Expression of a Functional IL-2 Receptor in Vascular Smooth Muscle Cells

Interleukin-2 is best known as a key regulator of T cell homeostasis, promoting the upregulation of immune responses through proliferation and differentiation of effector T cell subsets and downregulation via its role in survival and expansion of T regulatory (Treg) cells (1). Although well established as a soluble protein produced by T cells, our laboratory has focused on the association of IL-2 with perlecan, which is the main heparan sulfate proteoglycan in vascular basement membranes (2–5). We previously reported that IL-2–deficient mice lose smooth muscle cells over time, suggesting that IL-2 may support vascular smooth muscle cell (VSMC) survival and/or proliferation in vivo (5). Given that smooth muscle cells are not typically considered IL-2 responsive, these findings led us to ask whether smooth muscle cells express IL-2Rs. We subsequently reported the expression of IL-2Rβ (CD122) on murine and human aortic smooth muscle cells (5). Our current work expands our previous observations by assessing the expression of the IL-2Rα and -γ subunits by VSMCs.

The present work was prompted, in large part, by the lack of information in the literature regarding IL-2Rs and smooth muscle cells. IL-2R is made up of three subunits: α (CD25), β (CD122), and γ (CD132). β and γ are shared with IL-15R, and the α subunits are distinct. In an extensive literature search, we identified one paper documenting IL-15α, IL-2/IL-15β, and IL-2/IL-15γ expression via RT-PCR in isolated smooth muscle cells (6). This report also documented IL-15Rα expression on VSMC histologically. We found another report showing expression of IL-15α, -β, and -γ by immunohistochemistry on VSMC in sections of murine lung tissue (7). We have not identified any reports showing expression of IL-2Rα in smooth muscle cells.

Of even greater physiologic importance is whether IL-2 influences the function of VSMCs. There is very little known about the direct impact of IL-2 on VSMCs. One in vitro study reported that IL-2 has no direct physiologic impact on SMC biology, instead acting as a modifier in the presence of angiotensin (8). Another study from our laboratory showed that, in stark contrast to monomeric IL-2, a dimeric form of IL-2 isolated from mammalian arteries was directly cytotoxic to VSMC (9). Anti–IL-2Rβ Abs blocked this cytotoxic effect.

In vivo studies also suggest that IL-2 impacts the vasculature with no clear consensus as to its role and/or mechanism of action. In atherosclerotic animal models, systemic administration of IL-2 enhanced lesion growth, whereas local administration of IL-2 decreased atherosclerosis via an impact on Treg cells (10, 11). Also through Treg cells, IL-2 was found to limit the development of aneurysms in a murine model (12). Finally, in the cancer field, high-dose infusion of IL-2 is well known to cause a vascular leak syndrome, causing side effects such as pulmonary edema (13). This injury is thought to be an indirect effect of IL-2 on endothelial cells through neutrophils and lymphocytes (14).

Given the localization of IL-2 in proximity to VSMC, the phenotype of IL-2–deficient mice, and our prior observation that VSMC express IL-2Rβ, we asked whether human VSMC express all three subunits of the IL-2R and whether IL-2 influences the function of VSMC. We discovered that human VSMC express all three subunits of the IL-2R. Functional studies revealed that IL-2 promotes proliferation and migration of VSMC in vitro, and IL-2 expression is significantly upregulated in vivo under conditions, promoting intimal hyperplasia.

Small sections of human iliac artery were obtained from deceased organ donors. Consent for research was provided by next of kin. Human tissues were handled in accord with the Wright State University Institutional Biosafety Committee. The animal studies were approved by the Institutional Animal Care and Use Committee at the University of Florida, and conformed to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health Publication, revised 2011). Jurkat T cells, clone E6-1, were from American Type Culture Collection (Manassas, VA). Chemical reagents, unless otherwise indicated, were obtained from Sigma-Aldrich (St. Louis, MO). Abs used were as follows: rabbit anti-human polyclonal anti–IL-2Rα, rabbit anti-human polyclonal anti–IL-2Rβ, and rabbit anti-human polyclonal anti–IL-2Rγ were from Bioss Antibodies (Woburn, MA). Rabbit anti-human smooth muscle cell α-actin was from Novus Biologicals (Littleton, CO). Mouse anti-smoothelin (multiple species, clone R4A) was from MilliporeSigma (Temecula, CA). Rabbit anti-human smooth muscle myosin H chain (SM-MHC) was from Abcam (Cambridge, MA). Neutralizing anti–IL-2Rα and -β Abs for immunofluorescence studies were from R&D Systems (Minneapolis, MN). Neutralizing anti–IL-2Rα and -β Abs used to inhibit proliferation were from Thermo Fisher Scientific (Waltham, MA). Anti-human IL-2 Abs were from Rockland Immunochemicals (Limerick, PA). Culture materials used were as follows: DMEM, Ham’s F12, and epidermal growth factor were from Thermo Fisher Scientific. SMC media and a proprietary smooth muscle growth supplement containing 2 ng/ml insulin-like growth factor, 7.5 ng/ml insulin, and 2 ng/ml fibroblast growth factor (FGF) were from ScienCell (Carlsbad, CA). Insulin/selenium/transferrin supplement and FGF were from Sigma-Aldrich.

