IL-4 inhibits the TNF-α induced proliferation of renal cell carcinoma (RCC) and cooperates with TNF-α to induce apoptotic and cytokine responses by RCC: implications for antitumor immune responses

Abstract

Objective: While previous reports clearly demonstrated antiproliferative effects of IL-4 on renal cell carcinoma (RCC) in vitro, the administration of IL-4 to patients with metastatic RCC in clinical trials could not recapitulate the promising preclinical results. In the present study we wanted to examine the context of IL-4 action and to establish conditions of enhanced IL-4 efficacy. Methods: Primary and permanent human RCC cells were cultured in either serum-supplemented or chemically defined, serum-free culture medium in the presence or absence of cytokines. Cell proliferation was assessed as [³H]-thymidine incorporation. Cell apoptosis was measured using the fluorescent DNA intercalator 7-aminoactinomycin D and flow cytometry. In addition, culture media conditioned by RCC were subjected to cytokine antibody array and cytokine multiplex analysis. Results: Our results indicate that the previously reported antiproliferative effects of IL-4 are serum-dependent. Under serum-free conditions, IL-4 failed to exhibit growth-inhibitory effects or was even growth-stimulatory. In a chemically defined, serum-free medium (AIM-V), however, IL-4 inhibited the TNF-α induced proliferation of RCC. IL-4 and TNF-α synergistically induced apoptosis of RCC as well as a complex cytokine response by RCC, which included the synergistic upregulation of RANTES and MCP-1. Conclusions: IL-4 alone has little effect on the spontaneous proliferation of RCC but can prevent the enhancement of proliferation induced by growth promoters like FBS and TNF-α. The concomitant growth inhibitory, apoptosis-inducing, and cytokine-enhancing effects of IL-4 in combination with TNF-α on RCC support the view that Th2 cytokines may be required for productive immune responses against RCC.

Introduction

Renal cell carcinoma (RCC) accounts for 2–3% of all malignancies and is the most common cancer of the kidney [1, 2]. RCC represents a heterogenous group of neoplasms composed of clear cell, papillary, chromophobe, and collecting duct cell types. Clear-cell renal cancers, which derive from proximal tubular epithelial cells, comprise the great majority of RCC (>90%). RCC affects more than 30,000 individuals each year in Europe and has a peak incidence in the fifth and sixth decade of life. More than 40% of the patients have metastatic disease at diagnosis and 30–50% of initially localized RCCs eventually metastasize. Patients with metastatic RCC have a median survival time of about 10 months reflecting the lack of effective treatment for metastatic disease [1, 2].

Infiltration of RCC tissue by lymphocytes [3], macrophages [4], and dendritic cells [5] reflects an ongoing or past interaction between the tumor and the immune system. Moreover, occasional regressions and certain responsiveness to immunotherapeutic interventions have strengthened the view that RCC is an immunogenic tumor [1, 2]. Interleukin-2 (IL-2) and interferon-α (IFN-α) have demonstrated limited clinical activity in the treatment of metastatic RCC. While high-dose bolus IL-2 could induce durable responses in a small portion of highly selected patients, treatment with IFN-α was associated with a modest survival benefit [1, 2].

Renal cell carcinoma has also been shown to respond in vitro to various cytokines including tumor necrosis factor-α (TNF-α) [6], IL-6 [7], type 1 IFN [8], IFN-γ [9], IL-4 [10–12], and IL-13 [13]. The cytokine effects observed in these studies were either promoting growth (TNF-α, IL-6) or antiproliferative (type 1 IFN, IFN-γ, IL-4, and IL-13).

The expression of intermediate to high-affinity receptors for the closely related cytokines IL-4 and IL-13 by RCC has been clearly demonstrated [11, 13]. For both cytokines growth-inhibitory effects on RCC in vitro via binding of the cytokines to their specific receptors have been reported [11, 13]. However, despite these pronounced preclinical effects of IL-4 subsequent clinical trials of systemic IL-4 in patients with RCC and other tumors were rather disappointing [14–16].

In the present work we have re-examined the effects of IL-4 on RCC in vitro and found (1) that IL-4 effects on RCC proliferation are serum-dependent (2) that IL-4 inhibits the TNF-α induced RCC proliferation (3) that IL-4 plus TNF-α synergistically induce apoptosis of RCC and (4) that IL-4 and TNF-α co-operatively induce a complex cytokine response by RCC.

