A robust protocol to monitor neural populations by time-lapse video-microscopy followed by software-based post-processing is described. This method represents a powerful tool to identify biological events in a selected population during live imaging experiments.
Understanding the mechanisms that control critical biological events of neural cell populations, such as proliferation, differentiation, or cell fate decisions, will be crucial to design therapeutic strategies for many diseases affecting the nervous system. Current methods to track cell populations rely on their final outcomes in still images and they generally fail to provide sufficient temporal resolution to identify behavioral features in single cells. Moreover, variations in cell death, behavioral heterogeneity within a cell population, dilution, spreading, or the low efficiency of the markers used to analyze cells are all important handicaps that will lead to incomplete or incorrect read-outs of the results. Conversely, performing live imaging and single cell tracking under appropriate conditions represents a powerful tool to monitor each of these events. Here, a time-lapse video-microscopy protocol, followed by post-processing, is described to track neural populations with single cell resolution, employing specific software. The methods described enable researchers to address essential questions regarding the cell biology and lineage progression of distinct neural populations.
In order to develop new and more effective therapeutic strategies to regenerate neural populations, we must first understand the basic mechanisms that maintain cells with a regenerative neural potential. Pursuing this goal requires a comprehensive knowledge of the factors that regulate the balance between quiescence, proliferation/differentiation, the mode and timing of division, cell cycle length, migratory capacities, viability, etc. Although it is a technical approach that has been employed for many years1, live imaging and direct observation still remain the best option to monitor the events listed above. As opposed to many other approaches centered on end-point readouts, live imaging and single cell tracking provide information throughout the length of an experiment2,3,4,5,6. Thus, the addition of temporal resolution allows cell death, heterogeneous cell behavior, or cell fate decisions, as well as many other critical events to be identified that might otherwise pass unnoticed. Ideally, these features of cells should best be monitored at the single cell level in vivo, where both intrinsic (cell autonomous) and extrinsic (cell niche) cues are taken into account.
However, although in the in vitro situation events occur in an environment that does not reproduce the natural milieu, the low-density culture conditions typically used in these protocols are more suitable to reveal intrinsic characteristics of the cells. Moreover, a more simplistic control of the surrounding milieu, by simply modifying the growth medium, may constitute a valuable tool to investigate the individual role of each extrinsic factor that defines the neural niche, as well as environmental factors that may be induced in pathological scenarios7,8,9,10,11,12,13. Therefore, when correctly configured, as in the protocol proposed here, live imaging provides a feasible in vitro solution to address most of the questions previously enumerated.
In brief, this protocol describes the hardware, software, culture conditions, and the main steps required to successfully perform a live imaging experiment followed by single cell tracking. This approach offers valuable information that helps to reveal fundamental aspects of the biology, and of the lineage progression, of multiple neural populations.
The following sections describe the steps required to perform live imaging followed by single cell tracking of multiple neural populations (Figure 1). All the procedures involving animals described in this protocol must be carried out in accordance with the guidelines of the International Council for Laboratory Animal Science (ICLAS).
Figure 1. Scheme illustrating the principal experimental steps of the procedure, i.e.: cell culture, live imaging, PICC and data collection, single cell tracking, and the final outcome. The steps are numbered according to the work-flow of the protocol. Please click here to view a larger version of this figure.
1. Cell Culture: Isolation and Plating of the Selected Neural Population or Cell Lineage
NOTE: In conjunction with this protocol, examples of its application to distinct cell populations are given to validate its utility to analyze the biology of neural cells. These include: Adult Neural Stem Cells (aNSCs) derived from the mouse SubEpendymal Zone (SEZ) (for a detailed isolation protocol see14); Postnatal cortical astrocytes to study neuronal reprogramming (for a detailed isolation protocol see15); Postnatal cerebellar astrocytes (for a detailed isolation method see16); and Mouse Neuro-2a Neuroblastoma Cell line (N2a).
