Background
During the past two decades, therapeutic antibodies have started to make major contributions to the treatment of B cell malignancies. In 1997, rituximab (a genetically engineered monoclonal chimeric antibody directed against the CD20 B cell antigen) was approved by the US Food and Drug Administration (FDA) for the indication of follicular lymphoma and low-grade B cell non-Hodgkin’s lymphoma (B-NHL) [
1]. However, cancer relapse and metastasis often occurred after medical intervention with CD20-based therapy in patients with follicular lymphoma and acute lymphocytic leukemia [
2]. Thus, it is absolutely critical to develop new therapeutic regimens to overcome these challenges.
T cells are the most potent tumor-killing effector cells, but they cannot be recruited by conventional antibodies. However, several bispecific antibodies (bsAbs) that recruit T cells have been developed [
3], which may have the potential to circumvent this problem. To date, the most promising for the therapeutic application of this approach is blinatumomab, a tandem single-chain variable fragment (scFv) bsAb in a bispecific T cell engager (BiTE) format targeting CD19/CD3, which was approved by the FDA for the treatment of B-precursor acute lymphoblastic leukemia (B-ALL) [
4]. Although blinatumomab has impressive efficacy in the clinic and exhibits high clinical response rates in patients with relapsed or refractory B-ALL and B-NHL [
5,
6], it still has some limitations due to its low molecular weight (~55 kDa), which is below the glomerular filtration threshold [
6]. Therefore, blinatumomab is administered over a 28-day continuous infusion using a mini-pump in order to maintain steady drug concentration [
6], which results in inconvenience for patients and an increased possibility of treatment-related adverse event. To overcome the drawback of short half-life, Kipriyanov and his colleagues [
7] firstly designed a tetravalent bispecific tandem diabody (TandAb) with two binding sites for CD3 and two for CD19. Due to the TandAb’s large molecular weight (~105 kDa), it is not subject to glomerular filtration. In addition, the molecule exhibits stability properties with a half-life ranging from 18.4 to 22.9 h after intravenous administration in mice [
8].
Most of therapeutic antibodies are administrated by intravenous infusion, which results in a handful of antibodies that can reach the tumor sites. Thus, introducing an efficient and targeted delivery system for these therapeutic antibodies may enhance the efficacy of treatment for tumors, especially for minimal residual diseases (MRD) [
9]. Mesenchymal stromal cells (MSCs) are attractive cellular vehicles for the therapy of malignant diseases as they have the ability to migrate into tumors and track microscopic metastasis [
10,
11]. We have previously employed human umbilical cord-derived MSCs (UC-MSCs) as carriers for gene therapy [
12,
13] because UC-MSCs are easier to isolate and expand, and the harvesting procedure is more consistent and yields a greater number of relevant cells than other adult and feta tissues [
14]. These characteristics indicate that UC-MSC is a promising targeted delivery system with anticancer agents for a variety of cancers.
However, a study from Ribeiro and his colleagues shows that MSCs derived from different tissues possess different immunosuppression capabilities and their action varies with the immune cell type [
15]. Thus, MSCs are actually a double-edged sword when employed as carriers with agents triggering cytotoxicity of T cells for tumors. Although the exact mechanisms of MSC-mediated immunosuppression are still debated, many different factors are believed to be involved. It has been revealed that indoleamine 2,3-dioxygenase (IDO), the first and rate-limiting enzyme in the degradation of tryptophan [
16], plays a role in human MSCs to regulate immunity in tumor microenvironment [
17]. By depleting tryptophan locally and accumulation of its metabolites such as kynurenine and quinolinic acid, which could interact the aryl hydrocarbon receptor (AHR) on T cells, IDO seems to block the proliferation of T cells and inhibit T cell activity [
18,
19]. Because IDO is induced by inflammatory cytokines such as interferon-γ (IFN-γ) which act through JAK-STAT signaling pathways on interferon stimulatory response elements (ISRE) and γ-activating sequences (GAS) in the IDO promoter [
20], its expression is thought to be an endogenous feedback mechanism controlling excessive immune response [
21].
