IL-4 inhibits VLA-4 expression on Tc1 cells resulting in poor tumor infiltration and reduced therapy benefit


  • Kotaro Sasaki,

    1. Department of Dermatology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
    Search for more papers by this author
  • Angela D. Pardee,

    1. Department of Immunology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
    Search for more papers by this author
  • Hideho Okada,

    1. Department of Neurological Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
    2. University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA
    Search for more papers by this author
  • Walter J. Storkus

    Corresponding author
    1. Department of Dermatology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
    2. Department of Immunology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
    3. University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA
    • Department of Dermatology, University of Pittsburgh Medical Center, W1041.2 Biomedical Sciences Tower, 200 Lothrop Street, Pittsburgh, PA 15213, USA Fax: +1-412-383-5857
    Search for more papers by this author


We and others have previously demonstrated that IL-4-dependent Tc2 are inferior to Tc1-effector CD8+ T cells in regulating tumor progression in vivo. This functional disparity relates, in part, to the comparatively poor ability of Tc2 to migrate into diseased tissues. We now show that IL-4 treatment of committed Tc1 cells promotes the selective loss in the expression of very-late antigen (VLA)-4, without impacting the Tc1 cytokine production profile, cytotoxic activity, or expression of alternate cell surface markers. Down-regulation of VLA-4 expression on Tc1 cells was unique to treatment with IL-4 (i.e. Tc1IL-4) and did not occur in the presence of the Type-2 cytokine IL-13 or the regulatory cytokines IL-10 or TGF-β. Notably, the inhibitory effects of IL-4 on Tc1 expression of VLA-4 could be blocked by the presence of IL-12, but not IFN-γ. Predictably, Tc1IL-4 (but not Tc1 control) cells adhere poorly to plate-bound VCAM-1-Fc fusion protein and fail to be co-stimulated by VCAM-1 in vitro. They were also markedly impaired in their ability to traffic into intracranial melanoma lesions after adoptive transfer, yielding inferior therapeutic benefit to tumor-bearing mice. These results suggest a novel suppressive mechanism for IL-4 that limits Tc1 efficacy via preventing their recruitment into tumors.


T cell-mediated anti-tumor immune responses require the active trafficking of tumor-specific T cells in tumor tissues. In our previous study using mice bearing intracranial M05 melanomas, we showed that adoptively transferred OVA-specific polorized type-1 CD8+ T cell (Tc1) cells were superior to polorized type-2C8+ T cell (Tc2) cells in trafficking into the intracranial tumor lesions and in mediating potent therapeutic responses 1. Furthermore, when characterizing the expression of a panel of homing receptors on Tc1 and Tc2 cells, we found that very late antigen (VLA)-4 (CD49d/CD29) is preferentially expressed on Tc1 cells, but not on Tc2 cells, and that this integrin plays a critical role in the effective trafficking of therapeutic Tc1 cells into the CNS tumors 2. Moreover, we determined that IL-4 intrinsically inhibits VLA-4 expression on Tc2 cells 2.

Since IL-4 also induces Type-2 polarization in CD8+ T cells 3, this finding raises the question as to whether IL-4-mediated VLA-4 down-regulation is merely an indicator of the functional Type-2 skewing of T cells. Although polarized T-effector cells rarely interconvert, cytokines secreted by Type-1 and Type-2 T cells are known to cross-regulate the proliferation or cytokine secretion of committed cells of the opposite subset 4–6. IL-10 7, 8 and IL-4 9, 10 secreted by Th2 cells inhibit cytokine secretion by Th1 cells. Similarly, IFN-γ secreted by Th1 cells inhibits the proliferation of Th2, but not Th1 cells 11. Little is known, however, regarding the influence of polarizing (Type-2) cytokines on the trafficking capacity of committed CD8+ T cells.

We now show that, in addition to the well-established functional Type-2 polarizing effects of IL-4 on naïve T cells, IL-4 suppresses the expression of VLA-4 on differentiated CD8+ Tc1 cells, without altering their Type-1 polarized cytokine production profile. This effect could be negated by co-provision of IL-12 during, but not after, IL-4-treatment of Tc1 cells. Notably, while Tc1IL-4 retained strong and specific cytolytic reactivity in vivo and specifically produced IFN-γ and TNF-α in response to stimulation with tumor cells in vitro, these cells failed to migrate into tumor lesions after adoptive transfer and yielded poor protection as a therapeutic agent (in contrast to control Tc1 cells).


