Soluble OX40L favors tumor rejection in CT26 colon carcinoma model

Cytokine 84 (2016) 10–16

Contents lists available at ScienceDirect

Cytokine journal homepage: www.journals.elsevier.com/cytokine

Soluble OX40L favors tumor rejection in CT26 colon carcinoma model Ekaterina O. Serebrovskaya a,b,⇑, Diana V. Yuzhakova a,e, Alina P. Ryumina a,b, Irina N. Druzhkova a, George V. Sharonov c, Alexey A. Kotlobay b, Elena V. Zagaynova a, Sergey A. Lukyanov b,d, Marina V. Shirmanova a a

Nizhny Novgorod State Medical Academy, 603005 Minin and Pozharsky Sq., 10/1, Nizhny Novgorod, Russia Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry RAS, 117997, Miklukho-Maklaya St., 16/10, Moscow, Russia Moscow State University, Faculty of Medicine, 119192, Lomonosovsky pr., 31/5, Moscow, Russia d Pirogov Russian National Research Medical University, 117997, Ostrovityanova st., 1, Moscow, Russia e Lobachevsky State University of Nizhny Novgorod, 603950 Gagarina Ave., 23, Nizhny Novgorod, Russia b c

a r t i c l e

i n f o

Article history: Received 10 August 2015 Received in revised form 6 April 2016 Accepted 2 May 2016

Keywords: Soluble OX40L CT26 colon carcinoma Tumorigenicity Antitumor immune response Immunological memory

a b s t r a c t OX40 receptor-expressing regulatory T cells (Tregs) populate tumors and suppress a variety of immune cells, posing a major obstacle for cancer immunotherapy. Different ways to functionally inactivate Tregs by triggering OX40 receptor have been suggested, including anti-OX40 antibodies and Fc:OX40L fusion proteins. To investigate whether the soluble extracellular domain of OX40L (OX40Lexo) is sufficient to enhance antitumor immune response, we generated an OX40Lexo-expressing CT26 colon carcinoma cell line and studied its tumorigenicity in immunocompetent BALB/c and T cell deficient nu/nu mice. We found that soluble OX40L expressed in CT26 colon carcinoma favors the induction of an antitumor response which is not limited just to cells co-expressing EGFP as an antigenic determinant, but also eliminates CT26 cells expressing another fluorescent protein, KillerRed. Tumor rejection required the presence of T lymphocytes, as indicated by the unhampered tumor growth in nu/nu mice. Subsequent re-challenge of tumor-free BALB/c mice with CT26 EGFP cells resulted in no tumor growth, which is indicative of the formation of immunological memory. Adoptive transfer of splenocytes from mice that successfully rejected CT26 OX40Lexo EGFP tumors to naïve mice conferred 100% resistance to subsequent challenge with the CT26 EGFP tumor. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction OX40 is a member of the TNFR family of receptors. It has a costimulatory function in T cell activation [8,33]. OX40 is preferentially expressed on activated CD4 rather than CD8 T cells, suggesting a more prominent role for OX40 on CD4 T cells [31,35]. The exception is the situation of strong antigenic stimulation, when OX40 may also be expressed on CD8 T cells [4]. Importantly, OX40 is constitutively expressed on murine FoxP3 CD4 naturally occurring T-regulatory cells (nTreg), and is inducible on human Treg cells [29]. The OX40 ligand (OX40L) in turn is expressed on various cell types, including B cells [22], T cell [18] and activated

Abbreviations: OX40Lexo, extracellular domain of the OX40 ligand, 50–198 aa; OX40Lfull, the complete OX40 ligand. ⇑ Corresponding author at: Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry RAS, 117997, Miklukho-Maklaya St., 16/10, Moscow, Russia. E-mail address: [email protected] (E.O. Serebrovskaya). http://dx.doi.org/10.1016/j.cyto.2016.05.005 1043-4666/Ó 2016 Elsevier Ltd. All rights reserved.

DCs [8,25]. Signaling through the OX40 receptor activates the expression of pro-survival genes and is essential for the long-term survival of CD4 + T cells, however, the OX40L signal alone, without B7:CD28 interaction, is insufficient for full T cell activation [26]. Tregs are considered a major obstacle for the development of an effective antitumor immune response [38]. They are capable of direct contact inhibition of antigen-presenting and effector cells [7]. They also release anti-inflammatory cytokines such as IL-10 or transforming growth factor-b (TGF-b) [32]. Treg cells specific for a particular T cell receptor (TCR) have the ability to suppress several effector cells with distinct TCR specificities when colocalized on the same antigen-presenting cell [9,17,30]. Thymic development of nTreg cells does not require OX40, as the spleens of OX40-/- mice have nTreg cells, which are present at a reduced frequency suggesting that OX40 has a role in nTreg cell homeostasis [29]. In addition to nTreg cells, naïve T cells can become induced Treg cells (iTreg) when activated in the presence of transforming growth factor-beta (TGF-b) [37].

