Lentiviral delivery of novel fusion protein IL12/FasTI for cancer immune/gene therapy

Many of the cytokine-based cancer immunotherapies are hindered by the devastating side effects of systemic delivery of the cytokines. To address this problem, we previously described a novel approach to locally achieve high doses of interleukin-12 (IL-12) in tumors and demonstrated that bi-functional fusion protein mIL-12/FasTI expressed by stable clones of TC-1 cells efficiently suppressed tumor proliferation by activating natural killer (NK) cells and other cytolytic killer cells and sending apoptotic signals into tumor cells. In the present study, we employed a lentiviral vector-based gene delivery system to deliver this fusion construct directly into tumor cells. We show that lentiviral vector efficiently delivers the fusion constructs into Hela cells in vitro as assayed by RT-PCR and immunohistochemistry (IHC). We also confirm that fusion protein mIL-12/FasTI delivered by the viral vector significantly enhanced killer cell activation, increased caspase-3 activity and decreased tumor growth in vitro. This study offers a further step for fusion protein cancer therapy for cancer patients.


Introduction
Natural killer (NK) cells are specialized lymphocytes capable of targeting virus-infected cells and malignant cells such as tumor cells. Unlike cytotoxic T cells (CTL), NK cells recognize potential target cells without the need for immunization or pre-activation [1]. The regulation of NK cell activation requires a variety of cell surface activating and inhibitory receptors. Importantly, NK cells are inhibited by receptors that recognize MHC class I molecules; therefore, normal cells expressing high levels of MHC class I can be protected from NK cell-mediated cytotoxicity [2]. NK cell activation also requires engagement of pro-inflammatory cytokines, particularly interleukin-12 (IL-12), IL-15 and type I interferons (IFN), which enhance NK cell proliferation as well as promote NK cell cytotoxicity and/or production of IFN-γ [3,4]. Recent evidence suggests that NK cells, besides their direct cytolytic effect against tumor cells, may also shape the adaptive immune response toward a T helper1 (Th1) profile, which is thought to favor antitumor responses [5]. Despite the powerful cytotoxic activity of NK cells, the tumor microenvironment of advanced tumors can interfere with NK cell activation pathways or the complex receptor array that regulate NK cell activation and anti-tumor effects. One major challenge for effective cancer therapy is how to reshape the tumor microenvironment to favor NK cells and other killer cells. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 IL-12 is a heterodimeric protein composed of covalently linked p35 and p40, which is majorly derived from activated antigen presenting cells (APC). It was first recognized as "natural killer-stimulating factor" by Trinchieri and colleagues in 1989. By enhancing the production of IFN-γ, proliferation and cytotoxic ability of NK cells and T cells, IL-12 bridges innate and adaptive immunity. In addition, IL-12 induces an antiangiogenic program mediated by IFN-γ and lymphocyte-endothelial cell cross-talk [6,7]. The antitumor and anti-metastatic activities of IL-12 have been extensively demonstrated in many murine models and clinical studies. However, systemic administration of IL-12 in patients is limited by severe toxicity. Additionally, the immune suppression-dominated microenvironment in advanced tumors is a major obstacle limiting the efficacy of IL-12-based therapies.
Fas (CD95/APO-1) is a cell surface death receptor of tumor necrosis factor (TNF) receptor superfamily. Binding of Fas ligand (FasL) induces Fas trimerization, which results in recruitment of adapter protein Fas-associated death domain (FADD) through DD-DD interactions. The death-effector domain of FADD then recruits pro-caspase-8/10 to the receptor, resulting in the formation of death-inducing signaling complex (DISC) [8,9]. DISC activates effector caspases such as caspase-3 that in turn cleave a restricted set of targets proteins and are responsible for cell apoptosis.
In our previous study, we have demonstrated that mouse IL-12/FasTI, a bi-functional fusion protein containing mouse IL-12 and Fas transmembrane/intracellular domains (FasTI), can simultaneously enhance NK cell activity and induce apoptosis of tumor cells in a tumor microenvironment, resulting in effective tumor cell elimination in vitro [10]. To further confirm the antitumor efficacy of fusion protein IL-12/FasTI in a therapeutic setting, a high-efficient gene delivery method is in demand. Lentiviruses are a subgroup of the retrovirus family, which have been developed mostly from simian immunodeficiency virus (SIV) and human immunodeficiency virus type I (HIV-1) [11]. The design of lentivirus-based expression vectors allows high-level expression of recombinant proteins in dividing and non-dividing mammalian cells. In the present study, lentiviral vectors, containing the cDNA sequence of fusion gene mIL12/FasTI (pLenti7.3/ mIL12/FasTI) and the control gene (pLenti7.3/IL12, pLenti7.3/Fas), were established through advanced molecular cloning. The pLenti7.3/V5-TOPO lentiviral expression vector contains two new elements to yield cell-specific and high performance delivery: the WPRE (Woodchuck Posttranscriptional Regulatory Element) and cPPT (Polypurine Tract), which can produce at least a four-fold increase in viral titer [12,13]. The titer of each viral clone was determined by transducing 293 cells. The transient gene expression of IL-12/ FasTI, IL-12 and Fas via lenti-viral transduction was confirmed by both reverse-transcription PCR (RT-PCR) and immunohistochemistry (IHC). It was also demonstrated that the expression of IL-12/FasTI enhanced apoptosis levels, NK cell activity as well as overall cytotoxicity against tumor cells, comparing to IL-12 and Fas controls. Combined with high-efficient lentiviral expression system, our fusion protein strategies might serve as one potential option for cancer immuno/gene therapy in the future.

