RESEARCH ARTICLE Molecular Reproduction & Development (2015)
Multispectral Labeling of Embryonic Cells with Lipophilic Carbocyanine Dyes MARIA VOLNOUKHINy
AND
BRUCE P. BRANDHORST*
Department of Molecular Biology and Biochemistry, University Drive Simon Fraser University, Burnaby, British Columbia, Canada
SUMMARY Incubation of hatched Strongylocentrotus purpuratus sea urchin embryos or larvae with suspensions of the carbocyanine dyes DiI, DiO, and DiD resulted in the random labeling of membranes of some ectodermal epithelial cells and blastocoelar cells, producing a range of differentially colored cells that can be tracked during development. Simultaneous application of soluble Vybrant1 preparations of the three dyes resulted in similar labeling of each cell. Dye labeling of the ectoderm was nearly eliminated by deciliation and some ciliated squamous epithelial cells adjacent to labelled cells were refractory to Vybrant1 dye uptake irrespective of concentration or duration of treatment, together suggesting local variation in the properties of cell membranes or cilia. Furthermore, single cells possessing distinctive morphological features were detected.
Corresponding author: Department of Molecular Biology and Biochemistry 8888 University Drive Simon Fraser University Burnaby, BC Canada V3H3K3. E-mail: [email protected]
y
Deceased.
Grant sponsor: Natural Sciences and Engineering Research Council of Canada Discovery Grant; Grant number: 45977
Mol. Reprod. Dev. 2015. ß 2015 Wiley Periodicals, Inc. Received 16 September 2014; Accepted 4 March 2015
INTRODUCTION Marking blastomeres by the application of particles or organic dyes onto living embryos is critical for the generation of cell-lineage diagrams and fate maps in animals (reviewed by Kretzschmar and Watt, 2012). More recently, injection of dyes into embryonic blastomeres or the expression of marker proteins have been used to track cell lineages and fates; such analysis has been further refined by the expression of fluorescent reporter proteins driven by specific transgenic promoters. Combined with modern fluorescence microscopic techniques, these cell-labeling approaches allow for the tracking of patches or clones of cells in live embryos over time and space, although individual cells can be difficult to distinguish. The development of Brainbow for mouse embryos, for example, provides dis-
ß 2015 WILEY PERIODICALS, INC.
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mrd.22477
tinctive multicolor labeling of adjacent neural cells, whose individual fates and spatial interactions can then be tracked using fluorescence microscopy (Livet et al., 2007). Brainbow depends on Cre-mediated combinatorial expression of multiple recombinant fluorescent proteins in transgenic embryos; as such, this method has been modified and extended to provide multispectral marking of a range of cell types in Drosophila (Boulina et al., 2013; Worely et al., 2013), mice (Snippert et al., 2010), and zebrafish (Gupta and Poss, 2012; Pan et al., 2013), allowing clonal analyses at various developmental stages and facilitating axonal tracking in transparent embryos and larvae. While multispectral labeling of embryonic cells based on the induced recombination of fluorescent-protein transgenes is a powerful tool, it depends on the creation and
Molecular Reproduction & Development
manipulation of appropriate stable, transgenic lines. Such a genetic approach is not practical for some organisms, such as the sea urchin, because of long generation times and the difficult husbandry of juveniles. On the other hand, the sea urchin has a well-defined gene regulatory network for early embryonic development (Davidson, 2006; Peter and Davidson, 2010) that allows for the transient expression of fluorescent proteins in a cell-type-specific manner in living embryos, which are well-suited for 4-dimensional fluorescence microscopy (Barsi et al., 2014). When contiguous cells express the same fluorescent protein(s), however, distinguishing and tracking individual cells can be challenging. Here we report that the brief suspension of sea urchin embryos in combinations of lipophilic carbocyanine dyes suspended in seawater results in random labeling of individual cells in a range of colors. This approach allows for clonal analysis of the labelled cells and tracking of their descendents as well as identification of rare cell types and possible cellular interactions that may be difficult to detect using reporter-gene fusions. This simple method should be useful for multispectral, random labeling of embryonic cells in a range of organisms that are amenable to fluorescence microscopy.
RESULTS AND DISCUSSION Often used for axon tracking, lipophilic carbocyanine dyes diffuse rapidly to stain the entire cell membrane after being incorporated, but generally stay confined to the membrane of that cell and its descendants. These membrane dyes are relatively stable in storage, photostable, and non-toxic. When used as vital dyes, they are often applied to a cell as solid crystals or pastes by micromanipulation or by injection after being dissolved in oil. Their fluorescence intensity is also greatly enhanced after being incorporated into a membrane. Blastomeres of sea urchin embryos can be labelled with DiI for cell-lineage analysis by micro-application of crystals to the surface of embryos (Ruffins and Ettensohn, 1996). In order to obtain more general labeling of blastomeres, we suspended early Strongylocentrotus purpuratus pluteus larvae in 2 mg/ml of DiO in seawater (just after vortexing) for 1 hr. Live confocal microscopy of these larvae indicated that scattered cells had been labelled. Suspension of plutei for 1 hr in seawater including 2 mg/ml each of DiI, DiO, and DiD resulted in more scattered cells being labelled; furthermore, these cells exhibited either red, green, or blue fluorescence, respectively, indicative of having incorporated a single dye, as well as some cells showing labeling with two dyes (shades of yellow, purple, or teal upon merging images) or (rarely) all three dyes (white) (Fig. 1A-C). Cells of the aboral ectoderm, the ciliary band, and apical organ (including apparent neural cells) were frequently labelled in apparently random patterns, but many ectodermal cells were not labelled. In many embryos, some cells of the blastocoelar network (particularly those near the tips of arms or the posterior tip adjacent to skeletal spicules) were also labelled.