Human VSMCs were isolated from pieces of aorta using the explant technique (15). Tissues were washed with PBS, and all adipose was removed. The artery was then cut into small (∼5 mm × 5 mm) pieces and placed lumen side down into a 75-cm tissue culture flask (three to four pieces per flask). SMC media (SMC media with smooth muscle growth supplement and 10% FBS) was then carefully added to the flask so as not to dislodge the pieces of aorta. The tissue was maintained at 37°C, and SMCs were harvested in 3–4 wk. VSMCs were then passaged every 4–7 d and used after one passage. VSMC characteristics varied markedly with culture conditions (Fig. 2). VSMCs cultured in SMC media were highly proliferative, rhomboid in shape, SMC actin low, and smoothelin negative. VSMCs cultured in DMEM supplemented with 5% FBS, 0.5 μg/l basic FGF, 2 μg/l epidermal growth factor, 50 U/ml penicillin, and 100 μg/ml streptomycin proliferated slowly, were elongated, moderately actin positive, and occasional cells were smoothelin low. VSMCs cultured under serum-free conditions in 50:50 DMEM/Ham’s F12 with insulin/selenium/transferrin were quiescent, very thin and elongated, strongly actin positive, and smoothelin positive. The latter conditions have been demonstrated to yield quiescent, contractile VSMCs (16). Of note, primary dermal fibroblasts were smoothelin negative.

Human arteries from deceased organ donors were prepared as for smooth muscle culture. Single pieces of aorta of the same approximate size (3 × 3-mm square) were placed lumen side down, one per well, in a 24-well plate. Explants were cultured in DMEM with 1% FBS with/without IL-2 25 ng/ml. Fresh media was applied weekly, and explants were cultured for 4–6 wk based on degree of outgrowth. Explants serving as positive controls were cultured in SMC media, defined previously. At termination of the experiment, the explant tissue was removed and number of cells quantified using alamarBlue, as directed by the manufacturer (Thermo Fisher Scientific).

VSMC RNA was purified from aortic tissue using TRIzol (Life Technologies). Contaminating DNA was removed using TURBO DNAse-free (Thermo Fisher Scientific), and 0.5 μg/ml of purified RNA was then used for cDNA synthesis with random hexamers and oligo(dT). Human IL-2R component sequences were used to design intron-spanning primers with the following sequences: IL-2Rα forward: 5′-CCCACACGCCACATTCAAAG-3′, reverse: 5′-TTGTGTTCCGAGTGGCAGAG-3′; IL-2Rβ forward: 5′-ACCCACAGATGCAACATAAGCTGG-3′, reverse: 5′-TTGACCCGCACCTGAAACTCATAC-3′; and IL-2Rγ forward: 5′-GCTCCAGAGAACCTAACACTTCAC-3′, reverse: 5′-TTAAAGCGGCTCCGAACACGAAAC-3′. cDNA samples from VSMCs were combined with our primers for amplification of targeted regions. Products were separated by electrophoresis in a 0.5% agarose gel containing 0.5 μg/ml ethidium bromide and visualized using the ChemiDoc MP System (Bio-Rad Laboratories).

Amplified IL-2Rα, -β, or -γ products were purified using QIAquick PCR Purification Kit (Qiagen, Valencia, CA) and submitted for sequencing (Eurofins Genomics, Louisville, KY). The resulting sequences were compared with currently known human transcripts using the standard nucleotide BLAST from National Center for Biotechnology Information.