Materials and methods

Cell lines and culture

The A-498 cell line was purchased from the DSMZ (Braunschweig, Germany) [17–19]. Primary RCC cultures were established from fresh RCC tissue with clear-cell histology. Single cell suspensions were obtained by mechanic and enzymatic tissue disintegration and short-term cultures were initiated. Cells were cultured in RPMI 1640 (Cambrex, Bio Whittaker, Verviers, Belgium) supplemented with 50 U/ml penicillin (PAA Laboratories, Pasching, Austria), 50 μg/ml streptomycin (PAA Laboratories), 2 mM l-glutamine (Invitrogen, Gibro, Paisley, Scotland), 0.1 mM nonessential amino acids (Cambrex, Bio Whittaker), 1 mM pyruvate (Cambrex, Bio Whittaker), and 10 mM Hepes (Cambrex, Bio Whittaker). Where indicated RPMI 1640 was further supplemented with either 10% FBS (HyClone, Logan, UT, USA) or human AB serum (1 or 10%; local Institute of Blood Transfusion). Alternatively, cells were cultured in AIM-V, a chemically defined, serum-free culture medium designed for human ex vivo tissue and cell culture applications (Invitrogen, Gibco). RCC identity of the resulting cultures was proven by use of flow cytometric staining with a monclonal antibody (mAb; IgG2a) directed against G250, a renal adenocarcinoma specific antigen [20] (kindly provided by Egbert Oosterwijk, Nijmegen, The Netherlands).

Flow cytometric analyses

To determine surface expression of various RCC-associated antigens, cell suspensions (10⁵ cells in 50 μl) were labeled for 30 min on ice with primary mouse mAb in complete medium followed by FITC-conjugated F(ab’)2 fragments of goat anti-mouse Ig (Dako, Glostrup, Denmark). The following anti human mAbs were used: pan-cytokeratin (IgG1, Abcom, Cambridge, UK), G250 (IgG2a), HLA-A,B,C (IgG1, kappa, PharMingen, San Diego, California, USA), HLA-DR (IgG2b, kappa, PharMingen), ICAM-1 (CD54; IgG1 kappa), and EGF receptor (IgG2b, PharMingen). The samples were analyzed with a flow cytometer (FACSCalibur™) and CellQuest™ software from BD Biosciences (Mountain View, CA, USA).

Proliferation assay

Cells (A-498 or primary RCC) were plated in 96-well plates at 2×10⁴ cells/well in triplicates. The following cytokines were added as indicated: IL-4 (1,000 U/ml; CellGro, CellGenix, Freiburg, Germany), TNF-α (1,000 U/ml; R&D, Abingdon, UK), and IL-1ß (1,000 U/ml; R&D). The cultures were incubated at 37°C in a 5% CO₂ incubator for 24 h and then pulsed with 1 μCi/well of [³H]-thymidine (MP, Biomed Incorporation, Morgan, Irvine, CA, USA) for the last 16 h. Cells were washed with PBS, detached, and harvested onto glass fiber filter with the “Harvester 96, MACH III M, (TOMTEC). [³H]-thymidine incorporation was determined with the Multi Label detection platform CHAMELEON (HVD Life Science, Vienna, Austria). Data are presented as mean values of triplicate measurements ± SD.

Detection of apoptosis

Untreated or cytokine treated A-498 cells were harvested after 40 h and stained with 1 μg/ml nuclear dye 7-amino-actinomycin D (7-AAD; Sigma-Aldrich, Vienna, Austria). 7-AAD is a fluorescent intercalator that undergoes a spectral shift upon association with DNA. 7-AAD/DNA complexes are excited by the argon-ion laser and emit beyond 610 nm. Since 7-AAD is generally excluded from live cells, 7-AAD can be used for the flow cytometric detection of late apoptosis [21]. The percentage of 7-AAD+ A-498 cells was determined with a FACSCalibur flow cytometer (BD) and CellQuest software (BD).

Human cytokine antibody protein array

Serum-free AIM-V medium conditioned by untreated or cytokine-treated A-498 cells was collected after 40 h and used for the human cytokine antibody protein array (RayBio® Human Cytokine antibody array C series 1000, which detects 79 cytokines in one experiment: http://www.raybiotech.com/humanV.html). The array membrane was pretreated according to the manufacturer’s instruction, incubated with 2 ml of undiluted conditioned medium for 2 h, washed, incubated with biotin-conjugated anti-cytokine mix, washed, and then developed with streptavidin-HRP conjugate and subsequent ECL (Amersham).

