interleukin 4 on rat experimental gliomas. S Benedetti1, B Pirola1, PL Poliani1, L Cajola1, B Pollo1, R Bagnati2, L Magrassi3, P Tunici1 and G Finocchiaro1.
Gene Therapy (2003) 10, 188–192 & 2003 Nature Publishing Group All rights reserved 0969-7128/03 $25.00 www.nature.com/gt
BRIEF COMMUNICATION
Dexamethasone inhibits the anti-tumor effect of interleukin 4 on rat experimental gliomas S Benedetti1, B Pirola1, PL Poliani1, L Cajola1, B Pollo1, R Bagnati2, L Magrassi3, P Tunici1 and G Finocchiaro1 1
Istituto Neurologico Besta, 20133 Milan, Italy; 2 Department of Environmental Health Sciences, Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy; and 3 Policlinico S. Matteo, Department of Surgery, University of Pavia, Italy
Retroviral-mediated gene transfer of the IL-4 gene into experimental gliomas can cause tumor rejection, supporting the clinical use of this form of gene therapy for glioblastomas (GBM). In a clinical setting, the administration of dexamethasone (dex) is a standard procedure for GBM patients. This led us to examine the effects of dex on IL-4 gene therapy. We injected intracranially Fischer 344 rats with phosphatebuffered saline, 9L gliosarcoma cells mixed with E86.L4SN200 cells (retroviral producer cells, RPC, transducing IL-4 cDNA) and 9L cells mixed with PA317.STK.SBA cells (control RPC expressing the HSV-tk gene). The rats
from each group were treated with 0, 50, 100 or 250 mg dex/ kg/day released by osmotic pumps implanted subcutaneously. While 80–100% of rats receiving 9L cells mixed with IL-4 RPC and not treated by dex survived for at least 2 months following tumor injection, only 50% and 17% of rats receiving 50 or 100 mg/kg/day of dex, respectively, reached this time point. These results indicate that dex significantly diminished the anti-tumor effect of IL-4. Thus, in a clinical setting, IL-4 gene transfer should be performed when low levels of dex are administered or in the absence of dex. Gene Therapy (2003) 10, 188–192. doi:10.1038/sj.gt.3301863
Keywords: IL-4; glioma; dexamethasone
Glucocorticosteroids (GC) play an important role in the management of malignant brain tumors, either primary or secondary and perioperatively in brain surgery.1 Dexamethasone (dex) is the GC given in the majority of neuro-oncologic patients, at an empirically chosen dosage of 4 mg qid. Dex has a dramatic effect on symptoms in patients with brain tumors by decreasing the blood–tumor barrier permeability and the regional cerebral blood volume.2 Furthermore, dex may decrease brain-tumor-associated edema by counteracting the action of VEGF.3 Steroid medications, including dex, also have immunosuppressive functions. Studies based on flow cytometry to quantify the extent of inflammatory cell infiltration in the immunogenic rat C6 glioma model showed that a 7-day course of dex (100 mg/kg/day) resulted in a greater than 50% inhibition of microglia and lymphocyte (but not macrophage) infiltration into tumors.4 Dex may specifically inhibit the action of IL-4 by decreasing the expression of the IL-4 receptor alpha.5 These findings should be considered when experimental immunotherapeutic strategies are evaluated for clinical application. We and others previously found that viral-mediated transfer of the interleukin-4 (IL-4) gene may significantly inhibit the growth of C6, 9L and GL261 experimental gliomas.6–11 In these experiments IL-4 gene transfer was associated with tumor infiltration by Correspondence: G Finocchiaro, Istituto Nazionale Neurologico Besta, Biochemistry and Genetics, Laboratory of Neuro-Oncology and Gene Therapy, Via Celoria 11, 20133 Milan, Italy Received 13 November 2001; accepted 15 July 2002