IL-4 suppresses VLA-4 expression by Tc1 cells without altering their cytokine profile

OVA-specific effector CD8+ Tc1 cells were established by stimulating freshly isolated splenic CD8+ T cells with OVA257–264 peptide in the presence of syngenic APC, IL-2, IFN-γ, IL-12, and anti-IL-4 mAb. After 6 days in culture, the resulting Tc1 cells were cultured in media containing IL-2 and IL-4 in the presence or absence of OVA peptide stimulation. After an additional 6 days, responder T cells were harvested and VLA-4 expression was analyzed. Tc1 cells cultured without IL-4 expressed high levels of VLA-4 regardless of whether specific antigen was provided on day 6 of culture (Fig. 1A). In contrast, IL-4-treated Tc1 (Tc1IL-4) cells exhibited dramatically reduced levels of VLA-4 expression, as assessed at both the protein (Fig. 1A) and mRNA level (Supporting Information Fig. 1). Consistent with previous reports 9, IL-4 treatment of Tc1 cells (in the absence or presence of cognate antigen) did not result in a shift towards Tc2 functional phenotype, as Tc1IL-4 cells continued to secrete IFN-γ (Fig. 1B) and failed to secrete either IL-4 (Fig. 1B) or IL-5 (data not shown). Expression of other activation markers, such as CD25 and CD62L, did not change after treatment of Tc1 cells with IL-4, supporting the relative selectivity of IL-4's impact on VLA-4 expression on Tc1IL-4versus Tc1 cells (Supporting Information Fig. 2).

Figure 1.

IL-4 down-regulates VLA-4 expression on committed Tc1 cells without changing their cytokine production profile. Tc1 cells were induced from OT-I-derived naïve CD8+ T cells as described in the Materials and methods. Day 6 Tc1 cells were treated with 10 ng/mL of IL-4 with or without OVA stimulation for an additional 7 days to generate IL-4-treated Tc1 cells (Tc1IL-4). (A) Flow cytometric analyses of VLA-4 (CD49d/CD29) expression on Tc1 and Tc1IL-4 cells. (B) Intracellular cytokine expression in Tc1 (days 6 and 13) versus Tc1IL-4 (day 13) cells after in vitro re-stimulation with plate-bound anti-CD3 mAb. In panels (A) and (B), the inset numbers in histograms reflect the percentage of cells in a given quadrant. (C) Kinetics of VLA-4 (CD49d/CD29) expression down-modulation by IL-4 on Tc1IL-4 cells. The percentages of VLA-4+ (CD49d+/CD29+) cells on days 6, 7, 8, 10, 12 are indicated. All data are representative of three independent experiments performed.

To gain additional insight into how Tc1 expression of VLA-4 is regulated by IL-4, we next examined the kinetic time course of VLA-4 expression on Tc1 cells after IL-4 treatment. VLA-4 expression was consistently up-regulated/maintained on Tc1 cells in the absence of IL-4, whereas a gradual decrease in VLA-4 expression was observed over time in the presence of IL-4 (Fig. 1C).

IL-4's ability to suppress VLA-4 expression on Tc1 cells appears unique

To determine whether alternate Type-2 or regulatory cytokines could also promote the silencing of VLA-4 expression by Tc1 cells, we analyzed VLA-4 levels on Tc1 control cells versus Tc1IL-4, Tc1IL-10, Tc1IL-13 and Tc1TGF-β by flow cytometry. IL-4 was unique among all cytokines evaluated in this study in down-modulating VLA-4 expression on Tc1 cells (Fig. 2A). As Type-2 polarized CD4+ T cells are likely a major source of IL-4 in vivo, we next examined whether committed Th2 cells could suppress VLA-4 expression on Tc1 cells in vitro. OT-I-derived day 6 Tc1 cells were mixed with OT-II-derived day 8 Th2 cells, with cultures then stimulated with the (OT-II recognized) OVA323–339 Th peptide epitope. As shown in Fig. 2B, co-culture with Th2 cells significantly down-regulated VLA-4 expression on Tc1 cells. Inclusion of neutralizing anti-IL-4, but not anti-IL-10 or anti-IL-13, mAb ablated such inhibition, suggesting the requisite nature of Th2 elaborated IL-4 for suppression of VLA-4 expression by Tc1 cells.

