J Cancer Res Clin Oncol DOI 10.1007/s00432-014-1842-9

ORIGINAL ARTICLE - CANCER RESEARCH

Targeting VEGF and interleukin-6 for controlling malignant effusion of primary effusion lymphoma Hiroki Goto · Eriko Kudo · Ryusho Kariya · Manabu Taura · Harutaka Katano · Seiji Okada 

Received: 24 February 2014 / Accepted: 27 September 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract  Purpose  Primary effusion lymphoma (PEL) is an aggressive subtype of non-Hodgkin lymphoma that shows malignant effusion most commonly seen in advanced AIDS patients. In this study, we clarified the potential role of VEGF and IL-6 in PEL fluid retention and evaluated the efficacy of humanized anti-VEGF monoclonal antibody (mAb), bevacizumab, and humanized anti-IL-6 receptor mAb, tocilizumab, against PEL. Methods  The production of VEGF and IL-6, and the expression of IL-6Rα in PEL cell lines were examined. The antiproliferative effect of bevacizumab and tocilizumab on PEL cells was evaluated in vitro. The effect of tocilizumab on VEGF was also examined. An intraperitoneal xenograft mouse model was used for in vivo efficacy. Results  Although we found the production of VEGF and IL-6, and the expression of IL-6Rα in PEL cell lines, bevacizumab and tocilizumab did not inhibit the proliferation of PEL cells in vitro. Tocilizumab decreased VEGF mRNA and VEGF production by inhibiting Stat3 phosphorylation and Stat3 binding to VEGF promoter. In a PEL xenograft mouse model that showed profuse ascites, bevacizumab suppressed ascites formation completely, indicating the critical role of VEGF for PEL fluid retention. Tocilizumab also significantly inhibited ascites formation in vivo.

H. Goto · E. Kudo · R. Kariya · M. Taura · S. Okada (*)  Division of Hematopoiesis, Center for AIDS Research, Kumamoto University, 2-2-1 Honjo, Chuo-ku, Kumamoto 860-0811, Japan e-mail: [email protected] H. Katano  Department of Pathology, National Institute of Infectious Diseases, 1-23-1 Toyama, Shinjuku-ku, Tokyo 162-8640, Japan

Moreover, these mAbs improved the overall survival of treated mice. Conclusions  IL-6-VEGF axis contributed to fluid retention, and bevacizumab and tocilizumab could be effective molecular targeting therapies for PEL. Keywords  Fluid retention · IL-6 · Primary effusion lymphoma · VEGF

Introduction Primary effusion lymphoma (PEL) is an infrequent and aggressive subtype of non-Hodgkin lymphoma that occurs in an immunodeficient status, most commonly in human immunodeficiency virus (HIV)-infected patients (Greene et al. 2007), and is universally associated with infection by Kaposi’s sarcoma-associated herpes virus (KSHV)/human herpesvirus-8 (HHV-8) (Cesarman et al. 1995). PEL presents with pleural, peritoneal, or pericardial effusion without tumor masses. Malignant effusion helps tumor invasion and changes the blood concentration of anticancer agents. The accumulation of effusion is an important cause of morbidity and mortality in patients with PEL (Castillo et al. 2012). Moreover, PEL is generally resistant to conventional chemotherapy and has a poor prognosis (Boulanger et al. 2005). Thus, there is a need for novel agents targeting the characteristics of PEL, such as malignant effusion. PEL has been reported to secrete both vascular endothelial growth factor (VEGF) and interleukin-6 (IL-6) (Aoki and Tosato 1999; Drexler et al. 1999). These molecules are considered to be related to PEL progression. However, the biological activity of IL-6 signaling in PEL cells and whether VEGF and IL-6R-mediated signaling can be a therapeutic target for PEL have not been fully clarified. In addition,

13



no preclinical evaluation of the activity of humanized antiVEGF monoclonal antibody (mAb), bevacizumab, and humanized anti-IL-6 receptor (IL-6R) mAb, tocilizumab, in PEL is available. Therefore, the aim of this study was to identify the antitumor effect of both mAbs on PEL using a PEL xenotransplantation model and to clarify the potential role of VEGF and IL-6 in fluid retention. VEGF is an angiogenic factor that induces endothelial cell proliferation and angiogenesis (Ferrara 2002). Bevacizumab inhibits the angiogenesis of tumors and is now clinically used for the treatment of a variety of human cancers, including colorectal, non-small cell lung, ovarian, and metastatic renal cell carcinoma (Shih and Lindley 2006). VEGF also enhances vascular permeability and may play a role in the pathogenesis of certain ascites tumors (Senger et al. 1983; Nagy et al. 1995). IL-6 is a multifunctional cytokine that regulates immune and inflammatory responses (Kishimoto 2005). Serum IL-6 is elevated in inflammatory diseases and certain tumors and is correlated with disease progression (Trikha et al. 2003; Hong et al. 2007). A series of studies has shown that inhibition of IL-6 signaling by tocilizumab is therapeutically effective in rheumatoid arthritis, juvenile idiopathic arthritis, Castleman’s disease, Crohn’s disease, and IL-6-related malignancy, such as multiple myeloma (Tanaka et al. 2012). In this study, we investigated the potential role of VEGF and IL-6 in PEL fluid retention using specific humanized mAbs.