Construction of lentiviral vectors
Mouse IL-12/FasTI cDNA sequence was cloned from pcDNA3.1/IL-12/FasTI/Zeo(+) vector from previous study using ACCATGGGTCAATCACGCTACas 5' primer and CTCCAGACATT GTCCTTCATTTTC as 3' primer. The optimal sequences for translation initiation (Kozak sequences) were included in the sequence. To enable TA cloning with pLenti7.3/V5-TOPO vector, Taq DNA polymerase was used for PCR to generate extruding "A" at both ends of PCR product. Four microliter purified PCR product of IL-12/FasTI, 1μl Salt solution and 1μl pLenti-TOPO vector (Invitrogen, USA) were then mixed and incubated at room temperature for 5 min for TOPO cloning reaction. Six ampicillin-resistant colonies from transformation were picked, and plasmid DNA for each colony was isolated using a QIAprep Spin miniprep kit (Qiagen). The plasmid was analyzed by restriction enzyme digest and DNA Sanger sequencing to confirm the presence and orientation of insert as well as the integrity of the vector.
Mouse IL-12 cDNA sequence was cloned from pcDNA3.1/IL-12/Zeo(+) from previous study using ACCATGGGTCAATCACGCTAC as 5' primer and CGTGGCTTCTTCTGCCAAAGC ATG as 3' primer. Mouse Fas cDNA sequence was cloned from pCMV-mFAS-His purchased from Sino Biological Inc., using ACCATGGTGTGGATCTGG as 5' primer and CTCCAGACATT GTCCTTCATTTTC as 3' primer. PLenti7.3/IL-12 and pLenti7.3/Fas were constructed by the same methodology following manufacturer's instructions. pLenti7.3/IL-12 and pLenti7.3/Fas sequences were analyzed by DNA Sanger sequencing to confirm the presence and orientation of insert as well as the integrity of the vector.

Production of lentiviral particles in 293FT cells
293FT cells were co-transfected with pLenti7.3/mIL-12/FasTI, pLenti7.3/mIL-12, or pLenti7.3/mFas, respectively, as well as lentiviral packing mix (containing three packing plasmids, pLP1, pLP2 and pLP/VSVG) using Lipofectamine 2000 (Invitrogen) as directed by the manufacturer's instructions. Virus-containing supernatant of each pLenti expression construct was harvested 48-72 hours post-transfection directed by manufacturer's instruction. Meanwhile, pLenti7.3/V5-GW/LacZ was used as a control in co-transfection to optimize expression conditions. Before proceeding to transduction and expression experiments, lentiviral stock of each construct was concentrated by using Lenti-X concentrator (ClonTech) and the viral titer was determined for further analyses. Lentiviral titration was determined by fluorescence-based cytometry of GFP positive cells following manufacturer's instruction.

