Trafficking of high avidity HER-2/neu-specific T cells into HER-2/neu-expressing tumors after depletion of effector/memory-like regulatory T cells

Abstract

Background: Cancer vaccines are designed to activate and enhance cancer-antigen-targeted T cells that are suppressed through multiple mechanisms of immune tolerance in cancer-bearing hosts. T regulatory cell (Treg) suppression of tumor-specific T cells is one barrier to effective immunization. A second mechanism is the deletion of high avidity tumor-specific T cells, which leaves a less effective low avidity tumor specific T cell repertoire available for activation by vaccines. Treg depleting agents including low dose cyclophosphamide (Cy) and antibodies that deplete CD25-expressing Tregs have been used with limited success to enhance the potency of tumor-specific vaccines. In addition, few studies have evaluated mechanisms that activate low avidity cancer antigen-specific T cells. Therefore, we developed high and low avidity HER-2/neu-specific TCR transgenic mouse colonies specific for the same HER-2/neu epitope to define the tolerance mechanisms that specifically affect high versus low avidity tumor-specific T cells.

Methodology/principal findings: High and low avidity CD8(+) T cell receptor (TCR) transgenic mice specific for the breast cancer antigen HER-2/neu (neu) were developed to provide a purified source of naïve, tumor-specific T cells that can be used to study tolerance mechanisms. Adoptive transfer studies into tolerant FVB/N-derived HER-2/neu transgenic (neu-N) mice demonstrated that high avidity, but not low avidity, neu-specific T cells are inhibited by Tregs as the dominant tolerizing mechanism. High avidity T cells persisted, produced IFNγ, trafficked into tumors, and lysed tumors after adoptive transfer into mice treated with a neu-specific vaccine and low dose Cy to deplete Tregs. Analysis of Treg subsets revealed a Cy-sensitive CD4(+)Foxp3(+)CD25(low) tumor-seeking migratory phenotype, characteristic of effector/memory Tregs, and capable of high avidity T cell suppression.

Conclusion/significance: Depletion of CD25(low) Tregs allows activation of tumor-clearing high avidity T cells. Thus, the development of agents that specifically deplete Treg subsets should translate into more effective immunotherapies while avoiding autoimmunity.

Figure 1. Adoptively transferred RNEU_420–429-specific high avidity T cells enhance tumor clearance in FVB/N mice.

The minimal tumorigenic dose was given to FVB/N mice on day 0; Cy was given on day 2; vaccine or mock vaccine was injected on day 3; high avidity or low avidity T cells were adoptively transferred on day 4. Treatment groups are indicated in the figures. (A) Tumor growth after high avidity T cell adoptive transfer into FVB/N mice. (B) Tumor growth after low avidity T cell adoptive transfer into FVB/N mice. Data was graphed as the tumor area vs. days (figures 1 A–B) after adoptive transfer. * = p<0.05; ** = p<0.001. Experiments were repeated three times (n = 10 mice per group).

Figure 2. RNEU_420–429-specific high avidity T cells enhance tumor clearance in neu-N mice given Cy and vaccine.

Neu-N mice were treated as in Figure 1 and the Methods. (A) Tumor growth after high avidity T cell adoptive transfer into neu-N mice. (B) Tumor growth after low avidity T cell adoptive transfer into neu-N mice. (C) Twice the minimum tumor dose was given to neu-N mice (1×10⁵ cells/mouse) on day 0. Data was graphed as the tumor free probability vs. weeks (Figure 2C) or tumor size (mm²) vs. days (figures 2A–B) after adoptive transfer. * = p<0.05; ** = p<0.001. Experiments were repeated three times (n = 10 mice per group).

Figure 3. High avidity T cells expand preferentially after adoptive transfer into Cy and vaccine-treated neu-N mice.