Figure 2.

IL-4, but not other Type-2/regulatory cytokines, suppresses VLA-4 expression on Tc1 cells. (A) OT-I mouse-derived Tc1 cells were treated with 10 ng/mL of IL-4, IL-10, IL-13, or TGF-β in the presence of rmIL-2 beginning on day 6 of Tc1 culture. Cells were harvested on day 13 of culture and VLA-4 expression was assessed by flow cytometry. The inset numbers in histograms reflect the percentage of cells in the indicated quadrant. (B) OT-I mouse-derived day 6 Tc1 cells were mixed with OT-II mouse-derived day 6 Th2 cells and stimulated with the OVA323–339 Th peptide epitope in the presence or absence of 10 μg/mL of anti-IL-4 mAb, anti-IL-10 mAb, anti-IL-13 mAb, or control rat IgG. T cells were harvested at day 12 and VLA-4 expression in the CD8+ gated population assessed by flow cytometry. Data are reported as mean±SD of triplicate determinations. *p=0.0026, Tc1 cells alone versus Tc1 cells co-cultured with Th2 in the presence of PBS. **p=0.0028, Tc1 cells alone versus Tc1 cells co-cultured with Th2 in the presence of control rat IgG. ***p=0.0019, Tc1 cells co-cultured with Th2 in the presence of control rat IgG versus anti-IL-4 mAb. All data are representative of three independent experiments performed.

IL-12 but not IFN-γ prevents IL-4 suppression of Tc1 expressed VLA-4

Given the functional antagonism commonly observed for Type-1 (i.e. IL-12p70, IFN-γ) versus Type-2 (i.e. IL-4) cytokines, we next chose to investigate whether the inclusion of IL-12 or IFN-γ during the period of IL-4 co-culture would protect Tc1 cells from suppression in VLA-4 expression. While neither IL-12 nor IFN-γ promoted changes in constitutive expression of VLA-4 by Tc1 cells, we observed that IL-12, but not IFN-γ, was able to counter the inhibitory action of IL-4 on VLA-4 expression by Tc1 cells (Fig. 3A). These observations led us to further examine if IL-12 treatment could recover VLA-4 expression on Tc1IL-4 cells. As shown in Fig. 3B, neither IL-12 nor IFN-γ was competent to resurrect VLA-4 expression on Tc1IL-4 cells, suggesting that IL-4-mediated VLA-4 down-regulation may be irreversible.

Figure 3.

IL-12, but not IFN-γ, blocks IL-4-induced down-regulation of VLA-4 expression on committed Tc1 cells. (A) OT-I mouse-derived day 6 Tc1 cells were treated with 4 ng/mL of IFN-γ or IL-12 in the presence or absence of IL-4 (without antigen-stimulation). MFI of VLA-4 expression was assessed on cells harvested at day 13 (mean±SD for data obtained in three independent experiments). (B) Tc1IL-4 cells were treated for 7 days with 4 ng/mL of IL-12 or IFN-γ and analyzed for VLA-4 expression by flow cytometry. The inset numbers in histograms reflect the percentage of cells in the indicated quadrant. *p<0.003; No treatment (none) versus IL-4-treatment (none+IL-4) on Tc1 cells or no-treatment versus IL-4 and IFN-γ-treatment (IFN-γ+IL-4) on Tc1 cells. Data are representative of three independent experiments performed.

Tc1IL-4 bind poorly to, and are co-stimulated poorly by, VCAM-1 in vitro

To examine the functional significance of IL-4-mediated VLA-4 suppression on Tc1 cells, Tc1, Tc2 and Tc1IL-4 cells were evaluated for their ability to adhere to plate-bound VCAM-1-Fc fusion protein. As predicted based on their relative expression levels of VLA-4, Tc1 cells displayed specific adhesion to immobilized VCAM-1-Fc, while Tc2 and Tc1IL-4 cells showed only background level of cell adhesion (Fig. 4A, left). In contrast, all three CD8+ T cell types express comparable levels of VLA-5 and exhibit strong adherence to culture wells coated with the VLA-5 ligand fibronectin (Fig. 4A, right, and data not shown).

