Here, we present a protocol for the preparation and the stereotaxic administration of allogeneic human lymphocytes in immunodeficient mice carrying orthotopic human primary brain tumors. This study provides a proof-of-concept for both the feasibility and the antitumor efficacy of intrabrain-delivered cellular immunotherapies.
Glioblastoma multiforme (GBM), the most frequent and aggressive primary brain cancer in adults, is generally associated with a poor prognosis, and scarce efficient therapies have been proposed over the last decade. Among the promising candidates for designing novel therapeutic strategies, cellular immunotherapies have been targeted to eliminate highly invasive and chemo-radioresistant tumor cells, likely involved in a rapid and fatal relapse of this cancer. Thus, administration(s) of allogeneic GBM-reactive immune cell effectors, such as human Vϒ9Vδ2 T lymphocytes, in the vicinity of the tumor would represents a unique opportunity to deliver efficient and highly concentrated therapeutic agents directly into the site of brain malignancies. Here, we present a protocol for the preparation and the stereotaxic administration of allogeneic human lymphocytes in immunodeficient mice carrying orthotopic human primary brain tumors. This study provides a preclinical proof-of-concept for both the feasibility and the antitumor efficacy of these cellular immunotherapies that rely on stereotactic injections of allogeneic human lymphocytes within intrabrain tumor beds.
GBM (WHO grade IV astrocytoma), is the most frequent and aggressive primary brain cancer in adults. In spite of aggressive treatments that combine surgery and radio-chemotherapy, GBM remains associated with an extremely poor prognosis (median survival of 14.6 months and a 2-year-mortality > 73%)1. This evidences that few efficient therapeutic advances have been validated over the last decade2. Among candidates for the design of more effective therapeutic strategies3,4,5, immunotherapies6 are currently explored to track and eliminate highly invasive and radio/chemo-resistant tumor cells, suspected for their key contribution to rapid and fatal tumor relapse7. Various potential immunological targets were identified and proposed for immunotherapies, involving natural or modified αβ or ϒδ T lymphocytes such as GBM-specific tumor antigens or stress-induced molecules8,9,10. The possibility to administrate selected GBM-reactive immune cell effectors represents a unique opportunity to deliver elevated amounts of effector lymphocytes directly into the site of residual malignancy. To support this strategy, we have recently shown that models based on immunodeficient mice carrying orthotopic primary human GBM xenografts faithfully recapitulate the development of brain tumors in GBM patients9,11. Moreover, these models were used to demonstrate the strong antitumor efficiency of adoptively transferred allogeneic human Vϒ9Vδ2T lymphocytes.
This protocol describes the critical experimental steps for achieving stereotactic immunotherapies of brain tumors, such as GBM, based on the adoptive transfer of allogeneic T lymphocytes. The article shows: (i) the amplification of therapeutic allogeneic immune effector T lymphocytes, such as human Vϒ9Vδ2T lymphocytes; (ii) the preparation of these effector T lymphocytes for injection; (iii) the procedure for stereotactic administration within the brain, near the tumor. This article also provides insight into the behavior of these cellular effectors after stereotactic injection.
The therapeutic approach presented here is based on the injection of 20 x 106 effector cells per dose for each brain tumor-bearing immunodeficient mouse. An initial in vitro expansion step is required to produce large quantities of immune cells. Therefore, non-specific cell expansions are performed using phytohemagglutinin (PHA-L) and irradiated allogeneic feeder cells: peripheral-blood mononuclear cells (PBMCs) from healthy donors and Epstein Barr Virus (EBV)-transformed B-lymphoblastoid cell lines (BLCLs), derived from PBMCs by in vitro infection with EBV-containing culture supernatant from the Marmoset B95-8 cell line, in the presence of 1 µg/mL cyclosporin-A.