2
VOLNOUKHIN
AND
BRANDHORST
The carbocyanine dyes we used are not considered soluble in aqueous solution, although no obvious precipitate was observed when diluted into seawater to 2 6 mg/ ml. Passage of the seawater-dissolved dyes through 0.22 mm Amicon PVD membranes or low-binding nitrocellulose filters nearly eliminated the capacity to label embryos, implying that the differential labeling of individual blastomeres depends on the random solubilization of small particles of dye into cellular membranes upon contact. Vybrant1 forms of DiI, DiO, and DiD are described by the manufacturer as more-soluble forms of the dyes sold in a proprietary solution that permits uniform labeling of cells in tissue culture. Indeed, when the 3 equivalent Vybrant1 dyes were applied simultaneously to plutei after being diluted into seawater, cells were either labelled with all 3 dyes (white in merged images), or not at all (Fig. 1D-F). The numbers of cells per embryo that labelled with Vybrant1 dyes was generally more than for the suspended dyes, but the patterns of labeling were similar. This simple method to differentially label random blastomeres with lipophilic carbocyanine dyes can be applied when desired during development; it works particularly well when applied to hatched blastulae, labeling random epithelial cells. The fates of descendents of these cells, including differentially labelled neighbors, can then be tracked in living embryos because the dye is retained by daughter cells and does not spread to neighbouring cells (Fig 1G). Filopodial extensions of labelled cells can also be detected in living embryos. The fixable nature of the carbocyanine dyes also allows them to be used in conjunction with immunostaining, facilitating identification of cells and structures. For example, Figure 1H shows the arm of a pluteus larva labelled with DiD and then immunostained with antibodies against a serotonin receptor expressed in the network of blastocoelar cells (5-HT-hpr) (Katow et al., 2004) and acetyl-a-tubulin, which detects cilia. Figure 1I shows a cell of the ciliary band labelled with DiI that has an axon-like projection immunostained with a pan-neural antibody against synaptotagmin B; it appears to be similar to cells described by Nakajima, (1986). The uptake of DiI (from aqueous suspensions) into sensory neurons of the nematode Caenorhabditis elegans depends on the presence of cilia (Starich et al., 1995). Ruffins and Ettensohn (1996) also reported that labeling blastomeres of hatched sea urchin embryos by topical application of crystals of DiI requires the presence of cilia. Mesenchymal blastocoelar cells, which reside inside the ectodermal envelop, are not known to have cilia, and we have not detected any by immunostaining for acetyl-atubulin, a cilium marker. Yet some blastocoelar cells near the tips of skeletal spicules are reported to have cell processes protruding through the ectodermal epithelium (Katow et al., 2007), which may facilitate dye uptake and could thus account for the scattered labeling we observed. To directly test the relationship between cilia and dye uptake, we deciliated plutei by treatment with hypertonic seawater immediately before resuspending them in seawater containing 2 mg/ml DiI or Vybrant1 DiI for 30 sec or
Mol. Reprod. Dev. (2015)
MULTISPECTRAL LABELING
OF
EMBRYONIC CELLS WITH
Figure 1. Confocal micrographs of embryos labelled with carbocyanine dyes, sometimes followed by immunostaining. Columns show dyes or antibodies used, or merged and overlaid onto brightfield images. Rows A C: Live plutei labelled with DiI, DiO, and DiD showing differentially labelled cells. D F: Vybrant1 DiI, DiO and DiD dyes consistently labelled the same cells. A, D: Surface of the aboral ectoderm. B, E: Optical section of the tip of an anterolateral arm. C, F: Optical section of the apical organ including cells proposed to be ectoneural. G: Hatched blastulae were labelled for 1 hr with DiI, DiO, and DiD, and the archenteron was imaged during gastrulation. H, I: Plutei were labelled with DiD or DiI, fixed, and immunostained for acetyl-a-tubulin (Ac-Tubulin), serotonin receptor (5HTPR), or synaptotagmin B.H: shows anterolateral arm; I: shows the oral ciliary band. Scale bars, 20 mm.