In this study, we employed the TandAb platform to construct a new tandem tetravalent antibody targeting both CD3 and CD19 (Tandab (CD3/CD19)) using our own DNA sequences of scFvs. Then, we exploited the feasibility and efficacy of using genetically modified UC-MSCs which constitutively secreted Tandab (CD3/CD19) (MSC-Tandab) for the treatment of human B cell lymphoma. Our results showed that MSC-Tandab could induce specific lysis of CD19-positive Raji cells in the presence of T cells in vitro and reduce xenograft tumor growth in vivo in combination with
d-1-methyl-tryptophan (D-1MT), an IDO pathway inhibitor [
22].
Methods
Cell lines and cell culture
Human embryonic kidney cell-derived 293T cells (kindly provided by Professor Cheng Tao, PUMC) were maintained in DMEM (Invitrogen, USA) supplemented with 2 mM l-glutamine, 100 U/mL penicillin (Gibco, USA), 100 μg/mL streptomycin (Gibco, USA) and 10% FBS (Gibco, USA). The human acute T cell leukemia cell line Jurkat, human chronic myelogenous leukemia cell line K562, and human B cell lymphoma cell lines Raji, Daudi, and BJAB (Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Tianjin, China) were grown in RPMI-1640 medium (Invitrogen, USA) supplemented with 2 mM l-glutamine, 100 U/mL penicillin (Gibco, USA), 100 μg/mL streptomycin (Gibco, USA), and 10% FBS. The cells were incubated at 37 °C in a humidified atmosphere containing 5% CO2.
PBMC isolation
With informed consent, human peripheral blood mononuclear cells (PBMCs) were isolated from healthy volunteers. Details are provided in the Additional file
1.
Construction of lentiviral expression vectors
Plasmids pLH-T3a*-19a (containing a cysteine at position Ser-100 of V
L3) and pLH-19a-T3a* (containing a cysteine at position Gly-44 of V
H3) encoding for hybrid V
L3-V
H19 and V
L19-V
H3 scFvs [
23], respectively, were used for assembly of Tandab (CD3/CD19) genes. See details in Additional file
1.
Expression and purification of Tandab (CD3/CD19)
293T cells were transfected with pLentiR-Tandab (CD3/CD19) or pLentiR-EV using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer’s protocol. After 48 h of transfection, supernatants were collected by centrifugation at 500×g for 10 min at 4 °C to clear 293T cells. The soluble antibodies in the supernatants were purified by 6×His-tag affinity chromatography (GE Healthcare, Sweden) according to the manufacturer’s instruction. The purified preparations were quantified with His-tag ELISA detection kit (GenScript, USA) and were used for cell-binding assays and cytotoxicity assays in vitro. In addition, the unpurified or purified Tandab (CD3/CD19) were verified by Western blot analysis.
Cell-binding assay
The CD19-positive cell lines Raji, Daudi, and BJAB and the CD3-positive cell line Jurkat were employed for analysis of binding activity of Tandab (CD3CD19) by flow cytometry (LSRII, Becton Dickinson Bioscience, San Jose, CA). The CD19- and CD3-negative K562 cells were served as negative control. See details in Additional file
1.
Cytotoxicity assay
All cytotoxicity assays were performed with PBMC effector cells. And PBMCs were pre-activated with 50 IU/mL IL-2 for 3 days before cytotoxicity assays. CD19
+ cells (Raji, Daudi, and BJAB) and CD19
− cells (K562) were prepared as target cells. The specific lysis of target cells was detected by LDH release assay according to the manufacturer’s protocol. See details in Additional file
1.
