Microparticle-tagged image-based cell counting (ImmunoSpin) for CD4 + T cells

Abstract

Affordable point-of-care (POC) CD4 + T lymphocyte counting techniques have been developed as alternatives to flow cytometry-based instruments caring for patients with human immunodeficiency virus (HIV)-1. However, POC CD4 enumeration technologies can be inaccurate. Here, we developed a microparticle-based visual detector of CD4 + T lymphocytes (ImmunoSpin) using microparticles conjugated with anti-CD4 antibodies, independent of microfluidic or fluorescence detection systems. Visual enumeration of CD4 + T cells under conventional light microscope was accurate compared to flow cytometry. Microparticle-tagged CD4 + T cells were well-recognized under a light microscope. ImmunoSpin showed very good precision (coefficients of variation of ImmunoSpin were ≤ 10%) and high correlation with clinical-grade flow cytometry for the enumeration of CD4 + T cells (y = 0.4232 + 0.9485 × for the %CD4 + T cell count, R² = 0.99). At thresholds of 200 and 350 cells/µL, there was no misclassification of the ImmunoSpin system compared to the reference flow cytometry. ImmunoSpin showed clear differential classification of CD4 + T lymphocytes from granulocytes and monocytes. Because non-fluorescence microparticle-tags and cytospin slides are used in ImmunoSpin, they can be applied to an automatic digital image analyzer. Slide preparation allows long-term storage, no analysis time limitations, and image transfer in remote areas.

Keywords: CD4 + T cell; Human immunodeficiency virus; Image-based cell counting; ImmunoSpin; Microparticle.

Fig. 1

Process illustration for microparticle-tagged image-based cell counting (ImmunoSpin). Representation of specific labeling of CD4 + T cells from whole blood (WB) with the anti-CD4 antibody–microparticle complex and their subsequent RBC lysis and washing step (optional). Then, cytospin preparations were established for the enrichment of leukocytes, including anti-CD4 antibody–microparticle-labeled lymphocytes. Cytospin slides were counterstained with Wright (or methylene blue) stain. At least two cytospin slides were prepared from 100 µL WB and were analyzed

Fig. 2

Microscopy images indicating the CD4 + T cells by anti-CD4 antibody-conjugated microparticles and a granulocyte (dashed box) in the background of other leukocytes (above). Several morphologies of microparticle-tagged CD4 + T cells are illustrated (below)

Fig. 3

Comparison of ImmunoSpin with flow cytometry. Total WB samples (n = 45) were used for method comparison. CD4 + T cell proportions (above) and absolute CD4 T cell counts (below) measured on the ImmunoSpin were compared to the corresponding CD4 + T cell counting from the flow cytometer BD FACSCanto II (BD Biosciences). The regression equation was y = 0.4232 + 0.9485 × for %CD4 + T cell count (R² = 0.99) and y = 4.1001 + 0.9590 × for absolute CD4 + T cell count (R² = 0.99). The slopes were 0.95 (95%CI, 0.93–0.97) for %CD4 + T cell count and 0.96 (95%CI, 0.94–0.99) for absolute CD4 + T cell count. Absolute differences calculated with Bland–Altman plot analysis are plotted