Preservation of antigen-specific responses in cryopreserved CD4+ and CD8+ T cells expanded with IL-2 and IL-7

Abstract

Objectives: We sought to develop medium throughput standard operating procedures for screening cryopreserved human peripheral blood mononuclear cells (PBMCs) for CD4⁺ and CD8⁺ T cell responses to potential autoantigens.

Methods: Dendritic cells were loaded with a peptide cocktail from ubiquitous viruses or full-length viral protein antigens and cocultured with autologous T cells. We measured expression of surface activation markers on T cells by flow cytometry and cytometry by time of flight 24-72 h later. We tested responses among T cells freshly isolated from healthy control PBMCs, cryopreserved T cells, and T cells derived from a variety of T cell expansion protocols. We also compared the transcriptional profile of CD8⁺ T cells rested with interleukin (IL)7 for 48 h after 1) initial thawing, 2) expansion, and 3) secondary cryopreservation/thawing of expanded cells. To generate competent antigen presenting cells from PBMCs, we promoted differentiation of PBMCs into dendritic cells with granulocyte macrophage colony stimulating factor and IL-4.

Results: We observed robust dendritic cell differentiation from human PBMCs treated with 50 ng/mL GM-CSF and 20 ng/mL IL-4 in as little as 3 days. Dendritic cell purity was substantially increased by magnetically enriching for CD14⁺ monocytes prior to differentiation. We also measured antigen-dependent T cell activation in DC-T cell cocultures. However, polyclonal expansion of T cells with anti-CD3/antiCD28 abolished antigen-dependent upregulation of CD69 in our assay despite minimal transcriptional differences between rested CD8⁺ T cells before and after expansion. Furthermore, resting these expanded T cells in IL-2, IL-7 or IL-15 did not restore the antigen dependent responses. In contrast, T cells that were initially expanded with IL-2 + IL-7 rather than plate bound anti-CD3 + anti-CD28 retained responsiveness to antigen stimulation and these responses strongly correlated with responses measured at initial thawing.

Significance: While screening techniques for potential pathological autoantibodies have come a long way, comparable full-length protein target assays for screening patient T cells at medium throughput are noticeably lacking due to technical hurdles. Here we advance techniques that should have broad applicability to translational studies investigating cell mediated immunity in infectious or autoimmune diseases. Future studies are aimed at investigating possible CD8⁺ T cell autoantigens in MS and other CNS autoimmune diseases.

Keywords: Antigen stimulation; Dendritic cells; Monocytes; Polyclonal expansion; T cell.

Fig. 1

A) Schematic of strategy for DC differentiation from PBMCs or monocyte-enriched PBMCs prior to co-culture with autologous T cells and antigen recall. B) Representative flow cytometry plots showing % PBMCs expressing DC markers CD11c and CD1c after 7 days in the indicated conditions. C) Graphs show percent of leukocytes (e.g. gated as in Fig. 2a) staining for CD11c and CD1c at 3 and 7 days in vitro. **P < .01 for CD11c⁺CD1c⁺ treated vs. CD11c⁺CD1c⁺ untreated (no GM-CSF IL-4).

Fig. 2

Monocyte enrichment from PBMCs and differentiation into dendritic cells. Monocytes were enriched from PBMCs by magnetic activated cells sorting (MACS) using the Human Monocyte enrichment kit (BD Biosciences). Representative flow plots show percentage of cells exhibiting A) expression of monocyte markers before and after MACS or B) expression of dendritic cell markers among PBMCs or monocyte-enriched cells after 3 days differentiation in the indicated conditions.