E.O. Serebrovskaya et al. / Cytokine 84 (2016) 10–16

In the development of antitumor immune response, OX40L appears to be a double-edged sword rather than a magic bullet. The OX40 signal antagonizes the TGF-b- and antigen-mediated conversion of naïve CD25
Foxp3
CD4 T cells into CD25+Foxp3+ CD4 T cells [28]. The authors showed that in vitro stimulation with the anti-OX40 agonistic antibody OX86 can suppress Foxp3 expression in naïve CD4 T cells and antagonize the CD28 and IL2R signals that drive the expression of Foxp3. However, other evidence suggests that OX40 stimulation drives all T cell lineages and may result in Treg accumulation in the context of cytokine blockade [27]. There are numerous examples of OX40 engagement resulting in the inhibition of tumor growth. Injection of Fc:OX40L fusion protein or anti-OX40R in vivo during the initial stages of tumor growth results in a significant increase in the percentage of tumor-free survivors in different murine tumor models [19,34]. It has been shown by Bansal-Pakala et al. that a single injection of anti-OX40 antibody can break peptide-induced peripheral tolerance [3]. A subcutaneous CT26 tumor expressing both GM-CSF for APC stimulation and OX40L regressed in 85% of the mice tested [13,14]. Intraperitoneal injection of the mOX40L-mIgGFc fusion protein 3 days after inoculation with the CT26 cells almost completely inhibited tumor growth [1]. In T cell proliferation assay, this fusion protein was significantly superior to OX86 antibody. Human recombinant Fc: OX40L fusion protein linked via a coiled-coil trimerization domain had superior capacity to stimulate human T-cell proliferation compared to the agonist antibody [20]. An agonistic aptamer against OX40 was able to enhance an antigen-pulsed DC-mediated antitumor immune response in murine B16-F10.9 melanoma [10]. Treatment of tumor-bearing mice with an intratumoral injection of full-size murine OX40L-expressing adenovirus resulted in a significant suppression of tumor growth along with survival advantages [2]. APCs with OX40L promoted partial activation of naïve T cells with some IL-2 secretion, but were unable to enhance proliferation, unlike that with B7-1 [12]. The long-lived concept underlying these results is that both anti-OX40 antibody and OX40L:Fc fusions act as OX40 receptor agonists. However recent work suggests that at least part of the in vivo effects of OX40 agonists is mediated by an Fc-dependent cellular cytotoxicity mechanism that selectively depletes intratumoral Treg cells [5]. In this connection it would be beneficial to explore the in vivo effects of soluble OX40L without the Fc fragment. We propose a model system where a CT26 cell line is transduced to stably express either soluble murine OX40L (extracellular portion, 50–198 aa) in secretory form or full OX40Lexo in membrane-bound form. BALB/c mice have been inoculated with these OX40L-expressing cells and the tumor incidence and growth parameters individually monitored. This allowed us to explore the influence of OX40L in tumor development.

11

by Evrogen, Russia, was inserted into the NheI-EcoRI sites of the polylinker inframe with the leader sequence, producing a vector for the lentiviral expression of OX40Lexo under the control of the hPGK promoter and IRES-driven expression of EGFP. Similarly, for OX40L full construct, the sequence for the multiple cloning site, IRES and EGFP, was amplified with primers containing sites for BglII and XhoI, and inserted into the BamHI and SalI sites of pRRL.SIN. WPRE.hPGK.EGFP in place of EGFP. The OX40Lfull ORF was amplified with primers containing SalI and BamHI recognition sites and cloned into the pRRL.SIN.WPRE.hPGK.IRES.EGFP vector. 2.2. Production of lentiviral vector particles Lentiviral particles for mammalian cell infection were obtained according to standard procedure. In short, 24–48 h prior to transfection, HEK293T cells at a logarithmic growth stage were plated onto d = 6 cm cell culture dishes (SPL Life sciences, Korea) in DMEM supplemented with 10% fetal calf serum (FCS), 1% glutamine, 10 units/ml penicillin and 10 lg/ml streptomycin. The transfection with 4 lg pR8.91, 1.2 lg pMD.G and 5 lg of the OX40L-carrying plasmid was carried out using a calciumphosphate transfection kit (Molecular Probes, USA) according to the manufacturer’s protocol with a total of 10 lg DNA, when the HEK293T cells had reached about 90% confluency. The culture medium was changed 16 h after transfection and the lentiviral particles were harvested 24 h thereafter. Medium containing the viral particles was concentrated 20 fold using a Pierce concentrator with a 150 KDa cut-off (Thermo Scientific, USA). For lentiviral infection, the CT26 cells were plated onto d = 35 mm cell culture dishes (SPL Life sciences, Korea), at a density of 2.5  104 cells/dish, in DMEM with 10% FCS, 1% glutamine, 10 units/ml penicillin and 10 lg/ml streptomycin. After culturing for 24 h the medium was changed for the concentrated medium with viral particles. The fluorescence of the infected cells was analyzed 5–7 days post infection. 2.3. Western blot Cell culture medium after 72 h cultivation of HEK293 cells transiently expressing OX40Lexo was incubated with Ni sepharose excel (GE Healthcare) (1 mL of medium with 5 lL of Nisepharose slurry) for 1 h for selective binding of 6xHis-tagged protein. Then Ni-sepharose beads were washed with TBS (20 mM TrisHCl pH8 150 mM NaCl) and OX40Lexo was eluted with 500 mM imidazole in TBS. Cell culture medium after cultivation of non-OX40L-expressing HEK293 cells was treated in parallel and used as a control. Eluted protein was denatured in SDS loading buffer and subjected to 10–25% gradient SDS-PAGE. Proteins were transferred to a nitrocellulose membrane and probed with monoclonal anti-6xHis-Tag mouse antibodies (Abcam, ab18184). Secondary anti-mouse antibodies conjugated with horseradish peroxidase (Sigma) were detected using a standard luminolperborate chemiluminescent reaction.