Cytokine quantitation

Serum-free AIM-V medium conditioned by untreated or cytokine-treated A-498 cells was collected after 40 h, and RANTES and MCP-1 were determined using the specific Fluorokine MultiAnalyte Profiling (MAP) kits together with the human MAP base kit A (R&D, Biomedica) following the manufacturers’ instructions. After a final wash step, the beads are resuspended in buffer and read on the Luminex 100 Analyzer (Biomedica, Vienna, Austria) to determine the concentration of RANTES and MCP-1. All samples were tested in triplicate wells. Data analysis was performed using the Luminex 100 IS sofware version 2.3. The minimum detectable dose was 1.08 pg/ml (RANTES) and 0.95 pg/ml (MCP-1), respectively. Results are mean values ± SD.

Statistical analysis

Statistics were performed by analyzing data with the student’s t test by utilizing Microsoft Excel software. Results were considered statistically significant at P values≤0.05.

Results

Influence of the culture conditions on the growth-modulatory effects of IL-4

While previous reports clearly demonstrated antiproliferative effects of IL-4 on RCC in vitro [10–12], the administration of IL-4 to patients with metastatic RCC and other tumors in clinical trials could not recapitulate the promising preclinical results [14–16]. The purpose of the present study was to re-examine IL-4 effects on RCC in vitro in the context of varying culture conditions in order to provide an explanation for the discrepancy between the in vitro and in vivo results as well as to establish conditions of enhanced IL-4 efficacy.

The growth inhibitory effect of IL-4 on human RCC has been demonstrated both for short-term primary tumor cells [11] and for permanent RCC cell lines including the A-498 cell line [10]. In the present work we therefore used, besides primary RCC cells, the well-characterized A-498 cell line as a model system [17–19]. Figure 1 depicts a detailed phenotypic analysis of A-498 cells and demonstrates expression of the G250 antigen, which is one of the most reliable markers for RCC [20]. As epithelial cells, A-498 cells also expressed MHC class I antigens but lacked MHC class II antigens. MHC class II antigens could be induced on A-498 cells by IFN-γ (data not shown). Other epithelial markers expressed by A-498 cells include cytokeratins as detected by a pan-cytokeratin antibody and ICAM-1 (CD54) as well as the epidermal growth factor (EGF) receptor, altogether confirming that A-498 is a genuine, human RCC cell line, and thus a useful model to study the pathophysiology of human RCC.

Fig. 1.

Phenotype of the RCC cell line A-498. A-498 cells were subjected to flow cytometric analyses using antibodies against markers of RCC (G250) and epithelial cell-associated antigens. MHC class II, which is expressed exclusively on antigen-presenting cells, served as a negative control

All previous reports, which demonstrated growth-inhibitory effects of IL-4 used culture media that had been supplemented with 10% fetal bovine serum (FBS) [10–12]. In a first set of experiments we wanted to test the influence of serum on the growth-modulatory effects of IL-4 and therefore cultured A-498 cells in RPMI 1640 culture medium with or without FBS (10%), human serum (1 or 10%) or, alternatively, in a chemically defined, serum-free culture medium developed for the use of human ex vivo tissue and cell culture applications (AIM-V). Figure 2 shows A-498 cell proliferation as assessed by [³H]-thymidine incorporation. In accordance with published data, IL-4 induces growth inhibition in the presence of 10% FBS. A similar growth inhibition of A-498 cells by IL-4 was observed in the presence of lipoprotein-deficient FBS (data not shown). The antiproliferative effect of IL-4 was also observed in the presence of low concentrations of human serum (1%). However, when the concentration of human serum was raised to 10%, IL-4 failed to inhibit A-498 cell proliferation. Likewise, in the serum-free culture medium AIM-V IL-4 failed to exhibit antiproliferative effects. Unexpectedly, in RPMI 1640 without serum supplementation IL-4 exerted significant growth-promoting effects. These data indicate that the antiproliferative effects of IL-4 are serum-dependent and that IL-4 can exhibit both growth-inhibitory and growth-stimulatory effects depending on the culture conditions. To exclude serum effects AIM-V culture medium was used in all further experiments.