inflammatory cells and particularly T-lymphocytes and macrophages. Thus, in view of future clinical trials based on IL-4 gene transfer in gliomas we have evaluated the effects of dex administration on IL-4 gene therapy of 9L malignant gliomas. In experiment A, we tested the effects of 250 mg/kg/ day of dex administered for 4 weeks on the growth of 9L gliosarcomas mixed with IL-4 retroviral producer cells (RPC). Results are shown in Figure 1a. This amount, however, was severely toxic. All the rats treated by dex (including those only injected with phosphate-buffered saline, PBS) lost weight and eventually died by day 40. By that time rats injected with 9L and control RPC also died while 5/6 rats injected with 9L cells and IL-4 RPC survived to day 90. In experiment B, we therefore decreased the amount of dex to 100 mg/kg/day for 2 weeks (Figure 1b). This concentration was well tolerated and all rats treated by dex and injected with PBS survived for at least 2 months (an arbitrary endpoint). Rats injected with 9L cells and control RPC, with or without dex, all died by day 40, without significant differences between the two groups. In case of rats bearing IL-4 RPC, however, the difference was quite dramatic. Sixty days after tumor injection, only 1/6 rats treated by dex were alive while 5/6 rats were alive in the group without dex. In experiment C, we lowered the amount of dex used to 50 mg/kg/day (Figure 1c). Rats sham-injected with PBS and treated with dex did not show any abnormality until the end of the experiment (60 days). Rats bearing 9L gliosarcomas that were co-injected with control RPC (SBA cells) all died by day 47, independent of dex
Dexamethasone inhibits IL-4 action on gliomas S Benedetti et al
189
Figure 1 Kaplan Meier analysis of survival of rats with 9L gliomas treated by IL-4 gene transfer and receiving different amounts of dex. (a–c) Results of experiments performed using the amounts of dex indicated in the box. Each group of Fischer 344 rats is indicated by a different color. Rats injected with PBS and treated by Dex are indicated in yellow. Rats injected with 9L cells and IL-4 RPC are in black. Rats injected with 9L cells and control RPC are in green. Rats injected with 9L cells and IL-4 RPC and treated by dex are in red. Rats injected with 9L cells and control RPC and treated by dex are in blue. 9L gliosarcoma cells were obtained from Dr Karl Plate (Freiburg). RPC transducing murine IL-4 (E86.L4SN200 cells) were obtained as described.8 Dexamethasone sodium phosphate (Soldesam, Laboratorio Farmacologico Milanese, Varese, Italy) was used as a solution (4 mg/ml). Delivery of dexamethasone was obtained using Alzet osmotic pumps (Alza Corp, Palo Alto, CA, USA) set to release 2.5 ml/h for 28 days (model 2ML4) or 0.5 ml/h for 14 days (model 2002). Pumps were implanted subcute, dorsally, in anesthetized rats (Fischer 344, female, weight 160–180 g) using sterile procedures. During anesthesia, rats were injected with 6 104 9L cells mixed with the same number of IL-4 RPC or control RPC (SBA cells, expressing the HSV-tk gene7). Three sets of experiments were performed to test the effects of different amounts of dex on the anti-glioma effect of IL-4. In the first experiment (experiment A) we set pumps to release 250 mg/kg/day of dex for 4 weeks. Thirty-eight rats were studied in this experiment, divided into five groups as in A and B: group 1 (n¼3); group 2 (n¼12); group 3 (n¼11); group 4 (n¼6); group 5 (n¼6). In the second experiment (experiment B) osmotic pumps were set to release 100 mg/kg/day of dex for 2 weeks. Twenty-six rats were used in experiment B and divided into five groups with the same characteristics of experiment A (except that dex, when used, was twice the amount used in A): group 1 (n¼3); group 2 (n¼6); group 3 (n¼6); group 4 (n¼6); and group 5 (n¼5). In the third experiment (experiment C) osmotic pumps were set to release 50 mg/kg/day of dex. Twenty-seven rats were studied and divided into five groups: group 1 (n¼3) received an i.c. injection of PBS and dex, to test the toxicity of the drug; group 2 (n¼6) was injected with 9L cells mixed with E86.L4SN200 cells with dex; group 3 (n¼6) was injected with 9L cells mixed with control SBA cells with dex; group 4 (n¼6) was like group 2 but without dex; group 5 (n¼6) was like group 3 but without dex. Kaplan–Meier analysis was used to evaluate survival using software StatView 4.5 (Abacus, Berkeley, CA, USA).
