Rapamycin delays growth of Wnt-1 tumors in spite of suppression of host immunity

Wnt-1 tumor cells (1–2 × 10⁵) were obtained as described [27], and inoculated subcutaneously on the right flank or into the left inguinal mouse fat pad (MFP #4). The injection of cells in 50 μl of PBS was performed through the skin of anesthetized mice. Experiments were conducted either in intact non-irradiated naïve syngeneic recipients or lethally irradiated (1050 cGy) mice using a¹³⁷Cs gamma radiation source (gamma Cell 40; Atomic Energy of Canada). Irradiated mice were reconstituted with bone marrow (5 × 10⁶ cells/mouse) from syngeneic B6 mice administered intravenously in 200 μl of PBS. Wnt-1 cells were implanted on the same day following irradiation and bone marrow reconstitution (5 to 10 mice per group). A stock solution of Rapamycin (LP Laboratory, USA) was made in ethanol at 1 mg/ml. Mice were given daily intraperitoneal injections of 30 μg of Rapamycin in 200 μl of 0.2% carboxymethyl-cellulose (Sigma) used as a diluent. Rapamycin therapy was initiated on day 1 after tumor implantation and continued for indicated times. Control animals received injections with vehicle alone. Tumor size was measured with vernier calipers twice a week and calculated using the formula (W² × L)/2, where W and L corresponded to width and length of tumors.

Spleen and thymus cells were isolated by using stainless steel 40 micron wire mesh. Bone marrow (BM) was flushed from one femur and one tibia and made into single cell suspensions by passing through 25 gauge needle. Red cells were lysed by ACK buffer (Quality Biologicals, Gaithersburg, MD). Cells were washed twice in phosphate buffered saline (PBS) and transferred to complete medium (CM) consisting of RPMI 1640 (Mediatech, Herndon, VA) supplemented with 10% FCS (Gemini Bio-Products, West Sacramento, CA), pen-strep-glut, non-essential amino acids, and 2-ME 5 × 10^-5 M (all from Invitrogen Life Technologies, Carlsbad, CA).

To generate T cells that are resistant to Rapamycin, B-cells were depleted from splenocytes using goat anti-mouse magnetic particles (Polysciences, Warrington, PA). CD4 and CD8 cells were purified by CD4 enrichment kit (StemCell Technologies, USA) and cultivated separately to generate either Th1 or Tc1 cells as previously described [26]. We have included Tc1 cells which are more likely to mediate cytotoxic anti tumor responses and have persistent in vivo survival (ibid). Briefly, to obtain Rapamycin resistant T1 (1:1 Th1+Tc1) cells purified CD4⁺ or CD8⁺ T-cells were stimulated with CD3/CD28 beads in the presence of N-acetyl-cysteine (3.3 mM; Bristol-Myers Squibb, New York, NY), selective cytokines and 1 μM Rapamycin. Anti-CD3 and anti-CD28-coated beads (CD3/CD28 beads) were produced according to previously developed protocol [26] and used routinely in our laboratory at 3:1 (bead:cell) ratio. Conditioned medium was supplemented with recombinant murine IL-12 (2.5 ng/ml; R&D Systems, Minneapolis, MN), recombinant human (rh)IL-2 (20 IU/ml; National Cancer Institute (NCI)-Biologic Resource Branch (BRB) Repository), rhIL-7 (20 ng/ml; PeproTech, Rocky Hill, NJ), and anti-murine IL-4 (clone 11B.11 (10 μg/ml); NCI-BRB). Cytokine- and Rapamycin-containing medium was added on days 0, 2, and 6 to maintain 0.2–1.0 × 10⁶ cells/ml. Addition of rmIL-12 was performed only at day 0 of T1 culture. Before injection into mice, T1 cells were analyzed by flow cytometry for purity of preparation. Seven millions of Rapamycin resistant T1 cells (T1Rapa) were injected in 200 μl of PBS intravenously into orbital sinus of mice at indicated times.

