This study presents a protocol of live-3D-cell immunocytochemistry applied to a pediatric diffuse midline glioma cell line, useful to study in real-time the expression of proteins on the plasma membrane during dynamic processes like 3D cell invasion and migration.
Cell migration and invasion are specific hallmarks of Diffuse Midline Glioma (DMG) H3K27M-mutant tumors. We have already modeled these features using three-dimensional (3D) cell-based invasion and migration assays. In this study, we have optimized these 3D assays for live-cell immunocytochemistry. An Antibody Labeling Reagent was used to detect in real-time the expression of the adhesion molecule CD44, on the plasma membrane of migrating and invading cells of a DMG H3K27M primary patient-derived cell line. CD44 is associated with cancer stem cell phenotype and tumor cell migration and invasion and is involved in the direct interactions with the central nervous system (CNS) extracellular matrix. Neurospheres (NS) from the DMG H3K27M cell line were embedded into the basal membrane matrix (BMM) or placed onto a thin coating layer of BMM, in the presence of an anti-CD44 antibody in conjunction with the antibody labeling reagent (ALR). The live-3D-cell immunocytochemistry image analysis was performed on a live-cell analysis instrument to quantitatively measure the overall CD44 expression, specifically on the migrating and invading cells. The method also allows visualizing in real-time the intermittent expression of CD44 on the plasma membrane of migrating and invading cells. Moreover, the assay also provided new insights into the potential role of CD44 in the mesenchymal to amoeboid transition in DMG H3K27M cells.
The ability of tumor cells to evade and disseminate through the surrounding tissue is a hallmark of cancer1. In particular, tumor cell motility is a characteristic feature of malignant tumors, whether it is a metastatic tumor type such as breast2 or colorectal cancer3 or a locally invasive type such as diffuse glioma4,5.
Imaging has a central role in the investigation of many aspects of tumor cell phenotypes; however, live-cell imaging is definitely to be preferred when studying dynamic cellular processes such as migration and invasion, when changes in morphology and cell-cell interaction6,7 occur and can be more easily examined over time. For live-cell imaging, different optical microscopy systems can be used, from phase contrast to confocal fluorescent microscopes, and image acquisition performed over a short or long period of time on an inverted microscope equipped with a chamber for temperature and CO2 control, or in high-content image analysis systems which have built-in chambers, or alternatively in image systems that can sit in the incubator without the need to disturb the cells during the whole duration of the experiment. The choice of the system used is often dictated by a number of factors such as resolution needed, length of the overall acquisition time and time intervals, vessel type used and throughput of the assay (single chamber or multi-well plate), the sensitivity of the cells used (precious and/or rare cells) and phototoxicity of the cells if fluorophores are present.
With regard to fluorescent imaging in live mode, this can be achieved by transducing cells for the expression of fluorescent proteins either for stable expression or as an inducible system8, by transient cell transfection, or by using cell dyes which are now available for live-cell labeling7, for live-cell tracking as well as for labeling subcellular organelles9.
A useful approach has been recently developed for live-cell immunocytochemistry, where an antibody recognizing a surface marker of choice can be bound to a labeling reagent, and upon addition to the culture media, cells expressing the specific marker can be readily imaged in real-time by live-cell imaging. The visualization and quantification of marker expression using such a system can be easily achieved when cells are grown in two-dimensional (2D) culture conditions10.
In this study, we optimized protocols for live-3D-cell immunocytochemistry invasion and migration of pediatric diffuse midline glioma (DMG) patient-derived cells11,12. DMG are highly aggressive brain tumors affecting children, for the vast majority associated with the driver mutation K27M in histone H3 variants. DMG arise in the brain stem and the midline regions of the central nervous system (CNS) and are characterized by a highly infiltrative nature. This invasive capacity has been shown to be at least in part mediated by the intratumor heterogeneity and the cancer-stem-like features of DMG cells7.
To exemplify our assays, an antibody labeling reagent (ALR) was used in combination with an antibody for CD44. CD44 is a transmembrane glycoprotein and adhesion molecule expressed on stem-cell and other cell types, associated with cancer stem cell phenotype and tumor cell migration and invasion13. The protocols include the sample preparation, the image acquisition in brightfield and fluorescent mode, and the analysis on a live-cell analysis instrument that allowed to quantitatively measure in real-time the overall CD44 expression on the DMG cell membrane during 3D invasion and migration. The assays also allowed the possibility to visualize the intermittent fluorescent signal of CD44 on individual cells while migrating and invading. Interestingly an effect of the anti-CD44 antibody was also observed, which potentially acting as a blocking antibody, also seemed to reduce cell migration and invasion as well as to induce a switch of the invasion pattern from a collective mesenchymal-like to a more single-cell amoeboid-like phenotype.