MSCs preparation
MSCs were isolated from human umbilical cord Wharton’s jelly (WJ) as previous described [
24]. MSCs were cultured at a density of 8 × 10
3 cell/cm
2 in DF-12 medium (Invitrogen, USA) supplemented with 2 mM
l-glutamine and 10% FBS (Gibco, USA). When cells reached 80~90% confluence, they were detached using a 0.125% trypsin/1 mM EDTA solution and re-seeded using the same growth medium for subsequent passages. For all experiments, early passages MSCs (3P to 5P) were used.
Production of lentivirus
The lentiviral particles carrying Tandab (CD3/CD19) gene were packaged according to the SBI’s protocol. See details in Additional file
1.
Transduction of MSCs and viability of transduced MSCs
The transduction of MSCs was performed as previously reported [
12]. And viability of transduced MSCs was detected by MTT assays. See details in Additional file
1.
Immunophenotype profile and tri-lineage differentiation of MSCs
MSCs and transduced MSCs (including MSC-EV and MSC-Tandab) were trypsinized (0.125% trypsin-EDTA) and washed twice with PBS, then incubated with APC-labeled anti-human CD73, CD90, CD105, CD14, CD19, CD34, CD45, and HLA-DR (all from BD Biosciences) for 30 min. After washing with PBS, the expression level of these molecules was determined by flow cytometry.
To test the in vitro differentiation ability, MSCs or transduced MSCs were cultured in adipogenic, osteogenic, and chondrogenic differentiation medium, respectively. For adipogenic differentiation, the MSCs were maintained in medium containing 1 mM dexamethasone, 500 μM IBMX, 10 μg/mL insulin, and 60 μM indomethacin (all from Sigma). Three weeks later, the cells were fixed and stained with Oil Red O (Sigma). For osteogenic differentiation, cells were cultured in IMDM (Gibco) supplemented with 10% FBS, 100 nM dexamethasone, 50 μg/mL ascorbic acid, and 10 μM β-glycerophosphate (all from Sigma) for about 3 weeks. At the end of incubation, the cells were assayed by Alizarin Red S (Sigma) staining for calcium deposition. To induce chondrogenic differentiation, MSCs were maintained in medium with 100 nM dexamethasone, 50 μg/mL ascorbic acid, 40 μg/mL proline, 10 ng/mL TGF-β
3, 2 mM ITS, 53.5 μg/mL lindeic acid, and 12.5 μg/mL BSA (all from Sigma). Three weeks later, the cultured cells were stained with Alcian Blue (Sigma). In addition, RNA was isolated from cells after the induction, and the expression level of differentiation-related gene was determined by real-time PCR on an ABI Prism 7500 detection system (Applied Biosystems, USA). And primers used for real-time PCR were summarized in Table
1.
Table 1
Primer sequences for genes in real-time PCR
PPAR-r | Forward: GCTGGCCTCCTTGATGAATA Reverse: TGTCTTCAATGGGCTTCACA |
ADIPOQ | Forward: TGGTCCTAAGGGAGACATCG |
Reverse: TGGAATTTACCAGTGGAGCC |
RUNX2 | Forward: CTCACTACCACACCTACCTG |
Reverse: TCAATATGGTCGCCAAACAGATTC |
BGLAP | Forward: GGCGCTACCTGTATCAATGG |
Reverse: TCAGCCAACTCGTCACAGTC |
SOX9 | Forward: AATGGAGCAGCGAAATCAAC |
Reverse: CAGAGAGATTTAGCACACTGATC |
COL2A1 | Forward: GGCAATAGCAGGTTCACGTACA |
Reverse: CGATAACAGTCTTGCCCCACTT |
IDO | Forward: GCCCTTCAAGTGTTTCACCAA |
Reverse: CCAGCCAGACAAATATATGCGA |
CD98 | Forward: GCTGCTGCTCTTCTGGCTC |
Reverse: GCCAGTGGCATTCAAATAC |
Jumonji | Forward: GCTCAGGACTTACGGAAACA |
Reverse: TGTGGTTGACAGCGGAACTG |
GAPDH | Forward: GGTCTTACTCCTTGGAGGCCATGT |
Reverse: ACCTAACTACATGGTTTACATGTT |
In vitro and in vivo MSC migration assay
In vitro and in vivo MSCs migration assays were performed as previously reported [
12,
25]. Details are provided in Additional file
1.