RT-PCR
Three hundred thousand Hela cells were plated on 6-well-plate and incubated for 24 hour with Hela media. After the incubation, the media was removed and replaced with DMEM media containing 8μg/ml polybrene. Lentiviral particles of pLenti/IL-12/FasTI, pLenti/IL-12 and pLenti/Fas (Lent-IF, Lent-IL12, Lent-Fas) were added in plating media by1:10 dilution, and incubated for another 48 hour for transient transduction. Total RNA was extracted from each clone using an RNeasy Plus Mini kit following the manufacturer's directions. RT-PCR was run with 2μg of the total RNA using Phusion RT-PCR kit following the manufacturer's instruction (Thermo Fisher). To amplify a 1059-bp segment of IL-12/FasTI containing 522-bp of the mIL-12, the linker (GGTGGTGGTTCTGGTGGTGGTTCTGGTGGTGGTTCT),and the entire 495-bp of FasTI, a 5' primer GCAGTGACATGTGGAATGGC and a 3' primer CGGAATTCTCACTC CAGA CATTGTCCTTCATTTTC were used. To amplify 180-bp portion of IL-12, a 5' primer GGAAGC ACGGCAGCAGAATA and a 3' primer AACTTGAGGGAGAAGTAGGAATGG were used. To amplify 984-bp of Fas sequence, a 5' primer ACCATGGTGTGGATCTGG and a 3' primer CTCCAGACATTGTCCTTCATTTTC were used. All RT-PCR reactions included the β-actin housekeeping gene as loading control using a 5'primer ATGGGTCAGAAGGATTCCTATGTG and a 3' primer CTTCATGAGGTAGTCAGTCAGGTC amplifying a 488-bp fragment.

Immunohistochemistry (IHC)
Viral transduction was performed on Hela cells as described in RT-PCR. After transduction, cells of each group (Lent-IF, Lent-IL12, Lent-Fas) were cytospun onto glass slides by centrifuging at 2,000 rpm for 10 min at room temperature. After centrifugation the cells on the slides were fixed with 4% para-formaldehyde (PFA) solution for 10 min. The cells were then stained at room temperature for one hour with isotype rat IgG 2a antibody (R&D Systems, clone #54447) and/or anti-mouse IL-12/p35 antibody (R&D Systems, clone #45806), after a 1:100 dilution with sterile PBS. After washing, cells were treated with secondary antibody conjugated with HSS-HRP for 30 min at room temperature. DAB substrate-chromogen mixture was applied to the cells and incubated for 10-20 min. Slides were washed three times using fresh PBS, and then examined under a microscope (Olympus 1x70 fluorescent microscope).

Coculture of human NK cells with lentiviral transduced cells
Human NK92 cells were plated with Hela cells, or lentiviral transduced Hela cells expressing IL-12/FasTI, IL-12 or Fas on 96-well plate at a ratio of 5:1 in 200 μl NK media for 48 hours. After co-cultural incubation, 100 μl of supernatant was removed for analysis of IFN-γ production using human IFN-γ ELISA reagent kit (Thermo Fisher Scientific) following manufacturer's direction. The remaining media was removed, and suspension NK92 cells were carefully removed. The number of remaining tumor cells were determined using CellTiter 96 AQueous non-radioactive cell proliferation assay (Promega) following manufacturer's directions. The results were analyzed using a one way ANOVA with Tukey's post-test.

Tumor cell apoptosis
One million cells of Hela cells or lentiviral transduced Hela cells expressing IL-12/FasTI, IL12 or Fas were first plated on a 6-well-plate and incubated for 24 hour. At this point one million NK92 cells were added to all the cells and incubated overnight for coculture. After removing suspension NK92 cells, apoptosis of the tumor cells was determined by a caspase 3 assay kit (abcam) following the manufacturer's instruction.

Statistical analysis
GraphPad software package was used to plot graphs and run statistical analysis. The statistical significance was represented as Ã p<0.05, ÃÃ p<0.01, ÃÃÃ p<0.001.

Gene expression via lentiviral delivery system
Human cervical cancer Hela cells were transduced with Lent-IF, Lent-IL12, or Lent-Fas and RT-PCR and IHC were performed to determine the gene expressions. RT-PCR shows that after lentiviral transduction, all the three constructs, Lent-IF, Lent-IL12, and Lent-Fas are transcribed correctively (Fig 3; supporting information). IHC results demonstrated that IL12/Fas and IL-12 proteins are expressed on the cell surface (Fig 4). Fluorescent microscopic analysis showed that control Fas protein was also expressed on the cell surface (data not shown).

