primitive streak region delaminate through apical constriction
(leading to the so-called “bottle shape”) followed by retraction of
the apical process and extrusion on the basal side [4, 5]. Cells
subsequently acquire a mesenchymal shape with an array of filopodia and migrate away from the streak [6].
Technically, live imaging analysis of whole mouse embryos
during gastrulation remains challenging. Embryos have already
implanted, so culture conditions must compensate for the removal
from the maternal environment. Mouse embryos cannot be fully
constrained and are very photosensitive. Rapid embryo growth,
complex optical properties, and the requirement of optically scattering, auto-fluorescent serum for proper development are further
obstacles for high-resolution imaging.
Recent advances in fluorescent reporters, embryo culture, and
microscopy have nevertheless allowed visualization and tracking of
entire embryonic populations in real time thanks to improvements
in spatial and temporal resolution, physical coverage of the embryo,
and long-term imaging capability through reduction of phototoxicity [7, 8]. Laser scanning confocal and multiphoton microscopy
are the most common tools for imaging embryo development.
Both provide excellent spatial resolution; multiphoton microscopy
allows higher penetration depth, minimizes exposure as it relies on
low energy photons of high wavelength, and reduces the background signal due to multiphoton absorption. Light sheet imaging
further diminishes phototoxicity; in particular, an adaptive light
sheet imaging approach has recently allowed tracking single cell
behavior for 48 h between E6.5 and E8.5 [9].
Important insight has been obtained through qualitative analysis of live imaging data, but quantitative analysis remains a bottleneck. Nuclear reporters are best suited for cell tracking, and several
algorithms have been successfully developed to segment nuclei
[9]. Cell shape information requires the use of membrane reporters, and it is practically extremely complex to properly segment cell
contours when all cells are fluorescent. One way to circumvent this
issue is to simplify the model through single cell labeling. This can
be achieved through either injection or electroporation of nucleic
acids into individual or groups of cells [10], use of photomodulatable proteins that can be activated or converted in a region of
interest, or regulation of the expression of fluorescent proteins
through lineage-specific inducible promoters. Here we describe a
protocol allowing labeling a portion of epiblast cells at a chosen
ratio in order to observe cells’ shape changes as they undergo EMT
during mouse embryo gastrulation.
We use hydroxy (OH)-tamoxifen-induced recombination of
Rosa
mT/mG [11] through Sox2Cre-ER
T2 [12] (see below) activation to achieve mosaic labeling of the epiblast. Cells in the PS
acquiring either a round or a bottle shape can be segmented and
tracked to identify their destiny (within the epiblast or mesoderm
136
Wallis Nahaboo et al.
(leading to the so-called “bottle shape”) followed by retraction of
the apical process and extrusion on the basal side [4, 5]. Cells
subsequently acquire a mesenchymal shape with an array of filopodia and migrate away from the streak [6].
Technically, live imaging analysis of whole mouse embryos
during gastrulation remains challenging. Embryos have already
implanted, so culture conditions must compensate for the removal
from the maternal environment. Mouse embryos cannot be fully
constrained and are very photosensitive. Rapid embryo growth,
complex optical properties, and the requirement of optically scattering, auto-fluorescent serum for proper development are further
obstacles for high-resolution imaging.
Recent advances in fluorescent reporters, embryo culture, and
microscopy have nevertheless allowed visualization and tracking of
entire embryonic populations in real time thanks to improvements
in spatial and temporal resolution, physical coverage of the embryo,
and long-term imaging capability through reduction of phototoxicity [7, 8]. Laser scanning confocal and multiphoton microscopy
are the most common tools for imaging embryo development.
Both provide excellent spatial resolution; multiphoton microscopy
allows higher penetration depth, minimizes exposure as it relies on
low energy photons of high wavelength, and reduces the background signal due to multiphoton absorption. Light sheet imaging
further diminishes phototoxicity; in particular, an adaptive light
sheet imaging approach has recently allowed tracking single cell
behavior for 48 h between E6.5 and E8.5 [9].
Important insight has been obtained through qualitative analysis of live imaging data, but quantitative analysis remains a bottleneck. Nuclear reporters are best suited for cell tracking, and several
algorithms have been successfully developed to segment nuclei
[9]. Cell shape information requires the use of membrane reporters, and it is practically extremely complex to properly segment cell
contours when all cells are fluorescent. One way to circumvent this
issue is to simplify the model through single cell labeling. This can
be achieved through either injection or electroporation of nucleic
acids into individual or groups of cells [10], use of photomodulatable proteins that can be activated or converted in a region of
interest, or regulation of the expression of fluorescent proteins
through lineage-specific inducible promoters. Here we describe a
protocol allowing labeling a portion of epiblast cells at a chosen
ratio in order to observe cells’ shape changes as they undergo EMT
during mouse embryo gastrulation.
We use hydroxy (OH)-tamoxifen-induced recombination of
Rosa
mT/mG [11] through Sox2Cre-ER
T2 [12] (see below) activation to achieve mosaic labeling of the epiblast. Cells in the PS
acquiring either a round or a bottle shape can be segmented and
tracked to identify their destiny (within the epiblast or mesoderm
136
Wallis Nahaboo et al.