2. Materials and methods 2.4. Anti-OX40L staining and microscopy 2.1. Genetic manipulations A custom multiple cloning site and a sequence for the preprotrypsin leader peptide with a 6xHis tag and FLAG epitope were inserted into XhoI-EcoRI of a pIRES2-EGFP vector (Clontech, USA). The fragment containing the leader sequence, 6x-His tag, multiple cloning site, IRES and EGFP, was amplified by PCR and inserted into the BamHI-SalI sites of the pRRL.SIN. WPRE.hPGK.EGFP vector for lentiviral expression (kind gift by Prof. Terskikh, StanfordBurham Medical Research Institute). This intermediate product was then used as a control EGFP-only vector. The sequence for OX40L extracellular domain (OX40Lexo, aa 50–198), synthesized

Anti-OX40L staining of the CT26 OX40Lfull cell line was performed with rat mAB to CD252 (RM134L) and ab 95656 (Abcam), at a 1:250 dilution. The cells were washed 2 times with ice-cold PBS, and stained in PBS with 1% BSA for 1 h on ice. Then the cells were washed 3 times with PBS and used for the fluorescence microscopy analysis. Live cell imaging was performed in HEPES-buffered DMEM (PanEco, Russia) supplemented with 10% (v/v) FCS at 37 °C in an atmosphere of 5% CO2. For the fluorescence microscopy, a Leica AF6000 LX imaging system, based on a DMI 6000 B inverted microscope equipped with a Photo metrics CoolSNAP HQ CCD camera,

12

E.O. Serebrovskaya et al. / Cytokine 84 (2016) 10–16

was used. A 120WHXP short arc lamp (Osram) was used as the light source. A standard GFP filter set [excitation BP (band pass) 470/40 nm, emission BP 525/50 nm] and Tx2 filter set [excitation BP 560/40 nm, emission 645/75 nm] were used to acquire the green and red fluorescence, respectively. For whole tumor imaging, the tumors were excised from sacrificed animals and imaged immediately using an Olympus SZX16 zoom stereomicroscope equipped with an SDF PLAPO 0.5x objective and an Evolution QE monochrome 12 bit camera. The filter sets were an excitation BP 450–490 nm, emission LP 520 nm for the green channel and an excitation BP 540–580 nm, emission LP 620 nm for the red channel. 2.5. Flow cytometry and FACS For flow cytometric analysis of the fluorescence, the CT26 EGFP, CT26 OX40Lfull and CT26 OX40LexoEGFP cells were washed with PBS and analyzed with a Cytomics FC500 flow cytometer, equipped with an air-cooled argon-ion laser operating at 488 nm (Beckman Coulter). The following detection parameters were used: 6 mW laser power and a 525 nm band pass filter (FL1 channel). A minimum of 5000 events were collected for each sample. For sterile cell sorting, 2  106 cells were re-suspended in PBS at a density of 5  105 cells/ml. The cell suspension was then filtered through a 70 lm nylon mesh cell strainer. Using a FACS Aria cell sorter (BD Biosciences, USA), a minimum of 1.5  105 events were collected into sterile 2 ml tubes containing DMEM, 10% (v/v) FBS, 10 units/ml penicillin and 10 lg/ml streptomycin. OX40L in the culture medium was quantified using a commercially available DuoSet ELISA Kit (R&D Systems). A mass-calibrated recombinant OX40L standard was included in the kit. 2.6. Alamar blue reduction assay The cells were plated into the wells of 96-well plate in 200 lL DMEM with 10% FCS at a density of 1  104 cells/well. Alamar blue reduction was measured, in triplicate, 24, 48 and 72 h after plating for each cell line and time point. 10x Alamar blue stock solution (Life technologies, USA) was added to the wells and incubated for 4 h at 37 °C and 5% CO2. Then, absorbance of the reduced Alamar blue was measured with a NanoDrop spectrophotometer (Thermo Scientific, CIA) at 570 nm, and normalized to the absorbance at 600 nm. 2.7. Mice and tumor challenge Female BALB/c immunocompetent mice weighting 18–20 g were purchased from the Andreevka animal breeding centre (Russia). Female nu/nu mice aged 6–8 weeks were purchased from the Puschino animal breeding centre (Russia). All the experiments conducted on animals were approved by the Institutional Ethical Committee of either the Nizhny Novgorod State Medical Academy, Russia, or the Institute of Bioorganic Chemistry of the Russian Academy of Science. To generate tumors, mice were injected subcutaneously (s.c.) with 2  105 or 5  105 CT26 OX40LexoEGFP, CT26 OX40Lfull EGFP, CT26 EGFP, CT26 cells, or a 1:1 mixture of the abovementioned cells with H2B-KillerRedTandem expressing cells suspended in 100 lL sterile PBS in the hind limb (day 0). Tumor size was measured with a caliper three times a week. The tumor volume was calculated using the formula V = a ⁄ b ⁄ 1/2b, where a is the length and b is the width of the tumor. A tumor was considered palpable when the tumor volume reached 10 mm3. For survival experiments, tumor-bearing mice were euthanized when their tumors reached 1500 mm3 while tumor-free mice were considered cured

2 months after tumor regression. Kaplan–Meier curves were constructed for each group. Mice with completely regressed tumors were re-challenged with 5  105 CT26 EGFP cells s.c. into the opposite limb after they had been tumor free for 60 days. For EGFP-expressing tumors the mice were imaged in vivo every 2–3 days in the IVIS-Spectrum system (Caliper Life Sciences, USA). Fluorescence was excited at a wavelength of 465 nm (bandwidth 30 nm) and registered at 520 nm (bandwidth 20 nm).