2. Live Imaging by Time-lapse Video-microscopy
3. Post-imaging Immunocytochemistry (PICC), Data Collection, and Processing
4. Single Cell Tracking
5. Final Outcome
The method described enables critical questions regarding the cell biology of multiple neural populations to be resolved. For instance, it has been possible to monitor the progression of the neurogenic and oligodendrogliogenic lineage of aNSCs7,8,14,18. By tracking single aNSCs and their progeny (Figure 2A, B), it was possible to demonstrate that aNSCs isolated in vitro maintain their neurogenic nature, mostly generating neuroblasts, and that they follow a sequence proposed in vivo19 but not previously demonstrated at the single cell level. Moreover, this culture system allowed asymmetric cell divisions to be visualized for the first time in the aNSC lineage from the SEZ (Figure 2B), providing a unique model to study NSC self-renewal8,14. Likewise, and irrespective of the lineage analyzed, it was possible to obtain valuable data regarding cell growth, the rounds of division, cell viability, or cell cycle length.
Figure 2. Example of aNSCs isolated from the SEZ, and analyzed by live imaging and single cell tracking. Phase contrast images depict the progression of the clone at different time points (day-h: min). The final image corresponds to the post-imaging immunocytochemistry (PICC) for Glial fibrillary acidic protein (GFAP, red), βIII-tubulin (green) and 4',6-diamidino-2-phenylindole (DAPI, blue). (A) Analysis of symmetric neurogenic trees through different rounds of amplifying divisions to generate post-mitotic neuroblasts. Red arrows point to the cells included in the symmetric trees. On the right, the lineage trees corresponding to the clones and generated by the tTt software are displayed. (B) Example of a progenitor generating an asymmetric neurogenic tree, with one branch undergoing amplifying divisions to produce neuroblasts while the other gives rise to quiescent GFAP positive cells through a potential self-renewal event. On the right, the lineage tree generated by the tTt software is displayed. In all the lineage trees: "N" depicts post-mitotic neuroblasts; "G", quiescent GFAP-positive cells; "X", cell death; and "?" a lost cell. Scale bar represents 50 µm. Please click here to view a larger version of this figure.
Live imaging and single cell tracking analysis also provides an accurate readout of the migratory capacities of a neural population. Such information was obtained from postnatal cerebellar astrocytes submitted to a scratch wound assay20, generating information regarding the average distance traveled by the astrocytes when closing the wound (Figure 3). Moreover, it was possible to see that some of the astrocytes divided during the healing process, while others remain unaltered throughout the experiment. Strikingly, those that divided seem to exhibit more prolific migratory behavior than their non-dividing counterparts (traveling twice as far on average). This phenomenon suggests a very interesting heterogeneity in the astrocytes capacity to form a scar upon injury, which would have been diluted out in the read-out of a classical end-point analysis experiment.
Figure 3. Analysis of the migratory behavior of postnatal cerebellar astrocytes in a scratch wound assay. Phase contrast images depict the wound at different time points (day-h: min). Lineage trees, generated by tTt software, illustrate the representative behavior, in terms of cell division, of the astrocytes while closing the wound. Histogram shows the average distance traveled by the astrocytes analyzed by single cell tracking (mean ± S.E.M.). Scale bar represents 50 µm. Please click here to view a larger version of this figure.
Another interesting feature of the time-lapse video-microscopy experiments is the capacity to compare proliferation and differentiation in a cell population. We tested N2a cells plated under conditions that promote either proliferation (in the presence of 10% fetal bovine serum (FBS)) or differentiation (in the presence of 0.5% FBS + 10 µM arachidonic acid). It was possible to follow the lineage progression of these cells under proliferative conditions (Figure 4A), whereas differentiating cells do not proliferate and they form neurites (Figure 4B). Remarkably, single cell tracking allowed colonies with different proliferation capacities to be distinguished and neurite elongation (and retraction) to be evaluated, providing precise and quantitative data that can subsequently be exported.
Figure 4. The monitoring of N2a cell biology in proliferation (A) or differentiation conditions (B). Phase contrast images depicting the progression of the clone at different time points (day-h: min). The final image corresponds to the post-imaging immunocytochemistry (PICC) for α-tubulin (green). (A) Single cell tracking allows the rounds of division to be monitored, as well as the heterogeneity in the proliferative response of different cells. On the right, the lineage tree generated by the tTt software illustrates the proliferative behavior of N2a cells. (B) Cells under differentiation conditions exit the cell cycle and generate neurites, a process that can be effectively measured by post-imaging analysis. Single cell tracking, represented by the lineage tree on the right, illustrates how N2a cells exit cell cycle and stop the cell division under differentiation conditions. Scale bar represents 50 µm. Please click here to view a larger version of this figure.