Fig. 2.

Effects of IL-4 on the proliferation of A-498 cells: serum-dependence. A-498 cells were cultured in RPMI 1640 with or without serum supplemementation as indicated or in the chemically defined, serum-free culture medium AIM-V. Proliferation as assessed by [³H]-thymidine incorporation was determined in the presence or absence of IL-4 (1,000 U/ml). Results are mean values ± SEM of five independent experiments

Growth-modulatory effects of IL-4 in the context of pro-inflammatory cytokines

In a second set of experiments we wanted to evaluate the growth-modulatory effects of IL-4 in the context of other cytokines. Figure 3a demonstrates that TNF-α or IL-1ß could induce a threefold to fourfold increase in A-498 cell proliferation (P=0,0005/P=0.0056). Treatment of A-498 cells with both TNF-α plus IL-1ß resulted in further enhancement of cell proliferation (sixfold above spontaneous proliferation). Although IL-4 was unable to inhibit basic cell proliferation, IL-4 almost completely prevented the proliferation induced by TNF-α (P=0.0005), IL-1ß (P=0.0065) or TNF-α plus IL-1ß (P=0.0037; Fig. 3a). Likewise, TNF-α stimulated the proliferation of G250⁺ primary RCC cultures derived from fresh tumor tissue (P=0.0002; Fig. 3b) although the growth rate of primary cells was significantly lower as compared to the permanent A-498 cell line (Fig. 3a, b). IL-4 also inhibited the TNF-α induced proliferation of primary RCC cells (P=0.0004; Fig. 3b). In contrast to A-498 cells, in primary RCC cultures the effects of IL-1ß were inconsistent and not affected by IL-4 (data not shown).

Fig. 3.

Growth-modulatory effects of IL-4 on RCC in the context of pro-inflammatory cytokines. A-498 cells (a) or primary RCC cells (b) were cultured in the chemically defined, serum-free culture medium AIM-V. Proliferation as assessed by [³H]-thymidine incorporation was determined in the presence or absence of IL-4 (1,000 U/ml), TNF-α (1,000 U/ml), IL-1ß (10 ng/ml) or combinations. FACS histogram demonstrating G250 expression on primary RCC (b). Results are mean values of triplicate measurements ± SD. One out of three (a) and two experiments (b) with consistent results is shown

Induction of A-498 cell apoptosis by IL-4 and TNF-α

As a next step we wanted to further characterize the context of the IL-4 mediated inhibition of the TNF-α induced proliferation. To determine whether apoptosis was induced, untreated or cytokine treated, A-498 cells were stained with the nuclear intercalator 7-AAD and the percentage of 7-AAD⁺ cells was determined by flow cytometry. Figure 4 demonstrates that either cytokine alone increased the rate of apoptosis of untreated cells from 17.5 to 29.6% (IL-4) and 28.2% (TNF-α). The combination of IL-4 and TNF-α, however, induced apoptosis in 75.6 % of the cells, indicating that the two cytokines acted in a synergistic fashion.

Fig. 4.

Induction of apoptosis in A-498 cells by IL-4 and TNF-α. A-498 cells were cultured in the chemically defined, serum-free culture medium AIM-V in the presence or absence of IL-4 (1,000 U/ml) or TNF-α (1,000 U/ml) or a combination of IL-4 and TNF-α. The percentage of apoptotic cells was determined by flow cytometry using the fluoresecent DNA intercalator 7-AAD. One out two experiments with nearly identical results is shown

Cytokine profiles of A-498 cells and regulation by IL-4 and TNF-α

To further characterize the effects of IL-4 and TNF-α on human RCC we performed a detailed human cytokine protein array (RayBio) to screen for cytokines released into the culture medium during 48 h of treatment of A-498 cells with IL-4, TNF-α or IL-4 plus TNF-α. Figure 5 demonstrates that IL-4 and TNF-α indeed affect the cytokine profile of A-498 RCC cells. Under steady state conditions (Fig. 5a), A-498 cells appeared to produce IL-8 (J2), MCP-1 (E3), GDNF (H6), TIMP-1 (G8), and TIMP-2 (H8).

Fig. 5.