2.8. Adoptive transfer of immunity by spleen cells To assess the formation of immune protection against CT26EGFP tumors, a passive transfer of spleen cells was conducted from mice inoculated with 2  105 CT26 OX40Lexo EGFP, CT26EGFP or CT26 cells, and from naïve mice, as a control. The treatment protocol was adopted from [23]. For obtaining spleen cell suspensions, immediately after the animals were sacrificed, the spleens were removed using sterile technique and collected into d = 12 mm sterile Petri dishes with cold (4 °C) Hank’s balanced salt solution (HBSS) placed on ice (5 ml of HBSS per spleen). The organs were minced in the HBSS using sterile forceps, and then the cell suspensions were passed through a 10-ml syringe (needle No. 16) several times. The suspensions were placed into sterile 20 ml test tubes by syringe and centrifuged/washed 3 times at 150g for 5 min in HBSS. The cells were then re-suspended in HBSS to a concentration of 4  107 cells per 1 ml. The spleen cell suspensions were injected intraperitoneally (i.p.) to the recipient mice at a dose 2  107 cells in 0.5 ml HBSS. After 3 days the recipient mice were injected s.c. with 2  105 of CT26EGFP cells and examined for tumor formation over the next 60 days. Mice that developed tumors were sacrificed when the tumors reached a volume of 1500 mm3 according to our protocol.

3. Results 3.1. Cell line isolation and analysis To produce cell lines that stably express OX40L, we constructed lentiviral vectors encoding OX40L, followed by IRES element and EGFP ORF. Vectors encoding complete OX40L ORF (OX40Lfull) and OX40L extracellular domain (OX40Lexo) were produced. Then, using EGFP as a marker of successful transduction, we applied FACS to the mixed population of CT26 cells after viral transduction to select cells with high fluorescence intensity in the green channel. A no-OX40L control (EGFP only) was treated in parallel using the same gating settings. The resulting cell populations were analyzed for OX40L expression. OX40Lexo production by the CT26 OX40Lexo was 18 ± 0.23 ng/mL from 106 cells in 3 days, as shown by ELISA. Also we performed Western blot analysis of soluble OX40L with anti-His-tag antibodies after SDS-PAGE. We detected the His-tagged protein of the molecular weight that was higher than the theoretical molecular weight calculated for OX40L extracellular domain with His-tag (Fig. 1B). We attribute this difference to posttranslational modifications that are expected for OX40L. The CT26 OX40Lfull cells were immunostained with anti-OX40L antibody and analyzed using fluorescence microscopy (Fig. 1A). A majority of the cells showed both green fluorescence for EGFP, and red fluorescence for anti-OX40L-PE. The ability of OX40L-expressing cell lines to proliferate in vitro was tested using the Alamar blue reduction assay (Fig. 1C). The CT26 OX40Lfull and CT26 EGFP (no OX40L control) cell lines demonstrated comparable growth properties, whereas the CT26 OX40Lexo cell line proliferated at a higher rate.

E.O. Serebrovskaya et al. / Cytokine 84 (2016) 10–16

13

Fig. 1. Characterization of the OX40L-expressing cell lines. A – Anti-OX40L staining of the CT26 OX40Lfull EGFP cell line. Wide-field fluorescence microscopy, green fluorescence represents EGFP, red fluorescence represents anti-OX40L-PE staining. B – Western blot analysis of the soluble His-tagged OX40L variant after Ni-NTA isolation (mouse anti-His-tag antibody). C – Cell line proliferation 1–3 days after plating, Alamar blue reduction assay [mean ± SD]. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.2. The antitumor effect of OX40L To study the effect of OX40L on CT26 tumors, BALB/c mice were injected s.c. with 2  105 OX40Lexo EGFP CT26, or OX40Lfull EGFP CT26 cells. CT26 EGFP while parental CT26 tumors challenged with the same cell dose served as controls. In the two control groups the injection of cancer cells resulted in a palpable tumor in 9 of 10 animals (Fig. 2A and B). In these groups, 1 of 10 mice was tumor-free due to spontaneous tumor regression. In the OX40Lexo EGFP group, only 1 of 10 mice developed a tumor. 2 of the 9 tumorfree mice did not develop tumors during the whole period of observation, and, of the rest, 7 tumors started to grow but subsequently regressed by days 12–28 after inoculation (Fig. 2C). The OX40Lfull EGFP CT26 cells formed tumors in 4 of 10 animals, although 2 of these 4 tumors displayed a delayed onset (Fig. 2D). Therefore, OX40Lexo efficiently induced tumor rejection, and the rejection rate was significantly higher than for OX40Lfull. The percentage of tumor-free mice throughout the course of the experiment has also been shown as Kaplan-Meier curves in Fig. 3.