This protocol follows the guidelines of the institutions' human research ethics committees.
NOTE: This study was performed using Incucyte S3 and/or SX5 Live-Cell Analysis Instrument (referenced as live cell analysis instrument).
1. Generation of reproducibly sized tumor spheroids
NOTE: The protocol (section 1) described by Vinci et al. 20157,12, was used as reported below, with some modifications:
2. Preparation of the ALR/antibody complex and setup for the invasion assay
NOTE: For the antibody labeling procedure, the antibody labeling dyes protocol10 for live-cell Immunocytochemistry is used with some modifications, as reported below. For the invasion assay, the protocol previously described by Vinci et al. 201512 is followed.
3. Preparation of the ALR/antibody complex and setup for the migration assay
NOTE: For the antibody labeling procedure, the Labeling Dyes protocol10 for Live-Cell Immunocytochemistry is used.
4. Live-cell analysis instrument setting for image acquisition
5. Live-cell analysis instrument setting for image analysis
Live-3D-Cell Immunocytochemistry protocol for invasion and migration is summarized in a straightforward and reproducible workflow in Figure 1. By seeding the DMG cells in ULA 96-well round-bottom plates, reproducible sized NS are obtained and used in the steps displayed. When the NS have reached the ideal size of ~300 µm (approximately 4 days post-seeding) the invasion12 and migration14 assays are initiated. The addition of the ALR/antibody complex together with the background suppressor in the medium of the individual NS, allows following the specific marker expression on the cell membrane, in live imaging and over time. The surface marker expression during the cell invasion and migration is easily monitored at intervals starting from t = 0 up to 96 h using the live-cell analysis instrument. This imaging system allows a fully automated image analysis.
A primary patient-derived cell line, QCTB-R059, was used to exemplify the invasion and migration proprieties of pediatric DMG tumor dissemination. QCTB-R059 was originally indicated as a pediatric thalamic glioblastoma (GBM) cell line15. Later on, it has been indicated as H3-K27M thalamic glioma cell line16 or diffuse midline glioma (DMG) H3-K27M cell line11, following the 2016 World Health Organization classification of brain tumors with the introduction of DMG H3F3A K27M-mutant as a new entity17.
CD44, an adhesion molecule known to be involved in cell migration and invasion, was investigated. CD44 is expressed by QCTB-R059 cells as demonstrated by confocal images of immunofluorescent (IF) staining on 3D cell migration onto (Figure 2), and invasion into (Figure 3) BMM.
Taking into consideration that 3D invasion and migration are both non-static processes, we thought to investigate the expression of CD44 over time when cells are in movement. To do this we employed the live-cell immunocytochemistry assay and adapted the protocol for 3D assays. By using the ALR in complex with an anti-CD44 antibody, we are able to follow in real-time the expression of CD44 when the protein is expressed on the cell membrane while the cells evade the neurospheres and spread onto and into the BMM.
The live-3D-cell immunocytochemistry allows visualizing CD44 expression (Supplementary Video 1 and Supplementary Video 2). The representative frames of the time-lapse, for both migration and invasion (Figure 4A,B), show more in detail the intermittent expression of CD44 on the cell membrane. In particular, it is possible to see the green fluorescent signal to be on (green circle) and then off (black circle) on the same cell observed over time, suggesting that the expression of CD44 is on and off while cells are migrating and invading.
The migration and invasion processes are followed over 96 h, and as shown in Figure 5, QCTB-R059 cells show a high level of CD44, demonstrating that overall the expression observed with the live-cell immunocytochemistry is in line with the expression of CD44 obtained by IF shown on confocal images in Figure 2 and Figure 3. Interestingly though, we also noticed that when the anti-CD44 antibody is used on live cells, it affects cell morphology, inducing a transition from mesenchymal-like to amoeboid-like invasion. It induces a reduction of the invasive and migratory capacity of these cells (Supplementary Figure 1). We cannot exclude, though, that the reduction in cell migration and invasion observed is also in part due to an inhibition of cell proliferation.
