|Home | About | Journals | Submit | Contact Us | Français|
Murine CD4+ T cells cultured under Type-1 polarizing conditions selectively express significantly higher levels of the VLA-4 and VLA-6 integrins when compared to T cells cultured under Type-2 or non-polarizing (Type-0) conditions. This difference appears due to the action of IL-4, since loss of VLA-4/-6 expression on Th cells was prevented by inclusion of neutralizing anti-IL-4 mAb during the initial culture period. We also observed that CD4+ T cells deficient in Stat6, a critical component of the IL-4R signaling cascade, retained high levels of VLA-4 and VLA-6 expression, regardless IL-4 status in culture conditions. When applied to committed Th1 cells, rIL-4 readily inhibited VLA-4 and VLA-6 expression to levels observed for Th2 cells, without altering the Type-1 functional status of these cells. Conversely, low levels of VLA-4/-6 expressed by committed Th2 cells could not be resurrected by culture in the presence of the Th1-kines IL-12p70 and IFN-γ. Predictably, among the Th populations evaluated, Th1 cells alone adhered efficiently to, and were co-stimulated by, plate-bound VCAM-1 and laminin in a VLA-4- or VLA-6-dependent manner, respectively. Finally, adoptive-transferred Th1 (but not Th2) cells developed from OT-II mice were uniquely competent to traffick into OVA+ M05 melanoma lesions in vivo, thereby enhancing the therapeutic benefits associated with co-transferred OVA-specific Tc1 (OT-I) cells. These data suggest that treatment strategies capable of sustaining/enhancing VLA-4/-6 expression on Th1 effector cells may yield improved clinical efficacy in the cancer setting.
Primed CD4+ T cells may be subdivided into functional subsets based on their patterns of cytokine secretion in response to cognate antigen. Th1 cells produce IFN-γ and IL-2, and activate macrophages and cytotoxic T lymphocytes to promote cell-mediated immunity against intracellular pathogens and transformed cells, while Th2 cells produce IL-4, IL-5, and IL-13 and play important roles in the regulation of humoral and innate immunity against extracellular pathogens (1, 2). Other CD4+ T cell functional subsets include Th17 cells, that secrete IL-17 and promote inflammatory responses involving activated macrophages and neutrophils (3, 4), and Th3/Treg cells, that produce IL-10 and/or TGF-β while serving as “suppressor” T cells (5, 6).
One important variable in the maintenance of functionally-polarized immune responses appears to be the acquisition of specific homing receptors that direct the localization of the different types of effector T cells within relevant sites of antigenic surveillance/restimulation (7, 8). In this regard, signals delivered by distinct cytokines may imprint developing T cell subsets with specific homing patterns by “setting” expression levels of certain adhesion molecules and chemokine receptors (8, 9). For example, the preferential expression of P- and E-selectin ligands (10) and the chemokine receptors CXCR3 and CCR5 (11) on Th1 T cells, at least in part, explains their selective recruitment into sites of inflammation. However, despite the key role(s) that integrins play in leukocyte recruitment and function, their regulated expression by distinct, polarized Th cell subsets has remained understudied.
We have recently demonstrated that Very Late Antigen (VLA)-4 (a heterodimer of the α4 and β1 integrins) is preferentially expressed by Type-1 CD8+ cells (Tc1) when compared to Type-2 CD8+ T cells (Tc2), providing Tc1 cells with a differential ability to infiltrate into tumor lesions after adoptive T cell transfer (12, 13). Since integrins expressed by CD4+ T cells are similarly known to play critical role(s) in regulating inflammation in peripheral tissues (including tumors), we sought to determine whether certain integrins (including VLA-4) were also preferentially expressed by Type-1 CD4+ T cells (Th1 cells) and associated with the ability of these cells to trafficking into tumor lesions in vivo. In this report, we show for the first time that VLA-4 (CD49d/CD29) and VLA-6 (CD49f/CD29) are both preferentially expressed on murine Th1 cells, but not on Th2 cells, and that these integrins are crucial to the migration of Type-1 CD4+ T cells into tumor lesions.