Mol. Reprod. Dev. (2015)
3
Molecular Reproduction & Development
10 min. Labeling of ectodermal epithelial cells was nearly eliminated, although some neural cells in the ciliary band near tips of arms or in the apical organ were still labelled. Conversely, the extent of labeling of blastocoelar cells, while variable among embryos, did not appear to be reduced by deciliation. Suspension in dye 30 min after deciliation, as cilia were regenerating, restored the labeling of ectodermal epithelial cells. Therefore, cilia appear to be necessary for dye labeling of ectoderm cells; it is possible that the rare ectodermal cells that stained just after deciliation were refractory to deciliation. Embryo incubation with Vybrant1 DiI for times ranging from 30 sec to 2 hr or incubated with a tenfold higher concentration revealed that the pattern and frequency of cell staining did not depend on either incubation time or dye concentration, but the intensity of staining increased in a time and concentration-dependent manner. While all ectodermal cells have a cilium, some squamous epithelial cells of the aboral ectoderm adjacent to labelled cells were refractory to dye uptake, indicating that the presence of cilia
VOLNOUKHIN
AND
BRANDHORST
is not sufficient for dye incorporation into the membrane. These observations suggest local variation in the properties, and possibly function, of cellular membranes or cilia of neighbouring epithelial cells as these differentially labelled cells have no distinguishing features or locations, are derived from the same cell lineages, and are considered to be a uniform tissue of differentiated cells. The failure to label some neighbouring cells suggests unsuspected features perhaps physiological that distinguish these cell types. Together, our observations indicate that while cilia can greatly facilitate uptake of carbocyanine dyes by cellular membranes, they are neither necessary nor sufficient for it. The capacity to stain single, scattered cells allows for the detection of some rare cell types for which there are no convenient molecular markers. Based on electron microscopy, Nakajima, (1986) and Ryberg, (1977) described ectoneural cells of plutei associated with the ciliary band that have basal axons and 200 nm thick apical projections; we propose that these cells can be labelled with these carbocyanine dyes (Fig. 2A-C). They also described
Figure 2. a-c: Confocal sections showing putative ectoneural cells labelled with Vybrant1 dyes. Cells within the apical organ of a pluteus with basal (A) or apical (B) projections (arrowheads) or a cilium (arrow in C). D: Surface optical section of ectodermal cells with a cilium (arrow) and radial processes (arrowheads).
4
Mol. Reprod. Dev. (2015)
MULTISPECTRAL LABELING
ectoneural cells with a short, non-motile cilium and 10 20 fine processes that radiate from the cilium base and extend over the ectodermal epithelium under the hyaline layer. Again, we detected cells with such morphology that extend processes under the ectodermal epithelium (Fig. 2D). Cellular processes corresponding to these electron microscopy descriptions were observed in our study, especially when 5,50 -Ph2-DiIC18(3), a more lipophilic variant of DiI, was used for labeling plutei (results not shown). In summary, the simple method we described allows for the simultaneous, long term tracking of differentially labelled individual cells during development. It should be applicable to embryos of many different species.
MATERIALS AND METHODS
OF
EMBRYONIC CELLS WITH
(2010). Samples were mounted in EverBrite Mounting Medium (Biotium, Hayward, CA, 23001). Primary antibodies and dilutions used include: anti-synaptotagmin B (mouse monoclonal; Burke et al., 2006), 1:50; anti-5HThpr (mouse monoclonal; Katow et al., 2004), 1:100; and antiacetyl-a-tubulin (rabbit monoclonal, Cell Signaling Technology, Beverly, MA, Cat. No. 5335P), 1:800. Secondary antibodies included: goat anti-mouse Alexa Fluor 488 or anti-rabbit Alexa Fluor 568 (Invitrogen, Carlsbad, CA, Cat. No. A11001 or A11011, respectively), 1:500.
ACKNOWLEDGMENTS Antibodies against 5-HT-hpr and synaptotagmin B were gifts from Hideki Katow and Robert Burke, respectively. We thank Tim Heslip for help with microscopy.
Dyes DiI (DiIC18(3)), DiD (DiIC18(5)), DiO (DiOC18(3)), and 5,50 -Ph2-DiIC18(3) (kit L-7781) were purchased as solids from Life Technologies (Waltham, MA). Stock solutions were prepared by dissolving each dye at 2 mg/ml in dimethylformamide, then storing at 208C. To label embryos, each stock solution was diluted into seawater to 2 mg/ml. The Vybrant1 Multicolor Cell-Labeling Kit (V-22889), including DiI, DiD, and DiO, was purchased from Life Technologies; each dye was used at a final concentration of 2 mg/ml with labeling for 1 hr, unless noted otherwise.
Embryo Culture and Deciliation
REFERENCES Barsi JC, Tu Q, Davidson EH. 2014. General approach for in vivo recovery of cell type-specific effector gene sets. Genome Research 24:860 868. Bergeron KF, Xu X, Brandhorst BP. 2010. Oral-aboral patterning and gastrulation of sea urchin embryos depend on sulfated glycosaminoglycans. Mech Dev 128:71 89. Boulina M, Samarajeewa H, Baker JD, Kim MD, Chiba A. 2013. Live imaging of multicolor labelled cells in Drosophila. Development 140:1605 1613.