Finally, live imaging and single cell tracking is extremely useful to monitor morphological and molecular changes when cells are submitted to reprogramming. Live imaging of postnatal astrocytes transduced with Achaete-scute homolog 1 (Ascl1) provides valuable data regarding the morphological changes that occur during reprogramming or the blockage of cell division when astrocytes are being reprogrammed (see Figure 5). Moreover, when Ascl1 transduction is combined with the transduction of a construct encoding for Green fluorescence protein (GFP) under the control of the Double Cortin (DCX) promoter, it is possible to define the precise time point when neuronal specific markers begin to be expressed in the reprogrammed cells (Figure 5). Time-lapse video-microscopy also allows the number of cells that successfully complete reprogramming to be quantified and compared to the cells that die during this process. Monitoring such events led to the identification of the critical "checkpoints" in the cells that were successfully reprogrammed9.
Figure 5. Analysis of postnatal cortical astrocytes subjected to neuronal reprogramming. Reprogramming was induced by transduction with pro-neurogenic Ascl1-Red Fluorescence protein (RFP) vectors. Neuronal conversion was monitored by co-transduction with a vector encoding GFP under the control of a DCX promoter. Phase contrast images show the progression of reprogramming at different time points (day-h: min). Fluorescence images of RFP and GFP expression, respectively. Live imaging and single cell tracking allowed crucial events to be followed, such as morphological changes, the absence of cell division during reprogramming, cell death, and the precise time when reprogrammed cells start to express neuronal markers can be defined. Scale bar represents 80 µm. Please click here to view a larger version of this figure.
One of the most important values of live imaging is the possibility to perform accurate lineage tracing, elucidating critical aspects of lineage progression in a neural population. Lineage tracing is defined as the identification and monitoring of all the progeny of a single progenitor, from the founder of the clone to the subsequent clone formed21. Remarkably, alternative methods employed for lineage tracing (e.g., viral transduction or multicolor reporter constructs21) have a critical drawback, whereby the final outcome is based on still pictures and it does not necessarily constitute the whole sequence. This means that cell death, heterogeneity in the behavior of the cell population, dilution, spreading or poor efficiency of the markers, along with other important handicaps, lead to incomplete or incorrect read-outs of the results2. Furthermore, live imaging enables the researcher to analyze important features of the biology of neural populations, such as the mode and timing of cell division, cell growth, migration, proliferation versus differentiation, cell cycle length, neurite formation, complexity and length, cell fate selection (differentiation), or conversion (reprogramming).
In addition, live imaging can be easily complemented with other analysis intended to obtain data from single cells such as, for example, RNA sequencing. However, to achieve combined benefit from both live imaging and other techniques requires that those cells previously monitored in the movies are later re-identified and individually collected for the secondary analysis. This can be achieved by using microscopes that include positional coordinates, by applying fluorescent reporters for specific cells or analyzing the distribution of groups of cells as references. Indeed, the combination of the transcriptome profile and behavior of individual cells may represent a powerful route to elucidate new molecular cues involved in the biology of cells.
One of the main problems that can compromise a live imaging experiment is an inadequate cell culture density. As indicated previously, at high density the excess of debris or poor dissociation (clump formation) may affect the quality and spatial resolution of the images, making single cell tracking unfeasible. Therefore, the conditions of the distinct cell populations under study should be adjusted to the lowest number of cells possible without compromising the viability of the cell culture.
The frequency of image acquisition is also crucial and should be carefully adjusted, especially when fluorescence illumination is used. Over-exposure to transmitted and especially fluorescence light may compromise cell viability. Alternatively, an excessive delay between the capture of the images may interfere with the temporal resolution of the analysis.
Another critical step during the live imaging experiment is the periodic adjustment of focusing. Failure in the correct setting/re-setting of the focal distance may hinder single cell tracking. Moreover, it is necessary to carefully check that the incubation chamber preserves the adequate temperature, humidity, and CO2 levels, amending undesired variations that may induce cell death.
