This article describes surface labeling and ex ovo tissue culture in the early chick embryo. Techniques amenable to time-lapse bright field, fluorescence, and optical coherence tomography imaging are presented. Tracking surface labels with high spatiotemporal resolution enables kinematic quantities such as morphogenetic strains (deformations) to be calculated in both two and three dimensions.
Embryonic epithelia undergo complex deformations (e.g. bending, twisting, folding, and stretching) to form the primitive organs of the early embryo. Tracking fiducial markers on the surfaces of these cellular sheets is a well-established method for estimating morphogenetic quantities such as growth, contraction, and shear. However, not all surface labeling techniques are readily adaptable to conventional imaging modalities and possess different advantages and limitations. Here, we describe two labeling methods and illustrate the utility of each technique. In the first method, hundreds of fluorescent labels are applied simultaneously to the embryo using magnetic iron particles. These labels are then used to quantity 2-D tissue deformations during morphogenesis. In the second method, polystyrene microspheres are used as contrast agents in non-invasive optical coherence tomography (OCT) imaging to track 3-D tissue deformations. These techniques have been successfully implemented in our lab to studythe physical mechanisms of early head fold, heart, and brain development, and should be adaptable to a wide range morphogenetic processes.
1. General Experimental Preparation
2. DiI Labeling of HH Stage 5 Embryos using Magnetic Iron Particles
3. Polystyrene Microsphere Labeling for Optical Coherence Tomography (OCT) Imaging of HH11-12 embryos
4. Data Acquisition
* The following methods are well suited for the fluorescence labeling technique described above. However, with the microsphere labeling technique, beads can dislodge when transferring samples between incubators and imaging systems. In this case it is best to proceed to step 4.2 (which avoids sample movement/agitation altogether). In both methods, embryos are cultured while submerged in liquid culture medium. If the embryo is not completely submerged (as is the case with some conventional tissue culture techniques), tissue geometry is greatly altered by abnormally high surface tension loads and strain distributions will fail to accurately capture normal 3-D morphogenesis3, 4. Moreover, submerging the tissue prevents the bright contrast observed at the liquid-gas interface during OCT imaging from obscuring embryonic morphology and marker coordinates.
5. Representative Results:
Labels are tracked automatically (using Volocity, PerkinElmer) or manually (using the Manual Tracking plugin in ImageJ, NIH) in each experiment. In the fluorescent labeling technique, we use the Matlab routine gridfit to fit 2D surfaces through the marker coordinates, which enables morphogenetic surface strains to be calculated6, 7. Standard equations are used to transform these values into a relevant embryonic coordinate system. Alternatively, in the OCT technique, surfaces are generated from segmented image volumes (obtained via standard software such as Matlab or Caret8) and strains can be calculated in the direction of maximum or minimum curvature of the sample9.
We used our iron particle technique to label and track the motion of ectodermal cells during head fold formation in the early chick embryo. As depicted in Figure 2A, fluorescently labeled cells were distributed across the entire embryo. Bright field and fluorescent images of the embryo were captured at different intervals during ex ovo culture. The motions of the tracked labels (Fig. 2B,C) were then used to calculate the evolving morphogenetic strain distributions during head fold formation (Fig. 2D-F).
Similarly, our polystyrene microsphere technique was used to track tissue movements at the mid-hindbrain boundary of the early chick brain. As shown in Figure 3 B-D, bead motions were tracked for 6 hours and strains characterizing tissue deformation in the longitudinal and circumferential directions were calculated from marker coordinates. Note that this method is capable of handling distinctly 3-D deformations as beads tend to stick to all sides of the inner lumen of the brain (Fig. 3A).
Figure 1. Schematic of set-up for time-lapse tissue culture.