treatment. Rats with 9L-SBA-dex survived 36 7 10 days, rats with 9L-SBA without dex 34 7 10 days, but the difference was not significant. The rats injected with 9L cells and IL-4 RPC were healthy and all survived until the end of the experiment. However only 50% of the rats injected with 9L and IL-4 RPC and treated with dex survived to day 60. Levels of dex were measured in the serum of IL-4treated, control and naı¨ve rats 1 and 2 weeks after the implantation of the osmotic pumps. The data are presented in Table 1. Each value in the table, except for naı¨ve rats, is representative of one single animal. Blood samples at day 7 and 14 were taken from different animals. Overall, the results show a correlation between the amount of dex administered and blood measurements. Group A, treated by 250 mg/kg/day of dex, had 31.5 7 6.5 ng/ml. Group B (100 mg/kg/day) had 19.0 7 4.0 ng/ml. Group C (50 mg/kg/day) had 10.0 7 1.8 ng/ml (n¼6) of dex in the serum. Differences
between dex levels at 7 and 14 days or between different treatments were not significant. The results we have obtained by administering dex to rats with 9L experimental gliomas show that this steroid, by itself, does not significantly modify the lethality of the tumor. IL-4 gene transfer, on the contrary, has a relevant inhibitory effect on such experimental glioma, confirming previous data. Dex administration in IL-4-treated rats, however, significantly decreases the antiglioma effect of IL-4. This observation may suggest that dex inhibits the T-lymphocyte and macrophage infiltration at the tumor site that is triggered by the presence of IL-4 in the tumor environment. The histological examination of few IL-4 treated tumors, with and without dex appeared to support this explanation. To test this further, we performed another experiment with three groups of rats: one group (n¼10) was injected with 9L cells mixed with control RPC (SBA cells); a second group (n¼9) was injected with 9L cells mixed with IL-4 RPC; and a third Gene Therapy
Dexamethasone inhibits IL-4 action on gliomas S Benedetti et al
190
Gene Therapy
Dexamethasone inhibits IL-4 action on gliomas S Benedetti et al Table 1 Dex levels in the serum of na.ıve, control and IL-4-treated rats with 9L gliomas Rats
Exp. A (250 mg/kg/day) 9L+IL-4 RPC PBS Exp. B (100 mg/kg/day) 9L+IL-4 RPC PBS Exp. C (50 mg/kg/day) 9L+control RPC 9L+IL-4 RPC PBS Naive (no dex; n=2)
Dex levels (ng/ml) Day 7
Day 14
37.3 31.3
34.8 22.5
16.8 19.3
24.5 15.4
12.0 10.2 7.0 o1
10.0 11.7 9.1
Blood samples were drawn from rats in experiments A–C to determine dex levels in the serum 1 and 2 weeks after implantation of the pump. Analysis of dex was performed with an API 3000 triple quadrupole instrument (Applied Biosystems Sciex, Thornhill, Ontario, Canada), interfaced with two Series 200 micro LC pumps (Perkin-Elmer, Norwalk, CT, USA) through a standard Heated Nebulizer source. HPLC conditions were as follows: column: Chromspher C18, 5 mm, 3 100 mm (Chrompack, Middelburg, Netherlands); flow: 0.5 ml/min; eluent A: 10 mM ammonium formate in water; eluent B: methanol; gradient: from 35% B to 100% B in 8 min; loop 20 ml. The HPLC effluent was directly connected to the ion source. MS conditions were as follows: source: Heated Nebulizer, in negative ion mode and with heater gas at 4001C; orifice voltage: 60 V; ring voltage: 180 V; collision cell gas: nitrogen at a pressure of 2.6 105 torr; collision energy: 30 eV. Quantitative analysis of