Adult S. purparatus were collected on Vancouver Island. Embryos were cultured at 128C in filtered, natural seawater with stirring at 60 rpm. Deciliation of embryos was as described by Gong and Brandhorst, (1987). The fertilization envelope was removed (Foltz et al., 2004) to label embryos prior to hatching.
Burke RD, Osborne L, Wang D, Murabe N, Yaguchi S, Nakajima Y. 2006. Neuron-specific expression of a synaptotagmin gene in the sea urchin Strongylocentrotus purpuratus. J Comp Neurol 496:244 251.
Microscopy
KR Foltz, NL Adams, LL Runft. 2004. Echinoderm eggs and embryos: Procurement and culture. Met Cell Biol 74:39 74
For viewing fluorescent dyes in live embryos, samples were mounted on slides in seawater and visualized using 25x or 40x objective lenses on a WaveFX Spinning Disc Confocal System (Quorum Technologies, Ltd., Guelph, ON). DiO, DiI, and DiD signals were detected using 491, 561, and 647 nm lasers and GFP 515/30, Cy3 595/50, and Cy5 690/50 filters, respectively. For detecting Alexa Fluor 488 and 568, 491 nm and 568 nm lasers and GFP 515/30 and Cy3 (595/50) filters were used, respectively. Images were acquired using a Hamamatsu 9100-13 EMCCD camera (Hamamatsu Photonics) and processed with Volocity software V5.0 (Perkin Elmer).
Immunostaining Embryos were fixed in the dark for 2 hr at room temperature in 10 volumes of 4% paraformaldehyde, 50 mM EGTA in filtered seawater, and then washed 3 times in 10 volumes of phosphate-buffered saline for 15 min. Immunostaining of embryos was performed as described by Bergeron at al.
Mol. Reprod. Dev. (2015)
Davidson EH. 2006. The regulatory genome. Gene regulatory networks in development and evolution. Amsterdam:Elsevier.
Gong ZY, Brandhorst BP. 1987. Stimulation of tubulin gene transcription by deciliation of sea urchin embryos. Mol Cell Biol 7:4238 4246. Gupta V, Poss KD. 2012. Clonally dominant cardiomyocytes direct heart morphogenesis. Nature 484:479 484. Katow H, Yaguchi S, Kiyomoto M, Washio M. 2004. The 5-HT receptor cell is a new member of the the secondary mesenchyme cell descendents and forms a major blastocoelar network in sea urchin larvae. Mech Dev 121:325 337. Katow H, Yaguchi S, Kyozuka K. 2007. Serotonin stimulates Ca2þi elevation in ciliary ectodermal cells of echinoplutei through a serotonin receptor cell network in the blastocoel. J Exp Biol 210:403 412. Kretzschmar K, Watt FM. 2012. Lineage tracing. Cell 148:35 45. Livet J, Weissman TA, Kang H, Draft RW, Lu J, Bennis RA, Sanes JR, Lichtman JW. 2007. Transgenic strategies for combinatorial
5
Molecular Reproduction & Development
VOLNOUKHIN
AND
BRANDHORST
expression of fluorescent proteins in the nervous system. Nature 450:56 62.
Ryberg E. 1977. The nervous system of the early echinopluteus. Cell Tiss Res 179:157 167.
Nakajima Y. 1986. Development of the nervous system of sea urchin embryos:Formation of ciliary bands andthe appearance of twotypes of ectoneural cells in the pluteus. Dev Growth Differ 28:531 542.
Snippert HJ, van der Flier LG, Sato T, van Es JH, van den Born M, Kron-Veenboer C, Barker N, Klein AM, van Rheenen J, Simons BD, Clevers H. 2010. Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143:134 144.
Pan YA, Freundlich T, Weissman TA, Schoppik D, Wang XC, Zimmerman S, Ciruna B, Sanes JR, Lichtman JW, Schier AF. 2013. Zebrabow: Multspectral cell labeling for cell tracting and lineage analysis in zebrafish. Development 140:2835 2846. Peter IS, Davidson EH. 2010. The endoderm gene regulatory network in sea urchin embryos up to mid-blastula stage. Dev Biol 340:188 199. Ruffins SW, Ettensohn CA. 1996. A fate map of the vegetal plate of the sea urchin (Lytechinus variegatus) mesenchyme blastula. Development 122:253 263.
6
Starich TA, Herman RK, Kari CK, Yeh WH, Schackwitz WS, Schuyler MW, Collet J, Thomas JH, Riddle DL. 1995. Mutations affecting the chemosensory neurons of Caenorhabditis elegans. Genetics 139:171 188. Worely MI, Setiawan L, Harihan IK. 2013. TIE-DYE: A combinatorial marking system to visualize and genetically manipulate clones during development in Drosophila melanogaster. Development 140:3275 3284.
