We have established a cortical orthotopic glioblastoma model in mice for intravital two-photon microscopy that recapitulates the biophysical constraints normally at play during the growth of the tumor. A chronic glass window replacing the skull above the tumor enables the follow-up of the tumor progression over time by two-photon microscopy.
Glioblastoma multiforme (GBM) is the most aggressive form of brain tumors with no curative treatments available to date.
Murine models of this pathology rely on the injection of a suspension of glioma cells into the brain parenchyma following incision of the dura-mater. Whereas the cells have to be injected superficially to be accessible to intravital two-photon microscopy, superficial injections fail to recapitulate the physiopathological conditions. Indeed, escaping through the injection tract most tumor cells reach the extra-dural space where they expand abnormally fast in absence of mechanical constraints from the parenchyma.
Our improvements consist not only in focally implanting a glioma spheroid rather than injecting a suspension of glioma cells in the superficial layers of the cerebral cortex but also in clogging the injection site by a cross-linked dextran gel hemi-bead that is glued to the surrounding parenchyma and sealed to dura-mater with cyanoacrylate. Altogether these measures enforce the physiological expansion and infiltration of the tumor cells inside the brain parenchyma. Craniotomy was finally closed with a glass window cemented to the skull to allow chronic imaging over weeks in absence of scar tissue development.
Taking advantage of fluorescent transgenic animals grafted with fluorescent tumor cells we have shown that the dynamics of interactions occurring between glioma cells, neurons (e.g. Thy1-CFP mice) and vasculature (highlighted by an intravenous injection of a fluorescent dye) can be visualized by intravital two-photon microscopy during the progression of the disease.
The possibility to image a tumor at microscopic resolution in a minimally compromised cerebral environment represents an improvement of current GBM animal models which should benefit the field of neuro-oncology and drug testing.
Glioblastoma multiforme appears as the most aggressive form of brain tumor in adults with a median survival of 12 months and a 5-years survival rate of 5%. Clinical management relies on surgery, radiotherapy and chemotherapy often used in combination. However, the effects of these treatments remain palliative1-3.
Up to now, most of neuro-oncology studies rely on techniques that are only able to provide a static view and performed on large cohorts of tumor bearing animals sacrificed at different time-points (see for example4,5). The recent development of follow-up methods based on intravital imaging allows studying glioma growth and the interactions between tumor cells and their pathophysiological microenvironment on the same animal over time. This opens the way to exclusive piece of information that was so far unachievable6. Transgenic animals expressing fluorescent tags in cells of interest may be used to study specific interactions between tumor cells and e.g. neurons in this paper.
Over the past decade, intravital two-photon microscopy7 has become a gold standard in fundamental neuro-oncology studies and preclinical trials8,9 for its ability to perform deep intravital observation of mouse brain (>500 µm below the dura-mater) with a micrometric spatial resolution10. Using intravital two-photon microscopy with orthotopical animal models implanted with a chronic cranial window11, it is possible to follow the tumor progression over time on the same mouse9,12.
One of the major drawbacks of these previously published animal models is however that they do not mimic the physical constraints that govern tumor growth as the dura-mater is not sealed after the injection of the cell suspension9,13,14. Glioma cells may leak in the extradural space transforming an orthotopic glioma model into a heterotopic one.
The animal model presented here consists in the injection of a spheroid of fluorescent glioma cells in the cerebral cortex at a depth of 200 µm followed by the sealing of the dura-mater with a cross-linked dextran gel hemi-bead and histo-compatible glue. The tumor growth is then restricted to the brain parenchyma that maintains pathophysiological physical constraints. A chronic glass window implanted above the tumor allows an easy optical access for intravital two-photon microscopy. Using transgenic animals expressing fluorescent tags in cells of interest it is possible to perform a follow-up of the glioma growth over time and to study its interaction with its microenvironment (here with neurons and vasculature highlighted with fluorescent dextrans).
All experimental procedures were performed in accordance with the French legislation and in compliance with the European Community Council Directive of November 24, 1986 (86/609/EEC) for the care and use of laboratory animals. The research on animals was authorized by the Direction Départementale des Services Vétérinaires des Bouches-du-Rhône (license D-13-055-21) and approved by the ethical committee of Provence Cote d'Azur n°14 (Project 87-04122012).
1. Spheroids Preparation
2. Spheroid and Window Implantation
3. Post-operative Care and Preparation for Imaging
Once the surgical protocol is performed (Figure 1), animals can be observed by means of fluorescent microscopy over weeks until sacrifice. An inflammatory reaction may be observed after the surgery that disappears within one or two weeks. Tumor growth can be observed by various microscopy techniques including fluorescent macroscopy and two-photon microscopy (Figure 2). Example images depicted here were realized on a fluorescence macroscope and a two-photon microscope coupled to a femtosecond pulsed infrared tunable laser and home-modified to allow animal positioning below the 20X water immersion objective (1.0NA).
