Umbilical cord-mesenchymal stem cells induce a memory phenotype in CD4+ T cells

Abstract

Inflammation is a physiological state where immune cells evoke a response against detrimental insults. Finding a safe and effective treatment for inflammation associated diseases has been a challenge. In this regard, human mesenchymal stem cells (hMSC), exert immunomodulatory effects and have regenerative capacity making it a promising therapeutic option for resolution of acute and chronic inflammation. T cells play a critical role in inflammation and depending on their phenotype, they can stimulate or suppress inflammatory responses. However, the regulatory effects of hMSC on T cells and the underlying mechanisms are not fully elucidated. Most studies focused on activation, proliferation, and differentiation of T cells. Here, we further investigated memory formation and responsiveness of CD4⁺ T cells and their dynamics by immune-profiling and cytokine secretion analysis. Umbilical cord mesenchymal stem cells (UC-MSC) were co-cultured with either αCD3/CD28 beads, activated peripheral blood mononuclear cells (PBMC) or magnetically sorted CD4⁺ T cells. The mechanism of immune modulation of UC-MSC were investigated by comparing different modes of action; transwell, direct cell-cell contact, addition of UC-MSC conditioned medium or blockade of paracrine factor production by UC-MSC. We observed a differential effect of UC-MSC on CD4⁺ T cell activation and proliferation using PBMC or purified CD4⁺ T cell co-cultures. UC-MSC skewed the effector memory T cells into a central memory phenotype in both co-culture conditions. This effect on central memory formation was reversible, since UC-MSC primed central memory cells were still responsive after a second encounter with the same stimuli. The presence of both cell-cell contact and paracrine factors were necessary for the most pronounced immunomodulatory effect of UC-MSC on T cells. We found suggestive evidence for a partial role of IL-6 and TGFβ in the UC-MSC derived immunomodulatory function. Collectively, our data show that UC-MSCs clearly affect T cell activation, proliferation and maturation, depending on co-culture conditions for which both cell-cell contact and paracrine factors are needed.

Keywords: CD4 + T cells; cell contact; central memory; flow cytometry; immunomodulation; memory T cells; umbilical cord mesenchymal stem cells.

Figure 1

Gating strategy of CD4⁺ T cells. Representative gating strategy for assessment of CD25, Ki67, and memory T cell subsets in PBMC and UC-MSC cocultures (A) and CD4⁺ T cells and UC-MSC co-cultures were shown (B). Corresponding figures are from day 1 of co-cultures. Gate positions were selected by using unstimulated cells.

Figure 2

Proliferation, activation and memory formation of T cells in PBMC co-culture with UC-MSC. PBMC were stimulated with αCD3/CD28 beads (PBMC+Beads) in a 1:5 ratio (bead:cell). Unstimulated cells (PBMC Only) were used as negative control. 24 hours after stimulation, UC-MSC were added to PBMC+Beads cultures (PBMC+Beads+UC-MSC) in a 1:5 (UC-MSC : PBMC) ratio. Changes in cell phenotypes were acquired every other day for in total 5 days. Representative flow cytometry plots and gating strategy of the PBMC co-cultures can be found in Figure 1A . Frequencies of different T cell subsets are shown: CD4⁺CD25⁺ (A), CD4⁺Ki67⁺ (B), CD4⁺CD45RA^-CCR7^- (effector memory T cells; T_EM) (C), CD4⁺CD45RA^-CCR7⁺ (central memory T cells; T_CM) (D), and CD4⁺CD45RA⁺CCR⁺ (Naive) T cells (E). 7 biological replicates are used in the graphs and the paired-t-test was used to assess differences between each group; *p<0.05, **p<0.01, ***p<0.001.

Figure 3

Effect of UC-MSC on activation, proliferation and maturation of negatively selected CD4+ T cells. CD4+T cells were negatively selected from the PBMC pool. They were either stimulated with αCD3/CD28 beads (CD4⁺+Beads) or kept unstimulated (CD4+ Only). 24 hours after stimulation, UC-MSC were added to the culture system (CD4⁺+Beads+UC-MSC). Phenotypical changes were acquired every other day for 5 days. Representative flow cytometry plots and gating strategy of the CD4⁺ co-cultures can be found in Figure 1B . Percentages of CD4⁺CD25⁺ (A), CD4⁺Ki67⁺ (B), CD4⁺CD25^highCD127^-/low FOXP3⁺ (C), CD4⁺CD45RA^-CCR7^- (T_EM) (D), CD4⁺CD45RA^-CCR7⁺ (T_CM) (E), and CD4⁺CD45RA⁺CCR⁺ (Naive) T cells (F) are shown. 7 biological replicates are included and the paired-t-test was used to assess differences between each group; *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Figure 4