J Cancer Res Clin Oncol

measured by human IL-6 ELISA (e-Bioscience), following the manufacturer’s instructions. Flow cytometry Cells were stained with anti-IL-6Rα (CD126) PE (BioLegend, San Diego, CA, USA). After staining, cells were washed twice, resuspended in staining medium (PBS with 3 % FBS and 0.05 % sodium azide), and immediately analyzed on an LSR II flow cytometer (BD Bioscience, San Jose, CA, USA). Data were analyzed with FlowJo software (Tree Star, San Carlos, CA, USA). Tetrazolium dye methylthiotetrazole (MTT) assay The antiproliferative activities of bevacizumab and tocilizumab against PEL cell lines were measured by the methylthiotetrazole (MTT) method (Sigma-Aldrich, St. Louis, MO, USA). Briefly, 1 × 104 cells were incubated in triplicate in a 96-well microculture plate in the presence of different concentrations of bevacizumab or tocilizumab in a final volume of 0.1 ml for 72 h at 37 °C. Subsequently, MTT (0.5 mg/ml final concentration) was added to each well. After 3 h of additional incubation, 100 µl acidified isopropanol (HCl 34 µl/10 ml isopropanol) was added to dissolve the crystals. Absorption values at 595 nm were determined with an automatic ELISA plate reader (Multiskan; Thermo ElectronVantaa, Finland). Values were normalized to untreated (control) samples.

Materials and methods

RT-PCR analysis

Cell growth conditions

Total RNA was isolated from cells using RNAiso plus (Takara Bio, Ohtsu, Japan) according to the manufacturer’s instructions. Real-time quantitative reverse transcriptasepolymerase chain reaction (RT-PCR) analysis for human VEGF and internal control β2-microglobulin (β2 M) was carried out with SYBR Green Master Mix (Applied Biosystems, Carlsbad, CA, USA) following the manufacturer’s instructions. PCR amplifications were performed using the StepOne real-time PCR system (Applied Biosystems) with the following amplification conditions: 95 °C for 3 min, 40 cycles at 95 °C for 10 s, at 55 °C for 30 s. The Ct values for each gene amplification were normalized by subtracting the Ct value calculated for β2 M. The normalized gene expression values were expressed as the relative quantity of VEGF gene-specific messenger RNA (mRNA). The oligonucleotide primers used in this study are as shown below. VEGF-A 5′-ATGACGAGGGCCTGGAGTG TG-3′ and 5′-CCTATGTGCTGGCCTTGGTGAG-3′, β2  M 5′-CGGGCATTCCTGAAGCTGA-3′ and 5′-GGATGGA TGAAACCCAGACACATAG-3′.

BCBL-1, BC-1, BC-3, TY-1, and Raji cells were maintained in RPMI 1640 supplemented with 10 % fetal bovine serum (FBS), penicillin (100 U/ml), and streptomycin (100 µg/ml) in a humidified incubator at 37 °C and 5 % CO2. BC-2, BCP-1, and RM-P1 cells were maintained in RPMI 1640 supplemented with 20 % FBS, penicillin (100 U/ml), and streptomycin (100 µg/ml) in a humidified incubator at 37 °C and 5 % CO2. Reagents Bevacizumab and tocilizumab were obtained from Chugai Pharmaceutical Co. Ltd. (Tokyo, Japan). Measurement of VEGF and IL-6 VEGF was measured by human VEGF-A Platinum ELISA (e-Bioscience, San Diego, CA, USA), and IL-6 was

13

J Cancer Res Clin Oncol

a PEL

BL

b BCBL-1

BCBL-1

BC-1 BC-2 BC-3

BC-1 BC-2 BC-3

PEL

BCP-1

BCP-1

RM-P1 TY-1 Raji

RM-P1 TY-1 Raji

BL 0

1000

2000

3000

4000

0

250

c

d IL-6Rα relative MFl

BC-1 BC-2 BC-3 BCP-1

BL

RM-P1 TY-1 Raji 1.0

1.5

2.0

2.5

5000

4

BCBL-1 PEL

500

10000

IL-6 (pg/ml)

VEGF (pg/ml)

3.0

3.5

IL-6Rα relative MFI

R2=0.78 P < 0.05

3

2

1

0

2500

5000

7500

10000

IL-6 (pg/ml)

Fig. 1  PEL cell lines produce VEGF and IL-6, and express IL-6Rα. a Production of VEGF in PEL cell lines. b Production of IL-6 in PEL cell lines. Levels of VEGF and IL-6 in the culture supernatants were determined by ELISA. Results are the means of three independent experiments. c Expression of IL-6Rα on the surface of PEL cell lines.