Figure 4.

Tc1IL-4 cells lose the ability to adhere to VCAM-1 and fail to be co-stimulated by VLA-4 in vitro. (A) After pre-incubation with either anti-CD49d mAb (PS/2) (hatched bars), anti-CD29 mAb (Ha2/5) (white bars), or control rat IgG2b mAb (filled bars), Tc1, Tc1IL-4, and Tc2 cells were incubated in 96-well plates pre-coated with VCAM-1 ((A) left) or fibronectin ((A) right) for 30 min at 37°C. The plates were then washed twice and the number of cells adherent to the bottom of the wells was enumerated by flow cytometry, as described in the Materials and methods. *p=0.002, percentage adhesion of Tc1 versus Tc1IL-4 against VCAM-1-Fc; *p=0.0003, Tc1 versus Tc2. (B) Tc1 or Tc1IL-4 cells were added to 96-well plates coated with anti-CD3 mAb, together with either VCAM-1-Fc fusion protein or control human IgG1-Fc fusion protein. After incubation for 16 h at 37°C, supernatants were collected and IFN-γ levels (mean±SD) were measured by specific ELISA. *p=0.001, VCAM-1–stimulated versus control-stimulated Tc1 groups; n.s., not significant, VCAM-1–stimulated versus control-stimulated Tc1IL-4 groups. All data are reported as mean±SD of triplicate determinations and are representative of three independent experiments performed in each case.

Since the VLA-4/VCAM-1 interaction is known to provide co-stimulation to T cells 2, 12, 13, we next evaluated the co-stimulatory effects of VCAM-1 on Tc1 versus Tc1IL-4 cells using plate-bound VCAM-1-Fc or human IgG1-Fc control fusion protein in the presence of sub-mitogenic levels of immobilized anti-CD3 mAb. Tc1 cells showed dramatically enhanced production of IFN-γ upon stimulation with VCAM-1-Fc and anti-CD3 mAb, whereas Tc1IL-4 cells were poorly responsive to co-stimulation by control fusion protein (Fig. 4B).

IL-4 treatment does not impair cytotoxicity mediated by Tc1 cells

We next wanted to determine whether IL-4 alters the cytolytic function of Tc1 cells. We noted that in a comparison of Tc1 versus Tc1IL-4 cells that expression of the cytolytic molecule, granzyme B, was not differential (Supporting Information Fig. 3A). Similarly, both Tc1 and Tc1IL-4 cells developed from OT-I mice demonstrated comparable levels of specific cytotoxicity against OVA cDNA-transfected, but not parental, B16 melanoma cells in vitro (Supporting Information Fig. 3B). Furthermore, upon adoptive transfer into syngenic, naïve mice, Tc1 and Tc1IL-4 cells exhibited comparable in vivo killing capacity against OVA257–264 peptide-pulsed versus control splenocytes in CFSE-based analyses (Supporting Information Fig. 3C).

Tc1IL-4 cells fail to infiltrate tumors in vivo and are poorly therapeutic

We have previously demonstrated that the efficient trafficking of adoptively transferred (OT-I derived, anti-OVA) Tc1 cells into intracranial M05 (OVA cDNA-transfected B16) melanoma lesions in vivo was dependent upon the VLA-4/VCAM-1 interaction 2. Taking advantage of this tumor model system, we examined the comparative ability of Tc1IL-4 cells to accumulate within tumor sites in vivo. Adoptive-transferred, OT-I-derived Tc1IL-4 cells were inferior in their ability to infiltrate i.c. M05 melanoma lesions (Fig. 5A). Given their inability to localize within sites of disease, it was not surprising that treatment of M05-bearing mice with adoptively transferred Tc1IL-4 cells yielded minimal protection when compared to control Tc1 cells (Fig. 5B).

Figure 5.