VSMCs were grown to ∼80% confluence under various conditions as indicated in the figure or legend. VSMCs were washed in PBS and fixed in ice-cold methanol/2% acetic acid. Plates were then blocked for 2 h at room temperature with Odyssey Blocking solution (LI-COR Biosciences, Lincoln, NE). VSMCs were again washed, and primary Abs, diluted in TBS with 0.5% Tween, were applied overnight. Unbound Abs were removed by washing, and secondary Abs with an HRP conjugate were applied for 1 h at room temperature. The HRP conjugate was then detected with Alexa Fluor 488 or 555 tyramide signal amplification per manufacturer’s instructions (Thermo Fisher Scientific). Fluorescence was quantified using a Synergy H1 Multi-Mode Microplate Reader (BioTek, Winooski, VT). Readings were normalized to cell number obtained by nuclear staining using CyQUANT (Thermo Fisher Scientific).

VSMCs were grown to 60–80% confluence and stained using the in-cell Western protocol describe above. To stain unpermeabilized cells, the latter protocol was modified by fixing VSMCs in 4% paraformaldehyde, followed by heat-induced Ag retrieval in a citrate buffer (7.14 mM citric acid monohydrate and 34 mM sodium citrate dehydrate) (pH 6). Both primary and secondary Abs were diluted in TBS without Tween. Cells were imaged using an EVOS FL epifluorescence microscope (Thermo Fisher Scientific).

VSMCs at ∼30% confluence were cultured in serum-free DMEM for 48 h then treated with increasing concentrations of IL-2 for 72 h. For IL-2R neutralization experiments, anti–IL-2Rα and -β Abs were added 30 min prior to the addition of IL-2, each at 5 μg/ml based on product guidelines.

To measure proliferation, 5-ethynyl-2´-deoxyuridine (EdU) was added for the last 24 h of the treatment at a final concentration of 20 μM. VSMCs were then fixed in 4% paraformaldehyde and permeabilized with 0.5% Triton-X. Incorporated EdU was detected via Click chemistry using a commercially available kit (Click-iT EdU, Thermo Fisher Scientific) and quantified using a Synergy H1 Multi-Mode Microplate Reader. Select experiments used a modification of this kit, wherein the fluorescent component that binds EdU, picolyl azide Alexa Fluor 488, was exchanged for picolyl azide biotin, followed by streptavidin/HRP and then developed using SuperSignal Femto Chemiluminescent Substrate (Thermo Fisher Scientific). Luminescence was read on a Synergy plate reader.

Boyden Chambers (Greiner Bio-One, Monroe, NC) were placed in 24-well plates. VSMCs were plated at 5 × 10⁴ cells/well onto the upper Boyden Chambers, and the bottom chamber received DMEM. After 48 h, media in the upper chamber was replaced with fresh DMEM, whereas media in the lower chamber was replaced with DMEM containing various treatments (see figures). After 8 h, chambers were fixed in methanol with 2% acetic acid at −20°C. The upper chamber containing cells was scraped with a cotton swab to remove nonmigrating cells. Upper and lower chambers were washed with PBS, and nuclei were stained with SYBR Green. Cells were imaged using an EVOS FL epifluorescence microscope (Thermo Fisher Scientific) and then counted using Image J software from the National Institutes of Health.

VSMCs were cultured in serum-free DMEM for 48 h, then treated with increasing concentrations of IL-2 for 60 min. Cells were lysed using radioimmunoprecipitation assay buffer containing protease and phosphatase inhibitors, and protein concentrations were determined by bicinchoninic acid assay. Equal amounts of proteins from each sample were fractionated by SDS-PAGE on 4–12% (w/v) polyacrylamide-resolving gels. Proteins were transferred onto a polyvinylidene difluoride membrane (Bio-Rad Laboratories, Hercules, CA), probed with the indicated primary Ab, followed by corresponding secondary Ab conjugated with HRP. Abs [phospho-akt (ser 473), pan-akt, phospho-MAPK 44/42 (Thr202/204), and pan-MAPK 42/44] were from Cell Signaling Technology (Danvers, MA). The membrane was developed using an ECL substrate and visualized using the ChemiDoc MP Imaging System (Bio-Rad Laboratories).