dex was done by Multiple Reaction Monitoring (MRM), measuring the fragmentation product (m/ z 307) of the deprotonated pseudo-molecular ion (m/z 361). Flumethasone was used as internal standard, by measuring the corresponding MRM transition (m/z 379 -305). Calibration curves (typical rX0.990) were obtained by injecting standard solutions containing variable amounts of dex (1–40 ng/ml) and a fixed amount of flumethasone (20 ng/ml).Serum samples (0.05 ml) were spiked with 1 ng of flumethasone in 0.15 ml of ethanol, vortexed and centrifuged at 3000 rpm for 10 min. An aliquot of the supernatants (0.08 ml) was diluted with water (0.14 ml) and transferred to the HPLC autosampler vials (Perkin-Elmer Series 200) for HPLC-MS/ MS injections
group injected with 9L and IL-4 RPC cells was also bearing osmotic pumps releasing 50 mg/kg/day of dex for 14 days (n¼9). All the animals were killed on day 14 after tumor injection. Day 14 was selected because it is around that time that the fate of IL-4-treated and untreated tumors begins to diverge.8,12 The presence of macrophages and endothelial cells was tested by lectin staining.13 CD4+ and CD8+ T-lymphocytes were identi-
fied with specific antibodies. Figure 2 gives an overview of the results. IL-4-treated tumors, with and without dex, were smaller than controls. Tumors with dex/IL-4 were the smallest, possibly because of the anti-angiogenic effect of dexamethazone14,15 that, in this experimental setting, is delivered when the tumor is not yet established. A great number of necrotic areas, often surrounded by macrophages, was exclusively present in IL-4-treated rats without dex (Figure 2a-d). Infiltration by CD8+ Tlymphocytes was clear in IL-4-treated rats, scarce in controls and absent in IL-4/dex rats (Figures 2e and f). CD4+ T-lymphocytes were weakly present only in IL-4treated rats (Figures 2g and h). The vessel density was decreased in IL-4-treated rats, particularly in those treated by dex (Figure 2i-l). Reports on the anti-angiogenic role of IL-4 are somewhat contradictory.16,17 A specific anti-angiogenic action was demonstrated in C6 glioblastomas18 and our results in 9L tumors confirm this observation. However, in the presence of dex, when immune responses are weakened while the anti-angiogenic action is maintained, the intracranial growth of the tumor increases, supporting the notion that the immune system is playing a major role in the therapeutic action of IL-4. Thus, biologically, our findings confirm that the anti-tumor action of IL-4 is mostly immune-mediated. Other potential inhibitory actions based on anti-proliferative19 effects of IL-4 in gliomas may be relevant but were not evaluated here. Clinically, the results clearly suggest that immuno-gene therapy strategies for gliomas should be performed when low levels of dex are administered to the patients or, even better, when the clinical situation allows one to avoid its use.
191
Acknowledgements This study was supported by research grants from the Italian Minister of Health (‘Ricerca Finalizzata’) and from the Associazione Italiana per la Ricerca sul Cancro (AIRC) to GF.
References 1 Koehler PJ. Use of corticosteroids in neuro-oncology. Anticancer Drugs 1995; 6: 19–33. 2 Ostergaard L et al. Early changes measured by magnetic resonance imaging in cerebral blood flow, blood volume, and blood–brain barrier permeability following dexamethasone treatment in patients with brain tumors. J Neurosurg 1999; 90: 300–305.