Mol. Reprod. Dev. (2015)
Multispectral Labeling of Embryonic Cells with Lipophilic Carbocyanine Dyes MARIA VOLNOUKHINy
AND
BRUCE P. BRANDHORST*
Department of Molecular Biology and Biochemistry, University Drive Simon Fraser University, Burnaby, British Columbia, Canada
SUMMARY Incubation of hatched Strongylocentrotus purpuratus sea urchin embryos or larvae with suspensions of the carbocyanine dyes DiI, DiO, and DiD resulted in the random labeling of membranes of some ectodermal epithelial cells and blastocoelar cells, producing a range of differentially colored cells that can be tracked during development. Simultaneous application of soluble Vybrant1 preparations of the three dyes resulted in similar labeling of each cell. Dye labeling of the ectoderm was nearly eliminated by deciliation and some ciliated squamous epithelial cells adjacent to labelled cells were refractory to Vybrant1 dye uptake irrespective of concentration or duration of treatment, together suggesting local variation in the properties of cell membranes or cilia. Furthermore, single cells possessing distinctive morphological features were detected.
Corresponding author: Department of Molecular Biology and Biochemistry 8888 University Drive Simon Fraser University Burnaby, BC Canada V3H3K3. E-mail: [email protected]
y
Deceased.
Grant sponsor: Natural Sciences and Engineering Research Council of Canada Discovery Grant; Grant number: 45977
Mol. Reprod. Dev. 2015. ß 2015 Wiley Periodicals, Inc. Received 16 September 2014; Accepted 4 March 2015
INTRODUCTION Marking blastomeres by the application of particles or organic dyes onto living embryos is critical for the generation of cell-lineage diagrams and fate maps in animals (reviewed by Kretzschmar and Watt, 2012). More recently, injection of dyes into embryonic blastomeres or the expression of marker proteins have been used to track cell lineages and fates; such analysis has been further refined by the expression of fluorescent reporter proteins driven by specific transgenic promoters. Combined with modern fluorescence microscopic techniques, these cell-labeling approaches allow for the tracking of patches or clones of cells in live embryos over time and space, although individual cells can be difficult to distinguish. The development of Brainbow for mouse embryos, for example, provides dis-
ß 2015 WILEY PERIODICALS, INC.
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mrd.22477
tinctive multicolor labeling of adjacent neural cells, whose individual fates and spatial interactions can then be tracked using fluorescence microscopy (Livet et al., 2007). Brainbow depends on Cre-mediated combinatorial expression of multiple recombinant fluorescent proteins in transgenic embryos; as such, this method has been modified and extended to provide multispectral marking of a range of cell types in Drosophila (Boulina et al., 2013; Worely et al., 2013), mice (Snippert et al., 2010), and zebrafish (Gupta and Poss, 2012; Pan et al., 2013), allowing clonal analyses at various developmental stages and facilitating axonal tracking in transparent embryos and larvae. While multispectral labeling of embryonic cells based on the induced recombination of fluorescent-protein transgenes is a powerful tool, it depends on the creation and
Molecular Reproduction & Development
manipulation of appropriate stable, transgenic lines. Such a genetic approach is not practical for some organisms, such as the sea urchin, because of long generation times and the difficult husbandry of juveniles. On the other hand, the sea urchin has a well-defined gene regulatory network for early embryonic development (Davidson, 2006; Peter and Davidson, 2010) that allows for the transient expression of fluorescent proteins in a cell-type-specific manner in living embryos, which are well-suited for 4-dimensional fluorescence microscopy (Barsi et al., 2014). When contiguous cells express the same fluorescent protein(s), however, distinguishing and tracking individual cells can be challenging. Here we report that the brief suspension of sea urchin embryos in combinations of lipophilic carbocyanine dyes suspended in seawater results in random labeling of individual cells in a range of colors. This approach allows for clonal analysis of the labelled cells and tracking of their descendents as well as identification of rare cell types and possible cellular interactions that may be difficult to detect using reporter-gene fusions. This simple method should be useful for multispectral, random labeling of embryonic cells in a range of organisms that are amenable to fluorescence microscopy.
RESULTS AND DISCUSSION Often used for axon tracking, lipophilic carbocyanine dyes diffuse rapidly to stain the entire cell membrane after being incorporated, but generally stay confined to the membrane of that cell and its descendants. These membrane dyes are relatively stable in storage, photostable, and non-toxic. When used as vital dyes, they are often applied to a cell as solid crystals or pastes by micromanipulation or by injection after being dissolved in oil. Their fluorescence intensity is also greatly enhanced after being incorporated into a membrane. Blastomeres of sea urchin embryos can be labelled with DiI for cell-lineage analysis by micro-application of crystals to the surface of embryos (Ruffins and Ettensohn, 1996). In order to obtain more general labeling of blastomeres, we suspended early Strongylocentrotus purpuratus pluteus larvae in 2 mg/ml of DiO in seawater (just after vortexing) for 1 hr. Live confocal microscopy of these larvae indicated that scattered cells had been labelled. Suspension of plutei for 1 hr in seawater including 2 mg/ml each of DiI, DiO, and DiD resulted in more scattered cells being labelled; furthermore, these cells exhibited either red, green, or blue fluorescence, respectively, indicative of having incorporated a single dye, as well as some cells showing labeling with two dyes (shades of yellow, purple, or teal upon merging images) or (rarely) all three dyes (white) (Fig. 1A-C). Cells of the aboral ectoderm, the ciliary band, and apical organ (including apparent neural cells) were frequently labelled in apparently random patterns, but many ectodermal cells were not labelled. In many embryos, some cells of the blastocoelar network (particularly those near the tips of arms or the posterior tip adjacent to skeletal spicules) were also labelled.