C57BL/6 mice, BALB/c mice (5-9 weeks of age), OVA323-339-specific TCR transgenic OT-II mice (RAG-1−/− C57BL/6 background), Stat6−/− mice (C57BL/6 background) were purchased from the Jackson Laboratory (Bar Harbor, ME). OVA257-264-specific TCR transgenic OT-I mice (RAG-1−/− C57BL/6 background) were purchased from Taconic (Germantown, NY). Mice were maintained in a specific-pathogen-free animal facility at the University of Pittsburgh Cancer Institute. All animal work was done in accordance with an Institutional Animal Care and Use Committee (IACUC)-approved protocol.
Reagents used in this study included: rmIL-12 (Cell Sciences, Canton, MA), rhIL-2, rmIL-4, rmIL-6 and mTGF-β (all from Peprotech, Rocky Hill, NJ), mouse VCAM-1-Ig fusion protein (R&D Systems, Minneapolis, MN), purified anti-CD49d mAb (PS/2; Southern Biotech, Birmingham, AL), purified anti-CD49f mAb (Go H3; GeneTex, San Antonio, TX), and purified isotype-control rat IgG2b (RTK4530; BioLegend, San Diego, CA). Additional antibodies (all from BD-Pharmingen, San Diego, CA) included purified anti-mIL-12 mAb C15.6, anti-mIFN-γ mAbs R4-6A2 and XMG1.2, anti-mIL-4 mAb 11B11, anti-mCD3 mAb 145-2C11, anti-mCD29 mAb Ha2/5, fluorescein isothiocyanate (FITC)-conjugated anti-CD29 mAb HMβ1, FITC-anti-IFN-γ mAb, FITC-anti-CD69 mAb, FITC-anti-CD25 mAb, FITC-anti-CD54 mAb, FITC-anti-CD11a mAb and Phycoerythrin (PE)-conjugated anti-CD49d mAb, PE-anti-CD49f, PE-anti-IL-4 mAb, PE-anti-IL-17 mAb, PE-anti-CD44 mAb and PE-anti-α4β3 integrin mAb. The OVA257-264 and OVA323-339 peptides (>95% pure) were synthesized by N-(9-fluorenyl) methoxycarbonyl (fMOC) chemistry by the University of Pittsburgh Cancer Institute (UPCI) Peptide Synthesis Facility and confirmed for identity by the UPCI Protein Sequencing Facility (both Shared Resources).
Th1 and Th2 cells were induced from MACS-separated naïve CD4+ splenic T cells isolated from wild-type (or Stat6−/−) mice as described previously (14). Briefly, to generated Th1 cells, CD4+ T cells were stimulated with 5 μg/ml soluble anti-CD3 mAb in the presence of irradiated (3000 rad) C57BL/6 spleen cells as feeder cells, rhIL-2 (100 U/ml), rmIL-12 (4 ng/ml), rmIFN-γ (4 ng/ml) and anti-IL-4 mAb (10 μg/ml). Th2 cells were induced in anti-CD3/feeder cultures supplemented with rhIL-2 (100 U/ml), rmIL-4 (50 ng/ml), two anti-IFN-γ mAbs (R4-6A2 and XMG1.2; 10 μg/ml each), and anti-IL-12 mAb (C15.6; 10μg/ml). Where indicated, in some experiments, T cells were generated in the presence of rhIL-2 (100 U/ml) with or without various cytokines (rmIL-4, rmIL-12, rmIFN-γ) or neutralizing mAbs (anti-IL-4 mAb or anti-IL-4 mAb + anti-IL-12 mAb + anti-IFN-γ mAbs). rmIL-4-treated Th1 cells (Th1IL-4) were generated by culturing day 6 Th1 cells with rmIL-4 (10 ng/ml) and rhIL-2 (100 U/ml) for 6 days.