Numbers of IL-6Rα expression indicate the fold increase in the mean fluorescence intensity (MFI) by flow cytometry. Raji cells were used as a negative control. d Inverse correlation between IL-6 production and IL-6Rα expression in PEL cell lines

Western blot analysis

Chromatin immunoprecipitation (ChIP) assay

For whole cell extraction, BCBL-1 cells treated with 10 µM tocilizumab for 48 h were collected and washed in cold PBS before the addition of 100 µl cold lysis buffer (25 mM HEPES, 10 mM Na4P2O7·10H2O, 100 mM NaF, 5 mM EDTA, 2 mM Na3VO4, 1 % Triton X-100). After rotation for 2 h at 4 °C, whole cell extracts were obtained by centrifugation at 15 000 rpm for 15 min. Whole cell extracts (30 µg protein) were separated by 10 % SDS-PAGE and blotted onto a PVDF membrane (GE Healthcare, Tokyo, Japan). Blots were probed with the indicated antibodies and detected using Chemi-Lumi One Super (Nacalai Tesque, Kyoto, Japan). Primary antibodies were as follows: anti-IκBα (9242), anti-phospho(Ser32/36)-IκBα (9246), anti-Erk (4695), anti-phospho(Thr202/Tyr203)-Erk (4370) (Cell Signaling Technology, Danvers, MA, USA), anti- Stat3 (sc8019), anti-phospho(Tyr705)-Stat3 (sc-7993-R) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and antiHsc70 (SPA-815) (Stressgen Bioreagents, Ann Arbor, MI, USA).

To examine the binding of Stat3 to VEGF promoter in PEL cells, nuclear extracts of BCBL-1 cells were used for the ChIP assay. BCBL-1 cells were cross-linked using formaldehyde (1 % final concentration) added directly to the cell culture media at 37 °C for 15 min, and the reaction was stopped by adding glycine (0.125 M final concentration). Cells were rinsed with cold PBS and resuspended in cell lysis buffer, consisting of 5 mM PIPES [piperazine-N,N′bis(ethanesulfonic acid), pH 8.0], 85 mM KCl, 0.5 % NP-40, and 1 % protease inhibitor (PI) cocktail (Nacalai Tesque). This mixture was incubated on ice for 10 min and then homogenized. The nuclei were resuspended in nucleus lysis buffer [1 % sodium dodecyl sulfate (SDS), 10 mM EDTA, 50 mM Tris–HCl (pH 8.1), and 1 % PI cocktail] and incubated on ice for 10 min. The samples were sonicated on ice with the ultrasonic homogenizer VP-050 (TAITEC, Saitama, Japan) to an average length of up to 1,000 bp and microcentrifuged. The chromatin solution was precleared using Staphylococcus aureus protein A-positive cells (Pansorbin, #507862; Calbiochem). Precleared chromatin was incubated with 2 μg anti-Stat3 antibody (sc-8019) (Santa

13



J Cancer Res Clin Oncol

The mice were then treated with intraperitoneal injections of PBS, bevacizumab, or tocilizumab (100 µg/mouse, 3 times a week). Tumor burden was evaluated by measuring the volume of ascites on day 28. For assessment of overall survival, Kaplan–Meier analysis was performed and P values were determined by two-tailed analysis with the logrank test. Immunohistochemistry

Fig. 2  Bevacizumab (Bv) or tocilizumab (Toc) does not inhibit the proliferation of PEL cells. PEL cell lines (BCBL-1, BC-2, and BCP1) were incubated with 0, 0.3, 1, 3, 10 µM Bv or Toc for 72 h. A cell proliferation assay was carried out using MTT as described in “Materials and methods” section

Cruz Biotechnology) or mouse IgG at 4 °C for 24 h and microcentrifuged. After washing, we performed cross-link reversal and DNA extraction. PCR was performed using Ex Taq polymerase (Takara Bio) according to the recommended protocol. The primers recognized the putative Stat-binding sites located at −848 and −630 in the human VEGF promoter region (accession no. AF095785). The oligonucleotide primers used in this study were VEGF promoter 5′-TTGGTGCCAAATTCTTCTCC-3′ and 5′-CACACGTCCTCACTCTCGAA-3′ (Cheranov et al. 2008).

To investigate the expression of KSHV/HHV-8 ORF73 (LANA) protein, tissue samples were fixed with 10 % neutral-buffered formalin, embedded in paraffin, and cut into 4-µm sections. The sections were deparaffinized by sequential immersion in xylene and ethanol, and rehydrated in distilled water. They were then irradiated for 15 min in a microwave oven for antigen retrieval. Endogenous peroxidase activity was blocked by immersing the sections in methanol/0.6 % H2O2 for 30 min at room temperature. Affinity-purified PA1-73 N antibody, diluted 1:3000 in PBS/5 % bovine serum albumin (BSA), was then applied, and the sections were incubated overnight at 4 °C. After washing in PBS twice, the second and third reactions and the amplification procedure were performed using kits according to the manufacturer’s instructions (catalyzed signal amplification system kit; DAKO, Copenhagen, Denmark). The signal was visualized using 0.2 mg/ml diaminobenzidine and 0.015 % H2O2 in 0.05 mol/l Tris–HCl, pH 7.6. Statistical analysis Data are expressed as the mean ± SD. The statistical significance of the differences observed between experimental groups was determined using Student’s t test, and P < 0.05 was considered significant.