The tumor development was estimated over time using a fluorescence macroscope combining oblique illumination for bright field reflectance imaging and epifluorescence emission (Figures 2A-2C). Thresholding of the fluorescence easily gives access to the tumor area while superficial vasculature detach as dark structures out of the white background of the visible light image. On the other hand intravital two-photon microscopy allows z-sectioning and orthogonal reconstructions of large fields of view with subcellular resolution (see dendrites in Figure 2D). Using transgenic animals expressing fluorescent proteins in cells of interest (for example, Thy1-CFP mice that express the Cyan Fluorescent Protein in neurons, green pseudo-color in Figure 2D) it is possible to observe the tumor progression in its neuronal micro-environment. However, it must be noted that tumor cell density can limit the photon penetration depth as well as the emitted photon detection due to increased diffusion. This is clearly visible in Figure 2D as the spatial resolution starts to decrease at 200 µm below the dura-matter in the tumor but not in the healthy brain parenchyma. Moreover, the second-harmonic generation signal arising from organized patterns of non-centrosymetric molecules such as type I and III collagen provides a signature of the dura-mater (cyan pseudo-color in Figure 2D and arrows). It must be noted that with our surgical protocol, the tumor grows below the dura-mater in the brain parenchyma where its development is governed by the physical constraints of the brain. The sealing of the dura-mater with a cross-linked dextran gel hemi-bead prevents the escape of glioma cells in the extradural space.
When followed by intravital two-photon microscopy the tumor progression between different time-points can be compared at cellular resolution. Imaging the same area at D16 and D27 post-surgery (Figure 3A) we found some blood vessels as the one highlighted by arrows that were stable over a period of 11 days; on the other hand the tumor front clearly infiltrated into the healthy parenchyma over the same period. To evaluate tumor progression, its margin observed at D27 was outlined with a yellow dotted line and overlaid on the picture taken at D16. A 520 µm shift of the front edge was observed consistently with the reported proliferative profile of the GL261 glioma model15. The patterns of the vascular network inside the tumor were very different on D16 and D27 indicating that significant vascular remodeling occurs in pathological areas (Figure 3A). At the tumor margins, we also found cases of glioma cells detaching from the tumor core as expected from an invasive tumor. The microscopic resolution of our images showed not only that these cells emit cytoplasmic protrusions to sense the environment but also that they can use blood vessels as a matrix for their progression (arrowheads and z-reconstruction in Figure 3B). Two-photon microscopy thus allows to study cellular interactions in a truly physiological context. Using transgenic animals where each cell population of interest is identified by its own fluorescent reporter it becomes possible to look for the existence of physical interactions between tumor cells and their environment as key regulators of disease progression (Figure 3B, right).
Figure 1. Overview of the surgery. The craniotomy is performed on the left parietal bone (A). The limits of the left parietal bone are outlined with a dotted line and the removed part is filled in yellow (B). Once the craniotomy performed, the parietal cortex is exposed (C), the area for the injection of the spheroid is selected taking care to avoid main blood vessels (arrow in D) and a 26 gauge needle is used to open the dura-mater (arrow in E). The spheroid is injected and the hole in the dura-mater is clogged by a cross-linked dextran gel hemi-bead sealed with cyanoacrylate glue (arrow in F). A glass coverslip is glued to the bone while keeping contact with the brain. The coverslip is further cemented to the exposed skull to ensure long term solid attachment of the cranial window for chronic observations (G-I). The whole protocol is briefly schematized in J. Click here to view larger image.
Figure 2. Tumor growth is restricted to the brain parenchyma and can be monitored over time. A-C. Tumor growth (GL261 cell line expressing DsRed2) is observed by macroscopy on the day of surgery (DO, A) and at days 21 (B) and 27 (C). The arrowhead in C highlights a drop of dental cement that was not removed during step 2.3.5. D. Orthogonal reconstructions obtained from a 3 x 3 field acquired from 0 to 300 µm below the dura-mater in a Thy1-CFP mouse bearing a GL261 tumor by two-photon microscopy at D20 post-surgery. Tumor (red), dura-mater observed by second-harmonic generation (arrows, cyan) and neurons (green). Note that the tumor grows below the dura-mater in the brain parenchyma. Click here to view larger image.
Figure 3. Vascular remodeling and tumor cells invasion can be monitored over time with cellular spatial resolution by intravital two-photon microscopy. A. GL261 tumor (red) observed at D16 and D27 post-surgery after an intravenous injection of Cascade-blue dextran 70 kDa to highlight blood vessels. Dotted line: tumor margin at D27; arrows: example of blood vessels that can be used as landmarks to localize the same area at D16 and D27. B. Individual tumor cells (red) invade the healthy brain parenchyma (neurons; green) using blood vessels as a matrix for invasion (arrowheads). Images were taken at depth ranging from 100 to 150 µm below the dura-matter. Dotted line: region where the orthogonal reconstruction depicted in the bottom center panel was realized. Click here to view larger image.