Whole-cell extracts from VSMCs and Jurkat cells were fractionated by SDS-PAGE under denaturing and reducing conditions, using equal amounts of protein, and transferred to a polyvinylidene difluoride membrane. After incubation with 5% nonfat milk in TBST (10 mM Tris, [pH 8], 150 mM NaCl, and 0.5% Tween 20) for 60 min, the membrane was incubated with Abs against IL-2Rα (1:1000), IL-2Rβ (1:500), and IL-2Rγ (1:500) at 4°C for 12 h. Membranes were washed three times for 10 min and incubated with a 1:10,000 dilution of HRP-conjugated anti-rabbit Abs at room temperature for 2 h. Blots were washed with TBST three times and developed with the ECL system (Luminata Crescendo; MilliporeSigma) according to the manufacturer’s instructions.

Intimal hyperplasia was induced in rabbit jugular veins using an interposition model, as previously described (17, 18). Briefly, New Zealand white rabbits (3–3.5 kg) underwent carotid artery interposition grafting with the external jugular vein. Graft samples were collected both at the time of vein graft creation and 2 h and 3, 7, 14, and 28 d following implantation (n = 5 per time point). One hour prior to harvest, rabbits received an i.m. injection of BrdU (50 mg/kg). Upon harvesting, grafts were divided into two specimens, with one segment immediately stripped of the adventitia and endothelium and placed in RNAlater for genomic analysis and the second segment fixed in 10% formalin for histologic examination.

After storage at −80°C, RNAlater-preserved samples were homogenized in Buffer RLT (Qiagen, Valencia, CA) using a Miltenyi gentleMACS Dissociator, followed by QIAshredder for lysis (Qiagen). RNA isolation was performed using RNeasy Mini Kit (Qiagen), and quality was confirmed using an Agilent Bioanalyzer. cDNA was generated using Ovation Pico WTA kit (NuGEN, San Carlos, CA) and labeled using GeneChip WT Terminal Labeling (Affymetrix, Santa Clara, CA). Samples were hybridized to a proprietary rabbit array (Affymetrix); results were analyzed for changes in IL-2, SM-MHC, and smooth muscle cell α-actin expression. To identify actively proliferating cells, two histologic cross-sections (5-μm thickness) of formalin-fixed tissue were processed using an Invitrogen BrdU Staining Kit according to manufacturer’s recommendations. A series of histomicrographs (40× magnifications, 5–10 images per section) were analyzed using computer-aided morphometry to assess the density of proliferating cells within the graft intima.

Data are expressed as ± SD or SEM, as indicated in the figure legend. Statistical comparisons between two groups were performed with the unpaired Student t test with Welch correction using GraphPad Prism software (GraphPad, La Jolla, CA). Analysis of the time-dependent changes in mRNA IL-2 expression in vein grafts was performed via a one-way ANOVA, with Holm–Sidak post hoc testing to identify individual differences among groups (SigmaPlot, San Jose, CA).

To determine whether VSMCs have the potential to express the trimeric IL-2R, we first established whether human VSMCs express mRNA for each subunit of the IL-2R. To this end, mRNA was isolated from VSMCs and amplified by RT-PCR using intron-spanning primers specific to each component of the IL-2R. Separating the amplified cDNA on an agarose gel resulted in singular bands of expected sizes (Fig. 1). These PCR products were purified and sequenced, and the resulting sequences matched published sequences of IL-2R components within 98–100% identity (α, https://www.ncbi.nlm.nih.gov/nuccore/NM_000417; β, https://www.ncbi.nlm.nih.gov/nuccore/NM_000878.2; γ, https://www.ncbi.nlm.nih.gov/nuccore/NM_000206.2). Using the same primers, mRNA from Jurkat T cells amplified the same size products. These results indicate that human VSMCs express mRNA for IL-2Rα, -β, and -γ, consistent with our prior findings demonstrating expression of IL-2Rβ mRNA in murine VSMCs (5).

In light of our findings that human VSMCs express mRNA for all three subunits of the IL-2R, we next asked whether human VSMCs express the corresponding proteins. To assess IL-2R protein expression in VSMCs, we elected to use the in-cell Western developed by LI-COR Biosciences (Lincoln, NE). This assay is designed to detect intracellular and cell surface proteins on adherent cells, without the need to scrape or trypsinize, as required for flow cytometry (19). To enhance our signal with minimal increase in background, we incorporated tyramide signal amplification.