For the adoptive-transfer experiments, polarized Th1 and Th2 cells were generated from OT-II mice using specific peptide (OVA323-339; 5 μg/ml) stimulation instead of anti-CD3 mAb. OT-I-derived Tc1 cells were developed as previously described (12). In all experiments, rhIL-2 was maintained for the entire culture period. On days 6, 8 and 12 after primary in vitro antigenic stimulation, T cells were harvested and then restimulated with 5 μg/ml of plate-bound anti-CD3 mAb for 6 hrs in the presence of 10 μg/ml of brefeldin A (Sigma) over the final 2 hours of the incubation period. These cells were then analyzed for IFN-γ (for Th1) vs. IL-4 (for Th2) production by intracellular staining as monitored using flow cytometry (Supplemental Fig.1) in order to confirm their functional polarization status.
T cell adhesion to immobilized VCAM-1-Ig was assessed as described previously (12). Briefly, 96-well ELISA plates were coated with 10 μg/ml of mouse VCAM-1-Ig, mouse laminin, mouse fibronectin, control-human Ig or heat-denatured BSA. T cells were harvested at day 9 of culture, suspended in binding buffer (0.5% BSA, 2 mM CaCl2, 2 mM MgCl2 in PBS) and then added to the plate. For blocking experiments, cells resuspended in binding buffer were pre-treated with 20 μg/ml of anti-CD49d mAb (PS/2) or 20 μg/ml anti-CD29 mAb (Ha2/5), or two different doses of anti-CD49f mAb (Go H3) (20 μg/ml or 10 μg/ml) for 15 min at 37°, before being added to microwells. Plates were then centrifuged at 500 rpm for 1 min and cells allowed to secure their adhesions for 30 min at 23° with gentle shaking. Plates were then gently washed 3 times using binding buffer and the number of adherent cells enumerated by flow cytometry. % Adhesion was calculated as follows; % Adhesion = [(Number T cells adherent to VCAM-1-Ig) – (Number of T cells to adherent to heat-denatured BSA)]/Total number of input T cells.
Aliquots of cultured day 12 Th1 or Th2 cells suspended in serum-free RPMI containing 0.5% BSA were added to 96-well plates (2 × 105/well) that had been pre-coated with anti-CD3 mAb (1 μg/ml or 8 μg/ml for co-immobilization with VCAM-1-Ig, laminin, respectively) together with either 5 μg/ml of VCAM-1-Ig, 50 μg/ml of laminin or an equivalent dose of heat-denatured BSA. After incubation for 24h, supernatants were collected and IFN-γ production measured by specific ELISA (BD-Pharmingen), with results reported as mean +/- SD of triplicate determinations.
C57BL/6 mice received s.c. injections of 1 × 106 M05 (OVA-transfected B16 melanoma) cells in their right flank on day 0. On day 7, tumor-bearing mice received i.v. tail-vein injections with 2-5 × 107 Th1 or Th2 cells developed from OT-II mice, with or without, 6 × 106 Tc1 cells developed from OT-I mice. Treated animals were monitored daily for any therapy-associated toxicity and for tumor size (in mm2). Mice were sacrificed when tumors exceeded 400 mm2, if their lesions became ulcerated or if an animal displayed any signs of distress. Data are reported as percent survival over time and are displayed in Kaplan-Meier plots.
