamma release assay (IGRA). The TST is performed by injecting tuberculin-purified protein derivative into the skin and monitoring the presence of a localized reaction after 48-72 hours2. Although this method is low-cost and can be performed at the point-of-care, it is prone to both false positives and false negatives. Alternatively, the IGRA is the gold standard detection method for LTB, and is performed by measuring the release of interferon-gamma (IFN-γ) by T-cells in a blood sample. This method is sensitive, but is hindered by its high equipment cost, and need for specialized laboratories, making it impractical in resource-limited settings3. Alongside IFN-γ, tumor necrosis factor alpha (TNF-α) and interleukin-2 (IL-2) have been found to be biomarkers of LTB4.

Thus, LTB diagnosis necessitates the development of a test combining the affordability and ease-of-use of the TST with the sensitivity of the IGRA. Here, we report an inexpensive, high-throughput multiplexed, and point-of-care (POC) method of detection for IFN-γ, TNF-α, and IL-2 from a single drop of serum using enzymatic metallization on inexpensive polystyrene substrates which can be accurately quantified using a cellphone camera.

Materials and

Methods: Anti-human IFN-γ, TNF-α, or IL-2 antibodies were immobilized in microwells each capable of holding up to 3 µL of solution, which were defined by laser-cut polydimethylsiloxane (PDMS) on a standard gamma sterilized Petri dish made of polystyrene (Fig. 1a). To minimize non-specific binding, the wells were initially treated with a blocking solution of 1% bovine serum albumin (BSA) in phosphate-buffered saline with 0.05% Tween-20 (PBST). Serial dilutions of the human IFN-γ, TNF-α, or IL-2 antigen were subsequently added to the wells. Following that, biotinylated anti-human IFN-γ, TNF-α, or IL-2 antibody was introduced, followed by streptavidin-poly-HRP. Biotinyl tyramide was then added to react with the poly-HRP, resulting in the generation and attachment of more biotin molecules onto the stack; onto this, more streptavidin-poly-HRP was added. The resulting amplified signal was then developed using enzymatic silver metallization and captured using a cellphone camera. Finally, the acquired images were subjected to analysis to determine the darkness of the metallization in each well by calculating the average grayscale value. This procedure was utilized for detecting recombinant cytokines, stimulated primary peripheral blood mononuclear cell (PBMC) supernatant, and clinical LTB patient stimulated PBMC supernatant samples.

Results, Conclusions, and Discussions: The device exhibited dark metallization at the highest cytokine concentration, with diminishing signals in serial dilutions. Based on the measured signal of diluting buffer control samples, the limit of detection (LOD) for recombinant IFN-γ, TNF-α, and IL-2 was found to be 44pg/mL, 11pg/mL, and 15.625 pg/mL, respectively (Fig 2a, 3a, 4a). From the stimulated PBMC supernatant, the assay was able to differentiate between those stimulated with phytohemagglutinin-L (PHA-L) and those that were unstimulated to dilutions of up to 128X (Fig. 2b, 3b, 4b). Additionally, IFN-γ, TNF-α, and IL-2 levels were measured from pooled and individual clinical LTB patient samples. For pooled LTB+ patient samples, there was generally good differentiation between the antigen-stimulated samples and positive/negative controls (Fig. 2c, 3c, 4c); for pooled LTB- samples, the antigen-stimulated samples were at the level of the negative control (Fig. 2d, 3d, 4d). Similar trends were seen in the individual LTB+ and LTB- samples (Fig. 2e-f, 3e-f, 4e-f). Preliminary cross-reactivity testing has shown minimal cross-reactivitiy and that the cytokines can be differentiated when tested in parallel (Fig. 5). The miniaturization of this device drastically reduces reagent and sample volumes, making it potentially 4X less costly than ELISA, and utilizing up to 33X less patient sample volume. Moreover, this POC method eliminates the need for expensive specialized equipment, relying solely on a cellphone for signal readout and quantification. This indicates promising potential for effective and affordable latent TB diagnosis in resource-limited settings.

In this study, we have introduced a novel POC approach for the detection of IFN-γ, TNF-α, and IL-2 in latent tuberculosis, offering a cost-effective and high-throughput method tailored for resource-limited settings. The device successfully demonstrated LODs in the low pg/mL range, already falling within the reference range of the gold-standard conventional ELISA. Notably, it utilizes significantly lower sample volumes and less expensive materials compared to conventional techniques, while eliminating the need for specialized laboratory equipment. With its potential to impact TB diagnosis in resource-limited settings, this novel method for cytokine detection holds significant promise in contributing to improved TB management and reducing the burden of this global health challenge.

Acknowledgements (Optional):

References:
1. World Health Organization. (n.d.). Tuberculosis (TB). World Health Organization. https://www.who.int/news-room/fact-sheets/detail/tuberculosis
2. Centers for Disease Control and Prevention. (2020, November 2). Tuberculin Skin Testing Fact Sheet. Centers for Disease Control and Prevention. https://www.cdc.gov/tb/publications/factsheets/testing/skintesting.htm
3. Herrera, V., et al. Clinical Infectious Diseases, 2011.
4. Day, C. L., et al. Mycobacterium tuberculosis-specific cytokine responses according to HIV status among household contacts of people with TB. Tuberculosis, 2023.