Results Production of VEGF and IL-6 in PEL cell lines

Xenograft mouse model NOD Rag-2/Jak3 double-deficient (Rag-2−/−Jak3−/−) mice (NRJ mice) were established as described previously (Goto et al. 2012) and were housed and monitored in our animal research facility according to institutional guidelines. All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee at Kumamoto University. In an intraperitoneal xenograft mouse model, 10- to 12-week-old NRJ male mice were intraperitoneally inoculated with 7 × 106 BCBL-1 cells suspended in 200 µl phosphate-buffered saline (PBS).

13

Production of VEGF and IL-6 in culture supernatants of seven PEL cell lines (BCBL-1, BC-1, BC-2, BC-3, BCP-1, RM-P1, TY-1) and the Burkitt lymphoma cell line Raji was analyzed. After cell lines (5 × 105/ml) had been cultured for 48 h, the production of VEGF and IL-6 was quantified by enzyme-linked immunosorbent assay (ELISA). High levels of VEGF and IL-6 were detected in PEL cell lines (means of 489.7–2,861.7 pg/ ml and 1.4–8,777.7 pg/ml, respectively) compared with Raji (means of 504.6 pg/ml and 0.2 pg/ml, respectively) (Fig. 1a, b).

J Cancer Res Clin Oncol 1.47

1.6

1.2

1.00

1.6

0.94 0.73

0.8

0.4

Relative VEGF mRNA

a

Relative VEGF mRNA

Fig. 3  Tocilizumab (Toc) inhibits VEGF in BCBL-1 and BC-2. a BCBL-1 cells or BC-2 cells were treated with 10 µM Toc for 72 h in the absence or presence of 100 ng/ml IL-6. VEGF mRNA expression level was measured by quantitative RT-PCR. b 5 × 104/ml BCBL-1 cells or BC-2 cells were treated with 10 µM Toc and cultured for 72 h in the absence or presence of 100 ng/ml IL-6. Production of VEGF was measured by ELISA

0

1.2

1.16 0.97

1.00 0.75

0.8

0.4

0 Ctrl

Toc

IL-6

Ctrl

Toc+IL-6

Toc

Toc+IL-6

IL-6

Toc+IL-6

BC-2

BCBL-1

1250

1000

1000

VEGF (pg/ml)

b 1250

VEGF (pg/ml)

IL-6

750

500

250

750

500

250

0

0 Ctrl

Toc

IL-6

BCBL-1

Expression of IL-6Rα on the surface of PEL cell lines We examined PEL cell surface expression of IL-6Rα. As a negative control in flow cytometry analysis, Raji cells were used. As shown in Fig. 1c, PEL cell lines expressed IL-6Rα. Although the production of IL-6 was low in some PEL cell lines such as RM-P1 (1.4 pg/ml) (Fig. 1b) and the level of IL-6Rα expression varied among the cell lines, lower production of IL-6 was significantly correlated with higher IL-6Rα expression (R2 = 0.78, P < 0.05) (Fig. 1d). Some PEL cell lines may compensate for the low production of IL-6 by the high expression of IL-6R. Direct anti-proliferative effect of bevacizumab or tocilizumab on PEL cells We determined whether treatment with bevacizumab or tocilizumab leads to the inhibition of PEL cell proliferation using the MTT assay. Three PEL cell lines (BCBL-1, BC-2, BCP-1) were cultured with varying concentrations of

Toc+IL-6

Ctrl

Toc

BC-2

bevacizumab or tocilizumab (0, 0.3, 1, 3, 10 µM) for 72 h, and proliferation was analyzed by the MTT assay. Figure 2 shows that bevacizumab or tocilizumab did not inhibit the proliferation of PEL cells significantly. Tocilizumab reduces the level of VEGF in PEL cells To investigate the mechanism underlying the effect of tocilizumab on PEL, we assessed VEGF levels by quantitative RT-PCR and ELISA. As shown in Fig. 3a, tocilizumab reduced VEGF mRNA of BCBL-1 and BC-2 in the absence or presence of IL-6. The production of VEGF was also decreased by tocilizumab treatment (Fig. 3b). Tocilizumab suppresses IL-6-induced Stat3 activity and Stat3 binding to VEGF promoter in PEL cells Signal transduction through IL-6R is mediated by the JAK/STAT (Janus family tyrosine kinase/signal transducer and activator of transcription) and MAPK/ERK

13



J Cancer Res Clin Oncol

a

Ctrl

Toc

IL-6

IL-6 + Toc

pStat3

Stat3 pErk1/2 Erk1/2

In vivo effect of bevacizumab or tocilizumab in severely immunodeficient mice inoculated intraperitoneally with BCBL-1

pIκB IκB

Hsc70

b

Ctrl

Toc

IL-6

IL-6 + Toc

IP

Stat3

IgG

Input

Fig. 4  Tocilizumab (Toc) inhibits IL-6-induced Stat3 phosphorylation and Stat3 binding to VEGF promoter. a Inhibitory effects of Toc on IL-6-induced Stat3 phosphorylation. BCBL-1 cells were treated with 10 µM Toc for 48 h in the absence or presence of 100 ng/ml IL-6. Total proteins were extracted for Western blotting. b Suppression of Stat3 recruitment to VEGF promoter by Toc. BCBL-1 cells were treated with 10 µM Toc for 48 h in the absence or presence of 100 ng/ml IL-6. Stat3 binding on the VEGF promoter was determined by the ChIP assay using nuclear extracts of BCBL-1 cells