Comparison of the UC-MSC dependent mechanisms involved in T cell regulation. 24 hours after activating PBMC with αCD3/CD28 beads, activated PBMC were exposed to UC-MSC in either direct cell-cell contact (DCC), transwell (TW) or with UC-MSC derived condition medium (CM) with a final volume of 25% or 50% (CM25% or CM50%, respectively). Percentages of different T cell subsets at day1, 3 and 5 of co-culture are shown: CD4⁺CD25⁺ (A), CD4⁺Ki67⁺ (B), CD4⁺CD45RA^-CCR7^- (T_EM) (C), CD4⁺CD45RA^-CCR7⁺ (T_CM) (D), and CD4⁺CD45RA⁺CCR⁺ (Naive) (E). 7 biological replicates are included and the paired-t-test was used to assess differences between each group; *p<0.05, **p<0.01, ****p<0.0001.

Figure 5

Immunomodulatory effect of surface interaction between UC-MSC and CD4⁺ T cell. Negatively selected CD4⁺ T cells were stimulated 24 hours with αCD3/CD28 beads. CD4⁺ T cells were co-cultured with either UC-MSC (UC-MSC) or BPL treated UC-MSC (BPL-UC-MSC). 24 hours BPL treatment of UC-MSC was used to block paracrine factor secretion to evaluate the impact of surface proteins on immunomodulation. The effect of this blockade on T cell regulation was validated by measuring phenotypical changes of CD4⁺ T cells. Frequencies of different T cell subset were determined at day 1, 3 and 5 of co-culture: CD4⁺CD25⁺ (A), CD4⁺Ki67⁺ (B), CD4⁺CD25^highCD127^-/low FOXP3⁺ (C), CD4⁺CD45RA^-CCR7^- (T_EM) (D), CD4⁺CD45RA^-CCR7⁺ (T_CM) (E), and CD4⁺CD45RA⁺CCR⁺ (Naive) T cells (F). In all graphs, 7 biological replicates are included and the paired-t-test was used to assess differences between each group; *p<0.05, **p<0.01, ****p<0.0001.

Figure 6

Soluble factors were analyzed in UC-MSC co-cultures with either PBMC or purified CD4⁺ T cell. Soluble mediators were analyzed via Luminex immunoassay: IL-1β, IL-2, IL-6, IL-10, IL-12, IL-17, IFNγ, IFNγ-induced protein 10 (IP-10), macrophage inflammatory protein 1β (MIP-1β), TNFα, TNFβ, and Eotaxin. TGFβ levels were measured by ELISA. Differences between groups were expressed in (log10) fold change on the y-axis. Luminex assay was performed on co-culture supernatant of both PBMC (A) and purified CD4⁺ T (B) experiments. Similarly, TGFβ levels were measured for both PBMC (C) and pure CD4⁺ T (D) cell culture supernatant. In the conditions where the concentration was above or below the detection limit, maximum or minimum detectable concentration were used, respectively. Analytes that did not have proper standard curves were excluded from the analysis. 3 biological replicates are shown in all graphs. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 (data are compared using the paired-t-test between each group).

Figure 7

Neutralization of IL-6 and TGFβ in UC-MSC co-cultures with CD4⁺ T cell. Neutralizing IL-6 and TGFβ antibodies were added to 24h pre-activated CD4⁺ T cell cultures, 30 minutes prior to co-culture initiation. The effect of this neutralization on CD4⁺ T cells was validated by acquiring changes in (A) CD4⁺CD45RA^-CCR7^- (T_EM) and (B) CD4⁺CD45RA^-CCR7⁺ (T_CM) populations. 3 biological replicates are shown and the paired-t-test was used to assess differences between each group.

Figure 8

Transient effect of UC-MSC on CD4⁺T cells. The T_CM population obtained from day 3 culture of CD4⁺+Beads and CD4⁺+Beads+UC-MSC were FACS sorted for viable CD45⁺CD4⁺CD62L⁺ cells. Sorted cells were labeled with fluorescence dye CFSE, and stimulated with or without αCD3/CD28 beads in the absence of UC-MSC (A). Cell division was measured by flow cytometry 48-hour after labeling and with or without αCD3/CD28 beads re-stimulation (B). The phenotypical switch from central memory (C) to effector memory (D) was acquired by flow cytometry measurement. 4 biological replicates are shown and the paired-t-test was used to assess differences between each group; *p<0.05.

Figure 9

Schematic representation summarizing the immunomodulatory activity of UC-MSCs on CD4⁺ T cells. UC-MSC induce the expression of activation marker-CD25- on CD4⁺ T cells; while causing a delay in their proliferative state. UC-MSC induced a T_CM population in activated CD4⁺ T cells, and this phenotype was reversible when UC-MSC were removed from the co-culture and re-stimulated. This suggests a transient role of UC-MSC in modulating the immune response.