Fig. 3

Upregulation of CD69 on T cells upon coculture with dendritic cells presenting ubiquitous viral antigens. Patient T cells were rested and co-cultured with autologous DCs and control peptides from common pathogens (CEF = Cytomegalovirus, Epstein bar virus, influenza) or full-length viral antigens (Mumps = mumps viral protein extract) at the indicated concentration for 24–72 h. Activation was assessed by flow cytometry staining for CD69 as shown as well as CD44 and LFA-1 (representative flow plots shown in A; quantified in B). In some experiments, cells were given a second pulse of antigen (Ag boost) after 48 h and analyzed at 72 h. Bar graphs show percent of CD8⁺ or CD4⁺ T cells that were CD69⁺. Enriching for monocytes prior to DC differentiation increased the percentage of CD8⁺ T cells that upregulated CD69 in response to viral antigens (D) relative to DC generated from PBMCs (C). D) Treatment with TLR ligands poly (I:C) and LPS alone or in combination did not increase antigen-dependent T cell upregulation of CD69 as expected and caused elevated background levels of CD69⁺ especially among CD8⁺ T cells. poly (I:C) alone or in combination with LPS also abrogated antigen-dependent upregulation of CD69 in both CD4⁺ and CD8⁺ T cells. Error bars mean ^± SEM. *P < .05 vs control.

Fig. 4

T cells from cryopreserved human PBMCs retain responsiveness to ubiquitous viral antigens. A) PBMCs from healthy volunteers were stained for the indicated surface markers before or after cryopreservation. Results indicate little change in the frequency of monocytes, B cells, and T cells. B) Cryopreserved PBMCs were rested in IL-7 or differentiated into DCs over 3 days and subsequently cocultured with the indicated antigens for 72 h as in Fig. 3. Percent increase in frequency of CD69⁺ cells among CD4⁺ and CD8⁺ T cells is shown relative to media treated controls. C) Enriching for monocytes prior to DC differentiation increased the percentage of CD4⁺ and CD8⁺ T cells that upregulated CD69 in response to viral antigens. Error bars mean ^± SEM. *P < .05 vs control.

Fig. 5

Transcriptional profile of cryopreserved and anti-CD3/antiCD28 expanded human CD8⁺T cells. A) Standard operating procedure for collection, expansion, cryostorage, and transcriptional characterization of patient PBMC-derived CD8⁺ T cells. B) Percent of CD45⁺ PBMCs staining positive for the indicated markers in 5 control patients before cryopreservation, after cryopreservation, or after one that and subsequent T cell expansion with plate bound anti-CD3/antiCD28. C) Gating strategy for B cells (CD19), NK cells (CD16), monocytes (CD14), and T cells (CD3, CD8a, CD4) among CD45⁺ cells. D) Principal component analysis of RNAseq data shows clustering of T cells from each treatment group. E) Correlogram shows >90% correlation in gene expression for CD8⁺ T cells magnetically isolated from Thaw1 PBMCs and CD8⁺ T cells isolated form Thaw2 PBMCs. F) Heat map showing genes differentially expressed in CD8⁺ T cells magnetically purified from PBMCs after anti-CD3/anti-CD28 expansion. These genes showed similar expression in “Thaw1” and “Thaw2” CD8⁺ T cells. Error bars mean ^± SEM.

Fig. 6

Prolonged resting of anti-CD3/anti-CD28 human T cells in IL-2 family cytokines reduces baseline CD69⁺expression but fails to restore responsiveness to ubiquitous viral antigens. T cells were expanded from Thaw1 PBMCs with anti-CD3/anti-CD28 and subsequently underwent another freeze thaw cycle (Thaw2). A) Thaw2 CD4⁺ and CD8⁺ T cells show elevated baseline CD69 expression and reduced responsiveness to ubiquitous viral antigens relative to Thaw1 counterparts. B) Resting Thaw2 T cells in high concentrations of IL-2 and IL-7 causes further expansion and downregulation of CD69 over 7 days. C) Treatment of Thaw2 PBMCs with IL-2 causes downregulation of CD69 relative to IL-7 and IL-15 over 7 days. D) Thaw2 cells rested for 7 days in IL-2, IL-7, IL-15, IL-2 + IL-7, or IL-2 + IL-15 show variable responsiveness to ubiquitous viral antigens across patients. Cytometry by time of flight was performed on Thaw2 PBMCs rested as in (C). Cluster frequency changes for each treatment condition are summarized in (E). Error bars mean ^± SEM. *P < .05 **P < .01 vs media condition (A, D), vs 0 IL-2 0 IL-7 condition (B) or vs Day 3 (C).