To investigate the in vivo trafficking of T cells into M05 tumor bearing mice, cultured day 9 Th1 and Th2 cells generated from OT-II mice were labeled in vitro with 0.4 μM of Carboxyfluorescein Succinimidyl Ester (CFSE, Vybrant CFDA SE Cell Tracer kit; Molecular Probes) per the manufacturer's protocol. Seven days after tumor inoculation, mice received i.v. injections with 2 × 107 CFSE-labeled Th1 or Th2 cells. For VLA-4 or VLA-6 inhibition experiments, 50 μg/ml of anti-CD49d mAb (PS/2), anti-CD49f mAb (Go H3) or control rat IgG was used to pre-treat T cells prior to i.v. infusion. Additionally, the PS/2 (250 μg) or GoH3 (150 μg) mAbs or equivalent quantities of control rat IgG were administrated i.p. 4h prior to adoptive T cell transfers. To examine TIL, M05 tumors were resected, minced and then enzymatically-digested using a cocktail of 1% collagenase, 1% hyaluronidase and 0.1% DNAase for 40 min at 23° with gentle stirring. The resulting cell suspension was filtered through cell strainers (BD-Biosciences, San Jose, CA) and the flow through overlaid on an equal volume of lympholyte-M (Cedarlane Laboratories, NC) prior to centrifugation for 20 min at 800 × g. An enriched population of tumor-infiltrating lymphocytes (TIL) was then recovered at the gradient interface. Lung-infiltrating lymphocytes were similarly isolated by enzymatical digestion, and along with splenocytes, used as controls, where indicated. Enriched lymphocytes were counterstained with PE-anti-mouse CD8 mAb and the frequency of CD8+CFSE+ T cells determined using flow cytometry.
All inter-group comparisons 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.
In vitro cultured Th1 and Th2 cells developed from wild-type C57BL/6 mice were evaluated for expression of various activation and adhesion molecules by flow cytometry. We noted that the VLA-4 (CD49d/CD29) and VLA-6 (CD49f/CD29) integrins were preferentially expressed by Th1 vs. Th2 cells (Fig. 1A). In contrast, other activation markers (CD25, CD44, CD69) and adhesion molecules (CD11a, CD54, α4β3 integrin) were expressed equitably by both Th1 and Th2 cells (Supplementary Fig. 2). To better understand how VLA-4 and VLA-6 expression may be differentially regulated during the in vitro differentiation of Th1 and Th2 cells, we next examined the expression kinetics of these β1-integrins by Th1 vs. Th2 cells over their first 13 days in culture. Although both Th1 and Th2 cells initially displayed similar levels of VLA-4 (CD49d/CD29) and VLA-6 (CD49f/CD29) expression (through day 4 of polarizing cultures), thereafter, both molecules were consistently up-regulated/maintained on Th1 cells, while being coordinately down-modulated on Th2 cells (Fig. 1B, 1C).
Since our initial experiments employed heterogenous T cell populations activated using anti-CD3 mAb, we next evaluated whether these results could be generalized to OVA-specific Th1 and Th2 cells developed from OT-II (OVA-specific TCR transgenic) mice. As depicted in Supplemental Fig. 3, both VLA-4 and VLA-6 were preferentially expressed by Th1 vs. Th2 cells generated from OT-II mice, regardless of whether the “cognate” antigen used in vitro was provided in the form of OVA323-339 peptide-pulsed syngenic APC or immobilized, plate-bound anti-CD3. Since either antigen format yielded similar results, we believe that the differential expression of VLA-4 and VLA-6 by Th1 vs. Th2 cells was most likely due to the impact of specific cytokines added or neutralized by antibodies employed during the in vitro polarization culture period.