Following the primary OX40Lfull and OX40Lexo tumor rejection, when all the tumor-free mice in these groups were rechallenged with 5  105 CT26EGFPcells they showed resistance to the tumor challenge, indicating the development of immunological memory. The results of the adoptive immune transfer showed that all the mice (20 of 20) treated with spleen cells from donors previously challenged with CT26 OX40Lexo EGFP cells rejected the CT26 EGFP tumors. Tumor susceptibility was 47% (7 of 15) in the case of spleen cell transfer from naïve mice, and 50% (5 of 10) or 33% (2 of 6) where CT26 or CT26 EGFP bearing mice were the respective donors (Table 1). Additionally, to prove the involvement of the immune system in the development of the antitumor effect of OX40L, we inoculated nu/nu mice, lacking functional T-lymphocytes, with either OX40L-expressing or control CT26 cells. It was found that inoculation of all the cell lines resulted in tumor growth in 100% of the animals without any statistically significant differences in tumor sizes (Fig. 4A) on day 16. It is important to note that differences in the

14

E.O. Serebrovskaya et al. / Cytokine 84 (2016) 10–16

Fig. 2. Therapeutic effect of OX40L on CT26 tumors. A – CT26, B – CT26 EGFP, C – CT26 OX40Lexo EGFP, D – CT26 OX40Lfull EGFP. The tumors were generated by s.c. injection of 2  105 cells. The tumor sizes of individual mice are shown, with the tumor rejection rates being indicated in parentheses (n = 10 for each group).

Fig. 3. Kaplan–Meier curves of the percentage of tumor-free mice in the CT26 (}), CT26 EGFP (h), CT26 OX40Lfull EGFP (j) and CT26 OX40Lexo EGFP (▲) groups after s.c. injection of 2  105 cancer cells (n = 10 for each group).

Table 1 The susceptibility of BALB/c mice to challenge with CT26EGFP cells after spleen cell transfer. Donors of spleen cells

Tumor incidence

Naïve mice Mice with CT26 tumor Mice with CT26 EGFP tumor Mice rejected CT26 OX40Lexo EGFP tumor

7 5 2 0

(15), 46.7% (10), 50% (6), 33.3% (20), 0%*

Comparison between two means was carried out using Fisher’s exact test. P 6 0.05 was considered significant. * Statistically significant differences between group ‘‘Mice rejected CT26 OX40Lexo EGFP tumor” and all other groups, P 6 0.0462.

cell growth rate observed in vitro did not affect the ability to generate tumors in immunodeficient nu/nu mice. To examine whether the OX40L-mediated immune protection extended to another antigen besides EGFP, BALB/c mice were inoculated with a mixture of CT26 OX40Lexo EGFP, and CT26 cells expressing the red fluorescent protein KillerRed (1:1 ratio). Using fluorescence imaging we found that all growing tumors (n = 6) contained regions of both green (EGFP) and red (KillerRed) fluorescence (Fig. 4), and no single mouse had a tumor with only red fluorescence whereas the tumor-free mice (n = 4) developed an effective antitumor immune response, which eliminated both types of tumor cells. Therefore, the results of this study show that OX40Lexo elicits strong antitumor effects manifested in tumor rejection in immunocompetent BALB/c mice. The CT26 EGFP tumor rejection in BALB/c mice that had previously rejected CT26 OX40Lexo EGFP tumors, as well as in mice which received a spleen cell transfer from CT26 OX40Lexo EGFP inoculated donors, and the stable formation of OX40L-expressing tumors in immunodeficient nu/nu mice indicates that the generation of memory T cells provided protection against CT26 EGFP tumors. The elimination of tumors consisting of OX40Lexo-EGFP and KillerRed-expressing cells implies that immunity against another type of antigen is also generated. 4. Discussion OX40L is a well-studied molecule that has shown antitumor potential in several murine tumor models as an Fc-fusion protein. Fc fusion endows the hybrid molecule with a number of beneficial properties such as an increased plasma half-life [16], increased solubility due to independent Ig-domain folding and ease of purification using protein G-coupled sepharose [6]. Moreover, the Fc fragment is known to interact with Fc receptors on immune cells

E.O. Serebrovskaya et al. / Cytokine 84 (2016) 10–16

15

Fig. 4. A – Tumor growth in nu/nu mice. The animals were inoculated with 2  105 cells/animal and monitored for 16 days. Mean ± SD, n = 5 animals/group. B – Representative fluorescence image of tumor formed in BALB/c mouse as a result of inoculation with a 1:1 mixture of CT26 cells expressing OX40LexoEGFP and KillerRed (total 2  105 cells/mouse).