GBM-reactive effector immune cells are compared and selected from in vitro assays9. These effector cells are activated and amplified using standard protocols, according to their nature (e.g., human Vγ9Vδ2 T lymphocytes9 or human anti-herpes virus αβ T lymphocytes12) with a minimum purity of > 80%, as routinely checked by cytometric analysis. The cell expansion procedure detailed below applies to various human lymphocyte subsets.
The following procedure involving animal subjects was performed according to institutional guidelines (Agreement #00186.02; regional ethics committee of the Pays de la Loire [France]). Human PBMCs were isolated from the blood collected from informed healthy donors (Etablissement Français du Sang Nantes, France). All steps are performed under sterile conditions.
1. Non-specific Expansion of Cytotoxic Effector T Lymphocytes
2. Pre-operative Effector Cells Preparation
3. Stereotactic Injection
This study describes the strategy of adoptive transfers of cellular immune effector cells within the brain of tumor-carrying mice, based on stereotactic injections performed directly within the tumor bed.
To minimize any risk of brain injury associated with a large injection volume, the effector cell suspension needs to be concentrated (20 x 106 cells in 15 – 20 µL of PBS). To check whether this cell concentration step might affect the viability of the effector cells, these cells were prepared according to the described protocol and loaded into the microsyringe. The effector cells were collected immediately, or 10 min after loading them into the microsyringe. The cells were stained with propidium iodide (PI), a fluorescent DNA intercalating agent that is not permeant to live cells and analyzed by flow cytometry at different timepoints (0, 24, and 72 hours). The results show that the preparation and the loading into the microsyringe do not significantly affect the viability of effector cells for at least 24 hours (Figure 2A and 2B). At 72 hours, a slight, but non-significant, increase of PIpositive cells was observed (14% compared to 11% for unloaded cells). In a similar way, the antitumor reactivity of effector cells that were prepared and maintained on ice for 3 hours was analyzed. Effector cells were cocultured with brain tumor cells for 4 hours in presence of an anti-CD107a mAb. The upregulation of the activation marker CD107a, similar to the value obtained in the control conditions, indicates that the reactivity of effector cells is not affected by the preparation and the loading into the microsyringe (Figure 2C).
To evaluate whether effector cells survive and move within the brain parenchyma following their intra-tumoral implantation, 20 x 106 effector cells were injected into the tumor site of mice carrying a brain tumor (GBM-111). One week later, the brains were collected, fixed, sectioned, and stained for hematoxylin, eosin, and safran coloration (HES) and anti-human CD3 mAb (IHC). HES coloration identified the structure of the brain tumors (Figure 3, left panel). The CD3 staining identified and localized the effector immune T lymphocytes (here, human Vϒ9Vδ2 T cells) (Figure 3, right panel). Interestingly, effector T lymphocytes were detected around the tumor (Figure 3, upper right panel), in the tumor core (Figure 3, middle right panel), but also in the contralateral hemisphere (Figure 3, bottom right panel). Moreover, the function of human T lymphocytes isolated from the mouse brain 48 hours after their injection (4×106 αβ T cells), which represent 2% of the brain cells, was analyzed. The results indicate that collected brain-injected effector allogeneic αβ T cells expand and proliferate upon a non-specific PHA-feeder cells activation performed in the presence of IL-2 (Figure 4).
Together, these results show that effector T cells prepared and administrated under these procedures survive for hours in the brain and can patrol within the tumor and healthy brain tissues. These procedures have been used for assessing the antitumor efficiency of therapeutic stereotactic administrations of allogeneic human resting Vϒ9Vδ2 T cells to control the development of human GBM brain tumors9,11. These studies evidence that injections of allogeneic human effector T lymphocytes in the tumor bed significantly improve the survival of mice carrying brain tumors. Interestingly, surviving mice did not carry detectable tumor cells, thus indicating a complete tumor rejection.
Figure 1: Picture of the main anatomical landmarks on the mouse skull. This includes sagittal (red line) and coronal (blue line) sutures and their intersection (Bregma) used to orient the site of injection. The scale bar is indicated. Please click here to view a larger version of this figure.