2
VOLNOUKHIN
AND
BRANDHORST
The carbocyanine dyes we used are not considered soluble in aqueous solution, although no obvious precipitate was observed when diluted into seawater to 2 6 mg/ ml. Passage of the seawater-dissolved dyes through 0.22 mm Amicon PVD membranes or low-binding nitrocellulose filters nearly eliminated the capacity to label embryos, implying that the differential labeling of individual blastomeres depends on the random solubilization of small particles of dye into cellular membranes upon contact. Vybrant1 forms of DiI, DiO, and DiD are described by the manufacturer as more-soluble forms of the dyes sold in a proprietary solution that permits uniform labeling of cells in tissue culture. Indeed, when the 3 equivalent Vybrant1 dyes were applied simultaneously to plutei after being diluted into seawater, cells were either labelled with all 3 dyes (white in merged images), or not at all (Fig. 1D-F). The numbers of cells per embryo that labelled with Vybrant1 dyes was generally more than for the suspended dyes, but the patterns of labeling were similar. This simple method to differentially label random blastomeres with lipophilic carbocyanine dyes can be applied when desired during development; it works particularly well when applied to hatched blastulae, labeling random epithelial cells. The fates of descendents of these cells, including differentially labelled neighbors, can then be tracked in living embryos because the dye is retained by daughter cells and does not spread to neighbouring cells (Fig 1G). Filopodial extensions of labelled cells can also be detected in living embryos. The fixable nature of the carbocyanine dyes also allows them to be used in conjunction with immunostaining, facilitating identification of cells and structures. For example, Figure 1H shows the arm of a pluteus larva labelled with DiD and then immunostained with antibodies against a serotonin receptor expressed in the network of blastocoelar cells (5-HT-hpr) (Katow et al., 2004) and acetyl-a-tubulin, which detects cilia. Figure 1I shows a cell of the ciliary band labelled with DiI that has an axon-like projection immunostained with a pan-neural antibody against synaptotagmin B; it appears to be similar to cells described by Nakajima, (1986). The uptake of DiI (from aqueous suspensions) into sensory neurons of the nematode Caenorhabditis elegans depends on the presence of cilia (Starich et al., 1995). Ruffins and Ettensohn (1996) also reported that labeling blastomeres of hatched sea urchin embryos by topical application of crystals of DiI requires the presence of cilia. Mesenchymal blastocoelar cells, which reside inside the ectodermal envelop, are not known to have cilia, and we have not detected any by immunostaining for acetyl-atubulin, a cilium marker. Yet some blastocoelar cells near the tips of skeletal spicules are reported to have cell processes protruding through the ectodermal epithelium (Katow et al., 2007), which may facilitate dye uptake and could thus account for the scattered labeling we observed. To directly test the relationship between cilia and dye uptake, we deciliated plutei by treatment with hypertonic seawater immediately before resuspending them in seawater containing 2 mg/ml DiI or Vybrant1 DiI for 30 sec or
Mol. Reprod. Dev. (2015)
MULTISPECTRAL LABELING
OF
EMBRYONIC CELLS WITH
Figure 1. Confocal micrographs of embryos labelled with carbocyanine dyes, sometimes followed by immunostaining. Columns show dyes or antibodies used, or merged and overlaid onto brightfield images. Rows A C: Live plutei labelled with DiI, DiO, and DiD showing differentially labelled cells. D F: Vybrant1 DiI, DiO and DiD dyes consistently labelled the same cells. A, D: Surface of the aboral ectoderm. B, E: Optical section of the tip of an anterolateral arm. C, F: Optical section of the apical organ including cells proposed to be ectoneural. G: Hatched blastulae were labelled for 1 hr with DiI, DiO, and DiD, and the archenteron was imaged during gastrulation. H, I: Plutei were labelled with DiD or DiI, fixed, and immunostained for acetyl-a-tubulin (Ac-Tubulin), serotonin receptor (5HTPR), or synaptotagmin B.H: shows anterolateral arm; I: shows the oral ciliary band. Scale bars, 20 mm.