In vitro co-culture killing experiments
To assess the bioactivity of Tandab (CD3/CD19) secreted by MSCs, a co-culture system using transwell plates with 0.4-μm-pore membrane was established. The specific lysis of target cells was determined by FACS analysis according to the “calcein-loss” method [
26]. Briefly, MSC-Tandab, MSC-EV, or MSCs were seeded into 6-well culture plates at a density of 1 × 10
5 cells per well and incubated for 72 h. Then, Raji cells labeled with calcein-AM (Sigma, USA) and pre-activated PBMCs were added to the equilibrated inserts at an E:T ratio of 10:1. After co-cultured for 24 h, the cells in the inserts were harvested to be detected by flow cytometry. The expression of activation surface markers CD69 and CD25 of T cells was detected by flow cytometry in the same conditions of co-culture system with unlabeled Raji cells. And the supernatants in the inserts were collected for the assay of cytokines produced in the co-culture system, including IL-2, IFN-γ, and TNF-α using ELISA kits (R&D system, USA).
Because of expression of IDO in MSCs induced by IFN-γ in the co-culture system, we performed a co-culture experiment for a longer time in absence or presence of D-1MT (Sigma-Aldrich, USA), which is an IDO pathway inhibitor. As mentioned above, MSC-Tandab, pre-activated PBMCs, and Raji cells were added respectively into the co-culture system with or without 1 mM D-1MT and co-cultured for 24, 48, and 72 h. Cells in the upper chamber were harvested at the indicated time points. And the residual Raji cells were determined by flow cytometry with FITC-conjugated anti-CD19 antibodies. The supernatants were harvested for the measure of kynurenine, a metabolite of tryptophan in the IDO pathway. Repetitive wells were set for detection of CD98 and Jumonji in the messenger RNA (mRNA) level at different time points (24, 48, and 72 h) by real-time PCR. In addition, the proliferation of T cells in the co-culture system was detected by BrdU Flow Kit (BD Bioscences). Details are provided in Additional file
1.
Inducible expression of IDO in MSCs
MSCs were seeded into 6-well culture plates at a density of 1 × 105 cells per well in the absence or presence of 20 ng/mL of recombinant human IFN-γ (R&D system, USA). After incubation for 48 h, expression of IDO in the level of mRNA or protein upon treatment with IFN-γ was verified by real-time PCR and Western blot analysis.
Detection of kynurenine
Because IDO catalyzes the metabolism of tryptophan in the kynurenine pathway, the activity of IDO was determined by spectrophotometric assay for kynurenine [
17] in supernatants from co-culture system and cultures of MSCs with or without exogenous IFN-γ stimulation. See details in Additional file
1.
Cell viability assay
Cell viability assay was performed by MTT assay (Sigma-Aldrich, USA). See details in Additional file
1.
Western blot analysis
The expression of specific protein was detected by Western blot analysis. See details in Additional file
1.
Real-time PCR
Total RNA was extracted from corresponding cells using Trizol reagent (Invitrogen, USA) following the manufacturer’s protocol. The complementary DNA (cDNA) was generated using OligdT primers and M-MLV reverse transcriptase (Invitrogen, USA) with 2 μg total RNA. Real-time PCR was performed using ABI Prism 7500 real-time PCR system (Applied Biosystems, USA), in combination with SYBR Green (Takara, Dalian, China). Specific primers for each gene (Table
1) were selected using Primer Express (Applied Biosystems, USA). Relative transcript expression was normalized to that of GAPDH mRNA.