[24], triggering antibody-dependent cytotoxicity. For the OX40 activating antibody, OX86, it was recently shown that FccR binding may be an important factor contributing to Treg depletion and resulting in tumor elimination [5]. The fragment of IgG used in Fc:OX40L studies is capable of binding activating Fc receptor [5]. However, it is unclear whether the observed antitumor effect depends primarily on the activation of OX40 signaling or if FccR binding is also partially responsible. In this research we tested the hypothesis that the Fc component is not required for soluble murine OX40L functioning as a costimulating agent for the antitumor immune response. We found that the tumorigenicity of the OX40Lexo-secreting CT26 cell line is significantly reduced compared to the control CT26 cell line. Our results therefore indicate that soluble OX40L may influence the efficiency of the antitumor immune response at least if it is present at the tumor site during the initial stages of tumor growth in a murine syngeneic tumor model. OX40L activity is known to be enhanced by cell surface immobilization [21]. In our model system, we found the CT26 OX40Lfull EGFP cell line to be rejected with lower efficiency compared to the CT26 OX40Lexo EGFP cell line, but still more efficiently than the control cell lines. The reason for soluble OX40L superiority may be that the availability of OX40L for interaction with the tumor infiltrating immune cells was higher for soluble variant. This could be due to higher concentration of soluble ligand or due to spatial distribution (diffuse for soluble variant, or localized to the tumor cell membrane for full length variant). However, we did not directly compare the quantity of OX40L in solution and on the cell surface. So, the question of the relative efficiency of soluble and membrane-bound OX40L was out of the scope of this study. We also questioned whether only OX40L-expressing cells would be eliminated. Both EGFP and KillerRed are known to be immunogenic in BALB/c mice when expressed in transplantable tumors [11,36]. These two proteins share only 26% protein sequence identity, and KillerRed does not contain the epitope HYLSTQSAL identified for the EGFP. So we assume that the immunogenicity of KillerRed is driven by a different antigenic determinant. In our model two different fluorescent proteins served as markers for the cells that express OX40L (EGFP) and do not express OX40L (KillerRed). So, the result of co-inoculation with two different cell lines indicates that the immune response is mounted not only against the cells expressing OX40L, but also against the cells that do not express the ligand and do not share EGFP-derived antigenic epitope.

T lymphocytes are required for the rejection, as indicated by the lack of rejection in nu/nu mice. Subsequent re-challenge of tumorfree mice with CT26 EGFP cells resulted in no tumor growth, which was indicative of the formation of an immunological memory. Among the numerous examples of studies that involve OX40 receptor stimulation there are some that employ the same concept of the intratumoral expression of OX40L. Andarini et al. described the adenoviral delivery of OX40L into B16 melanoma, Lewis lung carcinoma and CT26 carcinoma tumors. In all the tumor models tested, treatment of tumor-bearing mice with AdOX40L resulted in a significant suppression of tumor growth [2]. Kaneko et al. found that the tumorigenicity of an OX40L-transfected EL4 cell line was lower than that of the parental EL4 cell line [15]. Gri et al. observed only a delay in tumor growth for an OX40L-transduced CT26 cell line, while CT26 expressing OX40L and GM-CSF regressed in 85% of mice [13,14]. Likewise, in our model EGFP serves as an immunogen, thereby synergizing with OX40L for the induction of an antitumor immune response and tumor rejection. The above findings and our results suggest that gene transfer of OX40L into tumor cells might be an eligible modality for the induction of anti-tumor immune response.

Acknowledgements This work was supported by the Presidium of the Russian Academy of Sciences program ‘‘Fundamental research for the development of biomedical technologies”. Mouse tumor xenograft experiments with CT26 OX40Lfull cell line were funded by the Russian Science Foundation (grant N 14-35-00105). This research was carried out using the equipment provided by IBCH core facility (CKP IBCH). E. Serebrovskaya is supported by the Russian President’s scholarship # CG-1653.2015.4.

References [1] S.a. Ali, M. Ahmad, J. Lynam, C.S. McLean, C. Entwisle, P. Loudon, E. Choolun, S. E.B. McArdle, G. Li, S. Mian, R.C. Rees, Anti-tumour therapeutic efficacy of OX40L in murine tumour model, Vaccine 22 (2004) 3585–3594, http://dx.doi. org/10.1016/j.vaccine.2004.03.041. [2] S. Andarini, T. Kikuchi, M. Nukiwa, P. Pradono, T. Suzuki, S. Ohkouchi, A. Inoue, M. Maemondo, N. Ishii, Y. Saijo, K. Sugamura, T. Nukiwa, Adenovirus vectormediated in vivo gene transfer of OX40 ligand to tumor cells enhances antitumor immunity of tumor-bearing hosts, Cancer Res. 64 (2004) 3281– 3287.