Mol. Reprod. Dev. (2015)
3
Molecular Reproduction & Development
10 min. Labeling of ectodermal epithelial cells was nearly eliminated, although some neural cells in the ciliary band near tips of arms or in the apical organ were still labelled. Conversely, the extent of labeling of blastocoelar cells, while variable among embryos, did not appear to be reduced by deciliation. Suspension in dye 30 min after deciliation, as cilia were regenerating, restored the labeling of ectodermal epithelial cells. Therefore, cilia appear to be necessary for dye labeling of ectoderm cells; it is possible that the rare ectodermal cells that stained just after deciliation were refractory to deciliation. Embryo incubation with Vybrant1 DiI for times ranging from 30 sec to 2 hr or incubated with a tenfold higher concentration revealed that the pattern and frequency of cell staining did not depend on either incubation time or dye concentration, but the intensity of staining increased in a time and concentration-dependent manner. While all ectodermal cells have a cilium, some squamous epithelial cells of the aboral ectoderm adjacent to labelled cells were refractory to dye uptake, indicating that the presence of cilia
VOLNOUKHIN
AND
BRANDHORST
is not sufficient for dye incorporation into the membrane. These observations suggest local variation in the properties, and possibly function, of cellular membranes or cilia of neighbouring epithelial cells as these differentially labelled cells have no distinguishing features or locations, are derived from the same cell lineages, and are considered to be a uniform tissue of differentiated cells. The failure to label some neighbouring cells suggests unsuspected features perhaps physiological that distinguish these cell types. Together, our observations indicate that while cilia can greatly facilitate uptake of carbocyanine dyes by cellular membranes, they are neither necessary nor sufficient for it. The capacity to stain single, scattered cells allows for the detection of some rare cell types for which there are no convenient molecular markers. Based on electron microscopy, Nakajima, (1986) and Ryberg, (1977) described ectoneural cells of plutei associated with the ciliary band that have basal axons and 200 nm thick apical projections; we propose that these cells can be labelled with these carbocyanine dyes (Fig. 2A-C). They also described
Figure 2. a-c: Confocal sections showing putative ectoneural cells labelled with Vybrant1 dyes. Cells within the apical organ of a pluteus with basal (A) or apical (B) projections (arrowheads) or a cilium (arrow in C). D: Surface optical section of ectodermal cells with a cilium (arrow) and radial processes (arrowheads).
4
Mol. Reprod. Dev. (2015)
MULTISPECTRAL LABELING
ectoneural cells with a short, non-motile cilium and 10 20 fine processes that radiate from the cilium base and extend over the ectodermal epithelium under the hyaline layer. Again, we detected cells with such morphology that extend processes under the ectodermal epithelium (Fig. 2D). Cellular processes corresponding to these electron microscopy descriptions were observed in our study, especially when 5,50 -Ph2-DiIC18(3), a more lipophilic variant of DiI, was used for labeling plutei (results not shown). In summary, the simple method we described allows for the simultaneous, long term tracking of differentially labelled individual cells during development. It should be applicable to embryos of many different species.
MATERIALS AND METHODS
OF
EMBRYONIC CELLS WITH
(2010). Samples were mounted in EverBrite Mounting Medium (Biotium, Hayward, CA, 23001). Primary antibodies and dilutions used include: anti-synaptotagmin B (mouse monoclonal; Burke et al., 2006), 1:50; anti-5HThpr (mouse monoclonal; Katow et al., 2004), 1:100; and antiacetyl-a-tubulin (rabbit monoclonal, Cell Signaling Technology, Beverly, MA, Cat. No. 5335P), 1:800. Secondary antibodies included: goat anti-mouse Alexa Fluor 488 or anti-rabbit Alexa Fluor 568 (Invitrogen, Carlsbad, CA, Cat. No. A11001 or A11011, respectively), 1:500.
ACKNOWLEDGMENTS Antibodies against 5-HT-hpr and synaptotagmin B were gifts from Hideki Katow and Robert Burke, respectively. We thank Tim Heslip for help with microscopy.
Dyes DiI (DiIC18(3)), DiD (DiIC18(5)), DiO (DiOC18(3)), and 5,50 -Ph2-DiIC18(3) (kit L-7781) were purchased as solids from Life Technologies (Waltham, MA). Stock solutions were prepared by dissolving each dye at 2 mg/ml in dimethylformamide, then storing at 208C. To label embryos, each stock solution was diluted into seawater to 2 mg/ml. The Vybrant1 Multicolor Cell-Labeling Kit (V-22889), including DiI, DiD, and DiO, was purchased from Life Technologies; each dye was used at a final concentration of 2 mg/ml with labeling for 1 hr, unless noted otherwise.
Embryo Culture and Deciliation
REFERENCES Barsi JC, Tu Q, Davidson EH. 2014. General approach for in vivo recovery of cell type-specific effector gene sets. Genome Research 24:860 868. Bergeron KF, Xu X, Brandhorst BP. 2010. Oral-aboral patterning and gastrulation of sea urchin embryos depend on sulfated glycosaminoglycans. Mech Dev 128:71 89. Boulina M, Samarajeewa H, Baker JD, Kim MD, Chiba A. 2013. Live imaging of multicolor labelled cells in Drosophila. Development 140:1605 1613.