To determine why VLA-4 and VLA-6 exhibited differential expression patterns on Th1 vs. Th2 cells, we next investigated the potential regulatory influence of the soluble factors employed in the polarizing cultures. CD4+ T cells cultured under neutral conditions (rIL-2 without other factors) displayed low levels of VLA-4 and VLA-6 expression that were comparable to those observed for Th2 cells (Fig. 2). Notably, the inclusion of neutralizing anti-IL-4 mAb prevented loss of VLA-4 and VLA-6 expression on “Th0” cells (Fig. 2A, 2B). This suggested the functional dominance of endogenous IL-4 production on CD4+ T cell expression of these β1-integrins. The addition of IL-12, but not IFN-γ to Th0 culture conditions resulted in partial maintenance of CD49f and CD29 (but not CD49d) expression on CD4+ T cells, suggesting that VLA-4 and VLA-6 expression are not regulated via absolutely identical mechanisms and that these Type-1 associated cytokines are not operationally dominant in determining levels of VLA-4 (and to a lesser extent VLA-6) expression on activated CD4+ T cells. This was further evidenced under neutral culture conditions depleted of IL-4, where the co-addition of anti-IL-12 and anti-IFN-γ mAbs did not result in any additional modulation of VLA-4 (or VLA-6) expression by CD4+ T cells (when compared to anti-IL-4 mAb only controls, Fig. 2A, 2B). CD4+ T cells cultured in the presence of neutralizing antibodies against IL-4, IL-12 and IFN-γ were poor producers of IFN-γ or IL-4, suggesting that up-regulation of VLA in CD4+ T cells occurs irrespective of polarization status (data not shown). We also observed that Th17 cells (producing IL-17 but not IFN-γ or IL-4) induced in vitro also expressed high levels of VLA-4 and VLA-6, suggesting that up-regulation of VLA-4/-6 is not a property unique to Th1 cells (data not shown). Given functional polarization biases in T cell responses that have been reported between various strains of mice (i.e. C57BL/6 are Th1-biased, while BALB/c mice are Th2-biased; ref. 15), we performed an analogous set of experiments using T cells isolated from BALB/c mice and observed similar results to those obtained in our B6 model (Supplemental Fig. 4).
Since IL-4 appeared to represent a dominant suppressor of VLA-4/-6 expression by CD4+ T cells cultured under non-Type-1 polarizing conditions, we next determined whether IL-4 could down-regulate the expression of these VLA molecules on committed Th1 cells. We observed that the treatment of committed Th1 cells with IL-4 (i.e. Th1IL-4) resulted in the dramatic down-regulation of both VLA-4 and VLA-6 expression (Fig. 3A). Interestingly, this phenotypic change occurred in the absence of any overt alteration in the Type-1 pattern of cytokine production by these T cells, which remained committed to IFN-γ (but not IL-4) production (Fig. 3B). In contrast, the addition of Type-1 cytokines (IFN-γ, IL-12) could not resurrect expression of VLA-4 and VLA-6 expression on committed Th2 cells (Supplemental Fig. 5), suggesting the possible irreversible (IL-4-dependent) nature of VLA-4 and VLA-6 loss of expression on Th2 cells.
Stat6 is an important down-stream component of the IL-4R-mediated signaling pathway (16, 17). Therefore, we next determined whether ablation of IL-4R-signaling in STAT6(-/-) mice would yield T cells refractory to IL-4-induced silencing of VLA-4 and VLA-6 expression. We observed that Stat6-deficient CD4+ T cells cultured under Th1, Th2, or neutral conditions displayed comparable, high levels of VLA-4 and VLA-6 expression (Fig. 4). These data are consistent with IL-4/IL-4R-mediated signaling dictating the expression pattern of these integrins by polarized subsets of Th cells.
Despite the typical redundancy in integrin recognition of a range of physiologic ligands, VLA-4 and VLA-6 expressed on T cells are primarily thought to recognize VCAM-1 and laminin, respectively (18-20). To examine the functional significance of differentially expressed VLA-4 and VLA-6 on Th subsets, Th1 and Th2 cells were tested for their abilities to adhere to plate-bound VCAM-1-Ig fusion protein or laminin. Consistent with the differential expression of VLA-4 and VLA-6 on Th1 and Th2 cells, Th1 cells exhibited specific adhesion to plate-coated VCAM-1-Ig and laminin, whereas Th2 cells displayed only background levels of adhesion to these ligands (Fig. 5A). In contrast, both Th1 and Th2 cells (which express comparable levels of the VLA-5 and αVβ3 integrins, Supplemental Fig. 2) adhered equitably to plates coated with fibronectin (Fig. 5A). Pretreatment of Th1 cells with specific anti-VLA mAbs or anti-CD29 (but not control Ig) completely abrogated the ability of Th1 cells to adhere to relevant ligand-coated plates (Fig. 5B, C). Consistent with the comparable levels of VLA-4 and VLA-6 expression found on all Th subsets developed from STAT6−/− mice, these T cells displayed a uniformly strong binding capacity to immobilized VCAM-1 and laminin (Supplemental Fig. 6A). Moreover, while wild-type CD4+ T cells cultured under neutral condition bound to VCAM-1 or laminin poorly, depletion of IL-4 from the culture media resulted in sustained VLA-4 and VLA-6 expression (Fig. 2) and the ability of these T cells to adhere to both immobilized VCAM-1 and laminin, respectively (Supplemental Fig. 6B).