Finally, once the PICC has been performed, it is important to properly retrieve the xyz zero position prior to the last round of image acquisition. Incorrect re-setting of the xyz zero position will make it difficult to match the phase-contrast and immunofluorescence images, impeding the identification of the cell progeny.
Although this approach has many positive facets, some limitations to the live imaging of neural populations still persist. For instance, the low cell density required to perform successful single cell tracking of aNSCs makes it impossible to employ biochemical assays, such as Western blotting14. Additionally, monitoring fast dividing populations like cerebellar astrocytes or N2a cells is temporally restricted as it is often too difficult to track cells as the cultures near confluence. Furthermore, many culture methods, as well as the inherent biological restrictions associated with the isolation of cells, often compromise cell viability over long periods, limiting the duration of the live imaging experiments. Finally, isolating cells from their natural environment has both positive and negative effects. Cells isolated from their physiological niche may fail to receive important signals that modulate their behavior, while at the same time, it represents a powerful means to test the effect of those signals individually in the lineage progression of specific neural populations.
Given the limitations described above, it is clear that the perfect methodological scenario would be to perform live imaging and single cell tracking experiments under normal physiological conditions in vivo. However, current techniques are unable to follow single cells for long periods of time in deep regions of the brain2. Therefore, the future of live imaging should focus on overcoming this limitation, aiming to fully analyze the cell biology of single cells in vivo with the most minor interference possible of the physiological environment3.
The authors have nothing to disclose.
We thank Beatriz Gascon for her assistance and art work in Figure 1. We also thank Dr. C. Norris for his assistance. The work presented here was supported by research grants, "Red de excelencia Consolider-Ingenio Spanish Ion Channel Initiative" (BFU2015-70067REDC), MEC (BFU2014-53654-P), BRADE-CM (S2013/ICE-2958), UCM-Santander (PR26/16-18B-3) and Fundación Ramon Areces Grant program (PR2018/16-02). Felipe Ortega acknowledges the Ramon y Cajal Program of the Spanish Ministry of Economy and Competitiveness (MEC: RYC-2013-13290).
Poly-D-lysine | Sigma | P0899 | Working solution 0.02 mg mL-1 |
24 wells plate | Falcon | 352047 | |
Dulbecco’s modified Eagle’s medium (DMEM): F12 Nutrient Mixture medium (-L Glutamine) | Invitrogen | 21331-020 | |
DMEM High Glucose medium | Sigma | D6546 | |
Bovine Serum Albumin | Sigma | A6003 | |
Triton X-100 | Merck | 11869 | non-ionic surfactant |
Mouse anti-β III Tubulin | Sigma | T8660 | |
Rabbit anti-GFAP | DakoCytomation | Z0334 | |
Mouse anti-α Tubulin | Sigma | T5168 | |
Anti-Mouse FITC | Jackson Laboratories | 715-095-150 | |
Anti-Rabbit Cy3 | Jackson Laboratories | 711-165-152 | |
Brightfield/Phase contrast/fluoresence microscope | Nikon | TE-2000-E | |
CFI PLAN FLUOR DLLL 10X objetives | Nikon | Ref 280MRH10101 | |
CFI SUPER PLAN FLUOR ELWD AMD 20X objetives | Nikon | Ref 280MRH48230 | |
pE-300 LED fluorescence | Cool LED | Ref Number 1981 | |
310M-201 Incubation system (temperature) | OKO-Lab | Serial Nº VOF007307 | |
Pro-ScanII Motorized stage system | Prior | Serial Nº 60018 | |
High precision microscope camera version 4.2 | ANDOR Zyla | VSC-03650 | |
Specifc software for live imaging with timelapse module: NIS-Elements AR4.5 | Nikon | NIS-Elements AR4.5 -Hasp ID: 13CE819E | |
OKO touch Incubation system (CO2) | OKO-lab | Serial number 1716 | |
Murine Neuro-2a Neuroblastoma Cell line | ATCC | ATCCCCL131 | |
HEPES buffer solution 1 M | Invitrogen | 15630-056 |