VSMCs are known to exhibit a wide range of phenotypes, ranging from quiescent/highly contractile, to proliferative and expressing low levels of contractile proteins (20). The latter phenotype is promoted by exposure to mitogens such as serum, in vitro, and by stimuli such as mechanical injury, oxidative stress, or high flow, in vivo. We assessed IL-2R expression in VSMCs with three distinct phenotypes we termed quiescent, intermediate, and proliferative (Fig. 2). These phenotypes were generated by different culture conditions, detailed in the 2Materials and Methods section. Quiescent VSMCs were large, elongated, nonproliferating cells expressing smooth muscle cell markers SMC actin, SM-MHC, and smoothelin. Although SMC actin can be found in fibroblasts, SM-MHC and smoothelin are specific to SMCs and highly differentiated SMCs, respectively (21, 22). Intermediate VSMCs were large, elongated, slowly proliferating cells that were moderately SMC actin positive, SM-MHC positive, and smoothelin low to negative. Proliferative VSMCs were small, rhomboid, dividing cells that were SMC actin low, SM-MHC positive, and smoothelin negative. The characteristics of these VSMCs are consistent with those in the literature describing quiescent versus proliferating VSMCs (20).

As seen in Fig. 3, cultured VSMCs expressed all three subunits of the IL-2R. Proliferative SMC, however, expressed lower levels of IL-2Rα compared with intermediate and quiescent phenotypes.

As a second means to identify IL-2R in VSMCs, we prepared cell lysates of VSMCs and assessed the presence of each receptor by Western blot analysis. VSMCs expressed each subunit of the IL-2R at the appropriate m.w. (23, 24). (https://www.bosterbio.com/anti-cd25-il-2sr-alpha-antibody-pa1747.html) (Fig. 3B). The Jurkat T cell line was used as a positive control. To rule out the presence of contaminating T cells as a source of IL-2R, we assessed our VSMC cultures for the presence of CD3. CD3 was not detectable in our VSMC cultures (Supplemental Fig. 1).

In addition to the above experiments, we examined stained VSMCs by fluorescent microscopy. Because the cells are permeabilized, the staining observed reflects membrane and cytoplasmic staining. Both would be expected given intracellular trafficking of IL-2R subunits (25). Localization of each subunit appeared to be grossly similar; however, expression of IL-2Rα was decreased in the proliferative phenotype, similar to the in-cell Western findings (Fig. 4A). To better understand the availability of the tripartite receptor for binding IL-2, we reassessed localization of IL-2Rα on nonpermeabilized proliferative and quiescent VSMCs. Interestingly, expression of IL-2Rα was localized to the cell body in quiescent cells, but extended to and, in some cases, concentrated in the filopodia of proliferating cells (Fig. 4B). These images suggest that the decrease in IL-2Rα expression may reflect, at least in part, a redistribution within the cell. Our results suggest that VSMCs express all three subunits of the IL-2R and that α expression and localization varies with phenotype.

Having determined that VSMCs express IL-2Rs, we asked whether VSMCs could take up IL-2 via the IL-2R. To this end, we treated VSMCs with IL-2 in the presence or absence of neutralizing IL-2Rα and -β Abs. IL-2 was detectable in the cytoplasm of VSMC following incubation with IL-2, and detection was decreased in the presence of anti–IL-2R Abs (Fig. 4C). In total, these findings suggest that VSMCs express an IL-2R and take up IL-2 in an IL-2R–dependent fashion.

Having demonstrated that VSMCs express all three subunits of the IL-2R and take up IL-2, we next explored the functional consequences of this expression and uptake. Because intimal hyperplasia is one of the main pathologies in vascular disease, we targeted SMC responses (proliferation and migration) that underlie this process. We first asked whether IL-2 impacts the outgrowth of VSMCs from aortic explants.

To this end, we used an explant model in which VSMCs were allowed to migrate from small pieces of cultured aorta. Tissues were cultured without growth factors and minimal sera (1% FBS/DMEM) with or without added IL-2. Tissues cultured in SMC growth factors (ScienCell) and 10% FBS were used as a positive control. Explants were cultured for 25–30 d with weekly media changes. As seen in Fig. 5A, explants treated with IL-2 exhibited significantly increased outgrowth of VSMCs when compared with media alone.

These results suggest that IL-2 promotes migration and/or proliferation of VSMCs. Because we cannot separate effects on migration versus proliferation in this assay, we went on to test these responses individually.