Since the VLA-4-VCAM-1 and VLA-6-laminin interactions are known to provide costimulatory signals to T cells (18, 20), we next evaluated the impact of providing both immobilized VCAM-1 or laminin along with a sub-mitogenic dose of anti-CD3 mAb on T cell production of cytokines. As expected, when compared with the anti-CD3 only treatment group, committed Th1 cells produced higher levels of IFN-γ, while committed Th2 cells failed to produce elevated levels of IL-4, upon co-stimulation with either VCAM-1-Ig or laminin (Fig. 5D).
We next sought to determine the physiologic significance of differential VLA-4 and VLA-6 expression by Th1 cells using an adoptive transfer approach applied to the M05 (B16.OVA) melanoma model. Mice bearing established day 7 s.c. melanomas received i.v. transfers of CFSE-labeled Th1 or Th2 cells developed from OT-II mice. Thirty-six hours later, tumor-infiltrating lymphocytes (TILs) were isolated and analyzed by flow cytometry. A high percentage (23%) of CSFE+ CD4+ TILs were identified from the cohort of animals receiving Th1 infusions, whereas, only very few CFSE+ CD4+ events (1%) were observed for the cohort of mice receiving Th2 cells (Fig. 6A, B). The number of CFSE+ cells accumulating in the spleens and lungs of mice treated with Th1 or Th2 infusions was comparable (Fig. 6B), suggesting the tumor-specific relevance of VLA-4 and/or VLA-6 to tumor homing/infiltration. As shown in Fig. 6C, pretreatment of OT-II Th1 cells with anti-VLA-4 mAb significantly inhibited their infiltration into M05 tumor lesions after adoptive transfer, but had little effect on their localization within the spleens of treated mice. While analogous experiments using the anti-VLA-6 mAb suggested a partial reduction in Th1 cell infiltration of the tumor (Fig. 6C), this difference did not reach statistical significance. These results were further confirmed in experiments in which anti-VLA-4 (PS/2), anti-VLA-6 (Go H3) or isotype control mAb were co-administered along with i.v. transferred Th1 cells in MO5-bearing mice (Fig. 6C).
We next assessed the therapeutic benefit of adoptively-transferred OT-II Th1 vs. Th2 cells applied to a day 7 established M05 (B16.OVA) melanoma model. Intravenous transfer of (up to 4 × 107) anti-OVA Th1 or Th2 cells in the absence of OT-I Tc1 transfer did not prolong the survival of treated animals (Fig. 6D, top). However, when combined with the co-transfer of OT-I Tc1 cells, Th1 (but not Th2) cells prolonged the survival of tumor-bearing mice vs. animals with Tc1 cells alone (p< 0.05; Fig. 6D, bottom).
While the vast majority of cancer immunotherapy studies have focused on the generation of anti-tumor CD8+ T cells, recent studies support the critical functions of CD4+ Th cells in productive anti-tumor immunity (5, 21-26). In this regard, CD4+ T cells have shown to assist in the priming of naïve, tumor-specific CD8+ T cells by directly providing IL-2 (27-29) or via the indirect CD40/CD40L-dependent “licensing” of DC (30-32). Recent studies also support an important role of tumor-specific CD4+ T cells in facilitating and sustaining CD8+ T effector and memory cell functions (21-23). In the former case, recruitment of tumor-specific CD4+ T cells into tumor lesions is envisioned to result in changes within the tumor microenvironment that are permissive for CD8+ T effector cell recruitment (via loco-regional production of inflammatory chemokines) and survival/proliferation (via production of IL-2 among other cytokines).