3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Figure 2 Histological analysis of rat brain gliomas developed 14 days after co-injection of 9L cells with IL-4 or control RPC. Figure 2 Hematoxylin–eosin staining of 9L tumors co-injected with control RPC (a) and IL-4 RPC (b). Necrotic areas are clearly visible in (b). Lectin histochemistry showing few macrophages in 9L-control RPC tumors (c) and numerous macrophages, mostly surrounding necrotic areas (d). Immunohistochemistry for CD8+ Tlymphocytes in a 9L-control RPC (e) and a 9L-IL-4 RPC tumor (f). Immunohistochemistry for CD4+ T-lymphocytes in a 9L-control RPC (g) and a 9L-IL-4 RPC tumor (h). Lectin histochemistry decorating vessels in 9L-IL-4 RPC (i) and 9L-IL-4-DEX tumor (l). For histological analysis rat brains were fixed in Carnoy and embedded in paraffin. Tissue sections were cut at 2 mm on a microtome and stained for histological examination. Hematoxylin and eosin staining was used for routine histological examination. For immunohistological and histochemical evaluations, deparaffined slides which underwent unmasking with buffer citrate (5 mM, pH 6) at 1001C for 6 min were rinsed in PBS–Triton 100 0.3% and incubated with 10% normal goat serum in PBS. Macrophages and blood vessels were stained using biotin-labelled BS-I isolectin B4 (Sigma, 1:50) while CD8 and CD4 T-lymphocyte subsets were stained using a mouse anti-CD8 (Cedarlane, 1:400) and a mouse anti-CD4 (Cedarlane, 1:400) antibodies revealed using biotin-labelled secondary anti-mouse antibody (Vector). Gene Therapy
Dexamethasone inhibits IL-4 action on gliomas S Benedetti et al
192 3 Heiss JD et al. Mechanism of dexamethasone suppression of brain tumor-associated vascular permeability in rats. Involvement of the glucocorticoid receptor and vascular permeability factor. J Clin Invest 1996; 98: 1400–1408. 4 Badie B et al. Dexamethasone-induced abolition of the inflammatory response in an experimental glioma model: a flow cytometry study. J Neurosurg 2000; 93: 634–639. 5 Mozo L et al. Glucocorticoids inhibit IL-4 and mitogen-induced IL-4R alpha chain expression by different posttranscriptional mechanisms. J Allergy Clin Immunol 1998; 102: 968–976. 6 Benedetti S, et al. Gene therapy of experimental brain tumors using neural progenitor cells. Nat Med 2000; 6: 447–450. 7 Benedetti S et al. Limited efficacy of the HSV-TK/GCV system for gene therapy of malignant gliomas and perspectives for the combined transduction of the interleukin-4 gene. Hum Gene Ther 1997; 8: 1345–1353. 8 Benedetti S et al. Eradication of rat malignant gliomas by retroviral-mediated, in vivo delivery of the interleukin 4 gene. Cancer Res 1999; 59: 645–652. 9 Wei MX et al. Effects on brain tumor cell proliferation by an adenovirus vector that bears the interleukin-4 gene. J Neurovirol 1998; 4: 237–241. 10 Saleh M et al. Effect of in situ retroviral interleukin-4 transfer on established intracranial tumors. J Natl Cancer Inst 1999; 91: 438– 445. 11 Okada H et al. Effective cytokine gene therapy against an intracranial glioma using a retrovirally transduced IL-4 plus HSVtk tumor vaccine. Gene Ther 1999; 6: 219–226.
Gene Therapy
12 Giezeman-Smits KM et al. Cytokine gene therapy of gliomas: induction of reactive CD4+ T cells by interleukin-4-transfected 9L gliosarcoma is essential for protective immunity. Cancer Res 2000; 60: 2449–2457. 13 Pennell NA, Streit WJ. Colonization of neural allografts by host microglial cells: relationship to graft neovascularization. Cell Transplant 1997; 6: 221–230. 14 Wolff JE, Molenkamp G, Hotfilder M, Laterra J. Dexamethasone inhibits glioma-induced formation of capillary like structures in vitro and angiogenesis in vivo. Klin Padiatr 1997; 209: 275–277. 15 Machein MR et al. Differential downregulation of vascular endothelial growth factor by dexamethasone in normoxic and hypoxic rat glioma cells. Neuropathol Appl Neurobiol 1999; 25: 104–112. 16 Volpert OV et al. Inhibition of angiogenesis by interleukin 4. J Exp Med 1998; 188: 1039–1046. 17 Fukushi J et al. Novel biological functions of interleukin-4: formation of tube-like structures by vascular endothelial cells in vitro and angiogenesis in vivo. Biochem Biophys Res Commun 1998; 250: 444–448. 18 Saleh M, Davis ID, Wilks AF. The paracrine role of tumourderived mIL-4 on tumour-associated endothelium. Int J Cancer 1997; 72: 664–672. 19 Iwasaki K, Rogers LR, Estes ML, Barna BP. Modulation of proliferation and antigen expression of a cloned human glioblastoma by interleukin-4 alone and in combination with tumor necrosis factor-alpha and/or interferon-gamma. Neurosurgery 1993; 33: 489–493; discussion 493-484.