The automated image analysis performed on the live-cell analysis instrument shows the quantification of CD44 expression and its increase over time, measured by the overall green fluorescent signal associated with the ALR (Figure 5B,C) for both migration and invasion. The quantifications are achieved as exemplified in Figure 5B and Figure 5C, with the automated image analysis set to segment all the area covered by the CD44 green-migrated cells (Figure 4B) and all the spread area covered by the CD44 green-invaded cells but excluding the neurosphere core (Figure 5C).
Figure 1: Schematic workflow of Live-3D-Cell Immunocytochemistry Assays. The workflow shows the steps involved in the 3D invasion and migration live imaging methods, including representative images of pediatric primary DMG patient-derived cells (QCTB-R059) after invasion into (top panel; t = 96 h) and migration onto (bottom panel; t = 96 h) BMM. Bars = 800 µm. Please click here to view a larger version of this figure.
Figure 2: CD44 expression in 3D tumor cell migration. Representative immunofluorescent confocal images of CD44 expression in primary DMG patient-derived cells (QCTB-R059) upon migration onto BMM. Timepoint = 96 h (red:CD44; blue: nuclei). Scale bars: 500 µm (upper panel) and 200 µm (lower panel). Please click here to view a larger version of this figure.
Figure 3: CD44 expression in 3D tumor cell invasion. Representative immunofluorescent confocal images of CD44 expression in primary DMG patient-derived cells (QCTB-R059) upon invasion into BMM. Time point = 96 h (red:CD44; blue: nuclei). Scale bars: 250 µm (upper panel) and 100 µm (lower panel). Please click here to view a larger version of this figure.
Figure 4: CD44 expression over time. Selected frames of QCTB-R059 migration (A) and invasion (B) time-lapse. Images were obtained on the live-cell imaging instrument. Green circle indicates the expression of CD44, black circle indicates no CD44 expression on the cell membrane of the same cell observed over time. Scale bars: 200 µm (A) and 100 µm (B). Please click here to view a larger version of this figure.
Figure 5: Live-3D-Cell Immunocytochemistry Assays for CD44: migration and invasion. (A) Representative brightfield, fluorescent (ALR with anti-CD44 antibody), and merge images of QCTB-R059 cell immunocytochemistry migration, and invasion (96 h) are shown. Scale bars: 400 µm. (B) Quantification of CD44 overall expression relatively to migration (B) and invasion (C), determined by ALR-anti-CD44 image analysis on the live-cell imaging instrument. The curves show the Green Mean Intensity of CD44 expression over time. Values are mean ± SD. The two small figures in the plots display the segmentation applied for the analysis of the migration (B) where all area was considered and the invasion (C) for which the NS core part was excluded. Please click here to view a larger version of this figure.
Supplementary Figure 1: Effect of anti-CD44 antibody on cell morphology and degree of cell motility. Representative images of QCTB-R059 invasion and migration assay show the effect of anti-CD44 antibody used for the live-3D-cell immunocytochemistry. Cells display a reduced invasion and migration capacity as well as the transition from a more mesenchymal-like to amoeboid-like invasion pattern between the negative control (without anti-CD44 antibody) and CD44 (plus anti-CD44 antibody). Lower panel shows higher power magnification displaying a more clear view on the morphological appearance of the cells in the absence and the presence of the anti-CD44 antibody (white arrows). Scale bars: 800 µm upper panel and 100 µm lower panel. Please click here to download this File.
Supplementary Video 1: Time-lapse video of QCTB-R059 3D cell migration on BMM in the presence of anti-CD44 antibody. Fluorescent green signal, representing the expression of CD44 on the cell membrane, is visualized over time by the conjugation of the anti-CD44 antibody with ALR. Please click here to download this Video.
Supplementary Video 2: Time-lapse video of QCTB-R059 3D cell invasion in BMM in the presence of anti-CD44 antibody. Fluorescent green signal, representing the expression of CD44 on the cell membrane, is visualized over time by the conjugation of the anti-CD44 antibody with ALR. Please click here to download this Video.
The live-3D-cell immunocytochemistry we have adopted here for pediatric DMG invasion and migration could be easily adapted also for other highly invasive tumor cell types, including breast and colon cancer cell lines.
Different from previously performed live-2D-cell immunocytochemistry assays10, when working in 3D, it is suggested to pay attention to some critical steps. In particular, for the invasion assay we describe, it is advised to add the ALR/antibody mix directly to the medium with NS in each well, prior to the addition of the BMM, and not in the BMM or on top of the BMM once gelled. This is to allow a good mix of the reagents with the medium and ensure more direct access of the reagents to the cell surface. Moreover, to ensure a better quality of the imaging, although the protocol includes the use of the BSR, we advise using phenol red-free medium and BMM.