Figure 2: Survival and activation of prepared effector T lymphocytes. Effector cells (here, resting human peripheral Vϒ9Vδ2 T cells) were amplified and prepared according to the described procedure. Staining of lymphocytes with propidium iodide (PI), a DNA intercalating fluorescent compound that is not permeant to live cells, and functional activation were performed. (A) This panel shows a representative plot of forward (FSC-H) versus side (SSC-H) scatter gating of effector lymphocytes (left panel). The histogram shows the PI staining of gated control lymphocytes (right panel). The percentage of PIpositive lymphocytes is indicated. (B) This panel shows the percentage of dead lymphocytes (PIpositive) collected immediately (light blue line) or after 10 min (dark blue line) in the microsyringe, measured at the indicated timepoints. As control, unloaded prepared cells were used (grey line). (n = 3; mean ± SD; ns = not significant.) (C) CD107a surface mobilization was measured by flow cytometry on either Vδ2positive control cells (unprepared) or prepared lymphocytes maintained on ice (prepared + 3 h) following a coculture with target human primary brain tumor cells. The percentage of lymphocytes is indicated in each cytometric quadrant. Please click here to view a larger version of this figure.
Figure 3: Detection and localization of effector T lymphocytes within the brain of tumor-bearing mice. One week after glioblastoma brain tumor implantation, effector immune cells (here, resting human peripheral Vϒ9Vδ2 T cells) were injected into the brains of mice. One week after the immunotherapeutic treatment, the brains were collected, fixed, and sectioned. Brain sections were colored for immunohistochemistry analysis (hematoxylin, eosin, and safran [HES] coloration) (left panel) or stained with anti-human CD3 antibody (right panel). The results shown are representative of three independent experiments. The scale bar is indicated.
Figure 4: Activation of effector T lymphocytes collected from the brain of treated mice. Resting human T lymphocytes (here, 4 x 106 human αβ T lymphocytes) were injected into the brains of NSG mice. After 48 hours, the brains were collected and dissociated. The percentage of human brain-infiltrating T lymphocytes was measured by flow cytometry using an anti-human CD3 antibody (bottom panel). The collected cells were seeded (10 cells/well) in 96-well U-bottomed plates and stimulated (PHA-feeder cells and IL-2) for 20 days (16 doubling cycles). Note, on day 10, 13.3 x 106 of human T lymphocytes were obtained (right panel).
An adoptive transfer of selected native or engineered immune effector cells represents a promising approach to efficiently treat tumors, such as infiltrative brain cancers, taking care of limiting reactivities against non-transformed cells15,16,17,18. However, the central nervous system, which comprises the brain, has a particular immune status, notably due to the existence of the blood-brain barrier and the lack of a classical lymphatic drainage system19,20. These physiological features affect tissue trafficking and might compromise systemic injections of immune cells. To overcome these hindrances, intraparenchymal injections have been explored on the principle that antitumor cells are rather locally delivered, closely to the tumor site, as for microspheres that contain pharmacological compounds21,22. On one hand, the limited dilution of lymphocytes within the organ might improve their antitumor efficiency, but it can also amplify deleterious mechanical or tumor adaptation effects, such as tissular compression, which is developing along brain tumor growth. This implies that this procedure requires small volumes of immunotherapeutic injections. This issue is even more critical in animal experimental models in which brain tumors cannot be surgically excised. This article describes a therapeutic approach for the preparation and the local delivery of brain tumor-specific cellular effectors, based on stereotactic injection(s) of allogeneic human T lymphocytes.