These observations suggest that the efficacy associated with adoptive immunotherapy approaches employing anti-tumor CD8+ T cells would likely be improved by the co-transfer of tumor-specific CD4+ T cells. Based on the results obtained in the current study, Type-1 CD4+ T cells would likely be most beneficial in such approaches, at least in part, due to their preferred capacity to effectively traffic into tumor sites in vivo. The differential ability of Th1 (vs. Th2) cells to become TIL appears critically dependent upon their expression of the β1-integrin VLA-4 and its interaction with VCAM-1 (expressed by activated vascular endothelial cells, such as those found in the tumor stroma; ref. 33). While there was some evidence for the ability of anti-VLA-6 mAb to limit tumor-infiltration by Th1 cells in vivo, which would appear to warrant prospective re-evaluation, this level of antagonism did not reach significance in the current studies. Regardless of the functional dominance of VLA-4 vs. VLA-6 in the ability of Th1 cells to migrate into MO5 tumors, these cells clearly facilitated the anti-tumor efficacy of co-transferred Tc1 cells, while exhibiting minimal anti-tumor impact when adoptively-transferred on their own.
Interestingly, while we observed that the Type-2 cytokine IL-4 was unable to alter the functional polarization of Th1 cells (based on cytokine profiling) in vitro, IL-4 was competent to suppress both VLA-4 and VLA-6 expression by Th1 cells via a STAT6-dependent mechanism. This finding suggests IL-4R/STAT6-signaling in committed Th1 cells may not induce expression of GATA-3 or c-Maf, which serve as central regulators of Type-2 cytokine genes in Th2 cells (34, 35). We are currently evaluating this possibility.
Given that neutralization of endogenous IL-4 resulted in enhanced VLA-4 and VLA-6 expression by CD4+ T cells activated under neutral polarizing conditions, and that Th2 cells developed from STAT6−/− mice express high levels of VLA-4/-6, these findings may suggest that therapeutic benefits would be anticipated from antagonizing the IL-4R/STAT6 signaling pathway in cancer patients. Under such conditions, one would expect improvements in Type-1 effector T cell function, recruitment of Th1 (and Tc1; ref. 12) into tumor lesions and corollary enhancement in anti-tumor efficacy.
The finding that antibodies directed against VLA-6 only modestly block Th1 cell recruitment into M05 tumors in vivo does not formally minimize the physiologic role of the VLA-6/laminin in Th1 cells. It remains possible that the VLA-6/laminin interaction plays an important role in retention of Th1 cells within laminin-rich tumor interstitium (36, 37) and/or that VLA-6 (along with VLA-4) may contribute key co-stimulatory signals to tumor-specific Th1 cells within the tumor microenvironment that may confer prolonged survival and effector function in vivo. Further studies in the current model, as well as alternate murine tumor models, will be needed in order to confirm or refute these possibilities and to determine the generality of these results and their translational utility.
Supplemental Figure 1. Cytokine production by CD4+ T cell subsets generated in vitro. Th1 and Th2 cells were generated from CD4+ T cells isolated from C57BL/6 spleen cells, as described in Materials and Methods. On day 8 following primary stimulation, T cells were harvested and restimulated with plate-bound anti-CD3 mAb for 6 hours before assessment of cytoplasmic expression of IFN-g and IL-4 using flow cytometry. Inset numbers indicate the percentage of cells in a given quadrant among total T cells analyzed. Data are representative of 3 independent experiments performed.