Cytokine production by A-498 cells in response to IL-4 and TNF-α. A-498 cells in serum-free medium (AIM-V) were either left untreated (control) or treated with IL-4 (1,000 U/ml), TNF-α (1,000 U/ml) or a combination of IL-4 and TNF-α. Cytokine protein array (Raybio) was performed with the conditioned media. GRO (J1), IL-8 (J2), IL-6 (H2), MCP-1 (E3), M-CSF (H3), RANTES (B4), TNF-α (G4), Eotaxin (I5), FGF-4 (A6), GDNF (H6), HGF (I6), NAP-2 (J7), TIMP-1 (G8), TIMP-2 (H8)

Overall, IL-4 treatment had little effect on the cytokine profile of A-498 cells (Fig. 5b). The most prominent effects of IL-4 were suppression of IL-8 (J2) as well as enhancement of TIMP-1 (G8). In contrast, treatment with TNF-α had dramatic effects on the cytokine profile of A-498 cells (Fig. 5c). Consistent with the proliferative response, TNF-α induced GRO in A-498 cells (J1). In addition, IL-6 and IL-8 were dramatically induced or upregulated (H2 and J2). MCP-1 expression was also clearly enhanced by TNF-α (E3). M-CSF, GDNF, HGF, and TIMP-1 also appeared to be enhanced by TNF-α (H3, H6, I6, G8). When A-498 cells were treated with a combination of TNF-α and IL-4 (Fig. 5d), levels of all cytokines induced by TNF-α such as GRO (J1), IL-8 (J2), and IL-6 (H2), remained high and were not inhibited by IL-4. Conversely, RANTES was induced in A-498 cells by TNF-α and IL-4 in a synergistic fashion (Fig. 5d, B4). MCP-1 (E3) and M-CSF (H3) also appeared to be enhanced in the presence of both TNF-α and IL-4 (Fig. 5d) as compared to the presence of TNF-α alone (Fig. 5c). The regulation of cytokine production in A-498 cells by TNF-α and IL-4 is summarized in Table 1.

Table 1.

Cytokine response of the A-498 RCC cell line

Cytokine	Abbreviation	Co-ordinates	Function (RCC related)/receptor	W/o	IL-4	TNF-α	IL-4 + TNF-α	Synergy
Growth-related oncogene	GRO	J1	IL-8 related cytokine chemotactic for neutrophils and basophils	–	–	+++	+++	No
Interleukin-8	IL-8	J2	Chemokine for neutrophils	+++	++	+++++	+++++	No
Interleukin-6	IL-6	H2	Pleiotropic, intracrine growth factor for RCC	–	+	+++++	+++++	No
Monocyte chemo-attractant protein-1	MCP-1	E3	Chemotactic for monocytes, pro-proliferative	+++	++(+)	++++	++++(+)	Yes
Macrophage colony-stimulating factor	M-CSF	H3	Growth, differentiation and survival of monocytes	(+)	(+)	+	++	?
RANTES	CCL-5	B4	Leukocyte recruitment into inflammatory sites/CCR3	–	(+)	(+)	+++	Yes
Eotaxin	Eotaxin	I5	Chemokine for eosinophils/CCR3	–	–	–	+(+)	Yes
Glial cell-derived neurotrophic factor	GDNF	H6	Kidney organogenesis, renal dysplasia	++	+++	+++	++(+)	No
Hepatocyte growth factor	HGF	I6	Kidney organogenesis, renal dysplasia	+	+	++(+)	++	No
Tissue inhibitor of metalloproteinase-1	TIMP-1	G8	Regulator of ECM production, inhibits tissue invasion, growth factor, correlates with poor prognosis in RCC	+(+)	+++	+++	+++(+)	No
Tissue inhibitor of metalloproteinase-2	TIMP-2	H8	Regulator of ECM production, inhibits tissue invasion, growth factor, correlates with poor prognosis in RCC	+++(+)	+++	++++	+++(+)	No

Regulation of RANTES and MCP-1 production in A-498 cell by IL-4 and TNF-α

To verify the results obtained by the cytokine protein array (Fig. 5), RANTES and MCP-1 were quantified using cytokine multiplex analysis technology based on the Luminex system. Figure 6 clearly demonstrates the synergistic induction of both RANTES and MCP-1 by TNF-α and IL-4 consistent with the cytokine protein array data shown in Fig. 5.