We first asked whether IL-2 affects migration of VSMCs. For these experiments, we used a traditional Boyden Chamber assay in which a cylindrical cell insert is placed inside the well of a tissue culture plate (26). The insert contains a polycarbonate filter on the bottom. Cells are plated on the filter and migrate toward potential chemoattractants in the tissue culture well. In our case, VSMCs were plated on the filter, and media containing IL-2 was added to the tissue culture wells. Platelet-derived growth factor (PDGF), a potent chemoattractant for VSMCs, was used as a positive control. As seen in Fig. 5B, IL-2 promoted the migration of VSMCs.

We next tested the impact of IL-2 on VSMC proliferation. Because VSMCs grow slowly, with doubling time of 48–60 h, we chose to increase the sensitivity of our assay by measuring DNA synthesis using incorporation of EdU (27). To this end, VSMCs were cultured in serum-free media (see 2Materials and Methods) for 48 h. Increasing concentrations of IL-2 were then added, and cells were cultured for 72 h with EdU added during the last 24 h. Incorporation of EdU was then detected by binding of an azide dye to EdU via Click chemistry. As seen in Fig. 5C, IL-2 promoted proliferation of VSMCs at lower concentrations, but not at higher concentrations. A similar response to IL-2 has been reported for colonic epithelial cells (28).

Compared with IL-2 concentrations used with lymphocytes, the concentration of IL-2 needed to promote proliferation was relatively high. Lower concentrations of IL-2 produced a trend toward a dose response but were not statistically significant (Supplemental Fig. 2). To ensure that proliferation was an IL-2–mediated process, we attempted to block proliferation with neutralizing IL-2Rα and -β Abs. Using both Abs effectively inhibited IL-2–induced proliferation (Fig. 5D). In total, these results suggest that the increased outgrowth of VSMCs from aortic tissue in the presence of IL-2 is a result of both enhanced migration and proliferation.

Because of the effects of IL-2 on VSMC migration and proliferation, we questioned what signals might be initiated by IL-2 to mediate these effects. In addition to the influence of IL-2 on VSMCs demonstrated in this study, we previously reported that IL-2–deficient mice lose SMC as they age (5), eventually causing widened esophagi and aortic aneurysms. The remaining cells are small compared with wild-type mouse cells. The latter results imply that IL-2 impacts cell survival and size.

Both the in vitro results reported in this study and the “SMC phenotype” of IL-2–deficient mice suggest that IL-2 affects proliferation, migration, cell size, and survival of VSMCs. The effect of IL-2 on survival and cell size implies that IL-2 might be signaling through the PI3K/akt/mTOR pathway (29). The influence of IL-2 on proliferation and migration indicates that IL-2 might be signaling through MAPK (30). The MAPK erk has been shown to have a significant influence on migration and proliferation of VSMCs in vitro and on intimal hyperplasia in vivo (30, 31). We, therefore, asked whether exposure of VSMCs to IL-2 activates akt and/or erk. Exposure of VSMCs to increasing concentrations of IL-2 for 1 h resulted in phosphorylation of erk and akt at lower concentrations of IL-2, with diminishing or no activation at higher concentrations, similar to the impact of IL-2 on proliferation (Fig. 6 compared with Fig. 5C).

Given the importance of the jak/stat pathway in IL-2 signaling, we explored whether the jak/stat pathway in VSMCs is activated through the IL-2R. To identify the best component to start with, we assessed levels of total Jak1/3 and Stat 5a/b by Western blot in VSMC lysates. None of these proteins yielded a strong enough signal to assess potential changes in phosphorylation. In looking at other receptors expressed on VSMCs, stimulation of the erythropoietin receptor, for example, initiates robust signals through the jak/stat pathway in endothelial cells, but weak or absent signals in smooth muscle cells (32, 33). Whether our findings reflect true low levels of expression, or a need to change Abs and/or culture conditions, requires further study.

These results demonstrate that the IL-2R on VSMCs is functional via initiating signaling pathways involving erk and akt. Additional blocking studies using small interfering RNA and biochemical inhibitors will be needed to determine the impact of these pathways on IL-2–mediated changes in proliferation, migration, cell size, and survival. These studies will be particularly important for akt given its low level of phosphorylation relative to PDGF.

In VSMCs, and many other cell types, survival is mediated by a balance between akt and the forkhead (FOXO) family of transcription factors. Akt is important to survival when cells are activated, but when cells are quiescent and akt is silent, FOXO transcription factors are critical for long-term survival of cells (34, 35). The gradual decline of VSMCs in IL-2–deficient mice suggests an impact on quiescent rather than activated cells. We, therefore, asked whether IL-2 impacts expression of FoxO3a, a forkhead transcription factor expressed by VSMCs. As seen in Fig. 7, incubation of VSMCs with IL-2 for 24 h increased expression of FoxO3a, suggesting a possible means by which IL-2 impacts the survival of VSMCs in vivo.