Supplemental Figure 2. Expression of activation/adhesion molecules by cultured Th1 and Th2 cells. T cells generated as in Supplemental Fig.1 were analyzed for their cell surface expression of activation (CD69, CD44, CD25) or adhesion (a4b3 integrin, aVb3 integrin, CD49e, CD54, CD11a) molecules by flow cytometry. Data are representative of 3 independent experiments performed.
Supplemental Figure 3. Differential expression of VLA-4 and VLA-6 in Th1 and Th2 cells is independent of the mode of primary stimulation (i.e. signal 1). Wild-type C57BL/6-derived Th1 and Th2 cells were generated by stimulating T cells with soluble anti-CD3 mAb in the presence of antigen-presenting cells or by stimulating with plate-bound anti-CD3 mAb in the absence of antigen-presenting cells. OT-II derived Th1 or Th2 cells were generated by stimulating with OVA323-339 peptide in the presence of syngenic APC or using plate-bound anti-CD3 as in the case of wild-type T cells. On day 12, the cultured T cells were harvested and VLA-4 and VLA-6 expression was analyzed by flow cytometry. Inset numbers reflect the percentage of CD49+CD29+ cells among total T cells analyzed. Data are representative of 3 independent experiments performed.
Supplemental Figure 4. IL-4-mediated VLA-4 and VLA-6 down-regulation in wild-type BALB/c-derived CD4+ T cells. BALB/c-derived CD4+ T cells were stimulated with anti-CD3 mAb under Th1- or Th2-inducing cytokine conditions or under neutral conditions in the absence or presence of anti-IL-4 mAb, as indicated. T cells were harvested on day 12 of culture, with VLA-4 and VLA-6 expression assessed by flow cytometry. Inset numbers reflect the percentage of CD49+CD29+ cells among total T cells analyzed. All data are representative of 3 independent experiments performed.
Supplemental Figure 5. Failure of Th1 cytokines to up-regulate VLA-4 and VLA-6 expression on committed Th2 cells. C57BL/6-derived Th2 cells were treated with IL-12 (4 ng/ml) and IFN-g (4 ng/ml) beginning on day 10 of culture. On day 18, the T cells were harvested and VLA-4 and VLA-6 expression analyzed by flow cytometry. Inset numbers reflect the percentage of CD49+CD29+ cells among total T cells analyzed. Data are representative of 3 independent experiments performed.
Supplemental Figure 6. IL-4 suppression of VLA-mediated CD4+ T cell adhesion and co-stimulation is Stat6-dependent. A., Splenic CD4+ T cells isolated from wild-type or Stat6-/- mice were stimulated with plate bound anti-CD3 mAb and cultured under Th1- or Th2-biasing cytokine conditions as described in Materials and Methods. T cells were harvested on day 10 of culture and assessed for their ability to bind to immobilized VCAM-1-Ig (A., top) or laminin (A., bottom). *P = 0.001, percentage adhesion of wild-type Th1 versus Th2 against VCAM-1-Ig; P = 0.0002, percentage adhesion of wild-type Th1 versus Th2 against laminin. B., T cells derived from either wild-type or Stat6-/- mice under Th1- vs. Th2-biasing conditions were harvested on day 12 of culture and assessed for their ability to bind to immobilized VCAM-1-Ig (B., top) or laminin (B., bottom). *P < 0.05, vs. adhesion of wild-type T cells cultured under neutral condition against VCAM-1-Ig; **P < 0.005, vs. adhesion of wild-type T cells cultured under neutral condition against laminin. All data are representative of 3 independent experiments performed.
The authors wish to thank Drs. Amy Wesa and Jennifer Taylor for helpful suggestions provided during the preparation of this manuscript.
2This work was supported by National Institutes of Health (NIH) grants P01 CA100327 and R01 CA63350 (to W.J.S.), PO1 CA101944 (to P.H.B.) and R01 NS055140 (to H.O.).
Financial Disclosure: The authors have declared there are no financial conflicts of interest in regards to this work.