(mitogen-activated protein kinase/extracellular-regulated kinase) pathways (Akira et al. 1990, 1994; Heinrich et al. 2003). Stat3, a member of the JAK/STAT pathway, and nuclear factor (NF)-κB are constitutively activated and related to the pathogenesis of PEL cells (Aoki et al. 2003; Keller et al. 2000); therefore, we examined whether tocilizumab inhibited the phosphorylated form of Stat3, Erk, and IκB. When BCBL-1 cells were treated with 10 µM tocilizumab for 48 h, treatment with tocilizumab reduced IL-6-induced Stat3 phosphorylation, whereas phosphoErk and phospho-IκB were not significantly changed (Fig.  4a), suggesting that IL-6 maintains Stat3 activation and exerts a beneficial effect on PEL cells, including the

13

production of VEGF. Next, we determined Stat3 binding to the VEGF promoter by the ChIP assay using nuclear extracts of BCBL-1 cells. When BCBL-1 cells were treated with 10 µM tocilizumab for 48 h, tocilizumab suppressed IL-6-induced Stat 3 binding to VEGF promoter in BCBL-1 cells (Fig. 4b). These results show that IL-6 modulates transcription factor Stat3, which directly binds to the promoter of VEGF, resulting in increased VEGF transcription in PEL cells.

Although bevacizumab or tocilizumab did not show direct antiproliferative effects of both mAbs on PEL, we assessed the in vivo effects of both mAbs using PEL xenograft NOD/Rag-2/Jak3-deficient (NRJ) mice (Goto et al. 2012). NRJ mice display not only complete deficiency in mature T/B lymphocytes and complement protein but also complete deficiency of NK cells, providing efficient engraftment of PEL cells. BCBL-1 cells (7 × 106/mouse) were inoculated intraperitoneally into NRJ mice. BCBL-1 xenograft mice showed profuse ascites within 4 weeks. As patients with PEL show lymphomatous effusion in body cavities without a definable tumor mass, these mice could be clinically equivalent to the PEL model. A dose of 100 µg/mouse bevacizumab, tocilizumab, or PBS alone was administrated via intraperitoneal injection on day 3 after cell inoculation and 3 times a week. Bevacizumabor tocilizumab-treated mice apparently seemed to stay healthy, and the body weight did not change, whereas the volume of ascites was significantly lower than in untreated mice on day 28 (0.0 ± 0.0, 0.4 ± 0.7, 2.6 ± 1.0 ml, respectively, n = 7, P < 0.001; Fig. 5a, b). Organ invasion by PEL cells on day 28 was evaluated by hematoxylin– eosin staining and LANA immunostaining. We found that mice inoculated intraperitoneally with BCBL-1 exhibited invasion into the liver and lungs without macroscopic lymphoma formation (Fig. 5c). The number of LANA-positive cells in treated mice was significantly reduced (0–1 cells per field magnification, 40×) compared with untreated mice (10–20 cells per field magnification, 40×). As shown in Fig. 6, treatment with bevacizumab or tocilizumab significantly prolonged the survival of the mice (48.4 ± 6.9 and 45.5 ± 7.7 days, respectively) compared with PBS (32.6 ± 3.8 days) (log-rank test, P < 0.01). Inhibiting the effect of human VEGF on mouse cells was considered to suppress vascular permeability. These results indicate that treatment with bevacizumab or tocilizumab inhibits the development of malignant effusion and provides a survival benefit. These mAbs could be potentially therapeutic agents in patients with PEL.

J Cancer Res Clin Oncol

a

b

5

*** ***

Ascites volume (ml)

4

3

2

1

PBS

Bv

Toc

0

PBS (n =7)

c

Lungs

Liver HE

Bv Toc (n =7) (n =7)

LANA

HE

LANA

PBS

Bv

Toc

Fig. 5  Treatment of NOD/Rag-2/Jak3-deficient mice with bevacizumab (Bv) or tocilizumab (Toc) suppressed the development of PEL in vivo. a Photograph of untreated and treated ascites-bearing mice 4 weeks after inoculation with BCBL-1 intraperitoneally. b The volume of ascites 4 weeks after inoculation with BCBL-1 cells in mice

is shown as the mean ± SD of 7 mice. ***P < 0.001 when compared with ascites volume. c Invasion of PEL cells into the organs of BCBL-1-inoculated mice on day 28. Hematoxylin–eosin staining and immunohistochemical staining using anti-LANA (PA1-73 N antibody) were performed to detect BCBL-1 in the liver and lungs

Discussion

therapeutic strategies for PEL such as inhibition of NF-κB (Keller et al. 2000), activating TRAIL-mediated apoptosis by IFN-α and azidothymidine (Toomey et al. 2001; Wu et al. 2005), and inducing lytic replication of HHV-8

PEL is a highly aggressive lymphoma that is resistant to conventional chemotherapy. Recent studies proposed new