This study shows the preparation and the stereotaxic delivery of allogeneic human Vϒ9Vδ2 T lymphocytes in immunodeficient NSG mice carrying human GBM xenografts. The first stage of this protocol describes a simple procedure for amplifying allogeneic T lymphocytes from PBMCs of healthy donors, using a standard nonspecific PHA-feeders-IL2 stimulation that produces large quantities of pure effector T lymphocytes, allowing their therapeutical utilization23,24. The second stage of this article focuses on the preparation of resting T lymphocyte suspensions on the day of the stereotaxic administration. A particular focus was placed on this important step that requires a high cellular density that should not affect the viability and the function of the selected effector T lymphocytes. Finally, regarding in vivo experiments, the preparation of effector T lymphocytes and their injection within the tumor core is associated with their dissemination, not only within the tumor but also in the surrounding brain tissues, highlighting their particular ability to patrol and to track invasive tumor cells. Importantly, these preparation and injection methods retain the ability of these T lymphocytes to be activated within the brain upon a specific recognition of brain tumor cells. Altogether, these compelling characteristics ensure the ability of T lymphocytes to specifically and efficiently target and eliminate deep infiltrative brain tumor cells which are a hallmark of GBM25. Of note, a special care has to be taken during the injection and the removal of the microsyringe to minimize any brain lesion or effector cell leaks.
In conclusion, this article describes an efficient procedure for delivering large amounts of allogeneic human anti-tumor lymphocytes, such as resting human Vϒ9Vδ2T lymphocytes, within the vicinity of brain tumors. Importantly, this therapeutical procedure is not accompanied with adverse effects either on the transferred T lymphocytes (e.g., viability, reactivity) or on the brain tissues. Recent studies, based on murine orthotopic models of primary human GBM, have demonstrated that Vϒ9Vδ2T lymphocytes efficiently target GBM cells, including tumor cells which have deeply infiltrated the brain parenchyma9,11. These elements open opportunities for the development of novel adoptive T lymphocytes transfer procedures that could be applied in the first instance in mice carrying orthotopic brain tumors and, then, in clinical studies in GBM patients.
The authors have nothing to disclose.
The authors thank the staff of the University Hospital animal facility (UTE) of Nantes for animal husbandry and care, the cellular and tissular imaging core facility of Nantes University (MicroPICell) for imaging, and the Cytometry facility (Cytocell) from Nantes for their expert technical assistance. This work was funded by INSERM, CNRS, Université de Nantes, Institut National du Cancer (INCa#PLBio2014-155), Ligue Nationale contre le Cancer (AO InterRegional 2017), and the European consortium ERA-Net Transcan2 (Immunoglio). The team is funded by the Fondation pour la Recherche Medicale (DEQ20170839118). This work was realized in the context of the LabEX IGO and the IHU-Cesti programs, supported by the National Research Agency Investissements d’Avenir via the programs ANR-11-LABX-0016-01 and ANR-10-IBHU-005, respectively. The IHU-Cesti project is also supported by Nantes Metropole and the Pays de la Loire Region. The authors thank Chirine Rafia for providing help in correcting the manuscript.
PBMCs | from 3 different healthy donors | ||
BLCLs | from 3 different donors | ||
Roswell Park Memorial Institute medium (RPMI) | Gibco | 31870-025 | |
FCS | Dutscher | S1810-500 | |
L-glutamine | Gibco | 25030-024 | |
penicillin/streptomycin | Gibco | 15140-122 | |
IL-2 | Novartis | proleukin | |
PHA-L | Sigma | L4144 | |
Stereotaxic frame | Stoelting Co. | 51600 | |
Mouse adaptator for stereotaxic frame | Stoelting Co. | 51624 | |
microsyringe pump injector | WPI | UMP3-4 | |
NanoFil Syringe | WPI | NF34BV-2 | |
NSG mice | Charles River | NSGSSFE07S | |
Ketamine | Merial | Imalgène 1000 | |
Xylazine | Bayer | Rompur 2% | |
Scissors | WPI | 201758 | |
Forceps | WPI | 501215 | |
OmniDrill 115/230V | WPI | 503598 | |
Vicryl 4-0 | Ethicon | VCP397H | |
Xylocaine | Astrazeneca | 3634461 |