The preceding results demonstrate that IL-2 promotes migration and proliferation of VSMCs, which are critical functions in the development of intimal hyperplasia. Given these findings, we began to ask whether IL-2 contributes to the development of intimal hyperplasia. To this end, we induced intimal hyperplasia in rabbit vein grafts using an interposition model (17, 18) and asked whether IL-2 expression changes over time and how this correlates to initiation of VSMC proliferation.

Following implantation of the vein into the arterial circulation, there was a burst of IL-2 mRNA expression within 2 h of graft creation (Fig. 8A). This increase began to return toward baseline levels as early as 3 d postimplantation, but remained elevated at ∼1.5-fold baseline levels through the 28-d experimental period. Coincident with this increase in IL-2 expression, and consistent with our in vitro observations, was a significant reduction in SM-MHC and SMC α-actin expression, confirming a conversion from a contractile to proliferative SMC phenotype (Fig. 8B, 8C). By day 3, these changes were associated with an increase in SMC proliferation within the developing intima, which was maintained through 2 wk postimplantation (Fig. 8D). These results show that, during the development of intimal hyperplasia, VSMC proliferation is preceded by a burst of IL-2 expression. Follow-up studies, inhibiting IL-2 through Abs or transgenic murine models, will be needed to determine the extent to which IL-2 contributes to SMC proliferation during the development of intimal hyperplasia.

IL-2 is well known for maintaining homeostasis in the immune system. The role of IL-2 in maintaining a delicate balance between too few and too many lymphocytes was demonstrated in IL-2–deficient mice, which develop a lymphoproliferative disorder. Although these mice demonstrated that other cytokines could substitute for IL-2 in promoting proliferation, their lymphoproliferative phenotype led to the discovery of a nonredundant role for IL-2 in downregulating immune responses via Treg cells and reactivation-induced cell death (36, 37).

In sharp contrast to the latter, well-described lymphoproliferative phenotype in IL-2–deficient mice, our laboratory previously reported that IL-2–deficient mice lose smooth muscle cells with increasing age. This loss is systemic, resulting in aortic aneurysms, ectatic esophagi, and decreased striae in the spleen (5) (L.E. Wrenshall and J.D. Miller, unpublished observations). This finding is the first, to our knowledge, implicating a role for a T cell cytokine in the maintenance of smooth muscle cells. These findings suggested that IL-2 impacts the survival and/or proliferation of SMCs, which prompted us to ask whether IL-2 has a direct impact on VSMCs. Our findings, reported in this review, demonstrate that IL-2 promotes the proliferation and migration of VSMCs and upregulates a prosurvival transcription factor, FoxO3a.

The lymphoproliferative phenotype of IL-2–deficient mice occurs, in large part, because of a decrease in peripheral Treg cells. IL-2 is critical to the survival of these cells (38). One significant difference between Treg and T effector cells is that Treg cells constitutively express high levels of IL-2Rα, whereas CD25 expression on T effector cells is transient and relatively low (36). Similar to Treg cells, our studies indicate that VSMCs constitutively express IL-2Rα, and that quiescent VSMCs express higher levels of CD25 than activated/proliferating VSMCs.

Although IL-2 promotes the proliferation and survival of both Treg and T effector cells, the predominant impact of IL-2 on Treg cells is survival, whereas in T effector cells, the predominant impact is proliferation. One of the main differences in IL-2R signaling dictating these outcomes is the level of akt activation. Activation of akt through the IL-2R is attenuated in Treg cells, but strong in T effector cells (36). Because one of the main targets of akt is FoxO3a, strong activation of akt leads to phosphorylation and inhibition of FoxO3a, whereas weak activation of akt allows for some activation of FoxO3a and, in Treg cells, expression of Foxp3 (39, 40). Comparing IL-2–mediated akt and FoxO3a signals in Treg and T effector cells, therefore, suggests that akt and FoxO3a are both activated at lower levels in Treg cells. In T effector cells, akt is strongly activated and FoxO3a inhibited.