16

E.O. Serebrovskaya et al. / Cytokine 84 (2016) 10–16

[3] P. Bansal-Pakala, A.G. Jember, M. Croft, Signaling through OX40 (CD134) breaks peripheral T-cell tolerance, Nat. Med. 7 (2001) 907–912, http://dx.doi.org/ 10.1038/90942. [4] P.R. Baum, R.B. Gayle, F. Ramsdell, S. Srinivasan, R.A. Sorensen, M.L. Watson, M. F. Seldin, E. Baker, G.R. Sutherland, K.N. Clifford, Molecular characterization of murine and human OX40/OX40 ligand systems: identification of a human OX40 ligand as the HTLV-1-regulated protein gp34, EMBO J. 13 (1994) 3992– 4001. [5] Y. Bulliard, R. Jolicoeur, J. Zhang, G. Dranoff, N.S. Wilson, J.L. Brogdon, OX40 engagement depletes intratumoral Tregs via activating FccRs, leading to antitumor efficacy, Immunol. Cell Biol. 92 (2014) 475–480, http://dx.doi.org/ 10.1038/icb.2014.26. [6] P.J. Carter, Introduction to current and future protein therapeutics: a protein engineering perspective, Exp. Cell Res. 317 (2011) 1261–1269, http://dx.doi. org/10.1016/j.yexcr.2011.02.013. [7] J.G. Chai, I. Bartok, P. Chandler, S. Vendetti, A. Antoniou, J. Dyson, R. Lechler, Anergic T cells act as suppressor cells in vitro and in vivo, Eur. J. Immunol. 29 (1999) 686–692, http://dx.doi.org/10.1002/(SICI)1521-4141(199902) 29:02<686::AID-IMMU686>3.0.CO;2-N. [8] A.I. Chen, A.J. McAdam, J.E. Buhlmann, S. Scott, M.L. Lupher, E.A. Greenfield, P.R. Baum, W.C. Fanslow, D.M. Calderhead, G.J. Freeman, A.H. Sharpe, Ox40-ligand has a critical costimulatory role in dendritic cell: T cell interactions, Immunity 11 (1999) 689–698. [9] Y. Chen, V.K. Kuchroo, J. Inobe, D.A. Hafler, H.L. Weiner, Regulatory T cell clones induced by oral tolerance: suppression of autoimmune encephalomyelitis, Science 265 (1994) 1237–1240. [10] C.C.M. Dollins, S. Nair, D. Boczkowski, J. Lee, J.M. Layzer, E. Gilboa, B.A. Sullenger, Assembling OX40 aptamers on a molecular scaffold to create a receptor-activating aptamer, Chem. Biol. 15 (2008) 675–682, http://dx.doi.org/ 10.1016/j.chembiol.2008.05.016.Assembling. [11] A. Gambotto, G. Dworacki, V. Cicinnati, T. Kenniston, J. Steitz, T. Tüting, P.D. Robbins, A.B. DeLeo, Immunogenicity of enhanced green fluorescent protein (EGFP) in BALB/c mice: identification of an H2-Kd-restricted CTL epitope, Gene Ther. 7 (2000) 2036–2040, http://dx.doi.org/10.1038/sj.gt.3301335. [12] I. Gramaglia, a.D. Weinberg, M. Lemon, M. Croft, Ox-40 ligand: a potent costimulatory molecule for sustaining primary CD4 T cell responses, J. Immunol. 161 (1998) 6510–6517. [13] G. Gri, E. Gallo, E. Di Carlo, P. Musiani, M.P. Colombo, OX40 ligand-transduced tumor cell vaccine synergizes with GM-CSF and requires CD40-Apc signaling to boost the host T cell antitumor response, J. Immunol. 170 (2003) 99–106. [14] G. Gri, E. Gallo, E. Di Carlo, P. Musiani, M.P. Colombo, OX40 ligand-transduced tumor cell vaccine synergizes with GM-CSF and requires CD40-Apc signaling to boost the host T cell antitumor response, J. Immunol. 170 (2003) 99–106, http://dx.doi.org/10.4049/jimmunol.170.1.99. [15] H. Kaneko, T. Hori, S. Yanagita, N. Kadowaki, T. Uchiyama, Introduction of OX40 ligand into lymphoma cells elicits anti-lymphoma immunity in vivo, Exp. Hematol. 33 (2005) 336–343, http://dx.doi.org/10.1016/j.exphem.2004.11.013. [16] R.J. Keizer, A.D.R. Huitema, J.H.M. Schellens, J.H. Beijnen, Clinical pharmacokinetics of therapeutic monoclonal antibodies, Clin. Pharmacokinet. 49 (2010) 493–507, http://dx.doi.org/10.2165/11531280000000000-00000. [17] A. Miller, O. Lider, H.L. Weiner, Antigen-driven bystander suppression after oral administration of antigens, J. Exp. Med. 174 (1991) 791–798. [18] S. Miura, K. Ohtani, N. Numata, M. Niki, K. Ohbo, Y. Ina, T. Gojobori, Y. Tanaka, H. Tozawa, M. Nakamura, Molecular cloning and characterization of a novel glycoprotein, gp34, that is specifically induced by the human T-cell leukemia virus type I transactivator p40tax, Mol. Cell. Biol. 11 (1991) 1313–1325. [19] A. Morris, J.T. Vetto, T. Ramstad, C.J. Funatake, E. Choolun, C. Entwisle, A.D. Weinberg, Induction of anti-mammary cancer immunity by engaging the OX40 receptor in vivo, Breast Cancer Res. Treat. 67 (2001) 71–80. [20] N.P. Morris, C. Peters, R. Montler, H. Hu, B.D. Curti, J. Urba, A.D. Weinberg, Development and characterization of recombinant human Fc:OX40L fusion protein linked via a coiled-coil trimerization domain, Mol. Immunol. 44 (2007) 3112–3121.