Fig. 6.

Regulation of RANTES and MCP-1 production in A-498 cell by IL-4 and TNF-α. A-498 cells in serum-free medium (AIM-V) were either left untreated (control) or treated with IL-4 (1,000 U/ml), TNF-α (1,000 U/ml) or a combination of IL-4 and TNF-α. Cytokine multiplex analysis was performed with the conditioned media using the commercial Fluorokine MAP kits for RANTES and MCP-1 as well as the Luminex 100 analyzer

Discussion

Previous work has demonstrated growth-inhibitory effects of IL-4 on human RCC in vitro [10–12]. However, clinical administration of IL-4 to patients with RCC or other tumors failed to reproduce the promising in vitro findings in vivo [14–16]. The antiproliferative effects of IL-4 on RCC in previous reports have always been observed in culture systems containing 10% FBS [10–12]. In our re-examination we found that the IL-4 mediated growth inhibition of RCC is indeed serum-dependent. In the presence of human serum the antiproliferative effect of IL-4 was either less pronounced (1% human serum) or absent (10% human serum; Fig. 2). In the absence of any serum supplementation (RPMI 1640 only), IL-4 even induced a more than twofold stimulation of RCC proliferation (Fig. 2). One possible interpretation of this finding is that under conditions of tumor starvation IL-4 might even stimulate tumor growth. To avoid serum effects we used a chemically defined, serum-free culture medium (AIM-V), which has been developed for human ex vivo tissue and cell culture applications. In AIM-V, IL-4 completely failed to affect RCC proliferation (Fig. 2). In subsequent experiments using AIM-V, we found that TNF-α strongly stimulates RCC proliferation and that IL-4 can indeed prevent the proliferative response induced by TNF-α both in A-498 cells (Fig. 3a) and in primary RCC (Fig. 3b). In summary, the findings presented in Figs. 2 and 3 indicate that IL-4 alone has little effect on the spontaneous proliferation of RCC but can inhibit or prevent the enhancement of proliferation induced by growth promoters like FBS and TNF-α.

In additional experiments we also found that IL-4 and TNF-α can cooperatively act upon RCC. IL-4 and TNF-α synergistically induced RCC apoptosis (Fig. 4) and, more surprisingly, induced a complex cytokine response in RCC cells (Fig. 5, Table 1).

In the absence of cytokine treatment, A-498 cells spontaneously produced IL-8, MCP-1, GDNF, TIMP-1, and TIMP-2. The expression of MCP-1 and IL-8 mRNA by human RCC has previously been demonstrated in vitro and in situ [22–24]. Moreover, immunohistochemical investigations of 153 RCC tissues revealed that TIMP-1 protein was present in 46% and TIMP-2 protein was present in 73% of the samples [25]. The detection of these cytokines (IL-8, MCP-1, TIMP-1, and TIMP-2) in media conditioned by A-498 cells (Fig. 5a) complements the phenotypic characterization of A-498 cells (Fig. 1) and can be considered additional validation for the usefulness of the A-498 cell line as a genuine RCC model system.

GDNF like HGF is known to be important during kidney organogenesis and promotes cell survival and tubulogenesis [26]. Moreover, GDNF expression is associated with renal dysplasia where it correlates with high proliferation and BCL-2 expression [27]. Our data now indicate for the first time that GDNF expression also occurs in human RCC (Fig. 5).

Stimulation of A-498 cells with TNF-α dramatically induced GRO and IL-6, and enhanced IL-8, MCP-1, HGF, TIMP-1, TIMP-2. (Fig. 5c). Intriguingly, IL-4 which is considered an anti-inflammatory cytokine did not inhibit the TNF-α induced cytokine production but instead enhanced production of several cytokines. The most striking finding was the synergistic induction of RANTES by TNF-α and IL-4 (Fig. 5d, B4), which was confirmed by cytokine multiplex analysis with the Luminex system (Fig. 6a). The synergistic induction of the chemokine RANTES, which can bind to the chemokine receptor CCR3, by TNF-α and IL-4 has been reported for bronchial epithelial cells and skin dermal fibroblasts and has been considered an allergic disease-specific mechanism that contributes to skin and lung inflammation by recruiting inflammatory cells in atopic individuals [28–31]. Our findings indicate that RCC cells similarly respond to the combined stimulation with TNF-α and IL-4. We have recently shown that RCC expresses CCR3 [19]. CCR3 was detected immunohistochemically in RCC tissue and on A-498 cells by flow cytometry. A-498 cells mounted an eotaxin-induced proliferative response indicating that A-498 cell-associated CCR3 has signaling competence [19]. Our present findings demonstrate that IL-4 inhibits the TNF-α induced proliferation (Fig. 3) despite the presence of an intact autocrine or paracrine loop consisting of RANTES (Fig. 5d) and its corresponding receptor CCR3 (data not shown) [17] indicating that the growth-inhibitory effect of IL-4 overrules the growth stimulatory effect of CCR3 signaling. CCR3 was not downregulated on A-498 cells but was instead modestly enhanced by treatment with TNF-α and IL-4 (data not shown). Cytokine multiplex analysis revealed that the enhancement of MCP-1 by TNF-α and IL-4 was also synergistic (Fig. 6b).