Adoptive-transferred Tc1IL-4 cells fail to migrate into intracranial tumor lesions and failed to provide therapeutic anti-tumor immunity. (A) Mice bearing established day 10 i.c. M05 tumors received i.v. transfer of Tc1 or Tc1IL-4 cells. BIL were harvested from enzymatically digested brain tissue 24 h after the adoptive transfer of cells and co-stained with PE-H-2Kb-OVA257–264 tetramers and FITC-anti-CD8 mAb, and then analyzed by flow cytometry to enumerate the total numbers (mean± SD) of CD8+/OVA-tetramer+ populations in the lymphocyte-gated populations. *p=0.03, Tc1 versus Tc1IL-4-transferred groups. (B) Mice bearing day 4 i.c. M05 received i.v. injections of 2×107 day 12 Tc1 or Tc1IL-4 cells, with survival subsequently monitored over time. Significance of differences: (log-rank test), p<0.0003 for i.v. Tc1 versus Tc1IL-4-treated groups. All cohorts contained 5 mice/group. Data are representative of two independent experiments performed.


Type-1 and Type-2 T cell responses often cross-regulate each other. IL-10 and IL-4 produced by Th2 and Tc2 cells inhibit IL-12 production from APC and subsequent differentiation of Th1 and Tc1 cells 14, 15. Conversely, IFN-γ produced by Th1 and Tc1 cells inhibits the proliferation of Th2 cells 11. VLA-4 and its counter-receptor VCAM-1 have been previously characterized as key mediators of type-1 T-cell entry into sites of inflammation, such as delayed type hypersensitivity skin, EAE, and other CNS inflammatory lesions 16–19. Therefore our finding that the Type-2 cytokine, IL-4, uniquely down-regulates VLA-4 expression on committed Tc1 cells without altering the functional Type-1 status of these cells suggests a novel mechanism of cross-regulation of Type-1 T cells by Type-2 T cells. In such a scenario, IL-4 elaborated from Th2/Tc2 T cells (particularly in a microenvironment devoid of IL-12) would likely suppress Tc1 (and Th1; our unpublished data) T-cell expression of VLA-4, thereby limiting effector Tc1 cell trafficking into peripheral sites of inflammation/tumor. Furthermore, Tc1 would no longer receive optimal co-stimulatory signals from VLA-4 ligands (i.e. VCAM-1) in situ, which could prove important to their survival and/or sustained effector function within hostile tissue microenvironments, such as that found in tumor lesions in vivo.

Interestingly, previous work by others 9, 10 suggests that IL-4 can restrict the ability of Tc1 cells to produce cytokines including IL-2, IFN-γ, and TNF under conditions of weak antigenic stimulation. Although we did not detect any significant reductions in IFN-γ or TNF-α production by Tc1IL-4versus Tc1 control cells in our study, this could reflect the comparably strong nature of re-stimulation provided by anti-CD3 mAb used in our experiments.

Another interesting finding in the current study is that the potent Type-1-biasing cytokine, IL-12, was able to protect Tc1 cells from IL-4-mediated VLA-4 down-regulation if these cytokines were provided concurrently. Since IL-12 alone was unable to enhance constitutive VLA-4 expression on Tc1 cells or resurrect VLA-4 on Tc1IL-4 cells, this suggests that IL-12 may interrupt or antagonize downstream IL-4-mediated signaling in Tc1 cells. Although the precise signaling events through which IL-4 and IL-12 may regulate VLA-4 expression are still under investigation, we noted that IL-4 (but not IL-13) treatment of IL-4Rα+Tc1 cells results in potent phosphorylation of STAT6 (Supporting Information Fig. 4) and that IL-4-mediated down-regulation of VLA-4 expression on CD8+ T cells is STAT6-dependent (our unpublished data), but independent of MAPK/ERK activity (Supporting Information Fig. 5). Recently, IL-4R/STAT6 signaling has been reported to be inhibited in Type-1 T cells via the action of the suppressor of cytokine signaling 5 (SOCS5) molecule 20. Notably, SOCS5 is up-regulated in Type-1 T cells by the IL-12/STAT4 signaling pathway 20. Therefore, IL-12 may inhibit IL-4R/STAT6 signaling in Tc1 cells via its effects on SOCS5, thereby blocking the IL-4-mediated down-regulation of VLA-4 expression by committed Tc1 cells. Indeed, recent in vivo studies have revealed a requirement for continuous IL-12 action (beyond the initial induction of Type-1 T-cell differentiation) for the maintenance of effective Type-1 T-cell responses against several pathogens 4. In the cancer setting, reports of Type-2 biasing in anti-tumor immunity 21, 22 may be consistent with defects in the tumor-specific CD8+ T-cell repertoire, which may be hypothetically incompetent to traffick into tumor lesions in vivo based on a deficiency in VLA-4 expression.

