Figure 2. Tracking Responders EXpanding (T-REX) identifies molecular signatures of CD4+ T cells that are expanded during acute rhinovirus infection and enriched for virus-specific cells.

A subject (RV001) was experimentally infected with rhinovirus (RV-A16) and CD4+ T cell signatures monitored by spectral flow cytometry in conjunction with tetramer staining during the course of infection. (A) Fold change in the number of tetramer+ cells (log₂) after rhinovirus challenge on day 0. (B) Data showing the percentage of tetramer+ cells in each cell’s k-nearest neighbors (KNN) region (where k = 60) plotted against the percentage change in its KNN region on day 7 vs. day 0. A statistical threshold of 80% or higher for the percentage change in KNN region corresponded to marked enrichment of tetramer+ cells at day 7. (C) Uniform Manifold Approximation (UMAP) plots with T-REX analysis of CD4+ T cells for day 7 vs. day 0 based on statistical thresholds of 90–95% change (left column) and ≥95% change (right column) in cell phenotypes. Pink and red colors denote regions of phenotypic change identified by T-REX. Numbers of tetramer+ cells within the cell’s KNN region captured in these areas of phenotypic change are denoted. Cells containing >5% tetramer+ virus-specific cells in the corresponding KNN region are labeled pink. Red cells denote a KNN region that was not enriched for tetramer+ cells, and purple cells denote a tetramer enriched region not captured by T-REX. Values in black indicate the actual number of tetramer+ cells in each circled hotspot of phenotypic change. Marker Enrichment Modeling (MEM) labels on the right indicate cell phenotypes of each hotspot.

Figure 2—figure supplement 1. Tracking Responders EXpanding (T-REX) identifies CD4+ T cell tetramer+ hotspot using all cells from RV001.

Uniform Manifold Approximation (UMAP) plot with T-REX analysis of all cells with day 7 vs. day 0 for ≥95% change in cell phenotypes. Numbers of tetramer+ cells and cells in hotspots denoted. Cells in regions of change and containing >5% tetramer+ virus-specific cells in the corresponding k-nearest neighbors (KNN) region are pink. Red cells denote a KNN region that was not enriched for tetramer+ cells, but were regions of great change. Marker Enrichment Modeling (MEM) labels on the right indicate cell phenotypes the hotspot found and the hotspot from Figure 2.

Figure 2—figure supplement 2. Tracking Responders EXpanding (T-REX) identifies regions of great change enriched for tetramers in infected individuals.

Subjects RV002 through RV008 were experimentally infected with rhinovirus and CD4+ T cell signatures monitored by spectral flow cytometry in conjunction with tetramer staining during the course of infection. (A) Fold change in the number of tetramer+ cells (log₂) after rhinovirus challenge on day 0. (B) Box and whisker plots show k-nearest neighbors (KNN) regions in terms of expansion during infection represented by percent change as well as percent of tetramer-+ cells for day 0 and day 7. (C) Uniform Manifold Approximation (UMAP) plots for percent change and tetramer percent cutoff denoted in upper-left corner in the left UMAP plot. Cell count in each region is in black as well as in the upper right of each UMAP plot for tetramer+ regions changing (red), tetramer– regions changing (pink), and tetramer+ regions with change below the expansion cutoff (purple). Marker Enrichment Modeling (MEM) labels are given for highly expanded and tetramer-enriched regions.

Figure 2—figure supplement 3. Marker Enrichment Modeling (MEM)-derived gating strategy for the enrichment of rhinovirus-specific CD4+ T cells.

MEM-gated cells are derived from the combination of all depicted gates (CD45R0+ CD38+ ICOS + CCR5+ PD-1+ CD95+ CD27+ CXCR3+). (Inset) Comparison of RV tetramer+ cell enrichment in ungated and MEM-gated cell populations.

Figure 2—figure supplement 4. T cell sorting strategy derived using Tracking Responders EXpanding (T-REX) effectively enriches for rhinovirus-specific cells in infected subjects.

(A) Precursor frequencies of total RV tetramer+ CD4+ T cells from all subjects on study day 7 (n = 8 subjects). Median ± interquartile range. (B) Artificial sorting was performed using unenriched day 7 samples. Consensus Marker Enrichment Modeling (MEM) markers were individually added to the sorting strategy according to MEM feature enrichment, and corresponding total tetramer+ cell frequencies assessed. All cells were pre-gated for total CD4+ T cells (Live, Dump [CD14, CD19, CD8a]–, CD3+, CD4+). Two controls were utilized: rhinovirus tetramer staining was performed in subjects who remained uninfected following challenge as an infection control (n = 2), and an unrelated influenza hemagglutinin tetramer was utilized in a rhinovirus-infected subject to confirm antigen specificity (n = 1) (right). Tetramer enrichment with the addition of each marker was compared to the total CD4+ T cell population using Friedman’s test with Dunn’s multiple comparisons correction. **p≤0.01; ***p≤0.001.