Tumor cell suspension was prepared as described for other organs. Briefly, tumors were excised at 1 gm of wet weight, cut into small pieces and tumor brei was prepared by pressing through 40 micron wire mesh. Single cell suspension was obtained by passing through 20–25 gauge needles. Red cells were lysed as above. Cells were washed twice in PBS and transferred into tissue culture plates (Nunc, Rochester, NY) in CM for in vitro studies. Primary cultures of tumor cells were depleted from contaminating lymphocytes by seeding cells on 100 mm culture plates in 10 ml of CM for 24–72 hours. When adherent confluent monolayer of tumor cells was formed, the plates were washed vigorously with PBS, trypsinized, and Wnt-1 cells were used for the analysis.

Wnt-1 cells obtained from tumors as above were seeded at low density onto 90 mm tissue culture dishes. Single colonies were picked using cloning cylinders, and transferred in CM to 96-well plate. Cells were incubated until confluence and transferred to 24-well plates. Two cell lines, W1204 and W1308, with slightly different cellular morphology were used for these studies.

Primary cultured Wnt-1 cells were seeded at 10⁴ cells/well in triplicates in 96-well plates. Serial dilutions of Rapamycin at 10–0.01 mM in CM were added to the cells 24 hours later. Cells were incubated for additional 96 hours and 1 μCi 3H-thymidin (Amersham) was added for the last 4 hours. Afterwards cells were harvested (Titertek, UK), transferred onto glass filters, and dried. 3H-thymidin incorporation was estimated by beta counter (BD, USA) using scintillation liquid. As a control, intact splenocytes (5 × 10⁴ cells/well), studied in the same way, were incubated in the presence of CD3/CD28 beads (1:1 ratio).

For the fluorescence-activated cell sorter (FACS) analysis cells were transferred to FACS buffer (PBS, 1% bovine serum albumin, 0.05% NaN₃). Three-color flow cytometry was performed using FACSCalibur instrument, CellQuest software (BD Biosciences) and the following antibodies: Ep-CAM-FITC, mouse anti-vimentin, rat-antimouse-FITC, anti-mouse CD3-FITC, CD4-PE, CD8-FITC, CD25-PE, CD19-PE, NK1.1-PE, CD11b-FITC, and Fas-PE (all from BD Pharmingen). Live events (5,000–10,000) were acquired with propidium iodide exclusion of dead cells.

Percentage of apoptotic cell was analyzed by AnnexinV and 3,3'-dihexyloxacarbocyanine iodide (DiOC(6)) double staining. The cationic lypophilic fluorochrome DiOC(6) (Invitrogen) was used to evaluate transmembrane potential in mitochondria [30]. Splenocytes or Wnt-1 cells were harvested, washed in pre-warmed PBS supplemented with 2% of FCS, resuspended at 10⁶ cells/ml in PBS/2%FCS and 40 μM DiOC(6) and incubated for 30 minutes at 37°C. Cells were washed with PBS/2% FCS and transferred to Annexin V binding buffer (Hepes buffer, 10 mM, pH 7.4, 150 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1.8 mM CaCl₂), stained with AnnexinV-APC and propidium iodide (PI), incubated in the dark for 15 min, and analyzed by flow cytometry.

Splenocytes were harvested from control or Rapamycin treated animals at indicated times and prepared as single cell suspension as above. Cells (10⁶ cells/ml) were plated in CM onto 24-well plates with or without CD3/CD28 beads. Supernatants were collected at 24 hours and cytokines were measured by Bio-Plex multiplex sandwich immunoassay (Bio-Rad) using Beadlyte Mouse Multi-Cytokine Beadmaster kit (Upstate, Lake Placid, NY).

Cell cycle was analyzed using DAPI-stained DNA. Two million cells were harvested at indicated time, washed in ice-cold PBS, fixed by the addition of 70% ethanol and left for 2 hours at 4°C. Thereafter, the cells were washed twice in PBS, stained with 5 μg/ml of DAPI (Sigma Chemical Co) in PBS and analyzed by FACS.

