Novel flow cytometry approach to identify bronchial epithelial cells from healthy human airways

Abstract

Sampling various compartments within the lower airways to examine human bronchial epithelial cells (HBEC) is essential for understanding numerous lung diseases. Conventional methods to identify HBEC in bronchoalveolar lavage (BAL) and wash (BW) have throughput limitations in terms of efficiency and ensuring adequate cell numbers for quantification. Flow cytometry can provide high-throughput quantification of cell number and function in BAL and BW samples, while requiring low cell numbers. To date, a flow cytometric method to identify HBEC recovered from lower human airway samples is unavailable. In this study we present a flow cytometric method identifying HBEC as CD45 negative, EpCAM/pan-cytokeratin (pan-CK) double-positive population after excluding debris, doublets and dead cells from the analysis. For validation, the HBEC panel was applied to primary HBEC resulting in 98.6% of live cells. In healthy volunteers, HBEC recovered from BAL (2.3% of live cells), BW (32.5%) and bronchial brushing samples (88.9%) correlated significantly (p = 0.0001) with the manual microscopy counts with an overall Pearson correlation of 0.96 across the three sample types. We therefore have developed, validated, and applied a flow cytometric method that will be useful to interrogate the role of the respiratory epithelium in multiple lung diseases.

Figure 1. Flow cytometric (pseudo-color/smooth) plots showing the gating strategy validating the identification of HBEC from cultured pHBEC.

Cells were gated based on size and granularity using FSC-A vs SSC-A to eliminate debris and clumped cells. Single cells were sub-gated using fixable viability dye eFluor 450 and subsequently live cells were discriminated by the expression of CD45 APC-Cy7. Exclusion of CD45 positive cells was followed by the examination of double expression of pan-cytokeratin FITC and EpCAM PerCP-Cy5.5. Values inside the plots represent the percentages from the parent gate. SSC-A: side scatter area, FSC-A: forward scatter area, FSC-H: forward scatter height. (a–e) Submerged culture and (f–j) ALI cultures.

Figure 2

Flow cytometric (pseudo-color/smooth) plots showing cultured pHBEC stained with (a) EpCAM only, (b) pan-CK only, and (c) EpCAM plus pan-CK antibodies. Samples also included the viability dye and CD45 fluorochrome-conjugated antibody. Values inside each plot represent the percentages of HBEC live single cells from a total of approximately 2 × 10⁴ cells acquired.

Figure 3. Flow cytometric staggered histograms of fully stained cultured pHBEC (red) and negative controls: unstained sample (green), isotype control (orange) and fluorescence minus-one (FMO) controls (blue).

N = 1 per condition. All plots were gated on single, viable, CD45 negative cells. (a,c) without blocker. (b,d) with blocker (mouse serum).

Figure 4. Flow cytometric (pseudo-color/smooth) plots showing the HBEC gating strategy and percentages from the respective parent gate of human BAL, BW and bronchial brushing samples.

Figure 5. May-Grünwald Giemsa-stained BAL, BW and bronchial brushing samples showing HBEC.

3 × 10⁴ cells were cytocentrifuged and fixed onto slides by dipping in methanol, once dry; fixed cells were stained with methylene blue and eosin solutions followed by a rinse with deionized water. A drop of Permount was added onto the slides, coverslips applied and air-dried for 24 hours until ready to be analyzed. Arrows are indicating examples of HBEC.

Figure 6. Correlation between manual cell counts and our flow cytometry panel.

Cell counts and flow values were plotted across the three sample types (Brushings, BW, BAL). A line of best fit was calculated and the R² value determined. Correlation is the square root of the R² value which equals 0.96.