Figure 2. Quantifying 2-D tissue deformations in the early chick embryo using tracked fluorescent labels. (A) Merged bright field/fluorescent image of HH stage 5 embryo after removal of iron particles. DiI-labeled ectodermal cells (red) are distributed across the entire embryo. (B,C) The motion of labeled cells was tracked using ImageJ. Label displacements were calculated in embryo coordinates (X,Y). Note that A and B are the same image. (D-F) Contour plots of evolving longitudinal strain distributions during head fold formation. Longitudinal Lagrangian strains were calculated relative to HH stage 5 (i.e., A and B) after (D) 75 min, (E) 120 min, and (F) 165 min of incubation. These strains characterize the length changes of line elements originally oriented along the Y-axis (in the stage 5 embryo). Further details on using tracked label coordinates to calculate morphogenetic strains can be found in Filas et al. (2007)6 and Varner et al. (2010)7. Note that C and F are the same image. Scale bar = 500 μm.
Figure 3. Quantifying 3-D tissue deformations in the early chick brain. (A) Polystyrene microspheres that adhere to the dorsal (black arrow) and ventral (white arrow) sides of the brain tube are easily discernible from surrounding tissues in transverse cross section.Ventral views of OCT reconstructions show microsphere locations near the mid-hindbrain boundary at (B) HH12 and after (C) 6 hr of incubation (M: midbrain, H: hindbrain). Several groups of beads are outlined to highlight the overall deformation of the tissue. (D) Longitudinal (Ezz) and circumferential (Eθθ) Lagrangian strains were calculated from these deformations. Positive longitudinal and negative circumferential strains at the mid-hindbrain boundary reflect an axial lengthening and circumferential shortening of this region during culture. Further details on calculating morphogenetic strains for complex surfaces during morphogenesis can be found in Filas et al. (2008)9. Scale bar = 200 μm.
Two tissue labeling techniques are presented for the ex ovo culture of early chick embryos. The first uses fluorescent lipophilic dyes delivered via magnetic iron particles to simultaneously label hundreds of cells. However, this method is currently not compatible with optical coherence tomography, as fluorescent dyes generally show little contrast from surrounding tissues using OCT10. Hence, we show an alternative technique using polystyrene microspheres to label tissues for time-lapse OCT analysis. This technique yields 3-D datasets but care must be taken not to dislodge the beads from the tissue. Using either one of these methods in the appropriate experimental setting should provide reliable tissue labeling and ex ovo development in early embryos. Both methods use readily available, relatively low-cost materials (e.g., iron powder, polystyrene microspheres) and enable tissue deformation to be quantified during morphogenesis. The tissue markers in both cases can be easily and reproducibly applied to embryonic tissues and do not require specialized equipment, making these experiments accessible to newcomers in the field. Visualizing and analyzing the resulting data (e.g., by computing morphogenetic strain maps) should help illuminate the mechanics of morphogenesis in your model system6, 7, 9, 11, 12.
The authors have nothing to disclose.
This work was supported by NIH grants R01 GM075200 and R01 HL083393 (LAT). We acknowledge fellowship support for BAF from NIH T90 DA022871 and the Mallinckrodt Institute of Radiology, and for VDV from grant 09PRE2060795 from the American Heart Association.
Material / Reagent | Company | Catalogue number | Comments |
DMEM – high glucose | Sigma-Aldrich | D5796 | |
Penicillin/Streptomycin/Neomycin | Sigma-Aldrich | P4083 | |
Chicken Serum | Invitrogen | 16110-082 | |
Dulbecco’s Phosphate Buffered Saline | Sigma-Aldrich | D1408 | 10X |
Whatman #2 Filter Paper | Whatman | 1002 090 | 90mm diameter |
Glass Micropipettes | World Precision Instruments (WPI) | TW150-6 | 1.5mm inner diameter |
DiI | Invitrogen | D-282 | |
Iron Reduced | Mallinckrodt Baker Inc. | 5320 | |
10 μm Diameter Microspheres (black) | Polysciences Inc. | 24294 | |
Delta T Dish (for time lapse culture) | Bioptechs | 04200415B | 0.17mm thick, black |
Delta T4 Culture Dish Controller | Bioptechs | 0420-4-03 | |
Mini-Pump Variable Flow Device | Fisher Scientific |