13



J Cancer Res Clin Oncol 100

PBS (n =17) Bv (n =17)

Survival (%)

80

Toc (n =17)

60 40 20 0

0

10

20

30

40

50

Days after inoculation

Fig. 6  Overall survival curve. Treatment with bevacizumab (Bv) or tocilizumab (Toc) prolonged survival in vivo

concomitantly with blocking virus production (Klass et al. 2005). These strategies are considered to be effective, but there is no proven standard therapy targeting specific molecules that are related to PEL pathogenesis. PEL has a unique clinical presentation with malignant effusion, causing treatment difficulty; therefore, targeting malignant effusion is a reasonable strategy in the treatment of PEL. We evaluated the efficacy of bevacizumab and tocilizumab in terms of controlling fluid retention. Although a direct antiproliferative effect of bevacizumab or tocilizumab on PEL cells was not observed in vitro (Fig. 2), both mAbs significantly suppressed in vivo ascites formation in a PEL mouse model. Treatment with mouse anti-human VEGF mAb has been reported to inhibit the development of ascites in SCID/beige mice inoculated intraperitoneally with PEL cells (Aoki and Tosato 1999). Our study evaluated the therapeutic effect of a humanized anti-VEGF mAb, bevacizumab, on PEL xenograft NRJ mice by not only the volume of ascites, but also the efficacy for organ invasion and overall survival. Furthermore, we assessed the effect of anti-IL-6 receptor mAb, tocilizumab, on PEL in vitro and in vivo for the first time. PEL cells produce VEGF and IL-6, and express IL-6Rα (Fig. 1); however, the direct anti-proliferative effect of bevacizumab or tocilizumab on PEL cells was not observed in vitro (Fig. 2). These results demonstrated that VEGF and IL-6 are not critical growth factors but other pathogenic factors in PEL cells. In vivo efficacy of bevacizumab and tocilizumab indicates the potential role of VEGF and IL-6 for fluid retention. IL-6 signaling is characterized by the binding of mammalian forms of IL-6 to membrane-bound IL-6R. The IL-6/ IL-6R complex binds and activates gp130, leading to downstream activation of signaling pathways such as the JAK/ STAT and MAPK/ERK pathways. Unlike normal cells, constitutively activated Stat3 is detected in a wide variety of human cancer cells, including PEL (Aoki et al. 2003; Al Zaid Siddiquee and Turkson 2008). As depicted in Fig. 3,

13

tocilizumab decreased VEGF mRNA and IL-6-induced VEGF production. Although Jak2 inhibitor AG490 directly suppressed Stat3 phosphorylation and induced apoptosis in PEL cells (Aoki et al. 2003), tocilizumab inhibited IL-6-mediated Stat3 phosphorylation (Fig. 4a) and Stat3 binding to VEGF promoter (Fig. 4b), inducing no growth inhibition (Fig. 2). We showed that IL-6 increased VEGF via additional Stat3 phosphorylation and Stat3 binding to VEGF promoter in PEL cells. The mechanism of tumor development in AIDS patients is a multistep and multifactorial process. Although the HIV-induced immunocompromised status is obviously involved in the development of PEL, cytokines may also contribute to the pathogenesis. Since the production of IL-6 is induced by HIV (Nakajima et al. 1989; Birx et al. 1990; Scala et al. 1994) and IL-6 increases in the plasma of HIV patients (Breen et al. 1990; Rieckmann et al. 1991), co-infection with HIV is considered to contribute to the pathogenesis of PEL, at least via the production of IL-6. HHV-8-infected cells secrete not only human IL-6 (hIL6) but also viral IL-6 (vIL-6). In contrast to hIL-6, vIL-6 does not require hIL-6R for receptor complex formation and signaling initiation (Molden et al. 1997; Osborne et al. 1999; Mullberg et al. 2000). vIL-6 has been also reported to promote VEGF secretion (Aoki et al. 1999); however, vIL-6 is mainly expressed not in latently infected cells but in the lytic lifecycle of HHV-8 infection (Nicholas et al. 1998), and the affinity of vIL-6 to gp130 is one thousand times lower than that of human IL-6 (Aoki et al. 2001). In addition, vIL-6 has been shown to cause the pathogenesis by inducing endogenous IL-6 expression in cell lines from patients with multicentric Castleman’s disease (MCD) (Mori et al. 2000) and in transgenic (Tg) mice that constitutively express vIL-6 under control of the MHC class promoter (Suthaus et al. 2012). The production of endogenous IL-6 but not vIL-6 is largely required for the development of the MCD-like phenotype in Tg mice (Suthaus et al. 2012). Taken together, endogenous IL-6 plays an important role in HHV-8-associated diseases and is considered to be a promising therapeutic target, even in the presence of vIL-6. In conclusion, we have shown the potent efficacy of bevacizumab and tocilizumab against PEL. Although inhibitory effects of tocilizumab on a PEL mouse model other than suppressing VEGF are expected because inhibition of VEGF was not complete in vitro, inhibition of IL-6R could be a promising therapeutic strategy for PEL from its in vivo effectiveness. Our data provide new insights into controlling fluid retention in PEL and the rationale for a clinical study in a single agent or in combination with conventional chemotherapy. Acknowledgments  We thank Ms. I. Suzu and Ms. S. Fujikawa for technical assistance and Ms. Y. Endo for secretarial assistance.