Primary cultures of Wnt-1 cells were grown in 24-well plates (Nunc, Rochester, NY) for 48–72 hours, then washed in FACS buffer and stained with anti-mouse ep-CAM-FITC antibodies. Wnt-1 cells were analyzed by laser scanning cytometry (CompuCyte Corp., Boston, MA). The fluorescence excitation was provided by a 488 nm argon laser beam. The green fluorescence from FITC was measured using a 530/30-nm band-pass filter and amplified using a photomultiplier.

After treatment with Rapamycin for indicated times, Wnt-1 primary cultured cells were washed twice with PBS and lysed in ice-cold lysis buffer (Cell Signaling Technology, Danvers, MA). Lysates were centrifuged at 12,000 × g for 10 min at 4°C, and protein concentration of the cleared cell lysates was measured using the Bio-Rad Protein Assay kit (Bio-Rad Laboratories, Hercules, CA). Thirty micrograms of protein were denatured in SDS-sample buffer, electrophoresed using 10% SDS-PAGE gels, transferred to nitrocellulose membranes, and blocked for 1 h at room temperature in TBS-T (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1% Tween-20) containing 5% non-fat milk. Membranes were then incubated overnight at 4°C with the indicated primary antibodies diluted 1:1000 in blocking solution. Antibodies against pp70S6K, S6K, pS6, p-Akt, and Akt were from Translational Control Sampler Kit (Cell Signaling, Beverly, MA). The appropriate secondary antibodies conjugated to horseradish peroxidase (Santa Cruz Biotechnology, CA) were used to visualize the bands (1 h incubation) with an enhanced chemiluminescence (ECL) visualization kit (Cell Signaling, Beverly, MA).

The effect of Rapamycin on growth of Wnt-1 tumors was examined in syngeneic C57BL/6 mice implanted with Wnt-1 tumor cells subcutaneously (s.c.) or into mouse fat pad #4 (MFP). For these experiments, as few as 1–2 × 10⁵cells are sufficient to generate synchronous tumors within 30 days. We used non-irradiated naïve mice or lethally irradiated and bone marrow-reconstituted animals. Rapamycin treatment for 20 days resulted in a significant delay in tumor growth evident by day 40 in naïve (non XRT) and irradiated (XRT) hosts (Fig. 1A–C). The differences in tumor growth rates between control and Rapamycin treated mice were statistically significant as determined by paired t-test. Similar results were obtained using subcutaneous (s.c.) implantation of tumor cells (10⁵ cells/mouse) and 30 days of treatment with Rapamycin (n = 8/group, Fig. 1A and 1D).

Bild vergrößern

When the effect of Rapamycin on tumor growth in non-irradiated and radiated animals was compared, it became evident that tumors grew faster in irradiated hosts. Figure 1D summarizes the results obtained on day 60 for tumors implanted s.c. and at day 50 for MFP tumors. Overall, Wnt-1 tumors grew faster in MFP than when implanted s.c. (Fig. 1D). Because growth of Wnt-1 tumors was also accelerated in irradiated mice, we hypothesized that the effect of Rapamycin could be related to its immunosuppressive action. To dissociate antitumor and immunosuppressive activities, we determined the effect of Rapamycin on Wnt-1 tumors and the immune system in vivo and in vitro.

To determine the level of immunosuppression induced by Rapamycin, lymphocytes from in vivo treated mice were analyzed at days 7 and 20 of treatment. At day 7, Rapamycin treated recipients had a substantial decrease in thymocytes and splenocytes (Fig. 2A). Although spleen cell numbers almost normalized by day 20, thymocyte counts remained severely depressed. There was no difference in the total number of bone marrow cells before and after Rapamycin treatment. Flow cytometry analysis on days 7 and 20 showed no significant difference in the proportion of splenic CD3⁺, CD4⁺, CD8⁺, CD4⁺CD25⁺, CD19⁺, NK1.1⁺, and CD11b⁺ cells (Additional file 1), demonstrating that different subpopulations of lymphocytes are sensitive to Rapamycin to the same extent.