Adult S. purparatus were collected on Vancouver Island. Embryos were cultured at 128C in filtered, natural seawater with stirring at 60 rpm. Deciliation of embryos was as described by Gong and Brandhorst, (1987). The fertilization envelope was removed (Foltz et al., 2004) to label embryos prior to hatching.
Burke RD, Osborne L, Wang D, Murabe N, Yaguchi S, Nakajima Y. 2006. Neuron-specific expression of a synaptotagmin gene in the sea urchin Strongylocentrotus purpuratus. J Comp Neurol 496:244 251.
Microscopy
KR Foltz, NL Adams, LL Runft. 2004. Echinoderm eggs and embryos: Procurement and culture. Met Cell Biol 74:39 74
For viewing fluorescent dyes in live embryos, samples were mounted on slides in seawater and visualized using 25x or 40x objective lenses on a WaveFX Spinning Disc Confocal System (Quorum Technologies, Ltd., Guelph, ON). DiO, DiI, and DiD signals were detected using 491, 561, and 647 nm lasers and GFP 515/30, Cy3 595/50, and Cy5 690/50 filters, respectively. For detecting Alexa Fluor 488 and 568, 491 nm and 568 nm lasers and GFP 515/30 and Cy3 (595/50) filters were used, respectively. Images were acquired using a Hamamatsu 9100-13 EMCCD camera (Hamamatsu Photonics) and processed with Volocity software V5.0 (Perkin Elmer).
Immunostaining Embryos were fixed in the dark for 2 hr at room temperature in 10 volumes of 4% paraformaldehyde, 50 mM EGTA in filtered seawater, and then washed 3 times in 10 volumes of phosphate-buffered saline for 15 min. Immunostaining of embryos was performed as described by Bergeron at al.
Mol. Reprod. Dev. (2015)
Davidson EH. 2006. The regulatory genome. Gene regulatory networks in development and evolution. Amsterdam:Elsevier.
Gong ZY, Brandhorst BP. 1987. Stimulation of tubulin gene transcription by deciliation of sea urchin embryos. Mol Cell Biol 7:4238 4246. Gupta V, Poss KD. 2012. Clonally dominant cardiomyocytes direct heart morphogenesis. Nature 484:479 484. Katow H, Yaguchi S, Kiyomoto M, Washio M. 2004. The 5-HT receptor cell is a new member of the the secondary mesenchyme cell descendents and forms a major blastocoelar network in sea urchin larvae. Mech Dev 121:325 337. Katow H, Yaguchi S, Kyozuka K. 2007. Serotonin stimulates Ca2þi elevation in ciliary ectodermal cells of echinoplutei through a serotonin receptor cell network in the blastocoel. J Exp Biol 210:403 412. Kretzschmar K, Watt FM. 2012. Lineage tracing. Cell 148:35 45. Livet J, Weissman TA, Kang H, Draft RW, Lu J, Bennis RA, Sanes JR, Lichtman JW. 2007. Transgenic strategies for combinatorial
5
Molecular Reproduction & Development
VOLNOUKHIN
AND
BRANDHORST
expression of fluorescent proteins in the nervous system. Nature 450:56 62.
Ryberg E. 1977. The nervous system of the early echinopluteus. Cell Tiss Res 179:157 167.
Nakajima Y. 1986. Development of the nervous system of sea urchin embryos:Formation of ciliary bands andthe appearance of twotypes of ectoneural cells in the pluteus. Dev Growth Differ 28:531 542.
Snippert HJ, van der Flier LG, Sato T, van Es JH, van den Born M, Kron-Veenboer C, Barker N, Klein AM, van Rheenen J, Simons BD, Clevers H. 2010. Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143:134 144.
Pan YA, Freundlich T, Weissman TA, Schoppik D, Wang XC, Zimmerman S, Ciruna B, Sanes JR, Lichtman JW, Schier AF. 2013. Zebrabow: Multspectral cell labeling for cell tracting and lineage analysis in zebrafish. Development 140:2835 2846. Peter IS, Davidson EH. 2010. The endoderm gene regulatory network in sea urchin embryos up to mid-blastula stage. Dev Biol 340:188 199. Ruffins SW, Ettensohn CA. 1996. A fate map of the vegetal plate of the sea urchin (Lytechinus variegatus) mesenchyme blastula. Development 122:253 263.
6
Starich TA, Herman RK, Kari CK, Yeh WH, Schackwitz WS, Schuyler MW, Collet J, Thomas JH, Riddle DL. 1995. Mutations affecting the chemosensory neurons of Caenorhabditis elegans. Genetics 139:171 188. Worely MI, Setiawan L, Harihan IK. 2013. TIE-DYE: A combinatorial marking system to visualize and genetically manipulate clones during development in Drosophila melanogaster. Development 140:3275 3284.
Mol. Reprod. Dev. (2015)