J Cancer Res Clin Oncol This work was supported by a Health and Labour Sciences Research Grant from the Ministry of Health, Labour, and Welfare of Japan (H25-AIDS-I-002), the Global COE program, “Global Education and Research Center Aiming at the Control of AIDS,” and Grants-in-Aid for Science Research (No. 25114711) from the Ministry of Education, Science, Sports, and Culture of Japan. Conflict of interest  The authors have declared that no conflict of interest exists.

References Akira S, Isshiki H, Sugita T, Tanabe O, Kinoshita S, Nishio Y, Nakajima T, Hirano T, Kishimoto T (1990) A nuclear factor for IL-6 expression (NF-IL6) is a member of a C/EBP family. EMBO J 9(6):1897–1906 Akira S, Nishio Y, Inoue M, Wang XJ, Wei S, Matsusaka T, Yoshida K, Sudo T, Naruto M, Kishimoto T (1994) Molecular cloning of APRF, a novel IFN-stimulated gene factor 3 p91-related transcription factor involved in the gp130-mediated signaling pathway. Cell 77(1):63–71 Al Zaid Siddiquee K, Turkson J (2008) STAT3 as a target for inducing apoptosis in solid and hematological tumors. Cell Res 18(2):254– 267. doi:10.1038/cr.2008.18 Aoki Y, Tosato G (1999) Role of vascular endothelial growth factor/ vascular permeability factor in the pathogenesis of Kaposi’s sarcoma-associated herpesvirus-infected primary effusion lymphomas. Blood 94(12):4247–4254 Aoki Y, Jaffe ES, Chang Y, Jones K, Teruya-Feldstein J, Moore PS, Tosato G (1999) Angiogenesis and hematopoiesis induced by Kaposi’s sarcoma-associated herpesvirus-encoded interleukin-6. Blood 93(12):4034–4043 Aoki Y, Narazaki M, Kishimoto T, Tosato G (2001) Receptor engagement by viral interleukin-6 encoded by Kaposi sarcoma-associated herpesvirus. Blood 98(10):3042–3049 Aoki Y, Feldman GM, Tosato G (2003) Inhibition of STAT3 signaling induces apoptosis and decreases survivin expression in primary effusion lymphoma. Blood 101(4):1535–1542. doi:10.1182/ blood-2002-07-2130 Birx DL, Redfield RR, Tencer K, Fowler A, Burke DS, Tosato G (1990) Induction of interleukin-6 during human immunodeficiency virus infection. Blood 76(11):2303–2310 Boulanger E, Gerard L, Gabarre J, Molina JM, Rapp C, Abino JF, Cadranel J, Chevret S, Oksenhendler E (2005) Prognostic factors and outcome of human herpesvirus 8-associated primary effusion lymphoma in patients with AIDS. J Clin Oncol 23(19):4372– 4380. doi:10.1200/JCO.2005.07.084 Breen EC, Rezai AR, Nakajima K, Beall GN, Mitsuyasu RT, Hirano T, Kishimoto T, Martinez-Maza O (1990) Infection with HIV is associated with elevated IL-6 levels and production. J Immunol 144(2):480–484 Castillo JJ, Shum H, Lahijani M, Winer ES, Butera JN (2012) Prognosis in primary effusion lymphoma is associated with the number of body cavities involved. Leuk Lymphoma 53(12):2378–2382. doi:10.3109/10428194.2012.694075 Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM (1995) Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med 332(18):1186–1191. doi:10.1056/NEJM199505043321802 Cheranov SY, Karpurapu M, Wang D, Zhang B, Venema RC, Rao GN (2008) An essential role for SRC-activated STAT-3 in 14, 15-EET-induced VEGF expression and angiogenesis. Blood 111(12):5581–5591. doi:10.1182/blood-2007-11-126680

