Comparison of polyclonal expansion methods to improve the recovery of cervical cytobrush-derived T cells from the female genital tract of HIV-infected women

Abstract

Cervical cytobrushing is a useful and non-invasive method for obtaining mucosal mononuclear cells from the female genital tract, but yields few cells. The aim of this study was to compare in vitro expansion protocols (anti-CD3, anti-CD3/CD28 or Dynal anti-CD3/CD28 beads) and cytokine combinations (IL-2, IL-7 and IL-15) to improve cervical T cell yields and viability. Eighteen HIV-infected women were included in this study to compare methods for polyclonal expansion of T cells from the female genital tract and blood. Comparison of T cell yields, viability and maturational status (by differential staining with CD45RO, CCR7 and CD27) was determined following 7 days of in vitro expansion. Anti-CD3 and IL-2 resulted in a 4.5-fold (range 3.7-5.3) expansion of cervical CD3+ T cells in 7 days compared to day 0. Inclusion of anti-CD28 or addition of IL-7 and IL-15 to this combination did not improve expansion. Culturing cells with Dynal beads (1:1) and IL-2, IL-7 and IL-15 gave rise to the highest yields after 7 days in both blood (7.1-fold) and cervix (5.6-fold). While expansion with anti-CD3 led to the accumulation of effector memory T cells (CD45RO+CCR7-CD27-), expansion with Dynabeads selected for accumulation of central memory T cells (CD45RO+CCR7+CD27+). We conclude that in vitro expansion with Dynabeads (1:1) in the presence of IL-2, IL-7 and IL-15 resulted in the greatest increase in viable T cells from both blood and cytobrush. Irrespective of the expansion method used, the T cell memory profile was altered following expansion.

Fig. 1

Comparison of polyclonal expansion methods for expanding T cells from blood. PBMC were expanded using seven different combinations of expansion protocols and cytokine cocktail: anti-CD3/IL-2 (dark blue), anti-CD3/anti-CD28/IL-2 (medium blue), anti-CD3/IL-2/IL-7/IL-15 (light blue), Dynal anti-CD3/CD28 beads (3:1)/IL-2 (yellow), Dynal anti-CD3/CD28 beads (1:5)/IL-2 (light orange), Dynal anti-CD3/CD28 beads (1:1)/IL-2 (dark orange), and Dynal anti-CD3/CD28 beads (1:1)/IL-2/IL-7/IL-15 (red). Bars represent the median of 5 donors' cell recovery after 3, 5 and 7 days. Error bars represent the interquartile range for measurements taken from five donors. The figures on top of the bars represent median fold expansion after culturing. Wilcoxon rank test was used for comparison of day 0 and day 7 yields.

Fig. 2

Impact of polyclonal expansion on T cell memory phenotype in blood. Representative plots showing (A) naive and antigen-experienced cells (total memory) based on differential staining with CD27 and CD45RO; and (B) differentiation markers expression (CD45RO, CD27, and CCR7) on CD8+ T cells from blood. Eight distinct memory subsets were defined from these markers: Central memory cells, (CD45RO⁺CD27⁺CCR7⁺), transitional memory cells (CD45RO⁺CD27⁺CCR7⁻), effector memory cells (CD45RO⁺CD27⁺CCR7⁺), CD45RO⁺CD27⁻CCR7⁺, naive T cells (CD45RO⁻CD27⁺CCR7⁺), intermediate memory cells (CD45RO⁻CD27⁺CCR7⁻), effector (CD45RO⁻CD27⁻CCR7⁻) and CD45RO⁻CD27⁻CCR7⁻cells. (C) Comparison of the frequency of total, effector memory and central memory subsets expressed as percentages of total CD4⁺ and CD8⁺ T cells in chronically HIV-infected individuals (n = 5) before and after expansion using seven different protocols. Each box and whisker plot shows the median (central line), IQR (outer lines of box) and 5–95% range (error bars) of 5 HIV-infected individuals. * indicates p < 0.05 while ** indicates p < 0.01 using Wilcoxon rank test.

Fig. 3

Comparison of polyclonal expansion methods for expanding cervical cytobrush-derived T cells. Cervical cells obtained from HIV-infected women were cultured with either anti-CD3/IL-2 (n = 5; blue bars), Dynal anti-CD3/CD28 beads (1:1)/IL-2 (n = 5; orange) and Dynal anti-CD3/CD28 beads (1:1)/IL-2/IL-7/IL-15 (n = 5; red). Bars represent the yield of cells obtained from each donor pre- and post-expansion. Wilcoxon rank test was used to compare day 0 and day 7 yields. Figures on top of the bars represent fold expansion relative to day 0.

Fig. 4

Impact of polyclonal expansion on T cell memory phenotype at the cervix. (A) Representative plots showing naive and antigen-experienced T cells (total memory) based on differential expression of CD27 and CD45RO. (B) Gating strategy used to define central memory (CD45RO⁺CD27⁺CCR7⁺), transitional memory cells (CD45RO⁺CD27⁺CCR7⁻), effector memory cells (CD45RO⁺CD27⁺CCR7⁺), CD45RO⁺CD27⁻CCR7⁺, naive T cells (CD45RO⁻CD27⁺CCR7⁺), intermediate memory cells (CD45RO⁻CD27⁺CCR7⁻), effector (CD45RO⁻CD27⁻CCR7⁻) and CD45RO⁻CD27⁻CCR7⁻ cells. (C) Comparison of the frequency of total, effector memory and central memory subsets expressed as percentages of total CD4⁺ and CD8⁺ T cells at the cervix of chronically HIV-infected individuals (n = 5) before and after expansion using three different protocols. Cervical cells were expanded using the methods that yielded best expansion in PBMC experiments [Dynal beads (1:1)/IL-2 and IL-2/IL-7/IL-15] and compared with anti-CD3/IL-2 alone. Each box and whisker plot shows the median (central line), IQR (outer lines of box) and 5–95% range (error bars) of 5 HIV-infected individuals. * indicates p < 0.05 while ** indicates p < 0.01 using Wilcoxon rank test.

Fig. 5

Summary of the phenotypic changes observed in distinct T cell subsets at the cervix and blood of HIV-infected women. Eight memory subsets were defined from these markers: Central memory cells, (CD45RO⁺CD27⁺CCR7⁺), transitional memory cells (CD45RO⁺CD27⁺CCR7⁻), effector memory cells (CD45RO⁺CD27⁺CCR7⁺), CD45RO⁺CD27⁻CCR7⁺, naive T cells (CD45RO⁻CD27⁺CCR7⁺), intermediate memory cells (CD45RO⁻CD27⁺CCR7⁻), effector (CD45RO⁻CD27⁻CCR7⁻) and CD45RO⁻CD27⁻CCR7⁻cells. Correlation between fresh and expanded cells memory phenotypes was performed using Spearman rank test and Spearman rho and p-values for each correlation are shown above each pie chart.