Our findings would suggest the importance of IL-12 provision (or IL-4 antagonism) in order to counteract such dysfunction and the promotion of enhanced Tc1 infiltration into sites of peripheral disease. Consistent with this paradigm, systemic IL-12 treatment has been reported to enhance VLA-4/VCAM-1-dependent trafficking of T cells into tumor lesions 23. IL-12 also promotes intratumoral production of Tc1-recruiting chemokines 24, 25, such as CXCL9 (Mig), CXCL10 (IP-10), and CXCL11 (I-TAC), that each bind to CXCR3. Interestingly, our recent preliminary studies have also suggested that IL-4 may silence Tc1 expression of CXCR3 (Supporting Information Fig. 6). Given these findings, we believe that further in vivo studies designed to elucidate the survival benefits of interrupting the IL-4R/STAT6 signaling pathway in tumor-bearing mice receiving (i) vaccines to promote Tc1 responses or (ii) adoptive transfer of Tc1 cells are clearly warranted.

Material and methods


C57BL/6 mice (5–9 wk of age), OVA257–264-specific TCR transgenic OT-I mice (RAG-1−/− C57BL/6 background) were purchased from Taconic (Germantown, NY). OVA323–339-specific TCR transgenic OT-II mice (RAG-1−/− C57BL/6 background) were purchased from Jackson Laboratories (Bar Harbor, ME). Mice were maintained in a specific pathogen-free animal facility at the Hillman Cancer Center. All animal work was done in accordance with an Institutional Animal Care and Use Committee-approved protocol.

Tumor cell lines

MO5 is an OVA cDNA transfectant of the murine (H-2b) B16 melanoma cell line 2. The B16 and MO5 cell lines were maintained in CM (RPMI-1640 supplemented with 10% heat-inactivated fetal bovine serum, 100 U/mL of penicillin, 100 μg/mL of streptomycin, and 10 mM L-glutamine; all reagents were from Invitrogen, Carlsbad, CA) in a humidified incubator at 5% CO2 and 37°C.


rmIL-12 was purchased from Cell Sciences (Canton, MA). rmIL-4, rmIL-10, rmIL-13, rhTGF-β1, and rhIL-2 were purchased from Peprotech (Rocky Hill, NJ). Mouse VCAM-1-Fc and human IgG1-Fc fusion proteins containing the same murine Fc domains were purchased from R&D Systems (Minneapolis, MN). Purified anti-CD49d mAb (PS/2) was purchased from Southern Biotech (Birmingham, AL). Purified isotype-control rat IgG2b (RTK4530) was purchased from BioLegend (San Diego, CA). Purified mAbs against IL-12 (C15.6), IFN-γ (R4-6A2 and XMG1.2), IL-4 (11B11), CD3 (145-2C11), CD29 (Ha2/5), FITC-conjugated anti-CD29 mAb(HMβ1), FITC-anti-IFN-γ mAb, FITC-anti-granzyme B mAb, and PE-conjugated anti-CD49d mAb, PE-anti-IL-4 mAb, PE-anti-IL-4 mAb, PE-anti-IL-4Rα mAb were all purchased from BD-Pharmingen (San Diego, CA). Purified, neutralizing anti-IL-10 mAb (JES5-16E3) was purchased from BioLegend. Purified, neutralizing anti-IL-13 mAb (1316H) was purchased from e-Bioscience (San Diego, CA). Purified mAbs against STAT6 and phosphorylated STAT6 were purchased from Cell Signaling Technology (Danvers, MA). PE-OVA-tetramer was purchased from Beckman Coulter (Fullerton, CA). OVA257–264 peptide and OVA323–339 (>95% pure) were synthesized by N-(9-fluorenyl) methoxycarbonyl chemistry in the University of Pittsburgh Cancer Institute's Peptide Synthesis facility.