Bild vergrößern

To determine whether Wnt-1 tumor implantation also had an effect on the immune system, an additional group of mice was treated with Rapamycin in the presence or absence of tumor. Implantation of tumors did not affect the number of cells in these groups (Additional file 1). An additional group of mice implanted with tumor cells but not treated with Rapamycin was also included. Only mice treated with Rapamycin showed a decrease in cell numbers. Thus, we concluded that immunosuppression was induced solely by Rapamycin treatment and transplantation of Wnt-1 cell did not have a detectable effect on the immune system in this model.

To determine whether the decrease in splenocyte numbers found at day 7 of Rapamycin treatment was associated with apoptosis, we stained freshly isolated splenocytes from control and Rapamycin treated animals with DiOC₆. In Rapamycin treated group, 30 to 60% of splenocytes were apoptotic as indicated by DiOC₆ staining (Fig. 2B). In contrast, control mice had only 10 to 18% apoptotic splenocytes (Fig. 2C). Similar results with 25 to 52% of splenocytes in apoptotic fraction were obtained at day 20 of treatment with Rapamycin (data not shown).

To evaluate the function of residual lymphocytes in Rapamycin treated animals, splenocytes were harvested at day 7 and 20 of therapy and co-stimulated with CD3 and CD28 antibodies. Cytokine production was found only in CD3/28 stimulated cultures. T cell cytokine secretion was completely blocked by Rapamycin on day 7 (Fig. 2D). However, by day 20 of therapy, splenic T cell cytokine secretion recovered probably due to generation of Rapamycin resistant T cells. Rapamycin did not induce a shift away from Th1-type cytokines, since IFN-gamma production was predominant in control (day 0) and 20 day treated groups (Fig. 2D).

As it was shown above, Rapamycin induced apoptosis in splenocytes. On the other hand XTR also accelerated Wnt-1 tumor growth. We hypothesized that injection of Rapamycin resistant T-cells could synergize with rapamycin in tumor control. T1Rapa cells are resistant to rapamycin, while host T cells undergo apoptosis after rapamycin therapy initiation. Besides, T1Rapa cells are fully differentiated effector cells of all specificities able to execute their function immediately after the contact with specific targets. Hypothetically this could provide some advantages, i.e.: i) in case immune response to tumor antigens is possible, some of these cells would proliferate quicker than naïve T-cells; ii) tumor antigens are presented by MHC class II molecules, which mostly stimulate Th1 or Th2 responses; while Tc1 cells are more likely to mediate cytotoxic anti tumor responses; iii) these cells are rather long living as it was determined in our previous paper [26].

To estimate the effect of Rapamycin resistant T1 cells (T1Rapa) on Wnt-1 tumor growth, irradiated and BM reconstituted mice were inoculated with tumor cells and injected either at day 5 or day 20 post transplant with 7 × 10⁶ cells/mouse of T1Rapa cells. Adoptive transfer of T1Rapa cells did not reduce the growth of Wnt-1 tumors (Fig. 3A).

Bild vergrößern

Although Rapamycin therapy delayed tumor growth, this effect was transient and tumor growth occurred after cessation of therapy. We tested whether adoptive transfer of T1Rapa cells at the end of Rapamycin treatment may delay tumor re-growth. Sequential Rapamycin therapy for 20 days followed by T1Rapa cell transfer injected on day 21 did not change Wnt-1 tumor growth as compared with Rapamycin alone (Fig. 3B). Thus, Wnt-1 tumor growth was inhibited by Rapamycin, but not by adoptive T1Rapa cell therapy.