Fig. 7

Mass cytometric characterization of Thaw2 cells rested in IL-2, IL-7, or IL-15 T cells were expanded from Thaw1 PBMCs with anti-CD3/anti-CD28 and subsequently underwent another freeze thaw cycle (Thaw2). Cytometry by time of flight was performed on Thaw2 PBMCs rested for 7 days in IL-2, IL-7, or IL-15. A) R-phenograph cluster show the indicated clusters in tSNE plots. Heat maps show T cell marker expression for each cluster (B). Representative tSNE plots for each condition are shown (C).

Fig. 8

Human T cells expanded with IL-2 and IL-7 retain responsiveness to ubiquitous viral antigens. Thaw1 PBMCs were expanded with 10 ng/mL of the indicated cytokines for 7 days. Subsequently, cells were stimulated with media or 10 μg/mL CEF for 72 h and % increase in CD69⁺ on CD4⁺ (A) and CD8⁺ T cells (B) was determined by flow cytometry. Results indicated that CD4⁺ and CD8⁺ T cells from Thaw1 PBMCs that were expanded with IL-2 + IL-7 retained responsiveness to CEF peptides, whereas cells expanded with IL-2 + IL-21 only retained responsiveness among CD8⁺ T cells. C-D) Correlation between antigenic responses in human T cells before and after expansion with IL-2 and IL-7. Unmanipulated cryopreserved PBMCs from healthy controls (n = 3) and patients with CNS autoantibodies (n = 3) were thawed (thaw1) and expanded with 10 ng/mL IL-2 and IL-7 for 7 days. In parallel treatments dendritic cells were differentiated from Thaw1 PBMCs using GMCSF (50 ng/mL) and IL-4 (20 ng/mL) for 7 days and then cryopreserved. These T cells and dendritic cells were cryopreserved for >7 days prior to thawing (thaw2). Antigen induced changes in CD69⁺ T cell frequency were determined for thaw1 as described in Supp. Fig. 5 in response to media, CEF, or cognate CNS neural antigens in patients and controls. For thaw2, separate vials of dendritic cells and T cells thawed and rested for 1 h in complete media prior to being combined for antigen stimulation. Changes in CD69⁺ T cell frequency were determined 72 h later by flow cytometry. Plots show correlation between Thaw1 and Thaw2 for antigen-induced change in CD69⁺ T cell frequency across all antigen conditions relative to vehicle treatment. Slope, Pearson Correlation, and P values are shown for CD4⁺ T cells (C) and CD8⁺ T cells (D). Linear regression fit (red line) and 90% prediction range (dotted lines) are shown. *P < .05 **P < .01 vs media condition. Error bars SEM.

Fig. 9

Human T cells expanded with IL-2 and IL-7 retain responsiveness to ubiquitous viral antigens. We performed CyTOF on Thaw1 PBMCs from 5 healthy controls before and after they were expanded for 7 days by treatment with 10 ng/mL IL-2 + IL-7. A) Representative tSNE plots for total, Thaw1 and IL-2 + IL-7 expanded PBMCs are shown alongside R-phenograph cluster map showing expansion-induced enrichment of clusters 1, 2, 13, and 18. tSNE plots for each sample are shown in (B). C) Expression levels for each marker protein are overlaid on R-phenograph cluster map. As expected, we observed increased expression of IL-2R (CD25) and IL-7R among clusters enriched by expansion with IL-2 + IL-7 relative to related clusters 3, 6, 18, 16 that were unenriched. Other markers upregulated in these clusters included CCR6 and CD45RO and CXCR3. In contrast CCR7 and CD45RA were downregulated among these clusters. Heat maps show protein marker expression for each cluster (D) and cluster frequency for each sample (E).