Growth inhibition of B cell lymphoma xenografts in vivo
All animal studies were performed in accordance with the guidelines under the Animal Ethics Committee of the Institute of Hematology and Hospital of Blood Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College. Raji cells (2 × 107 cells per mouse) were implanted subcutaneously into the right flank of each BALB/c nude mice (female, 5–6 weeks of age; PUMC, China) 1 day after the application of total body irradiation (300 cGy). One week later when tumor size reached 100–200 mm3, the mice were treated intravenously with MSC-Tandab and pre-activated PBMCs and D-1MT in the drinking water. Then, the mice were sacrificed for analysis of organ damages in the indicated time. In tumor therapy experiments, the mice were randomly divided into six groups (five mice for each group) as follows: (a) MSC + PBMC; (b) MSC-EV + PBMC; (c) MSC-Tandab + PBMC; (d) MSC + PBMC + D-1MT; (e) MSC-EV + PBMC + D-1MT; (f) MSC-Tandab + PBMC + D-1MT. MSCs (1 × 106 cells per mouse) were injected intravenously at day 0, followed by pre-activated PBMCs (5 × 106 cells per mouse) via the vein 2 days later, every 7 days for 2 weeks. The mice were treated with or without D-1MT (2 mg/mL in drinking water) from the beginning to the end of the treatment. At day 21 after treatment started, the mice were sacrificed by cervical dislocation under anesthesia. Then, tumor tissues were harvested and weighted for treatment evaluation.
Statistical analysis
Data are represented as mean ± SD. Statistical analysis was performed using GraphPad Prism 6 or Microsoft Excel software. Significance was assayed by an unpaired two-tailed Student t test or ANOVA (*P < 0.05, **P < 0.01, ***P < 0.001).
Discussion
Application of MSCs to serve as vehicles for delivering TandAb to tumor has not been reported so far. In this study, we investigated the therapeutic effects of gene-modified MSCs with Tandab (CD3/CD19) for B cell lymphoma. The hypothesis is that engineered MSCs which injected intravenously into the tumor-bearing mice would specifically migrate to tumor site and secrete Tandab (CD3/CD19) which recruits T cells to exhibit a potent antitumor immunity in combination with an IDO pathway inhibitor D-1MT. And results presented here support the feasibility of this strategy.
Currently, two approaches have been developed to harness T cells for killing tumor cells. The first approach utilizes gene-modified T cells with an engineered chimeric antigen receptor (CAR) combining a desired antigen-recognition fragment of a monoclonal antibody against T cell activation domains from the T cell receptor complex, such as the ζ chain, with co-stimulatory molecules, from CD28, 4-1BB, or OX40 [
35]. The other approach depends on T cell recruitment via bispecific antibodies (bsAbs) [
36] which bind one arm to T cell activation domain and bind the other arm to a tumor-associated antigen on the target cell. We previously constructed an anti-CD3 × anti-CD19 bispecific antibody, which could efficiently redirect T cells to lysis both B cell lymphoma cell lines and patient-derived B-ALL cells [
23,
37]. In this study, we successfully constructed a new tandem diabody with two binding sites for CD3 and two for CD19 using the DNA sequences from our anti-CD3 × anti-CD19 bispecific antibody. The relative amount of tandem diabodies proved to be dependent on the length of the linker in the middle of the chain [
7]. It was reported that the middle linker should not be too short or too long to force the four domains to interact with complementary ones of another molecule with the formation of an eight-domain Tandab [
7]. We therefore chose a linker with 15 amino acid residues in our construct. In addition, our earlier data have suggested that the introduced disulfide covalent bond between V
H3 and V
L3 at rational positions could enhance the stability of fusion proteins [
23,
38]. Since the two peptides of TandAb are held together by non-covalent associations of the corresponding V
H and V
L domains, an extra disulfide bond can prevent the peptide diffusing away from the other one. Results from Western blot analysis suggested that nearly no monomers were detectable in the non-reducing condition. However, we also observed a band in 212 kDa in the non-reducing condition (Figs.
1c, d and
2d), which should be dimers of Tandab (CD3/CD19) with intermolecular forces. And the amount of them decreased after purification (Fig.