Drexler HG, Meyer C, Gaidano G, Carbone A (1999) Constitutive cytokine production by primary effusion (body cavity-based) lymphoma-derived cell lines. Leukemia 13(4):634–640 Ferrara N (2002) VEGF and the quest for tumour angiogenesis factors. Nat Rev Cancer 2(10):795–803. doi:10.1038/nrc909 Goto H, Kariya R, Shimamoto M, Kudo E, Taura M, Katano H, Okada S (2012) Antitumor effect of berberine against primary effusion lymphoma via inhibition of NF-kappaB pathway. Cancer Sci 103(4):775–781. doi:10.1111/j.1349-7006.2012.02212.x Greene W, Kuhne K, Ye F, Chen J, Zhou F, Lei X, Gao SJ (2007) Molecular biology of KSHV in relation to AIDS-associated oncogenesis. Cancer Treat Res 133:69–127 Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, Schaper F (2003) Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J 374(Pt 1):1–20. doi:10.1042/BJ20030407 Hong DS, Angelo LS, Kurzrock R (2007) Interleukin-6 and its receptor in cancer: implications for translational therapeutics. Cancer 110(9):1911–1928. doi:10.1002/cncr.22999 Keller SA, Schattner EJ, Cesarman E (2000) Inhibition of NF-kappaB induces apoptosis of KSHV-infected primary effusion lymphoma cells. Blood 96(7):2537–2542 Kishimoto T (2005) Interleukin-6: from basic science to medicine–40 years in immunology. Annu Rev Immunol 23:1–21. doi:10.1146/annurev.immunol.23.021704.115806 Klass CM, Krug LT, Pozharskaya VP, Offermann MK (2005) The targeting of primary effusion lymphoma cells for apoptosis by inducing lytic replication of human herpesvirus 8 while blocking virus production. Blood 105(10):4028–4034. doi:10.1182/ blood-2004-09-3569 Molden J, Chang Y, You Y, Moore PS, Goldsmith MA (1997) A Kaposi’s sarcoma-associated herpesvirus-encoded cytokine homolog (vIL-6) activates signaling through the shared gp130 receptor subunit. J Biol Chem 272(31):19625–19631 Mori Y, Nishimoto N, Ohno M, Inagi R, Dhepakson P, Amou K, Yoshizaki K, Yamanishi K (2000) Human herpesvirus 8-encoded interleukin-6 homologue (viral IL-6) induces endogenous human IL-6 secretion. J Med Virol 61(3):332–335 Mullberg J, Geib T, Jostock T, Hoischen SH, Vollmer P, Voltz N, Heinz D, Galle PR, Klouche M, Rose-John S (2000) IL-6 receptor independent stimulation of human gp130 by viral IL-6. J Immunol 164(9):4672–4677 Nagy JA, Meyers MS, Masse EM, Herzberg KT, Dvorak HF (1995) Pathogenesis of ascites tumor growth: fibrinogen influx and fibrin accumulation in tissues lining the peritoneal cavity. Cancer Res 55(2):369–375 Nakajima K, Martinez-Maza O, Hirano T, Breen EC, Nishanian PG, Salazar-Gonzalez JF, Fahey JL, Kishimoto T (1989) Induction of IL-6 (B cell stimulatory factor-2/IFN-beta 2) production by HIV. J Immunol 142(2):531–536 Nicholas J, Zong JC, Alcendor DJ, Ciufo DM, Poole LJ, Sarisky RT, Chiou CJ, Zhang X, Wan X, Guo HG, Reitz MS, Hayward GS (1998) Novel organizational features, captured cellular genes, and strain variability within the genome of KSHV/HHV8. J Natl Cancer Inst Monogr 23:79–88 Osborne J, Moore PS, Chang Y (1999) KSHV-encoded viral IL-6 activates multiple human IL-6 signaling pathways. Hum Immunol 60(10):921–927 Rieckmann P, Poli G, Kehrl JH, Fauci AS (1991) Activated B lymphocytes from human immunodeficiency virus-infected individuals induce virus expression in infected T cells and a promonocytic cell line, U1. J Exp Med 173(1):1–5 Scala G, Ruocco MR, Ambrosino C, Mallardo M, Giordano V, Baldassarre F, Dragonetti E, Quinto I, Venuta S (1994) The expression of the interleukin 6 gene is induced by the human immunodeficiency virus 1 TAT protein. J Exp Med 179(3):961–971

13

Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS, Dvorak HF (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science 219(4587):983–985 Shih T, Lindley C (2006) Bevacizumab: an angiogenesis inhibitor for the treatment of solid malignancies. Clin Ther 28(11):1779–1802. doi:10.1016/j.clinthera.2006.11.015 Suthaus J, Stuhlmann-Laeisz C, Tompkins VS, Rosean TR, Klapper W, Tosato G, Janz S, Scheller J, Rose-John S (2012) HHV-8-encoded viral IL-6 collaborates with mouse IL-6 in the development of multicentric Castleman disease in mice. Blood 119(22):5173– 5181. doi:10.1182/blood-2011-09-377705 Tanaka T, Narazaki M, Kishimoto T (2012) Therapeutic targeting of the interleukin-6 receptor. Annu Rev Pharmacol Toxicol 52:199– 219. doi:10.1146/annurev-pharmtox-010611-134715

13

J Cancer Res Clin Oncol Toomey NL, Deyev VV, Wood C, Boise LH, Scott D, Liu LH, Cabral L, Podack ER, Barber GN, Harrington WJ Jr (2001) Induction of a TRAIL-mediated suicide program by interferon alpha in primary effusion lymphoma. Oncogene 20(48):7029–7040. doi:10.1038/sj.onc.1204895 Trikha M, Corringham R, Klein B, Rossi JF (2003) Targeted antiinterleukin-6 monoclonal antibody therapy for cancer: a review of the rationale and clinical evidence. Clin Cancer Res 9(13):4653–4665 Wu W, Rochford R, Toomey L, Harrington W Jr, Feuer G (2005) Inhibition of HHV-8/KSHV infected primary effusion lymphomas in NOD/SCID mice by azidothymidine and interferon-alpha. Leuk Res 29(5):545–555. doi:10.1016/j.leukres.2004.11.010