High-Throughput Multiplexed Antibody Fc Profiling and Biomarker Discovery for Endemic Infections Using a Flexible And Reconfigurable Integrated Microfluidic Microarray Platform

Introduction: Diagnosis of endemic infectious diseases is impeded by lack of sensitive and specific biomarkers. Direct pathogen detection via microscopy or biomarker-based tests remains complex and costly for many infections (e.g. Tuberculosis) due to low target abundance, complex sample acquisition and preparation (e.g. sputum) and varying pathogen sequestration and dynamics over its life cycle. Antibody-based tests, on the other hand are inexpensive, sample-sparing and highly sensitive due to their amplified amounts produced by the host immune system. However, they cannot distinguish current from past infection and thus have low specificity, especially for endemic infections. While antigen-specific antibody titer alone is inadequate for diagnosis, antibody Fc region properties, especially subclass, glycosylation and Fc receptor affinity has been recently shown to correlate with disease state in various infections (Figure 1). Current assays for Fc properties, such as Mass Spectrometry (MS), are sample and labor-intensive especially for large numbers of antigen-specific antibodies. To achieve sample-sparing, high-throughput antibody Fc biomarker discovery at reduced cost for endemic infectious diseases, we have developed a simple and reconfigurable multiplexed microfluidic chip integrating antigen microarrays that simultaneously tests 14 samples against 25 antigens and 4 Fc probes in one experiment, generating >1000 antibody couple Fab and Fc measurements per chip for downstream machine-learning based analytics and biomarker discovery. Here we report the development and operation of this chip and its use in two applications: antibody profiling in COVID-19 and discovery of biomarkers for Schistosomiasis, a helminth-mediated neglected tropical disease with high burden in tropical resource-poor settings.

Materials and

Methods: Commercial Poly-L-Lysine slides were laser-engraved to create alignment marks. Protein antigens were diluted in 5% glycerol at 100µg/ml then printed on the slides with sciFLEXARRAYER microarray printer (Figure 2A) and stored at 4°C until use. To perform the assay, the slide was blocked in 1% Bovine Serum Albumin for 40 minutes. After washing and drying, two laser-cut layers of PDMS were attached on the top of the slide to form vertical channels for 14 samples (Figure 2A). Alignment was performed visually using the alignment marks aided by 3D printed fixtures. 15µl of diluted samples were added to each channel (Figure 2B) and incubated for 1 hour. After PDMS removal, washing and drying, another two layers of PDMS were attached on the top of the slide to form horizontal channels for 4 Fc probes (Fig2A). 110µl of fluorescent probes were added to each channel (Figure 2C) and incubated for 30 minutes. After PDMS removal, washing and drying, the slide was scanned with Innoscan microarray scanner and image was analyzed by ImageJ. COVID-19 samples were purchased from RayBiotech. Healthy serum was purchased from Millipore Sigma. Glycosylation modification was performed by incubating the serum with α2-3,6,7,8,9 Neuraminidase overnight at 37°C to remove sialic acid on Fc region. Schistosomiasis samples were provided by the CDC. PE/Rhodamine-labeled Fc probes targeting antibody glycans (sialic acid)/FcReceptor2A/IgG/IgM/IgA were purchased from VectorLab/SouthernBiotech. PDMS sheets were purchased from Green’s Rubber.

Results, Conclusions, and Discussions: Fluidic integrity of the assembled chip (Figure 2A) was verified by comparing mean fluorescence intensity (MFI) (Figure 3A, B & C) between positive and negative control antigens printed in adjacent channels. Improvement in morphology and repeatability of printing was observed by adding glycerol, but with decreasing MFI (Figure 3D). Thus, an optimal glycerol concentration (5%) was chosen (Figure 3E). To test immunoassay performance, serial dilutions of Anti-SARS-CoV-2 Spike monoclonal IgG were added to a chip printed with SARS-CoV-2 antigens. Sample dilution-dependent variation in Spike and Spike-RBD-specific IgG was observed, while N-specific IgG remained low (Figure 3F), showing quantitative and specific antibody titer measurement. COVID-19 patient, vaccinated and pre-pandemic healthy serum samples were next tested against 8 SARS-CoV-2 antigens (Figure 3G), revealing different antigen-specific IgG fingerprints for them.
For glycoprofiling, a fluorescent-tagged lectin (SNA) was chosen as a sialic acid probe (Figure 4A). A linear relationship between percent sialylation of antibodies, generated via in-vitro enzymatic glycoengineering, and corresponding SNA-binding MFI was observed (Figure 4B). This correlated well with gold-standard MS measurement (Figure 4B) and a bead-based binding assay (Figure 4C). Serum containing either unmodified antibodies or asialated antibodies was tested against 4 antigens showing IgG MFI remains unchanged while SNA MFI decrease significantly for asialated antibodies (Fig4D).
Finally, the chip was used to test Schistosomiasis patient serum samples (n=12), including three sample classes: current infection (SEA+Egg+), past infection (SEA+Egg-) and uninfected controls (SEA-Egg-), against 7 Schistosoma Mansoni antigens and 4 Fc probes (IgG, IgG4 and FcR2A, Figure 5A). We found that SmSEA-specific IgG4 is a biomarker (Fig5B) that indicates current infection. Although binding with FcR2A was found to not differ between sample classes in this case, the MFI was highly correlated with that measured in a bead-based assay (Fig5C).
Overall, we demonstrated that the chip can be used to discover coupled antibody Fab and Fc biomarkers including antigen-specificity, titer, subclass, glycosylation and Fc receptor binding in a high-throughput, sample-sparing and cost-effective manner. The use of rapid prototyping tools, such as laser cutting, instead of cleanroom-base fabrication and the flexible and reconfigurable format of the chip enhance its applicability in resource-poor settings.

Acknowledgements (Optional): We thank Dr Lei Li for providing the microarray printer, and the Emory Glycomics and Molecular Interactions Core (EGMIC)(RRID:SCR_023524) for providing the microarray scanner.
Key citations include:
Kim, D., Biancon, G., Bai, Z., VanOudenhove, J., Liu, Y., Kothari, S., Gowda, L., Kwan, J.M., Buitrago‐Pocasangre, N.C., Lele, N. and Asashima, H., 2023. Microfluidic Immuno‐Serolomic Assay Reveals Systems Level Association with COVID‐19 Pathology and Vaccine Protection. Small Methods, 7(10), p.2300594.
Swank, Z., Michielin, G., Yip, H.M., Cohen, P., Andrey, D.O., Vuilleumier, N., Kaiser, L., Eckerle, I., Meyer, B. and Maerkl, S.J., 2021. A high-throughput microfluidic nanoimmunoassay for detecting anti–SARS-CoV-2 antibodies in serum or ultralow-volume blood samples. Proceedings of the National Academy of Sciences, 118(18), p.e2025289118.
Peddireddy, S.P., Rahman, S.A., Cillo, A.R., Vijay, G.M., Somasundaram, A., Workman, C.J., Bain, W., McVerry, B.J., Methe, B., Lee, J.S. and Ray, P., 2022. Antibodies targeting conserved non-canonical antigens and endemic coronaviruses associate with favorable outcomes in severe COVID-19. Cell Reports, 39(13).