This approach allows the use of optical imaging methods to monitor over days and weeks the growth of an orthotopically implanted glioma. The same animal can subsequently be subjected to virtually any brain imaging modality during the course of the pathology; yet the two-photon microscopy specific preparation offers the unique opportunity to achieve subcellular resolution inside the brain of the living animal. Our protocol presents the advantage to enforce the tumor growth into the cerebral parenchyma rather than extra-durally as it happens if glioma cells can leave the central nervous system through the damaged dura. In absence of physical constraints exerted by the brain environment, tumor cell can indeed proliferate faster as they do not need to infiltrate the parenchyma16. Previously published methods did not attempt to seal the dura mater after grafting which was all the more problematic than they injected suspension of cells rather than spheroids9,13,14. Suspensions of cells are indeed impossible to apply focally as individual cells are systematically seeded along the injection tract when withdrawing the pipette from the depth to the dura-mater.
The current protocol can be applied on various mouse strains including transgenic and immunosuppressed animals. Various glioma cell lines can be grafted and visualized, including cells directly derived from human biopsies as long as they are transfected to stably express a fluorescent reporter prior grafting. Surgical protocol can be typically performed in 1 hr and cohorts of animals can quickly be obtained. This approach can thus be used to study vascular remodeling, the effects of pharmacological treatments on angiogenesis and tumor growth, as well as the recruitment of immune cells into the tumor. When needed glioma cell migration and dynamic interactions between glioma cells and the microenvironment can be observed in real time for several hours.
As imaging can be done over depths of 500 µm it is possible to cover most of the tumor volume initially implanted 200 µm below the dura-mater. To ensure maximal imaging depth special care must be taken during the surgery to create physical contact between the dura-mater and the glass coverslip. This indeed reduces the formation of scar tissue that would otherwise compromise optical clarity. Absolute sterility of the surgical environment should also be observed to limit the post-surgery inflammation that transiently blurs the window. We anyway advise to wait 15 days between the surgery and the first observation session to obtain optimal imaging conditions. Thereafter, window was found to remain clear up to one year after the surgery on sham operated animals that underwent the whole protocol except the injection of the tumor spheroid.
In conclusion, the protocol described here represents an improvement of existing orthotopic glioma mouse model dedicated to chronic intravital two-photon microscopy. The superficial injection of a tumor spheroid, the clogging of the injection track with a cross-linked dextran gel hemi-bead and the sealing of the dura-mater enforce tumor development in the brain environment that truly recapitulates the physical constraints normally governing disease progression. Not only this method will serve fundamental studies on glioma pathophysiology but it will also improve the relevance of pharmacological results during preclinical trials.
The authors have nothing to disclose.
The authors warmly thank Dr. KK Fenrich, Dr. M-C. Amoureux, P. Weber and A. Jaouen for helpful discussions; M. Hocine, C. Meunier, M. Metwaly, S. Bensemmane, J. Bonnardel, the staff of the animal facility at IBDML and the staff of the PicSIL imaging platform at IBDML for technical support. This work was supported by grants from Institut National du Cancer (INCA-DGOS-INSERM6038) to GR, Agence Nationale de la Recherche (ANR JCJC PathoVisu3Dyn), Fédération pour la Recherche sur le Cerveau (FRC) to FD, by fellowships from the Fédération de la Recherche Médicale and Cancéropole PACA to CR.
Drill | Dremel (Germany) | 398 | any high quality surgical bone drill would suffice |
Drill burr (#1/4 Carbide Round Burr) | World Precision Instruments (USA) | 501860 (#1/4) | also sold by Harvard Apparatus |
Tissue scissors | World Precision Instruments (USA) | 14395 | |
Dumont tweezers M5S | World Precision Instruments (USA) | 501764 | |
Dental cement | GACD (USA) | 12-565 & 12-568 | |
Cyanoacrylate | Eleco-EFD (France) | Cyanolit 201 | |
Glass capillaries without filament | Clark Electromedical Instruments (UK) | GC100-15 | |
Microliter syringe (25 µl) | Hamilton (USA) | 702 | |
Micromanipulator | World Precision Instruments (USA) | Kite-R | |
T derivation (3-way stopcock – Luer lock) | World Precision Instruments (USA) | 14035-10 | |
Stereotactic frame (mouse adaptor) | World Precision Instruments (USA) | 502063 | |
Glass coverslips | Warner Instruments (USA) | CS-5R (64-0700) | |
Cross-linked dextran gel (Sephadex) G50 Coarse 100-300 µm beads | Available from various suppliers including Sigma (Germany) | ||
Eye ointment | TVM (France) | Ocry-gel | |
Fluorescence macroscope | Leica MZFLIII (Germany) | also sold by other companies | |
Two-photon microscope | Zeiss LSM 7MP (Germany) | also sold by other companies (Nikon, …) | |
Infrared tunable femtosecond laser (Maï-Taï) | Spectra Physics (USA) | also sold by other companies |