[21] N. Müller, A. Wyzgol, S. Münkel, K. Pfizenmaier, H. Wajant, Activity of soluble OX40 ligand is enhanced by oligomerization and cell surface immobilization, FEBS J. 275 (2008) 2296–2304, http://dx.doi.org/10.1111/j.17424658.2008.06382.x. [22] K. Murata, N. Ishii, H. Takano, S. Miura, L.C. Ndhlovu, M. Nose, T. Noda, K. Sugamura, Impairment of antigen-presenting cell function in mice lacking expression of OX40 ligand, J. Exp. Med. 191 (2000) 365–374. [23] H.B. Neel, R.E. Ritts, Transfer of tumor-specific immunity with syngeneic spleen cells and serum from mice that have large tumors and metastases, Cancer 40 (1977) 1643–1649. [24] F. Nimmerjahn, J.V. Ravetch, Fcgamma receptors as regulators of immune responses, Nat. Rev. Immunol. 8 (2008) 34–47, http://dx.doi.org/10.1038/ nri2206. [25] Y. Ohshima, L.P. Yang, T. Uchiyama, Y. Tanaka, P. Baum, M. Sergerie, P. Hermann, G. Delespesse, OX40 costimulation enhances interleukin-4 (IL-4) expression at priming and promotes the differentiation of naive human CD4(+) T cells into high IL-4-producing effectors, Blood 92 (1998) 3338–3345. [26] P.R. Rogers, J. Song, I. Gramaglia, N. Killeen, M. Croft, OX40 promotes Bcl-xL and Bcl-2 expression and is essential for long-term survival of CD4 T cells, Immunity 15 (2001) 445–455. [27] C.E. Ruby, M.a. Yates, D. Hirschhorn-Cymerman, P. Chlebeck, J.D. Wolchok, A.N. Houghton, H. Offner, A.D. Weinberg, Cutting edge: OX40 agonists can drive regulatory T cell expansion if the cytokine milieu is right, J. Immunol. 183 (2009) 4853–4857, http://dx.doi.org/10.4049/jimmunol.0901112. [28] T. So, M. Croft, Cutting edge: OX40 inhibits TGF-b- and antigen-driven conversion of naive CD4 T cells into CD25+Foxp3+ T cells, J. Immunol. 179 (2007) 1427–1430, http://dx.doi.org/10.4049/jimmunol.179.3.1427. [29] I. Takeda, S. Ine, N. Killeen, L.C. Ndhlovu, K. Murata, S. Satomi, K. Sugamura, N. Ishii, Distinct roles for the OX40-OX40 ligand interaction in regulatory and nonregulatory T cells, J. Immunol. 172 (2004) 3580–3589, http://dx.doi.org/ 10.4049/jimmunol.172.6.3580. [30] P. Verginis, K.A. McLaughlin, K.W. Wucherpfennig, H. von Boehmer, I. Apostolou, Induction of antigen-specific regulatory T cells in wild-type mice: visualization and targets of suppression, Proc. Natl. Acad. Sci. USA 105 (2008) 3479–3484, http://dx.doi.org/10.1073/pnas.0800149105. [31] J.T. Vetto, S. Lum, A. Morris, M. Sicotte, J. Davis, M. Lemon, A. Weinberg, Presence of the T-cell activation marker OX-40 on tumor infiltrating lymphocytes and draining lymph node cells from patients with melanoma and head and neck cancers, Am. J. Surg. 174 (1997) 258–265. [32] H. Von Boehmer, C. Daniel, Therapeutic opportunities for manipulating T(Reg) cells in autoimmunity and cancer, Nat. Rev. Drug Discov. 12 (2013) 51–63, http://dx.doi.org/10.1038/nrd3683. [33] T.H. Watts, TNF/TNFR family members in costimulation of T cell responses, Annu. Rev. Immunol. 23 (2005) 23–68, http://dx.doi.org/10.1146/annurev. immunol.23.021704.115839. [34] A.D. Weinberg, M.M. Rivera, R. Prell, a. Morris, T. Ramstad, J.T. Vetto, W.J. Urba, G. Alvord, C. Bunce, J. Shields, Engagement of the OX-40 receptor in vivo enhances antitumor immunity, J. Immunol. 164 (2000) 2160–2169. [35] A.D. Weinberg, A.T. Vella, M. Croft, OX-40: life beyond the effector T cell stage, Semin. Immunol. 10 (1998) 471–480, http://dx.doi.org/10.1006/ smim.1998.0146. [36] D.V. Yuzhakova, M.V. Shirmanova, E.O. Serebrovskaya, K.A. Lukyanov, I.N. Druzhkova, B.E. Shakhov, S.A. Lukyanov, E.V. Zagaynova, CT26 murine colon carcinoma expressing the red fluorescent protein KillerRed as a highly immunogenic tumor model, J. Biomed. Opt. 20 (2015) 88002, http://dx.doi. org/10.1117/1.JBO.20.8.088002. [37] S.G. Zheng, J. Wang, P.P. Wang, J.D. Gray, D.A. Horwitz, IL-2 is essential for TGFb to convert naive CD4+CD25- cells to CD25+Foxp3+ regulatory T cells and for expansion of these cells, J. Immunol. 178 (2007) 2018–2027, http://dx.doi.org/ 10.4049/jimmunol.178.4.2018. [38] W. Zou, Regulatory T cells, tumour immunity and immunotherapy, Nat. Rev. Immunol. 6 (2006) 295–307, http://dx.doi.org/10.1038/nri1806.