1d). Furthermore, data from cell-binding assays and cytotoxicity assays in vitro indicated that our Tandab (CD3/CD19) could bind to both of CD3-positive cells and CD19-positive cells and induce the specific lysis of CD19-positive cells.
MSCs hold great promise for clinical applications in the treatment of various diseases owing to their multi-lineage differentiation potential and immunosuppressive properties. Furthermore, the tumor tropism property of MSCs has led to the utilization of them as attractive delivery vehicles for a spectrum of antitumor agents against cancers [
39‐
43]. Several previous investigations reported by our laboratory also support feasibility of this strategy [
12,
13,
25,
44]. In the present study, we engineered the human umbilical cord-derived MSCs through transduction by lentivirus to secrete Tandab (CD3/CD19). We demonstrated that the MSC-Tandab could maintain their properties after lentiviral infection, including their surface markers, tri-lineage differentiation, proliferation, and the capacity to migrate towards tumor cells in vitro. Additionally, the tropism towards tumor-bearing mice with Raji cells was confirmed by in vivo imaging system with luciferase-labeled MSCs (Fig.
3e). These findings indicate that UC-MSCs can be used as ideal vehicles to deliver Tandab (CD3/CD19) to tumor.
Furthermore, to mimic the therapeutic process with MSC-Tandab in vivo, we designed a co-culture system using transwell plates mentioned above, in which Tandab (CD3/CD19) secreted from MSC-Tandab could migrate to the upper insert to trigger cytotoxicity of T cells against CD19-positive cells. Results from the specific lysis of Raji cells and released cytokines confirmed our initial hypothesis. However, it has been reported that various factors are believed to be involved in the mechanisms of MSC-mediated immunosuppression, including inducible nitric oxide synthase (iNOS), IDO, tumor necrosis factor-inducible gene 6 (TSG6), CC-chemokine ligand 2 (CCL2), IL-10, and prostaglandin E2 (PGE2) [
45]. Although the immune-regulatory effects of MSCs do not occur spontaneously, it should be taken into account in this study because of the released cytokines, such as IFN-γ. The expression of IDO in MSCs induced by IFN-γ might contribute to the immunosuppressive microenvironment owing to tryptophan depletion and accumulation of its metabolites. Additionally, MSC-Tandab shows basal increased levels of kynurenine (Fig.
5f), which is associated to an enhanced immunosuppression ability. This aspect indicates that we should pay attention to genetic modification of MSCs since they could be responsible of an augmented pro-tumoral effect. Therefore, we employed D-1MT, an IDO pathway inhibitor which differed from direct enzymatic inhibitors of IDO, to overcome the inhibition effect of IDO. Data from the co-culture killing experiments showed that D-1MT could increase the ratio of lytic cells at a certain degree. No significant changes on the level of kynurenine were observed, which suggested that D-1MT could not inhibit the enzymatic activity of IDO directly. Then, we confirmed that the expression of the T cell anergy-associated genes CD98 and Jumonji were decreased in the presence of D-1MT. And the proliferation capacity of T cells was also increased by D-1MT. Furthermore, the antitumor growth effect of MSC-Tandab in vivo was enhanced significantly in combination with D-MT. Although IDO is known to be overexpressed in several human cancers, including prostate, breast, brain, and hematologic malignancies [
46], it is not detectable in Raji cells even with the stimulation of IFN-γ [
47]. While a study about IDO expression by circulating leukemic cells from 5 B-CLL patients showed that IDO could not be detected in the PBMCs of any of them, but in four of the five, IDO was dramatically up-regulated when cells were cultured with IFN-γ for 24 h [
47]. This may be one of the reasons for drug resistance during the treatment. Thus, our treatment with MSC-Tandab in combination with IDO pathway inhibitor D-1MT indicates a new treatment strategy for refractory and relapse B cell malignancies.
Acknowledgements
Not applicable.