Another point to consider for the live-cell immunocytochemistry is that any antibody binding an extracellular membrane protein on live cells may affect the protein function by altering its conformation or by occupying the binding site of a ligand or of a protein on an adjacent cell, therefore acting as a “blocking agent”18,19. While this approach may be useful as a therapeutic strategy19, it may not be the primary goal of any experimental setup. Therefore, prior to performing a large set of experiments, different antibodies binding distinct epitopes of the same protein should be tested to also verify any potential “blocking” effect. In this study, we used a specific antibody to follow in real-time the expression of CD44 on the cell membrane of a highly aggressive pediatric DMG cell line in 3D invasion and migration. The protocol used allowed us to quantitatively measure the expression of CD44 over time on cells invading and migrating. Interestingly, in the presence of the anti-CD44 antibody, we also noted a reduction in cell motility in comparison to the cells with the ALR but in the absence of the antibody. We cannot exclude though also an inhibitory effect on cell proliferation. The acquisition of a different invasion pattern with a switch from mesenchymal-like to amoeboid-like cell morphology20 was also observed in the presence of anti-CD44 antibodies. These unexpected results suggest that blocking CD44 may contribute to mesenchymal to amoeboid transition in pediatric DMG.
With regard to the limitations of this protocol, taking into consideration the resolution of the CCD camera of the Incucyte Live-Cell Analysis Instrument and its limited z-stack capability, the setup we present for the live-3D-cell immunocytochemistry assays may be used as a preliminary approach, on a large scale multi-well format, before moving on to a more in-depth analysis using either more powerful fluorescent imaging systems (e.g., confocal microscopes and high-content imaging system with a confocal modality such as the Operetta CLS or the Opera Phoneix) or a more refined approach for studying in real-time the expression of a surface protein via a reporter assay21.
A broader application of the live-3d-cell immunocytochemistry presented here as a monoculture, could be a 3D co-culture assay established to image and analyze in real-time direct cell-cell interactions. In this case, two different ALR could be used with different fluorophores to bind proteins specifically expressed on the cell membrane on different cell types (e.g., tumor cell and immune cells). In this case, direct cell-cell contact may be analyzed with live imaging by the co-localization of the two different ALR/antibody complexes.
The authors have nothing to disclose.
We would like to thank Dr. Silvia Soddu and Dr. Giulia Federici (Unit of Cellular Networks and Molecular Therapeutic Targets, IRCCS-Regina Elena National Cancer Institute, Rome, Italy) for access to the IncuCyte S3 Live Cell Imaging System in the preliminary set up of the imaging protocol. Moreover, we thank Bernadett Kolozsvari for the technical advice. The study was supported by the Children with Cancer UK grant (16-234) and The Italian Ministry of Health Ricerca Corrente. M Vinci is a Children with Cancer UK Fellow. R Ferretti is a recipient of Fondazione Veronesi Fellowship (2018 and 2019). The authors acknowledge Fondazione Heal for their support and the Children's Hospital Foundation for funding the Queensland Children's Tumor Bank.
96 Well TC-Treated Microplates | Corning | 3595 | size 96 wells, polystyrene plate, flat bottom |
Accutase | Euroclone | ECB3056D | solution for neurosphere dissociation |
Burker chamber | Mv medical | FFL16034 | cell counting chamber |
CD-44 (156-3C11) | Cell Signaling Technology | 3570 | Mouse mAb IgG2a |
Corning Matrigel Matrix | Corning | 356237 | Basement Membrane Matrix (BMM), Phenol Red-free, LDEV-free |
Fabfluor-488 Antibody Labeling Dye | Incucyte | 4743 | Antibody labelling reagent (ALR): Mouse IgG2a 488 antibody for Live-Cell Immunocytochemistry |
Incucyte S3 and/or SX5 Live-Cell Analysis Instrument | Sartorius | – | The Incucyte S3 and/or SX5 Instrument is used for real-time cell monitoring and surveillance, cell health and viability, migration and invasion, plus a wide range of phenotypic cell-based assays. |
Inverted Microscope | – | any inverted microscope | |
Opti-Green Background Suppressor Reagent | Incucyte | 6500-0045 | Backgroung suppressor reagent (BSR) |
Tumor stem cell (TSM) medium | – | – | growth cell medium (see reference in the text for details) |
Ultra-Low Attachment Multiple Well Plate | Corning Costar | 7007 | size 96 well, round bottom clear |