Mice were injected with 1×10⁶ tumor cells/mouse one week prior to Cy and vaccine treatment. High avidity or low avidity T cells were adoptively transferred one day after vaccination. (A) The absolute # of Thy1.2 T cells were calculated in the tumor draining lymph nodes (TDNs) of neu-N mice on day 3 (White bars), 5 (Gray bars), and 8 (Black bars) after adoptive transfer (n = 9 mice per group). (B) The absolute # of Thy1.2 T cells were calculated in the vaccine draining lymph nodes (VDNs) on day 3 (White bars), 5 (Gray bars), and 8 (Black bars) after adoptive transfer (n = 9 mice per group). Treatment groups are as indicated. High/low = High or low avidity T cells, respectively. Cy = Cytoxan treated. Vaccine = 3T3neuGM vaccine, Mock = 3T3GM control vaccine. Each experiment was repeated three times with similar results.

Figure 4. Enhanced cytokine secretion in high avidity T cells post-transfer into Cy and vaccine-treated neu-N mice.

On day 3 and day 5 after adoptive transfer, lymphocytes isolated from the non-draining lymph nodes (NDNs, White bars), VDNs (Black bars), and TDNs (Gray bars) were analyzed for the ability to secrete IFNγ, TNFα, and IL-2 secretion by ICS, and the absolute # of activated Thy1.2 cells was calculated (n = 3 mice per group). Treatment groups are as indicated. High/low = High or low avidity T cells, respectively. Cy = Cytoxan treated. Vaccine = 3T3neuGM vaccine, Mock = 3T3GM control vaccine. Each experiment was repeated three times with similar results.

Figure 5. Cy plus vaccine enhances VLA-4, VLA-6, and CXCR3 expression on high avidity T cells.

On day 3, TDN lymphocytes from neu-N mice were stained for (A) β1 integrin expression, (B) α4 integrin, or (C) α6 integrin, and the Mean Fluorescent Intensities (MFI) calculated. (n = 3mice/group). Isotype control MFIs were β1 integrin ∼20, α4 integrin ∼50–100, α6 integrin ∼30. (D) TDN lymphocytes from neu-N mice on day 5 were stained for CXCR3 and the %CXCR3⁺ of Thy1.2 cells was calculated (n = 5 mice/group). MFI isotype for CXCR3 is ∼10. * = p<0.05, ** = p<0.001. High/low = High or low avidity T cell transfer, respectively. Cy = Cyclophosphamide treatment. Vaccine = 3T3neuGM vaccine, Mock = 3T3GM control vaccine. Each experiment was repeated three times with similar results.

Figure 6. Cy-mediated Treg depletion increases high avidity T cell tumor trafficking in neu-N mice.

Neu-N mice were treated as in Figure 3 . (A and B) On day 3 (White bars) and day 5 (Gray bars) after adoptive transfer, the absolute number of Thy1.2⁺ cells/mg of tumor was calculated for each treatment group (n = 3 mice/group). (C) On day 5 after adoptive transfer the ratio of Thy1.2⁺ cells/CD4⁺Foxp3⁺ cells was calculated (n = 3 mice/group). (D) Lymphocytes from tumors on day 5 after adoptive transfer were stimulated as described in the methods and the %Thy1.2⁺ cells producing IFNγ after RNEU_420–429 stimulation was calculated after NP stimulation was subtracted (n = 5mice/group). * = p<0.05. High/low = High or low avidity T cell transfer, respectively. Cy = Cyclophosphamide treatment. Vaccine = 3T3neuGM vaccine, Mock = 3T3GM control vaccine. Each experiment was repeated three times with similar results.

Figure 7. Confirmation of increased high avidity T cell tumor trafficking by in vivo imaging.