Generation of OVA-specific effector T cells

Tc1, Tc1IL-4, and Tc2 cells were induced from MACS-separated naïve CD8+ splenic T cells isolated from OT-1 mice, as previously described 2. Purified CD8+ cells were stimulated with 5 μg/mL of OVA257–264 peptide in the presence of irradiated (3000 rad) C57BL/6 spleen cells as feeder cells, 2 ng/mL of rmIL-12, 1 μg/mL of anti-IL-4 mAb, and 100 U/mL of rhIL-2 for Tc1 development. After 48 h, cells were re-stimulated under the same conditions. rhIL-2 was maintained for the entire culture period. Th2 cells were induced from MACS-separated naïve CD4+ cells isolated from OT-II mice. Purified CD4+ cells were stimulated with 5 μg/mL of OVA323–339 peptide in the presence of irradiated C57BL/6 spleen cells as feeder cells, 50 ng/mL of IL-4, 10 μg/mL of two anti-IFN-γ mAb (R4-6A2 and XMG1.2), 10 μg/mL of anti-IL-12 mAb (C15.6), and 100 U/mL of rhIL-2. IL-4-treated Tc1 cells (Tc1IL-4) were generated by culturing day 6 Tc1 cells with IL-4 (10 ng/mL) and IL-2 (100 U/mL) in the presence or absence of 5 μg/mL of OVA257–264 peptide for six additional days. In some experiments, day 6 Tc1 cells were cultured for six additional days in the presence of IL-2 (100 U/mL) along with rmIL-10 (10 ng/mL), rmIL-13 (10 ng/mL) or rhTGF-β (10 ng/mL).

T-effector cell functional analyses

On days 6 and 12 post-stimulation, T cells were harvested in order to measure specific IFN-γ and IL-4 production. Briefly, T cells were re-stimulated with 5 μg/mL of plate-bound anti-CD3 mAb for 6 h, with 10 μg/mL of Brefeldin A (Sigma-Aldrich, St. Louis, MO) and then added for the last 2 h of the incubation period. Intracellular staining for cytokines was then performed using directly PE-conjugated anti-mIFN-γ and anti-mIL-4 mAbs (BD-Pharmingen) and monitored by flow cytometry to confirm functional T-cell polarization status. In some experiments, specific CTL activity was also determined using standard 4 h 51Cr-release assays against M05 (OVA cDNA transduced B16) or B16 parental melanoma target cell lines, as previously described 1.

Cell adhesion assay

T-cell adhesion to immobilized VCAM-1- Fc was assessed as described previously 2. Briefly, 96-well ELISA plates were coated with 10 μg/mL of mouse VCAM-1-Fc, mouse fibronectin, or human IgG1-Fc (all purchased from R&D Systems). Tc1, Tc1IL-4, and Tc2 cells were harvested on day 12 of culture, suspended in binding buffer (0.5% BSA, 2 mM CaCl2, 2 mM MgCl2 in PBS), and then 2×105 cells were added to each well of the plate. For blocking experiments, cells in binding buffer were pre-treated with 20 μg/mL of anti-CD49d mAb (PS/2) or anti-CD29 mAb (Ha2/5) for 15 min at 37°C and then added to the plate. Plates were centrifuged at 500 rpm for 1 min and cells allowed to adhere for 30 min at 23°C with gentle shaking. The plate was then gently washed three times using the binding buffer and the number of adherent cells was enumerated by flow cytometry. Percent cell adhesion was calculated as follows: % adhesion=(number of adherent cells to the VCAM-1-Fc fusion protein−number of adherent cells to human IgG1-Fc fusion protein)/number of total input cells.

Co-stimulation of Tc cells with VCAM-1-Fc

Aliquots of Tc1 or Tc1IL-4 cells were suspended in 0.5% BSA in serum-free RPMI-1640 (GIBCO-BRL) and then added to 96-well plates (2×105 cells/well) that were pre-coated with a low-dose anti-CD3 mAb (0.5 μg/mL) together with either 10 μg/mL of the VCAM-1-Fc or human IgG1-Fc fusion proteins. After incubation for 12 h, supernatants were collected and IFN-γ production was measured by specific ELISA (BD-Pharmingen). The results represent mean±SD of triplicate determinations.