In VSMCs, the balance between akt and FoxO3a are important factors in the life and death of the cell (34). FoxO3a signals are thought to promote longevity and survival through cell cycle arrest and activation of reactive oxygen species detoxification genes (41). Under conditions of stimulation, akt signaling predominates and cells proliferate (34, 42). Under conditions of quiescence, akt will not be activated and FoxO signaling predominates (35). Our data indicate that, compared with PDGF, IL-2 initiates a weak activation of akt at lower concentrations of IL-2, which becomes absent at higher concentrations (Fig. 6). IL-2 increases expression of FoxO3a at both lower and higher concentrations. These findings suggest that, similar to Treg cells, the net impact of IL-2–mediated changes in phospho-akt and FoxO3a in VSMCs is to promote survival.

Our findings that IL-2 increases FoxO3a is particularly interesting because little is known about the upregulation of FoxO3a expression. FoxO3a is upregulated by itself and by the transcription factor E2F-1 (42, 43). Growth factors known to target VSMCs, specifically insulin-like growth factor, FGF, and PDGF, downregulate expression of FoxO3a (42). VSMC growth factors, therefore, push the akt/FoxO balance toward akt and proliferation, whereas our initial findings suggest that IL-2 pushes the akt/FoxO balance toward FoxO and survival. This prosurvival role for IL-2 may explain the loss of SMCs in IL-2 knockout mice.

In a normal blood vessel, smooth muscle cells are quiescent, nonproliferating, stationary cells with a contractile phenotype. When activated by stretch, abnormal flow, or injury, VSMCs decrease their expression of contractile proteins, change shape from elliptical to rhomboid, proliferate, and migrate. Proliferation and migration are the two key behaviors of VSMCs that lead to intimal hyperplasia, which can cause the occlusion of venous bypass grafts and is a precursor lesion to atherosclerosis. Our laboratory has accrued several findings that implicate IL-2 in the pathogenesis of intimal hyperplasia. These findings include the following: 1) IL-2 surrounds VSMCs in vivo through binding to perlecan, 2) expression of IL-2 is increased following transition to a high-flow arterial circulation, (3) IL-2 causes proliferation and migration of VSMCs, and 4) IL-2 signals via erk and akt in VSMC, both of which contribute to intimal hyperplasia. These findings, combined with the loss of SMC in IL-2–deficient mice, suggest that IL-2 contributes to the survival of VSMC under normal, quiescent conditions, but promotes intimal hyperplasia under abnormal conditions, such as high flow, that increase IL-2 expression. Whether the biphasic response to IL-2 elicited in vitro helps control this proliferative response or becomes dysregulated under proinflammatory conditions, remains to be determined. Increased expression of IL-2Rα in filopodia of proliferating cells may be another means by which the reactivity of VSMC to IL-2 is enhanced during inflammation. Although publications examining an effect of IL-2 on intimal hyperplasia are limited, we identified one study demonstrating that an IL-2 fusion toxin inhibited intimal hyperplasia in a rabbit model, which supports our findings (44).

In light of our studies, it fairly easy to hypothesize how T cells, the primary producers of IL-2, impact VSMCs in vivo. Inflammatory conditions promote IL-2 production in T cells, and this IL-2 then binds to perlecan in blood vessels. Our previous studies support this scenario, as IL-2 from both local and systemic sources was identified in murine arteries (5). IL-2 may then be released in large amounts by heparanase and other matrix-degrading enzymes that are upregulated during inflammation. Under noninflammatory conditions, smaller amounts of IL-2 are likely continuously released as part of natural turnover, and may, under these conditions, promote survival. What is less obvious is whether IL-2R expression on VSMCs impacts T cells. Like dendritic cells, it is possible that VSMCs “share” CD25 with primed T cells, which would promote further T cell expansion under inflammatory conditions (45).

Similar to balancing the number of lymphocytes in the immune system, regulation of proliferation, death, and survival is critical to maintaining optimal numbers of VSMCs in blood vessels. Our studies suggest that, analogous to its role in T cells, IL-2 contributes to the balance of VSMCs in the vasculature. These findings imply a new connection between the immune and vascular systems, with implications not only for atherosclerosis, but for vasculitis and autoimmune diseases as well. Further studies, both in vitro and in vivo, will be needed to define the mechanism(s) and full impact of this connection.

We thank Victoria Wong, Chris Davies, and Bharath Ramini for technical assistance. We also thank Dr. Thomas Brown for help with primer design.

The authors have no financial conflicts of interest.