A prevailing view in tumor immunology is that T helper (Th)1 responses involving CD8⁺ cytotoxic T cells (CTL) are required for tumor destruction. While killing by CTLs must still be considered a major tumoricidal effector mechanism, the central role of CD4⁺ T cells in the antitumor immune response has been recognized [32]. In addition, the simultaneous induction of both Th1 (IL-2, IFN-γ) and Th2 responses (IL-4, IL-5) appears to be required for maximal systemic antitumor immunity [32–34]. Our present data confirm this view, since IL-4 inhibited the TNF-α induced proliferation (Fig. 3). Moreover, IL-4 and TNF-α synergistically induced apoptosis (Fig. 4) and cytokine production by RCC (Figs. 5, 6). Thus, IL-4 and TNF-α not only directly inhibit RCC but may also promote the recruitment of immune effectors cells by inducing multiple chemokines (IL-8, GRO, MCP-1, M-CSF, RANTES; Figs. 5, 6). IL-8 and the related chemotactic cytokine GRO activate neutrophils (IL-8), and neutrophils and basophils, respectively (GRO). Neutrophils are indeed important for the rejection of IL-4-producing tumors [35]. Monocytes are attracted by MCP-1, and M-CSF promotes their differentiation and survival. RANTES is chemotactic for T-cells, human eosinophils and basophils, and plays an active role in recruiting leukocytes into inflammatory sites. Eotaxin induces the substantial accumulation of eosinophils, which have previously been implicated in the rejection of RCC [36].

Our present findings would suggest the combined administration of IL-4 and TNF-α in the immunotherapy of metastatic RCC. However, the systemic administration of IL-4 plus TNF-α to RCC patients is probably impossible since either cytokine alone has substantial side effects, which are likely to be potentiated due to the synergistic effects of IL-4 and TNF-α (Figs. 4, 5, 6). Instead, antitumor vaccination attempts are an attractive option to induce the desired immune response in situ. A frequently used approach in the therapeutic vaccination against RCC is based on dendritic cells [37–41]. The results obtained in dendritic cell-based immunotherapy, however, indicate that immune responses induced by cultured dendritic cells are extremely biased towards Th1 and virtually lack Th2 components (in particular IL-4) [38, 39, 41]. Thus, future clinical studies should exploit means of immune manipulation that shift the bias to generate more balanced immune responses consisting of both Th1 and Th2 components [42].

Abbreviations

CCR

Chemokine (C-C motif) receptor 3

CTL

Cytotoxic T lymphocyte

ECL

Enhanced chemoluminescence

ECM

Extracellular matrix

EGF

Epidermal growth factor

FACS

Fluorescence activated cell sorting

FBS

Fetal bovine serum

FGF

Fibroblast growth factor

GDNF

Glial cell derived neurotrophic factor

GRO

Growth-related oncogene

HGF

Hepatocyte growth factor

HRP

Horse radish peroxidase

IFN

Interferon

IL

Interleukin

MAP

MultiAnalyte profiling

MCP-1

Monocyte chemoattractant protein

M-CSF

Macrophage colony-stimulating factor

MHC

Major histocompatibility complex

NAP

Neutrophil activating protein

RANTES

Regulated upon activation, normal T-cell expressed, and presumably secreted

RCC

Renal cell carcinoma

Th

T helper

TNF

Tumor necrosis factor

TIMP

Tissue inhibitor of metalloproteinase