(A) Neu-N mice were tumor challenged and treated as in the Methods. One day after T cell adoptive transfer mice were treated with In-111 labeled anti-Thy1.2, the tumors isolated and the radiation content analyzed as described in Methods. Isotype = Isotype control treatment group. Tumor radioactivity concentration was reported as the average % injection dose/g of tumor (n = 3mice/group). (B) neu-N mice treated as in 7A, and underwent SPECT/CT imaging. Coronal (left panel) and transverse (right panel) cross-sections through the tumor are shown for each treatment. (C) Tumors were collected on day 5 after adoptive transfer and immunofluorescence stained for Thy 1.2. * = p<0.05. High/low = High or low avidity T cell transfer, respectively. Cy = Cyclophosphamide treatment. Vaccine = 3T3neuGM vaccine. Mock = 3T3GM Control Vaccine. These experiments were repeated 2–3 times with similar results.

Figure 8. Cy preferentially depletes CD25^lowFoxp3⁺ Tregs.

Neu-N mice were treated as in Figure 3 and the Methods. Lymphocytes from spleens were collected on day 3. (A) CD4⁺Foxp3⁺ Tregs were analyzed for CD25 expression. Definition of CD25^high vs. CD25^low is based on isotype control. CD25^low cells = gray fill under the histogram. (B) Absolute # CD25^lowCD4⁺Foxp3⁺ Tregs calculated in spleens. (C) Absolute number of CD25^highCD4⁺Foxp3⁺ Tregs calculated in spleens. High = high avidity T cell transfer, Cy = Cyclophosphamide treatment, Vaccine = 3T3neuGM vaccine treatment. * = p<0.05. These studies were repeated three times with similar results (n = 3 mice/group).

Figure 9. CD25^lowFoxp3⁺ Tregs demonstrate an effector phenotype.

(A) CD4⁺Foxp3⁺CD25^high and CD4⁺Foxp3⁺CD25^low Tregs from high avidity T cell and vaccine treated mice analyzed for expression of ICOS, CTLA-4, GITR, and CD44. (B) CD25^high and ^low Tregs were analyzed for β1 integrin, LFA-1, CD62L, and CXCR3. CD25^high histograms = Black line. CD25^low histograms = Dotted line. These studies were repeated three times with similar results (n = 3 mice/group).

Figure 10. CD25^lowFoxp3⁺ Tregs suppress high avidity T cells and traffic preferentially to the tumor.

(A) CD25^low and CD25^high CD4⁺ Treg cells suppress high avidity T cell activation in vitro. CD25^low or CD25^high CD4⁺ FoxP3^gfp+, DC, and CFSE-labeled high avidity Thy1.2⁺ CD8⁺ neu-specific T cells were mixed as described in the Methods. CFSE dilution was measured on days 3 and 5. CD4⁺ T cells from FVB/N mice were used as controls. High+DCs = High avidity T cells+DCs. High+DCs+CD4s = High avidity T cells+DCs+FVB/N CD4⁺ T cells, (B) CD25^lowCD4⁺FoxP3⁺ Tregs trend toward suppression in vivo. neu-N/FoxP3^gfp mice were given tumor and vaccinated as in the Methods. CD4⁺CD25^low Tregs were isolated 1 week later and transferred to FVB/N-TgN(TIE2GFP)287Sato mice which were then vaccinated. CD8⁺ T cells were isolated 2 weeks later and tested for the ability to secrete IFNγ in response to RNEU_420–429. Data are presented as % RNEU_420–429-specific, IFNγ⁺ of CD8⁺ T cells. This experiment was done twice with similar results. (C) CD25^low Tregs traffic preferentially to the tumor. neu-N mice received tumor and one week later received Cy, vaccine, 6×10⁶ high avidity T cells, and 5×10⁵ CD25^low or CD25^high CD4⁺GFP⁺ Tregs. On day 5 CD25^low or CD25^high CD4⁺GFP⁺ Tregs recovered from tumors, spleens, TDNs, and VDNs were enumerated as a fraction of the total # of FoxP3⁺ T cells at each site (n = 3mice/group). * = p<0.05. This experiment was repeated three times with similar results.