In vivo CTL assay

C57BL/6 mice received i.v. infusions of day 13 cultured Tc1 or Tc1IL-4 cells (2×107 cells/mouse). Erythrocyte-depleted splenocytes from naïve C57BL/6 mice were either (i) pulsed with 10 μg/mL of OVA-peptide, incubated at 37°C for 1 h, and labeled with a high concentration of CFSE (2 μM) (CFSEhigh cells) or (ii) pulsed with no peptide and labeled with a low concentration of CFSE (0.2 μM) (CFSElow cells). Then, equal numbers of cells from each population were mixed (2×107 cells total/mouse) and then infused i.v. into mice that were not pre-treated or that had received i.v. adoptive transfer of Tc1, Tc1IL-4 cells 24 h previously. At 7 h after i.v. infusion, harvested splenocytes were evaluated for the ratio of CFSElow to CFSEhigh cells by flow cytometry. To calculate specific lysis 26, the following formula was used: ratio=(percentage CFSElow/percentage CFSEhigh); percentage specific lysis=(1−(ratio naïve mice)/(ratio T-cell transferred mice)×100).

Therapy of i.c. M05-bearing mice with i.v. adoptive transfer with Tc1 and Tc1IL-4

Preparation of i.c. tumor-bearing mice occurred as previously described 2. Briefly, 5×103 M05 cells were stereotactically injected through an entry site at the bregma 3 mm to the right of the sagittal suture and 4 mm below the surface of the skull of anesthetized mice using a stereotactic frame (Kopf). On day 6, mice received i.v. injections containing 2×107 Tc1 or Tc1IL-4 cells. Animals were monitored daily after treatment for any manifestations of pathology. All cohorts contained 5 mice/group, with two independent experiments performed.

Isolation of brain-infiltrating lymphocytes (BIL)

Mice were sacrificed by CO2 asphyxia and then perfused through the left cardiac ventricle with PBS. Brains were enzymatically digested using a cocktail of 1% collagenase, 1% hyluronidase, and 0.1% DNAase (all purchased from Sigma-Aldrich) for 40 min at 23°C with gentle shaking. The isolated cells obtained from each brain were then resuspended in 70% percoll, overlaid with 37% and 30% percoll, and then centrifuged for 20 min at 500g. Enriched BIL populations were recovered at the 70 to 37% percoll interface for consequent flow cytometry analyses.

Western blot analysis

Tc1 and Tc2 cells were harvested at day 6 after primary in vitro stimulation. After washed twice, cells (1×106) in 1 mL of RPMI medium were stimulated with or without 10 ng/mL of IL-4 for 30 min at 37°C. Tc1 and Tc2 cells (5–10×106) were analyzed for STAT6 and phosphorylated STAT6 (p-STAT6) expression via Western blots using specific anti-p-STAT6 and anti-STAT6 antibodies. Cell pellets were lysed using 200 μL of 1% NP40 in PBS containing protease inhibitors (Complete; Boehringer Mannheim, Indianapolis, IN) for 1 h on ice. After centrifugation at 13 500g for 30 min, the supernatant was mixed 1:1 with SDS-PAGE running buffer and proteins separated on 10% PAGE gels, before electroblotting onto nitrocellulose membranes (Millipore, Bedford, MA). Blots were imaged on Kodak X-Omat Blue XB-1 film (NEN Life Science Products, Boston, MA) using horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin (Bio-Rad, Hercules, CA) and the ECL chemiluminescence detection kit (NEN Life Science Products).

Tc1/Th2 cell co-cultures

OT-I-derived day 6 Tc1 cells were mixed with OT-II-derived day 6 Th2 cells at a 2:1 ratio and stimulated with 5 μg/mL of OVA323–339 peptide and 100 U/mL of hrIL-2 with or without 10 μg/mL of anti-IL-4 mAb in the presence of irradiated C57BL/6 splenocytes as feeder cells. Five days later, the percentage of CD49d+cells was assessed by flow cytometry.

Statistical analysis

Survival data were compared using a log-rank test. All intergroup comparisons of means obtained from other data were assessed with one-sided, equal variance t tests. Before testing, percentage data were logit transformed and all other data were log transformed. p-Values<0.05 were considered significant.


The authors thank Dr. Per Basse for his kind gift of OT-I mice, Ms. Lisa Bailey and Drs. Xi Zhao and Yanyan Qu for their technical assistance. The authors also thank Dr. Amy Wesa for her careful review and helpful comments provided during the generation of this paper. This work was supported by National Institutes of Health (NIH) grants R01 CA63350 (to W.J.S.) and R01 NS055140 (to H.O).

Conflict of interest: The authors declare no financial or commercial conflict of interest.