Application of user-guided automated cytometric data analysis to large-scale immunoprofiling of invariant natural killer T cells

Abstract

Defining and characterizing pathologies of the immune system requires precise and accurate quantification of abundances and functions of cellular subsets via cytometric studies. At this time, data analysis relies on manual gating, which is a major source of variability in large-scale studies. We devised an automated, user-guided method, X-Cyt, which specializes in rapidly and robustly identifying targeted populations of interest in large data sets. We first applied X-Cyt to quantify CD4(+) effector and central memory T cells in 236 samples, demonstrating high concordance with manual analysis (r = 0.91 and 0.95, respectively) and superior performance to other available methods. We then quantified the rare mucosal associated invariant T cell population in 35 samples, achieving manual concordance of 0.98. Finally we characterized the population dynamics of invariant natural killer T (iNKT) cells, a particularly rare peripheral lymphocyte, in 110 individuals by assaying 19 markers. We demonstrated that although iNKT cell numbers and marker expression are highly variable in the population, iNKT abundance correlates with sex and age, and the expression of phenotypic and functional markers correlates closely with CD4 expression.

Keywords: automated analysis; flow cytometry.

Fig. 1.

Schematic of X-Cyt’s analytical process (synthetic samples). (A) In a few representative samples, the user adjusts analytical parameters and evaluates the clustering outcome. Adjustable parameters include the differentiation markers to be used, the number of clusters in mixture modeling (g), distribution type, and SD cutoffs for continuous markers. The user selects one optimal set of parameters that most accurately identifies the cell populations of interest (here the blue cluster using g = 3). The clustering result of the representative sample is chosen as the template (dashed circles in the g = 3 panel). (B) X-Cyt applies the template to guide the partitioning of all samples in the study. The population of interest (shown in red dots and blue dashed circle) is consistently identified across all samples. (C) Downstream to population extraction, random samples are pooled to establish the distribution of phenotypic/functional markers. The percentage of cells positive for each marker is reported based on either mixture modeling (top) or SD cutoff (bottom).

Fig. 2.

CD4⁺ memory T-cell subset identification. (A) An optimal model of seven clusters using CD45RA, CD45RO, and CD62L identified T_EM (red, cluster1) and T_CM (blue, cluster2) cells. Other clusters include naive and intermediate CD4⁺ T cells, as well as impurities. (B) X-Cyt consistently identified the T_EM (red) and T_CM (blue) populations in all samples. Four random samples are shown here. (C) X-Cyt and manual gating in FlowJo returned highly concordant proportions of T_EM and (D) T_CM in 236 samples.

Fig. 3.

iNKT cell identification. (A) The partitioning scheme: FSC and SSC were clustered into three components to identify the lymphocyte population (red). Lymphocytes were subsequently clustered using CD3ε and CD4 into three components; namely, the CD3ε⁻, CD3ε⁺CD4⁺, and CD3ε⁺CD4⁻ populations. A cutoff of five SDs above the mean in α-galactosylceramide-loaded CD1d tetramer isolated the tetramer⁺ iNKT cells (either CD4⁺ or CD4⁻). (B) X-Cyt returned iNKT cell proportions highly concordant with manual gating (Pearson’s r = 0.99). (C) The CD4⁺ proportion of iNKT cells correlates negatively with total iNKT abundance. Randomly sampled CD3ε⁺ cells from all 110 samples were pooled to establish the intensity distribution of each phenotypic marker. Fitted distributions of (D) NKG2D (bimodal) and (E) CCR4 (trimodal) are shown. (F) The first principal component of expression levels of the nine surface markers, which captured 31.8% of total variation, correlates strongly with the proportion of CD4⁺ iNKT cells. (G) The heat map shows the correlations (also indicated by Pearson’s r on top) of the nine surface markers’ expression with CD4⁺ proportion in iNKT cells. Each row represents one sample.