Human NK cell activation
In order to determine if the bi-functional fusion protein IL-12/FasTI delivered by the lentiviral system can activate killer cells, human NK92 cells were cocultured with Hela cells transduced with Lent-IF, Lent-IL12, or Lent-Fas for 48 hrs. One hundred microliter supernatant was collected from each well and screened for IFN-γ production by ELISA. NK92 cells cocultured with Hela cells transduced with Lent-IF and Lent-IL12 showed significantly increased production of IFN-γ comparing to the controls (Fig 5A). While the IFN-γ production by NK92 cells cocultured with Hela cells transduced with Lent-IF is not statistically higher than that of NK92 cells cocultured with Hela cells transduced with Lent-IL12, the enhanced trend is clear.
Interestingly, NK92 cells cocultured with Hela cells transduced with Lent-Fas produced significantly higher IFN-γ than control Hela cells. The cytotoxicity of tumor cells after co-culturing with NK92 cells were measured with MTS assay. NK92 cells co-cultured with Hela cells transduced with Lent-IF and Lent-IL12 killed much more tumor cells than the controls (Fig 5B). More importantly, Lent-IF is significantly more effective than Lent-IL12.

Apoptosis induction
NK92 cells were cocultured with Hela cells transduced with Lent-IF, Lent-IL12 or Lent-Fas for 24 hours to induce apoptosis within the target cells. Caspase 3 assay was performed to analyze the apoptosis in tumor cells. Lent-IF induced significantly higher caspase 3 activity in the transduced Hela cells than all the controls (Fig 6).

Discussion
The role of the immune system in recognition and elimination of tumor cells has been well recognized. NK92 cells and other killer cells are key players in anti-tumor immune response. NK92 cell activation requires cytokine IL-12. In our previous study, we built a bi-functional fusion gene IL-12/FasTI containing mouse IL-12 sequence and Fas transmembrane and intracellular domains and demonstrated its designed activities in vitro.
To examine the bi-functional fusion protein's anti-tumor activity in a therapeutic setting, a lentivirus-based gene delivery system was utilized to deliver the fusion gene construct into human tumor cells. While the raw transfection supernatants contain relatively less viral particles, after concentration, a titer of over 1x10 7 /ml was achieved (Fig 2). After transduction, Hela cells functionally expressed the transduced genes (Fig 3 and Fig 4).
The major purpose of building the bi-functional fusion protein is to achieve local high level expression of IL-12 to activate NK92 cells simultaneously send apoptotic signal to tumor cells. Therefore, human NK92 cells were co-cultured with Hela cells transduced with Lent-IF, Lent-IL12 or Lent-Fas. The activation of NK92 cells after co-culture was determined by the production of IFN-γ (Fig 5A). It is not a surprise that NK92 cells co-cultured with Hela cells transduced with Lent-IF and Lent-IL12 produced a similar amount of IFN-γ because the IL-12 portion of the fusion protein is identical to mouse IL-12. Further studies are needed to find out if Fas signaling is involved in IFNγ production via NK cells or other cells. Fusion protein IL-12/FasTI clearly demonstrated enhanced overall elimination against tumor cells comparing to IL12 control, while Fas control showed no significant effect on anti-tumor cytotoxicity. (Fig  5B) Apart from NK92 cell activation, the fusion gene design also involves Fas-induced apoptosis signaling in transmembrane and intracellular domains. The data demonstrated that strong apoptotic signals was sent into tumor cells that express the fusion protein IL12/Fas, not cells that express IL-12 or Fas, when co-cultured with NK92 cells (Fig 6). The data also confirmed a previous study that mouse Fas is able to send apoptotic signals in human cells [14].
In conclusion, lentivirus-based gene expression system was successfully used to deliver fusion gene construct IL12/FasTI into tumor cells with potential efficiency, which provided a further step for fusion protein strategies in cancer immune/gene therapies. Furthermore, the fusion gene product IL-12/FasTI produced by lentiviral transduction effectively activated killers as indicated by the enhanced production IFN-γ and tumor cell cytotoxicity.