To evaluate the cellular mechanisms operational during Rapamycin induced inhibition of Wnt-1 growth we obtained purified primary tumor cells in vitro. Tumor cells were plated in culture medium for 2–3 days, and non-adherent cells were removed. More than 90% of the remaining adherent cells had epithelioid morphology (Fig. 4A) and were positive for epithelial cell Ep-CAM marker as determined by scanning cytometry (Fig. 4B). Additional characterization included identification of vimentin-positive myoepithelial cells which constituted less than 2% (data not shown).

Bild vergrößern

The effect of Rapamycin on primary Wnt-1 tumor cell proliferation was determined in vitro on cells obtained from individual mouse tumors. Rapamycin inhibited proliferation of Wnt-1 cells, as well as normal lymphocytes, in a wide range of concentrations (0.1 to 50 μM), and was toxic at a concentration above 100 μM (Fig. 4C and 4D). Inhibition of Wnt-1 cell proliferation by Rapamycin was 30–50%, and growth inhibition of splenocytes was 50–90%. There was no difference in in vitro Rapamycin sensitivity between in vivo Rapa-treated or vehicle-treated cells (data not shown).

The effect of Rapamycin on the mTOR pathway was further examined in short-term primary cultures of Wnt-1 tumor cells and in two clonal cell lines established from these tumors. Phosphorylated Akt kinase, which activates Akt and directly phosphorylates mTOR, and expression of mTOR downstream messengers were present in all tumors, but their intensity varied in primary cells from different individual mice (Fig. 5A). Nine primary tumors were analyzed. Among others, 3 were like culture #1, and #2 last were like cultures #2 and #3, accordingly. We can see in samples #2 and #3 increased level of phosphorylated Akt kinase, while decreased amount of mTOR products. The reason for such variability is non known. This could be due to variable response of primary cells to tissue culture conditions. Phosphorylation of mTOR associated proteins was reduced by Rapamycin in 5 of 9 cultured tumors. We also generated two stable cell lines from two different primary tumors, and tested their response to Rapamycin after ten passages in vitro. Both cell lines were sensitive to Rapamycin with decreased phosphorylation of p70S6K and S6 ribosomal protein (Fig. 5B).

Bild vergrößern

Rapamycin has been shown to inhibit the proliferation of T cells and some tumors by inducing cell cycle arrest in G1 followed by apoptosis [31, 32]. We examined whether a similar process occurs in Wnt-1 tumor cells. Wnt-1 primary cultured cells (n = 6) were incubated with Rapamycin for 24 h. Freshly isolated splenocytes were used as controls (n = 5). At 24 h, nearly 30% of splenocytes and Wnt-1 cells were apoptotic in cultures exposed to media alone (Fig. 6). Rapamycin increased the percent of apoptotic splenocytes to 76% (Fig. 6A), but did not augment apoptosis of Wnt-1 cells (Fig. 6B). Fig. 6C summarizes data for Rapa-induced apoptosis in splenocytes and Wnt-1 cells.

Bild vergrößern

To test whether the failure of Rapamycin to induce apoptosis in Wnt-1 cells could be due to lack of Fas expression, we examined its expression on ep-CAM⁺ primary cultures of Wnt-1 cells. Fas expression was found in 2% to 10% of Wnt-1 cells (Fig. 6D) while in 90% of activated splenocytes (Fig. 6E). Thus, it is possible that reduced apoptotic response of Wnt-1 cells could be due to low Fas expression.

We also determined the effect of Rapamycin on cell cycle progression of primary Wnt-1 cultures and CD3/28 co-stimulated splenocytes. Cells were incubated with Rapamycin for 72 hours and then collected. There was no difference between control and Rapamycin treated Wnt-1 cells, among them 80–88% of cells were found in G1 phase, and 12–18% in G2/S phase. In contrast, Rapamycin induced cell cycle arrest in activated splenic T cells where the percentage of cells in G1 phase increased from 56 to 77%, while in S phase the portion of cells decreased from 33 to 10% (Fig. 7). This result demonstrates that Rapamycin does not induce cell cycle arrest in Wnt-1 cells.

Bild vergrößern